Show More
@@ -1,313 +1,313 b'' | |||||
1 | # filelog.py - file history class for mercurial |
|
1 | # filelog.py - file history class for mercurial | |
2 | # |
|
2 | # | |
3 | # Copyright 2005-2007 Olivia Mackall <olivia@selenic.com> |
|
3 | # Copyright 2005-2007 Olivia Mackall <olivia@selenic.com> | |
4 | # |
|
4 | # | |
5 | # This software may be used and distributed according to the terms of the |
|
5 | # This software may be used and distributed according to the terms of the | |
6 | # GNU General Public License version 2 or any later version. |
|
6 | # GNU General Public License version 2 or any later version. | |
7 |
|
7 | |||
8 |
|
8 | |||
9 | from .i18n import _ |
|
9 | from .i18n import _ | |
10 | from .node import nullrev |
|
10 | from .node import nullrev | |
11 | from . import ( |
|
11 | from . import ( | |
12 | error, |
|
12 | error, | |
13 | revlog, |
|
13 | revlog, | |
14 | ) |
|
14 | ) | |
15 | from .interfaces import ( |
|
15 | from .interfaces import ( | |
16 | repository, |
|
16 | repository, | |
17 | util as interfaceutil, |
|
17 | util as interfaceutil, | |
18 | ) |
|
18 | ) | |
19 | from .utils import storageutil |
|
19 | from .utils import storageutil | |
20 | from .revlogutils import ( |
|
20 | from .revlogutils import ( | |
21 | constants as revlog_constants, |
|
21 | constants as revlog_constants, | |
22 | rewrite, |
|
22 | rewrite, | |
23 | ) |
|
23 | ) | |
24 |
|
24 | |||
25 |
|
25 | |||
26 | @interfaceutil.implementer(repository.ifilestorage) |
|
26 | @interfaceutil.implementer(repository.ifilestorage) | |
27 | class filelog: |
|
27 | class filelog: | |
28 | def __init__(self, opener, path, try_split=False): |
|
28 | def __init__(self, opener, path, try_split=False): | |
29 | self._revlog = revlog.revlog( |
|
29 | self._revlog = revlog.revlog( | |
30 | opener, |
|
30 | opener, | |
31 | # XXX should use the unencoded path |
|
31 | # XXX should use the unencoded path | |
32 | target=(revlog_constants.KIND_FILELOG, path), |
|
32 | target=(revlog_constants.KIND_FILELOG, path), | |
33 | radix=b'/'.join((b'data', path)), |
|
33 | radix=b'/'.join((b'data', path)), | |
34 | censorable=True, |
|
34 | censorable=True, | |
35 | canonical_parent_order=False, # see comment in revlog.py |
|
35 | canonical_parent_order=False, # see comment in revlog.py | |
36 | try_split=try_split, |
|
36 | try_split=try_split, | |
37 | ) |
|
37 | ) | |
38 | # Full name of the user visible file, relative to the repository root. |
|
38 | # Full name of the user visible file, relative to the repository root. | |
39 | # Used by LFS. |
|
39 | # Used by LFS. | |
40 | self._revlog.filename = path |
|
40 | self._revlog.filename = path | |
41 | self.nullid = self._revlog.nullid |
|
41 | self.nullid = self._revlog.nullid | |
42 | opts = opener.options |
|
42 | opts = opener.options | |
43 | self._fix_issue6528 = opts.get(b'issue6528.fix-incoming', True) |
|
43 | self._fix_issue6528 = opts.get(b'issue6528.fix-incoming', True) | |
44 |
|
44 | |||
45 | def get_revlog(self): |
|
45 | def get_revlog(self): | |
46 | """return an actual revlog instance if any |
|
46 | """return an actual revlog instance if any | |
47 |
|
47 | |||
48 | This exist because a lot of code leverage the fact the underlying |
|
48 | This exist because a lot of code leverage the fact the underlying | |
49 | storage is a revlog for optimization, so giving simple way to access |
|
49 | storage is a revlog for optimization, so giving simple way to access | |
50 | the revlog instance helps such code. |
|
50 | the revlog instance helps such code. | |
51 | """ |
|
51 | """ | |
52 | return self._revlog |
|
52 | return self._revlog | |
53 |
|
53 | |||
54 | def __len__(self): |
|
54 | def __len__(self): | |
55 | return len(self._revlog) |
|
55 | return len(self._revlog) | |
56 |
|
56 | |||
57 | def __iter__(self): |
|
57 | def __iter__(self): | |
58 | return self._revlog.__iter__() |
|
58 | return self._revlog.__iter__() | |
59 |
|
59 | |||
60 | def hasnode(self, node): |
|
60 | def hasnode(self, node): | |
61 | if node in (self.nullid, nullrev): |
|
61 | if node in (self.nullid, nullrev): | |
62 | return False |
|
62 | return False | |
63 |
|
63 | |||
64 | try: |
|
64 | try: | |
65 | self._revlog.rev(node) |
|
65 | self._revlog.rev(node) | |
66 | return True |
|
66 | return True | |
67 | except (TypeError, ValueError, IndexError, error.LookupError): |
|
67 | except (TypeError, ValueError, IndexError, error.LookupError): | |
68 | return False |
|
68 | return False | |
69 |
|
69 | |||
70 | def revs(self, start=0, stop=None): |
|
70 | def revs(self, start=0, stop=None): | |
71 | return self._revlog.revs(start=start, stop=stop) |
|
71 | return self._revlog.revs(start=start, stop=stop) | |
72 |
|
72 | |||
73 | def parents(self, node): |
|
73 | def parents(self, node): | |
74 | return self._revlog.parents(node) |
|
74 | return self._revlog.parents(node) | |
75 |
|
75 | |||
76 | def parentrevs(self, rev): |
|
76 | def parentrevs(self, rev): | |
77 | return self._revlog.parentrevs(rev) |
|
77 | return self._revlog.parentrevs(rev) | |
78 |
|
78 | |||
79 | def rev(self, node): |
|
79 | def rev(self, node): | |
80 | return self._revlog.rev(node) |
|
80 | return self._revlog.rev(node) | |
81 |
|
81 | |||
82 | def node(self, rev): |
|
82 | def node(self, rev): | |
83 | return self._revlog.node(rev) |
|
83 | return self._revlog.node(rev) | |
84 |
|
84 | |||
85 | def lookup(self, node): |
|
85 | def lookup(self, node): | |
86 | return storageutil.fileidlookup( |
|
86 | return storageutil.fileidlookup( | |
87 | self._revlog, node, self._revlog.display_id |
|
87 | self._revlog, node, self._revlog.display_id | |
88 | ) |
|
88 | ) | |
89 |
|
89 | |||
90 | def linkrev(self, rev): |
|
90 | def linkrev(self, rev): | |
91 | return self._revlog.linkrev(rev) |
|
91 | return self._revlog.linkrev(rev) | |
92 |
|
92 | |||
93 | def commonancestorsheads(self, node1, node2): |
|
93 | def commonancestorsheads(self, node1, node2): | |
94 | return self._revlog.commonancestorsheads(node1, node2) |
|
94 | return self._revlog.commonancestorsheads(node1, node2) | |
95 |
|
95 | |||
96 | # Used by dagop.blockdescendants(). |
|
96 | # Used by dagop.blockdescendants(). | |
97 | def descendants(self, revs): |
|
97 | def descendants(self, revs): | |
98 | return self._revlog.descendants(revs) |
|
98 | return self._revlog.descendants(revs) | |
99 |
|
99 | |||
100 | def heads(self, start=None, stop=None): |
|
100 | def heads(self, start=None, stop=None): | |
101 | return self._revlog.heads(start, stop) |
|
101 | return self._revlog.heads(start, stop) | |
102 |
|
102 | |||
103 | # Used by hgweb, children extension. |
|
103 | # Used by hgweb, children extension. | |
104 | def children(self, node): |
|
104 | def children(self, node): | |
105 | return self._revlog.children(node) |
|
105 | return self._revlog.children(node) | |
106 |
|
106 | |||
107 | def iscensored(self, rev): |
|
107 | def iscensored(self, rev): | |
108 | return self._revlog.iscensored(rev) |
|
108 | return self._revlog.iscensored(rev) | |
109 |
|
109 | |||
110 | def revision(self, node): |
|
110 | def revision(self, node): | |
111 | return self._revlog.revision(node) |
|
111 | return self._revlog.revision(node) | |
112 |
|
112 | |||
113 |
def rawdata(self, node |
|
113 | def rawdata(self, node): | |
114 |
return self._revlog.rawdata(node |
|
114 | return self._revlog.rawdata(node) | |
115 |
|
115 | |||
116 | def emitrevisions( |
|
116 | def emitrevisions( | |
117 | self, |
|
117 | self, | |
118 | nodes, |
|
118 | nodes, | |
119 | nodesorder=None, |
|
119 | nodesorder=None, | |
120 | revisiondata=False, |
|
120 | revisiondata=False, | |
121 | assumehaveparentrevisions=False, |
|
121 | assumehaveparentrevisions=False, | |
122 | deltamode=repository.CG_DELTAMODE_STD, |
|
122 | deltamode=repository.CG_DELTAMODE_STD, | |
123 | sidedata_helpers=None, |
|
123 | sidedata_helpers=None, | |
124 | debug_info=None, |
|
124 | debug_info=None, | |
125 | ): |
|
125 | ): | |
126 | return self._revlog.emitrevisions( |
|
126 | return self._revlog.emitrevisions( | |
127 | nodes, |
|
127 | nodes, | |
128 | nodesorder=nodesorder, |
|
128 | nodesorder=nodesorder, | |
129 | revisiondata=revisiondata, |
|
129 | revisiondata=revisiondata, | |
130 | assumehaveparentrevisions=assumehaveparentrevisions, |
|
130 | assumehaveparentrevisions=assumehaveparentrevisions, | |
131 | deltamode=deltamode, |
|
131 | deltamode=deltamode, | |
132 | sidedata_helpers=sidedata_helpers, |
|
132 | sidedata_helpers=sidedata_helpers, | |
133 | debug_info=debug_info, |
|
133 | debug_info=debug_info, | |
134 | ) |
|
134 | ) | |
135 |
|
135 | |||
136 | def addrevision( |
|
136 | def addrevision( | |
137 | self, |
|
137 | self, | |
138 | revisiondata, |
|
138 | revisiondata, | |
139 | transaction, |
|
139 | transaction, | |
140 | linkrev, |
|
140 | linkrev, | |
141 | p1, |
|
141 | p1, | |
142 | p2, |
|
142 | p2, | |
143 | node=None, |
|
143 | node=None, | |
144 | flags=revlog.REVIDX_DEFAULT_FLAGS, |
|
144 | flags=revlog.REVIDX_DEFAULT_FLAGS, | |
145 | cachedelta=None, |
|
145 | cachedelta=None, | |
146 | ): |
|
146 | ): | |
147 | return self._revlog.addrevision( |
|
147 | return self._revlog.addrevision( | |
148 | revisiondata, |
|
148 | revisiondata, | |
149 | transaction, |
|
149 | transaction, | |
150 | linkrev, |
|
150 | linkrev, | |
151 | p1, |
|
151 | p1, | |
152 | p2, |
|
152 | p2, | |
153 | node=node, |
|
153 | node=node, | |
154 | flags=flags, |
|
154 | flags=flags, | |
155 | cachedelta=cachedelta, |
|
155 | cachedelta=cachedelta, | |
156 | ) |
|
156 | ) | |
157 |
|
157 | |||
158 | def addgroup( |
|
158 | def addgroup( | |
159 | self, |
|
159 | self, | |
160 | deltas, |
|
160 | deltas, | |
161 | linkmapper, |
|
161 | linkmapper, | |
162 | transaction, |
|
162 | transaction, | |
163 | addrevisioncb=None, |
|
163 | addrevisioncb=None, | |
164 | duplicaterevisioncb=None, |
|
164 | duplicaterevisioncb=None, | |
165 | maybemissingparents=False, |
|
165 | maybemissingparents=False, | |
166 | debug_info=None, |
|
166 | debug_info=None, | |
167 | delta_base_reuse_policy=None, |
|
167 | delta_base_reuse_policy=None, | |
168 | ): |
|
168 | ): | |
169 | if maybemissingparents: |
|
169 | if maybemissingparents: | |
170 | raise error.Abort( |
|
170 | raise error.Abort( | |
171 | _( |
|
171 | _( | |
172 | b'revlog storage does not support missing ' |
|
172 | b'revlog storage does not support missing ' | |
173 | b'parents write mode' |
|
173 | b'parents write mode' | |
174 | ) |
|
174 | ) | |
175 | ) |
|
175 | ) | |
176 |
|
176 | |||
177 | with self._revlog._writing(transaction): |
|
177 | with self._revlog._writing(transaction): | |
178 |
|
178 | |||
179 | if self._fix_issue6528: |
|
179 | if self._fix_issue6528: | |
180 | deltas = rewrite.filter_delta_issue6528(self._revlog, deltas) |
|
180 | deltas = rewrite.filter_delta_issue6528(self._revlog, deltas) | |
181 |
|
181 | |||
182 | return self._revlog.addgroup( |
|
182 | return self._revlog.addgroup( | |
183 | deltas, |
|
183 | deltas, | |
184 | linkmapper, |
|
184 | linkmapper, | |
185 | transaction, |
|
185 | transaction, | |
186 | addrevisioncb=addrevisioncb, |
|
186 | addrevisioncb=addrevisioncb, | |
187 | duplicaterevisioncb=duplicaterevisioncb, |
|
187 | duplicaterevisioncb=duplicaterevisioncb, | |
188 | debug_info=debug_info, |
|
188 | debug_info=debug_info, | |
189 | delta_base_reuse_policy=delta_base_reuse_policy, |
|
189 | delta_base_reuse_policy=delta_base_reuse_policy, | |
190 | ) |
|
190 | ) | |
191 |
|
191 | |||
192 | def getstrippoint(self, minlink): |
|
192 | def getstrippoint(self, minlink): | |
193 | return self._revlog.getstrippoint(minlink) |
|
193 | return self._revlog.getstrippoint(minlink) | |
194 |
|
194 | |||
195 | def strip(self, minlink, transaction): |
|
195 | def strip(self, minlink, transaction): | |
196 | return self._revlog.strip(minlink, transaction) |
|
196 | return self._revlog.strip(minlink, transaction) | |
197 |
|
197 | |||
198 | def censorrevision(self, tr, node, tombstone=b''): |
|
198 | def censorrevision(self, tr, node, tombstone=b''): | |
199 | return self._revlog.censorrevision(tr, node, tombstone=tombstone) |
|
199 | return self._revlog.censorrevision(tr, node, tombstone=tombstone) | |
200 |
|
200 | |||
201 | def files(self): |
|
201 | def files(self): | |
202 | return self._revlog.files() |
|
202 | return self._revlog.files() | |
203 |
|
203 | |||
204 | def read(self, node): |
|
204 | def read(self, node): | |
205 | return storageutil.filtermetadata(self.revision(node)) |
|
205 | return storageutil.filtermetadata(self.revision(node)) | |
206 |
|
206 | |||
207 | def add(self, text, meta, transaction, link, p1=None, p2=None): |
|
207 | def add(self, text, meta, transaction, link, p1=None, p2=None): | |
208 | if meta or text.startswith(b'\1\n'): |
|
208 | if meta or text.startswith(b'\1\n'): | |
209 | text = storageutil.packmeta(meta, text) |
|
209 | text = storageutil.packmeta(meta, text) | |
210 | rev = self.addrevision(text, transaction, link, p1, p2) |
|
210 | rev = self.addrevision(text, transaction, link, p1, p2) | |
211 | return self.node(rev) |
|
211 | return self.node(rev) | |
212 |
|
212 | |||
213 | def renamed(self, node): |
|
213 | def renamed(self, node): | |
214 | return storageutil.filerevisioncopied(self, node) |
|
214 | return storageutil.filerevisioncopied(self, node) | |
215 |
|
215 | |||
216 | def size(self, rev): |
|
216 | def size(self, rev): | |
217 | """return the size of a given revision""" |
|
217 | """return the size of a given revision""" | |
218 |
|
218 | |||
219 | # for revisions with renames, we have to go the slow way |
|
219 | # for revisions with renames, we have to go the slow way | |
220 | node = self.node(rev) |
|
220 | node = self.node(rev) | |
221 | if self.iscensored(rev): |
|
221 | if self.iscensored(rev): | |
222 | return 0 |
|
222 | return 0 | |
223 | if self.renamed(node): |
|
223 | if self.renamed(node): | |
224 | return len(self.read(node)) |
|
224 | return len(self.read(node)) | |
225 |
|
225 | |||
226 | # XXX if self.read(node).startswith("\1\n"), this returns (size+4) |
|
226 | # XXX if self.read(node).startswith("\1\n"), this returns (size+4) | |
227 | # XXX See also basefilectx.cmp. |
|
227 | # XXX See also basefilectx.cmp. | |
228 | return self._revlog.size(rev) |
|
228 | return self._revlog.size(rev) | |
229 |
|
229 | |||
230 | def cmp(self, node, text): |
|
230 | def cmp(self, node, text): | |
231 | """compare text with a given file revision |
|
231 | """compare text with a given file revision | |
232 |
|
232 | |||
233 | returns True if text is different than what is stored. |
|
233 | returns True if text is different than what is stored. | |
234 | """ |
|
234 | """ | |
235 | return not storageutil.filedataequivalent(self, node, text) |
|
235 | return not storageutil.filedataequivalent(self, node, text) | |
236 |
|
236 | |||
237 | def verifyintegrity(self, state): |
|
237 | def verifyintegrity(self, state): | |
238 | return self._revlog.verifyintegrity(state) |
|
238 | return self._revlog.verifyintegrity(state) | |
239 |
|
239 | |||
240 | def storageinfo( |
|
240 | def storageinfo( | |
241 | self, |
|
241 | self, | |
242 | exclusivefiles=False, |
|
242 | exclusivefiles=False, | |
243 | sharedfiles=False, |
|
243 | sharedfiles=False, | |
244 | revisionscount=False, |
|
244 | revisionscount=False, | |
245 | trackedsize=False, |
|
245 | trackedsize=False, | |
246 | storedsize=False, |
|
246 | storedsize=False, | |
247 | ): |
|
247 | ): | |
248 | return self._revlog.storageinfo( |
|
248 | return self._revlog.storageinfo( | |
249 | exclusivefiles=exclusivefiles, |
|
249 | exclusivefiles=exclusivefiles, | |
250 | sharedfiles=sharedfiles, |
|
250 | sharedfiles=sharedfiles, | |
251 | revisionscount=revisionscount, |
|
251 | revisionscount=revisionscount, | |
252 | trackedsize=trackedsize, |
|
252 | trackedsize=trackedsize, | |
253 | storedsize=storedsize, |
|
253 | storedsize=storedsize, | |
254 | ) |
|
254 | ) | |
255 |
|
255 | |||
256 | # Used by repo upgrade. |
|
256 | # Used by repo upgrade. | |
257 | def clone(self, tr, destrevlog, **kwargs): |
|
257 | def clone(self, tr, destrevlog, **kwargs): | |
258 | if not isinstance(destrevlog, filelog): |
|
258 | if not isinstance(destrevlog, filelog): | |
259 | msg = b'expected filelog to clone(), not %r' |
|
259 | msg = b'expected filelog to clone(), not %r' | |
260 | msg %= destrevlog |
|
260 | msg %= destrevlog | |
261 | raise error.ProgrammingError(msg) |
|
261 | raise error.ProgrammingError(msg) | |
262 |
|
262 | |||
263 | return self._revlog.clone(tr, destrevlog._revlog, **kwargs) |
|
263 | return self._revlog.clone(tr, destrevlog._revlog, **kwargs) | |
264 |
|
264 | |||
265 |
|
265 | |||
266 | class narrowfilelog(filelog): |
|
266 | class narrowfilelog(filelog): | |
267 | """Filelog variation to be used with narrow stores.""" |
|
267 | """Filelog variation to be used with narrow stores.""" | |
268 |
|
268 | |||
269 | def __init__(self, opener, path, narrowmatch, try_split=False): |
|
269 | def __init__(self, opener, path, narrowmatch, try_split=False): | |
270 | super(narrowfilelog, self).__init__(opener, path, try_split=try_split) |
|
270 | super(narrowfilelog, self).__init__(opener, path, try_split=try_split) | |
271 | self._narrowmatch = narrowmatch |
|
271 | self._narrowmatch = narrowmatch | |
272 |
|
272 | |||
273 | def renamed(self, node): |
|
273 | def renamed(self, node): | |
274 | res = super(narrowfilelog, self).renamed(node) |
|
274 | res = super(narrowfilelog, self).renamed(node) | |
275 |
|
275 | |||
276 | # Renames that come from outside the narrowspec are problematic |
|
276 | # Renames that come from outside the narrowspec are problematic | |
277 | # because we may lack the base text for the rename. This can result |
|
277 | # because we may lack the base text for the rename. This can result | |
278 | # in code attempting to walk the ancestry or compute a diff |
|
278 | # in code attempting to walk the ancestry or compute a diff | |
279 | # encountering a missing revision. We address this by silently |
|
279 | # encountering a missing revision. We address this by silently | |
280 | # removing rename metadata if the source file is outside the |
|
280 | # removing rename metadata if the source file is outside the | |
281 | # narrow spec. |
|
281 | # narrow spec. | |
282 | # |
|
282 | # | |
283 | # A better solution would be to see if the base revision is available, |
|
283 | # A better solution would be to see if the base revision is available, | |
284 | # rather than assuming it isn't. |
|
284 | # rather than assuming it isn't. | |
285 | # |
|
285 | # | |
286 | # An even better solution would be to teach all consumers of rename |
|
286 | # An even better solution would be to teach all consumers of rename | |
287 | # metadata that the base revision may not be available. |
|
287 | # metadata that the base revision may not be available. | |
288 | # |
|
288 | # | |
289 | # TODO consider better ways of doing this. |
|
289 | # TODO consider better ways of doing this. | |
290 | if res and not self._narrowmatch(res[0]): |
|
290 | if res and not self._narrowmatch(res[0]): | |
291 | return None |
|
291 | return None | |
292 |
|
292 | |||
293 | return res |
|
293 | return res | |
294 |
|
294 | |||
295 | def size(self, rev): |
|
295 | def size(self, rev): | |
296 | # Because we have a custom renamed() that may lie, we need to call |
|
296 | # Because we have a custom renamed() that may lie, we need to call | |
297 | # the base renamed() to report accurate results. |
|
297 | # the base renamed() to report accurate results. | |
298 | node = self.node(rev) |
|
298 | node = self.node(rev) | |
299 | if super(narrowfilelog, self).renamed(node): |
|
299 | if super(narrowfilelog, self).renamed(node): | |
300 | return len(self.read(node)) |
|
300 | return len(self.read(node)) | |
301 | else: |
|
301 | else: | |
302 | return super(narrowfilelog, self).size(rev) |
|
302 | return super(narrowfilelog, self).size(rev) | |
303 |
|
303 | |||
304 | def cmp(self, node, text): |
|
304 | def cmp(self, node, text): | |
305 | # We don't call `super` because narrow parents can be buggy in case of a |
|
305 | # We don't call `super` because narrow parents can be buggy in case of a | |
306 | # ambiguous dirstate. Always take the slow path until there is a better |
|
306 | # ambiguous dirstate. Always take the slow path until there is a better | |
307 | # fix, see issue6150. |
|
307 | # fix, see issue6150. | |
308 |
|
308 | |||
309 | # Censored files compare against the empty file. |
|
309 | # Censored files compare against the empty file. | |
310 | if self.iscensored(self.rev(node)): |
|
310 | if self.iscensored(self.rev(node)): | |
311 | return text != b'' |
|
311 | return text != b'' | |
312 |
|
312 | |||
313 | return self.read(node) != text |
|
313 | return self.read(node) != text |
@@ -1,2079 +1,2079 b'' | |||||
1 | # repository.py - Interfaces and base classes for repositories and peers. |
|
1 | # repository.py - Interfaces and base classes for repositories and peers. | |
2 | # coding: utf-8 |
|
2 | # coding: utf-8 | |
3 | # |
|
3 | # | |
4 | # Copyright 2017 Gregory Szorc <gregory.szorc@gmail.com> |
|
4 | # Copyright 2017 Gregory Szorc <gregory.szorc@gmail.com> | |
5 | # |
|
5 | # | |
6 | # This software may be used and distributed according to the terms of the |
|
6 | # This software may be used and distributed according to the terms of the | |
7 | # GNU General Public License version 2 or any later version. |
|
7 | # GNU General Public License version 2 or any later version. | |
8 |
|
8 | |||
9 |
|
9 | |||
10 | from ..i18n import _ |
|
10 | from ..i18n import _ | |
11 | from .. import error |
|
11 | from .. import error | |
12 | from . import util as interfaceutil |
|
12 | from . import util as interfaceutil | |
13 |
|
13 | |||
14 | # Local repository feature string. |
|
14 | # Local repository feature string. | |
15 |
|
15 | |||
16 | # Revlogs are being used for file storage. |
|
16 | # Revlogs are being used for file storage. | |
17 | REPO_FEATURE_REVLOG_FILE_STORAGE = b'revlogfilestorage' |
|
17 | REPO_FEATURE_REVLOG_FILE_STORAGE = b'revlogfilestorage' | |
18 | # The storage part of the repository is shared from an external source. |
|
18 | # The storage part of the repository is shared from an external source. | |
19 | REPO_FEATURE_SHARED_STORAGE = b'sharedstore' |
|
19 | REPO_FEATURE_SHARED_STORAGE = b'sharedstore' | |
20 | # LFS supported for backing file storage. |
|
20 | # LFS supported for backing file storage. | |
21 | REPO_FEATURE_LFS = b'lfs' |
|
21 | REPO_FEATURE_LFS = b'lfs' | |
22 | # Repository supports being stream cloned. |
|
22 | # Repository supports being stream cloned. | |
23 | REPO_FEATURE_STREAM_CLONE = b'streamclone' |
|
23 | REPO_FEATURE_STREAM_CLONE = b'streamclone' | |
24 | # Repository supports (at least) some sidedata to be stored |
|
24 | # Repository supports (at least) some sidedata to be stored | |
25 | REPO_FEATURE_SIDE_DATA = b'side-data' |
|
25 | REPO_FEATURE_SIDE_DATA = b'side-data' | |
26 | # Files storage may lack data for all ancestors. |
|
26 | # Files storage may lack data for all ancestors. | |
27 | REPO_FEATURE_SHALLOW_FILE_STORAGE = b'shallowfilestorage' |
|
27 | REPO_FEATURE_SHALLOW_FILE_STORAGE = b'shallowfilestorage' | |
28 |
|
28 | |||
29 | REVISION_FLAG_CENSORED = 1 << 15 |
|
29 | REVISION_FLAG_CENSORED = 1 << 15 | |
30 | REVISION_FLAG_ELLIPSIS = 1 << 14 |
|
30 | REVISION_FLAG_ELLIPSIS = 1 << 14 | |
31 | REVISION_FLAG_EXTSTORED = 1 << 13 |
|
31 | REVISION_FLAG_EXTSTORED = 1 << 13 | |
32 | REVISION_FLAG_HASCOPIESINFO = 1 << 12 |
|
32 | REVISION_FLAG_HASCOPIESINFO = 1 << 12 | |
33 |
|
33 | |||
34 | REVISION_FLAGS_KNOWN = ( |
|
34 | REVISION_FLAGS_KNOWN = ( | |
35 | REVISION_FLAG_CENSORED |
|
35 | REVISION_FLAG_CENSORED | |
36 | | REVISION_FLAG_ELLIPSIS |
|
36 | | REVISION_FLAG_ELLIPSIS | |
37 | | REVISION_FLAG_EXTSTORED |
|
37 | | REVISION_FLAG_EXTSTORED | |
38 | | REVISION_FLAG_HASCOPIESINFO |
|
38 | | REVISION_FLAG_HASCOPIESINFO | |
39 | ) |
|
39 | ) | |
40 |
|
40 | |||
41 | CG_DELTAMODE_STD = b'default' |
|
41 | CG_DELTAMODE_STD = b'default' | |
42 | CG_DELTAMODE_PREV = b'previous' |
|
42 | CG_DELTAMODE_PREV = b'previous' | |
43 | CG_DELTAMODE_FULL = b'fulltext' |
|
43 | CG_DELTAMODE_FULL = b'fulltext' | |
44 | CG_DELTAMODE_P1 = b'p1' |
|
44 | CG_DELTAMODE_P1 = b'p1' | |
45 |
|
45 | |||
46 |
|
46 | |||
47 | ## Cache related constants: |
|
47 | ## Cache related constants: | |
48 | # |
|
48 | # | |
49 | # Used to control which cache should be warmed in a repo.updatecaches(β¦) call. |
|
49 | # Used to control which cache should be warmed in a repo.updatecaches(β¦) call. | |
50 |
|
50 | |||
51 | # Warm branchmaps of all known repoview's filter-level |
|
51 | # Warm branchmaps of all known repoview's filter-level | |
52 | CACHE_BRANCHMAP_ALL = b"branchmap-all" |
|
52 | CACHE_BRANCHMAP_ALL = b"branchmap-all" | |
53 | # Warm branchmaps of repoview's filter-level used by server |
|
53 | # Warm branchmaps of repoview's filter-level used by server | |
54 | CACHE_BRANCHMAP_SERVED = b"branchmap-served" |
|
54 | CACHE_BRANCHMAP_SERVED = b"branchmap-served" | |
55 | # Warm internal changelog cache (eg: persistent nodemap) |
|
55 | # Warm internal changelog cache (eg: persistent nodemap) | |
56 | CACHE_CHANGELOG_CACHE = b"changelog-cache" |
|
56 | CACHE_CHANGELOG_CACHE = b"changelog-cache" | |
57 | # Warm full manifest cache |
|
57 | # Warm full manifest cache | |
58 | CACHE_FULL_MANIFEST = b"full-manifest" |
|
58 | CACHE_FULL_MANIFEST = b"full-manifest" | |
59 | # Warm file-node-tags cache |
|
59 | # Warm file-node-tags cache | |
60 | CACHE_FILE_NODE_TAGS = b"file-node-tags" |
|
60 | CACHE_FILE_NODE_TAGS = b"file-node-tags" | |
61 | # Warm internal manifestlog cache (eg: persistent nodemap) |
|
61 | # Warm internal manifestlog cache (eg: persistent nodemap) | |
62 | CACHE_MANIFESTLOG_CACHE = b"manifestlog-cache" |
|
62 | CACHE_MANIFESTLOG_CACHE = b"manifestlog-cache" | |
63 | # Warn rev branch cache |
|
63 | # Warn rev branch cache | |
64 | CACHE_REV_BRANCH = b"rev-branch-cache" |
|
64 | CACHE_REV_BRANCH = b"rev-branch-cache" | |
65 | # Warm tags' cache for default repoview' |
|
65 | # Warm tags' cache for default repoview' | |
66 | CACHE_TAGS_DEFAULT = b"tags-default" |
|
66 | CACHE_TAGS_DEFAULT = b"tags-default" | |
67 | # Warm tags' cache for repoview's filter-level used by server |
|
67 | # Warm tags' cache for repoview's filter-level used by server | |
68 | CACHE_TAGS_SERVED = b"tags-served" |
|
68 | CACHE_TAGS_SERVED = b"tags-served" | |
69 |
|
69 | |||
70 | # the cache to warm by default after a simple transaction |
|
70 | # the cache to warm by default after a simple transaction | |
71 | # (this is a mutable set to let extension update it) |
|
71 | # (this is a mutable set to let extension update it) | |
72 | CACHES_DEFAULT = { |
|
72 | CACHES_DEFAULT = { | |
73 | CACHE_BRANCHMAP_SERVED, |
|
73 | CACHE_BRANCHMAP_SERVED, | |
74 | } |
|
74 | } | |
75 |
|
75 | |||
76 | # the caches to warm when warming all of them |
|
76 | # the caches to warm when warming all of them | |
77 | # (this is a mutable set to let extension update it) |
|
77 | # (this is a mutable set to let extension update it) | |
78 | CACHES_ALL = { |
|
78 | CACHES_ALL = { | |
79 | CACHE_BRANCHMAP_SERVED, |
|
79 | CACHE_BRANCHMAP_SERVED, | |
80 | CACHE_BRANCHMAP_ALL, |
|
80 | CACHE_BRANCHMAP_ALL, | |
81 | CACHE_CHANGELOG_CACHE, |
|
81 | CACHE_CHANGELOG_CACHE, | |
82 | CACHE_FILE_NODE_TAGS, |
|
82 | CACHE_FILE_NODE_TAGS, | |
83 | CACHE_FULL_MANIFEST, |
|
83 | CACHE_FULL_MANIFEST, | |
84 | CACHE_MANIFESTLOG_CACHE, |
|
84 | CACHE_MANIFESTLOG_CACHE, | |
85 | CACHE_TAGS_DEFAULT, |
|
85 | CACHE_TAGS_DEFAULT, | |
86 | CACHE_TAGS_SERVED, |
|
86 | CACHE_TAGS_SERVED, | |
87 | } |
|
87 | } | |
88 |
|
88 | |||
89 | # the cache to warm by default on simple call |
|
89 | # the cache to warm by default on simple call | |
90 | # (this is a mutable set to let extension update it) |
|
90 | # (this is a mutable set to let extension update it) | |
91 | CACHES_POST_CLONE = CACHES_ALL.copy() |
|
91 | CACHES_POST_CLONE = CACHES_ALL.copy() | |
92 | CACHES_POST_CLONE.discard(CACHE_FILE_NODE_TAGS) |
|
92 | CACHES_POST_CLONE.discard(CACHE_FILE_NODE_TAGS) | |
93 |
|
93 | |||
94 |
|
94 | |||
95 | class ipeerconnection(interfaceutil.Interface): |
|
95 | class ipeerconnection(interfaceutil.Interface): | |
96 | """Represents a "connection" to a repository. |
|
96 | """Represents a "connection" to a repository. | |
97 |
|
97 | |||
98 | This is the base interface for representing a connection to a repository. |
|
98 | This is the base interface for representing a connection to a repository. | |
99 | It holds basic properties and methods applicable to all peer types. |
|
99 | It holds basic properties and methods applicable to all peer types. | |
100 |
|
100 | |||
101 | This is not a complete interface definition and should not be used |
|
101 | This is not a complete interface definition and should not be used | |
102 | outside of this module. |
|
102 | outside of this module. | |
103 | """ |
|
103 | """ | |
104 |
|
104 | |||
105 | ui = interfaceutil.Attribute("""ui.ui instance""") |
|
105 | ui = interfaceutil.Attribute("""ui.ui instance""") | |
106 | path = interfaceutil.Attribute("""a urlutil.path instance or None""") |
|
106 | path = interfaceutil.Attribute("""a urlutil.path instance or None""") | |
107 |
|
107 | |||
108 | def url(): |
|
108 | def url(): | |
109 | """Returns a URL string representing this peer. |
|
109 | """Returns a URL string representing this peer. | |
110 |
|
110 | |||
111 | Currently, implementations expose the raw URL used to construct the |
|
111 | Currently, implementations expose the raw URL used to construct the | |
112 | instance. It may contain credentials as part of the URL. The |
|
112 | instance. It may contain credentials as part of the URL. The | |
113 | expectations of the value aren't well-defined and this could lead to |
|
113 | expectations of the value aren't well-defined and this could lead to | |
114 | data leakage. |
|
114 | data leakage. | |
115 |
|
115 | |||
116 | TODO audit/clean consumers and more clearly define the contents of this |
|
116 | TODO audit/clean consumers and more clearly define the contents of this | |
117 | value. |
|
117 | value. | |
118 | """ |
|
118 | """ | |
119 |
|
119 | |||
120 | def local(): |
|
120 | def local(): | |
121 | """Returns a local repository instance. |
|
121 | """Returns a local repository instance. | |
122 |
|
122 | |||
123 | If the peer represents a local repository, returns an object that |
|
123 | If the peer represents a local repository, returns an object that | |
124 | can be used to interface with it. Otherwise returns ``None``. |
|
124 | can be used to interface with it. Otherwise returns ``None``. | |
125 | """ |
|
125 | """ | |
126 |
|
126 | |||
127 | def canpush(): |
|
127 | def canpush(): | |
128 | """Returns a boolean indicating if this peer can be pushed to.""" |
|
128 | """Returns a boolean indicating if this peer can be pushed to.""" | |
129 |
|
129 | |||
130 | def close(): |
|
130 | def close(): | |
131 | """Close the connection to this peer. |
|
131 | """Close the connection to this peer. | |
132 |
|
132 | |||
133 | This is called when the peer will no longer be used. Resources |
|
133 | This is called when the peer will no longer be used. Resources | |
134 | associated with the peer should be cleaned up. |
|
134 | associated with the peer should be cleaned up. | |
135 | """ |
|
135 | """ | |
136 |
|
136 | |||
137 |
|
137 | |||
138 | class ipeercapabilities(interfaceutil.Interface): |
|
138 | class ipeercapabilities(interfaceutil.Interface): | |
139 | """Peer sub-interface related to capabilities.""" |
|
139 | """Peer sub-interface related to capabilities.""" | |
140 |
|
140 | |||
141 | def capable(name): |
|
141 | def capable(name): | |
142 | """Determine support for a named capability. |
|
142 | """Determine support for a named capability. | |
143 |
|
143 | |||
144 | Returns ``False`` if capability not supported. |
|
144 | Returns ``False`` if capability not supported. | |
145 |
|
145 | |||
146 | Returns ``True`` if boolean capability is supported. Returns a string |
|
146 | Returns ``True`` if boolean capability is supported. Returns a string | |
147 | if capability support is non-boolean. |
|
147 | if capability support is non-boolean. | |
148 |
|
148 | |||
149 | Capability strings may or may not map to wire protocol capabilities. |
|
149 | Capability strings may or may not map to wire protocol capabilities. | |
150 | """ |
|
150 | """ | |
151 |
|
151 | |||
152 | def requirecap(name, purpose): |
|
152 | def requirecap(name, purpose): | |
153 | """Require a capability to be present. |
|
153 | """Require a capability to be present. | |
154 |
|
154 | |||
155 | Raises a ``CapabilityError`` if the capability isn't present. |
|
155 | Raises a ``CapabilityError`` if the capability isn't present. | |
156 | """ |
|
156 | """ | |
157 |
|
157 | |||
158 |
|
158 | |||
159 | class ipeercommands(interfaceutil.Interface): |
|
159 | class ipeercommands(interfaceutil.Interface): | |
160 | """Client-side interface for communicating over the wire protocol. |
|
160 | """Client-side interface for communicating over the wire protocol. | |
161 |
|
161 | |||
162 | This interface is used as a gateway to the Mercurial wire protocol. |
|
162 | This interface is used as a gateway to the Mercurial wire protocol. | |
163 | methods commonly call wire protocol commands of the same name. |
|
163 | methods commonly call wire protocol commands of the same name. | |
164 | """ |
|
164 | """ | |
165 |
|
165 | |||
166 | def branchmap(): |
|
166 | def branchmap(): | |
167 | """Obtain heads in named branches. |
|
167 | """Obtain heads in named branches. | |
168 |
|
168 | |||
169 | Returns a dict mapping branch name to an iterable of nodes that are |
|
169 | Returns a dict mapping branch name to an iterable of nodes that are | |
170 | heads on that branch. |
|
170 | heads on that branch. | |
171 | """ |
|
171 | """ | |
172 |
|
172 | |||
173 | def capabilities(): |
|
173 | def capabilities(): | |
174 | """Obtain capabilities of the peer. |
|
174 | """Obtain capabilities of the peer. | |
175 |
|
175 | |||
176 | Returns a set of string capabilities. |
|
176 | Returns a set of string capabilities. | |
177 | """ |
|
177 | """ | |
178 |
|
178 | |||
179 | def get_cached_bundle_inline(path): |
|
179 | def get_cached_bundle_inline(path): | |
180 | """Retrieve a clonebundle across the wire. |
|
180 | """Retrieve a clonebundle across the wire. | |
181 |
|
181 | |||
182 | Returns a chunkbuffer |
|
182 | Returns a chunkbuffer | |
183 | """ |
|
183 | """ | |
184 |
|
184 | |||
185 | def clonebundles(): |
|
185 | def clonebundles(): | |
186 | """Obtains the clone bundles manifest for the repo. |
|
186 | """Obtains the clone bundles manifest for the repo. | |
187 |
|
187 | |||
188 | Returns the manifest as unparsed bytes. |
|
188 | Returns the manifest as unparsed bytes. | |
189 | """ |
|
189 | """ | |
190 |
|
190 | |||
191 | def debugwireargs(one, two, three=None, four=None, five=None): |
|
191 | def debugwireargs(one, two, three=None, four=None, five=None): | |
192 | """Used to facilitate debugging of arguments passed over the wire.""" |
|
192 | """Used to facilitate debugging of arguments passed over the wire.""" | |
193 |
|
193 | |||
194 | def getbundle(source, **kwargs): |
|
194 | def getbundle(source, **kwargs): | |
195 | """Obtain remote repository data as a bundle. |
|
195 | """Obtain remote repository data as a bundle. | |
196 |
|
196 | |||
197 | This command is how the bulk of repository data is transferred from |
|
197 | This command is how the bulk of repository data is transferred from | |
198 | the peer to the local repository |
|
198 | the peer to the local repository | |
199 |
|
199 | |||
200 | Returns a generator of bundle data. |
|
200 | Returns a generator of bundle data. | |
201 | """ |
|
201 | """ | |
202 |
|
202 | |||
203 | def heads(): |
|
203 | def heads(): | |
204 | """Determine all known head revisions in the peer. |
|
204 | """Determine all known head revisions in the peer. | |
205 |
|
205 | |||
206 | Returns an iterable of binary nodes. |
|
206 | Returns an iterable of binary nodes. | |
207 | """ |
|
207 | """ | |
208 |
|
208 | |||
209 | def known(nodes): |
|
209 | def known(nodes): | |
210 | """Determine whether multiple nodes are known. |
|
210 | """Determine whether multiple nodes are known. | |
211 |
|
211 | |||
212 | Accepts an iterable of nodes whose presence to check for. |
|
212 | Accepts an iterable of nodes whose presence to check for. | |
213 |
|
213 | |||
214 | Returns an iterable of booleans indicating of the corresponding node |
|
214 | Returns an iterable of booleans indicating of the corresponding node | |
215 | at that index is known to the peer. |
|
215 | at that index is known to the peer. | |
216 | """ |
|
216 | """ | |
217 |
|
217 | |||
218 | def listkeys(namespace): |
|
218 | def listkeys(namespace): | |
219 | """Obtain all keys in a pushkey namespace. |
|
219 | """Obtain all keys in a pushkey namespace. | |
220 |
|
220 | |||
221 | Returns an iterable of key names. |
|
221 | Returns an iterable of key names. | |
222 | """ |
|
222 | """ | |
223 |
|
223 | |||
224 | def lookup(key): |
|
224 | def lookup(key): | |
225 | """Resolve a value to a known revision. |
|
225 | """Resolve a value to a known revision. | |
226 |
|
226 | |||
227 | Returns a binary node of the resolved revision on success. |
|
227 | Returns a binary node of the resolved revision on success. | |
228 | """ |
|
228 | """ | |
229 |
|
229 | |||
230 | def pushkey(namespace, key, old, new): |
|
230 | def pushkey(namespace, key, old, new): | |
231 | """Set a value using the ``pushkey`` protocol. |
|
231 | """Set a value using the ``pushkey`` protocol. | |
232 |
|
232 | |||
233 | Arguments correspond to the pushkey namespace and key to operate on and |
|
233 | Arguments correspond to the pushkey namespace and key to operate on and | |
234 | the old and new values for that key. |
|
234 | the old and new values for that key. | |
235 |
|
235 | |||
236 | Returns a string with the peer result. The value inside varies by the |
|
236 | Returns a string with the peer result. The value inside varies by the | |
237 | namespace. |
|
237 | namespace. | |
238 | """ |
|
238 | """ | |
239 |
|
239 | |||
240 | def stream_out(): |
|
240 | def stream_out(): | |
241 | """Obtain streaming clone data. |
|
241 | """Obtain streaming clone data. | |
242 |
|
242 | |||
243 | Successful result should be a generator of data chunks. |
|
243 | Successful result should be a generator of data chunks. | |
244 | """ |
|
244 | """ | |
245 |
|
245 | |||
246 | def unbundle(bundle, heads, url): |
|
246 | def unbundle(bundle, heads, url): | |
247 | """Transfer repository data to the peer. |
|
247 | """Transfer repository data to the peer. | |
248 |
|
248 | |||
249 | This is how the bulk of data during a push is transferred. |
|
249 | This is how the bulk of data during a push is transferred. | |
250 |
|
250 | |||
251 | Returns the integer number of heads added to the peer. |
|
251 | Returns the integer number of heads added to the peer. | |
252 | """ |
|
252 | """ | |
253 |
|
253 | |||
254 |
|
254 | |||
255 | class ipeerlegacycommands(interfaceutil.Interface): |
|
255 | class ipeerlegacycommands(interfaceutil.Interface): | |
256 | """Interface for implementing support for legacy wire protocol commands. |
|
256 | """Interface for implementing support for legacy wire protocol commands. | |
257 |
|
257 | |||
258 | Wire protocol commands transition to legacy status when they are no longer |
|
258 | Wire protocol commands transition to legacy status when they are no longer | |
259 | used by modern clients. To facilitate identifying which commands are |
|
259 | used by modern clients. To facilitate identifying which commands are | |
260 | legacy, the interfaces are split. |
|
260 | legacy, the interfaces are split. | |
261 | """ |
|
261 | """ | |
262 |
|
262 | |||
263 | def between(pairs): |
|
263 | def between(pairs): | |
264 | """Obtain nodes between pairs of nodes. |
|
264 | """Obtain nodes between pairs of nodes. | |
265 |
|
265 | |||
266 | ``pairs`` is an iterable of node pairs. |
|
266 | ``pairs`` is an iterable of node pairs. | |
267 |
|
267 | |||
268 | Returns an iterable of iterables of nodes corresponding to each |
|
268 | Returns an iterable of iterables of nodes corresponding to each | |
269 | requested pair. |
|
269 | requested pair. | |
270 | """ |
|
270 | """ | |
271 |
|
271 | |||
272 | def branches(nodes): |
|
272 | def branches(nodes): | |
273 | """Obtain ancestor changesets of specific nodes back to a branch point. |
|
273 | """Obtain ancestor changesets of specific nodes back to a branch point. | |
274 |
|
274 | |||
275 | For each requested node, the peer finds the first ancestor node that is |
|
275 | For each requested node, the peer finds the first ancestor node that is | |
276 | a DAG root or is a merge. |
|
276 | a DAG root or is a merge. | |
277 |
|
277 | |||
278 | Returns an iterable of iterables with the resolved values for each node. |
|
278 | Returns an iterable of iterables with the resolved values for each node. | |
279 | """ |
|
279 | """ | |
280 |
|
280 | |||
281 | def changegroup(nodes, source): |
|
281 | def changegroup(nodes, source): | |
282 | """Obtain a changegroup with data for descendants of specified nodes.""" |
|
282 | """Obtain a changegroup with data for descendants of specified nodes.""" | |
283 |
|
283 | |||
284 | def changegroupsubset(bases, heads, source): |
|
284 | def changegroupsubset(bases, heads, source): | |
285 | pass |
|
285 | pass | |
286 |
|
286 | |||
287 |
|
287 | |||
288 | class ipeercommandexecutor(interfaceutil.Interface): |
|
288 | class ipeercommandexecutor(interfaceutil.Interface): | |
289 | """Represents a mechanism to execute remote commands. |
|
289 | """Represents a mechanism to execute remote commands. | |
290 |
|
290 | |||
291 | This is the primary interface for requesting that wire protocol commands |
|
291 | This is the primary interface for requesting that wire protocol commands | |
292 | be executed. Instances of this interface are active in a context manager |
|
292 | be executed. Instances of this interface are active in a context manager | |
293 | and have a well-defined lifetime. When the context manager exits, all |
|
293 | and have a well-defined lifetime. When the context manager exits, all | |
294 | outstanding requests are waited on. |
|
294 | outstanding requests are waited on. | |
295 | """ |
|
295 | """ | |
296 |
|
296 | |||
297 | def callcommand(name, args): |
|
297 | def callcommand(name, args): | |
298 | """Request that a named command be executed. |
|
298 | """Request that a named command be executed. | |
299 |
|
299 | |||
300 | Receives the command name and a dictionary of command arguments. |
|
300 | Receives the command name and a dictionary of command arguments. | |
301 |
|
301 | |||
302 | Returns a ``concurrent.futures.Future`` that will resolve to the |
|
302 | Returns a ``concurrent.futures.Future`` that will resolve to the | |
303 | result of that command request. That exact value is left up to |
|
303 | result of that command request. That exact value is left up to | |
304 | the implementation and possibly varies by command. |
|
304 | the implementation and possibly varies by command. | |
305 |
|
305 | |||
306 | Not all commands can coexist with other commands in an executor |
|
306 | Not all commands can coexist with other commands in an executor | |
307 | instance: it depends on the underlying wire protocol transport being |
|
307 | instance: it depends on the underlying wire protocol transport being | |
308 | used and the command itself. |
|
308 | used and the command itself. | |
309 |
|
309 | |||
310 | Implementations MAY call ``sendcommands()`` automatically if the |
|
310 | Implementations MAY call ``sendcommands()`` automatically if the | |
311 | requested command can not coexist with other commands in this executor. |
|
311 | requested command can not coexist with other commands in this executor. | |
312 |
|
312 | |||
313 | Implementations MAY call ``sendcommands()`` automatically when the |
|
313 | Implementations MAY call ``sendcommands()`` automatically when the | |
314 | future's ``result()`` is called. So, consumers using multiple |
|
314 | future's ``result()`` is called. So, consumers using multiple | |
315 | commands with an executor MUST ensure that ``result()`` is not called |
|
315 | commands with an executor MUST ensure that ``result()`` is not called | |
316 | until all command requests have been issued. |
|
316 | until all command requests have been issued. | |
317 | """ |
|
317 | """ | |
318 |
|
318 | |||
319 | def sendcommands(): |
|
319 | def sendcommands(): | |
320 | """Trigger submission of queued command requests. |
|
320 | """Trigger submission of queued command requests. | |
321 |
|
321 | |||
322 | Not all transports submit commands as soon as they are requested to |
|
322 | Not all transports submit commands as soon as they are requested to | |
323 | run. When called, this method forces queued command requests to be |
|
323 | run. When called, this method forces queued command requests to be | |
324 | issued. It will no-op if all commands have already been sent. |
|
324 | issued. It will no-op if all commands have already been sent. | |
325 |
|
325 | |||
326 | When called, no more new commands may be issued with this executor. |
|
326 | When called, no more new commands may be issued with this executor. | |
327 | """ |
|
327 | """ | |
328 |
|
328 | |||
329 | def close(): |
|
329 | def close(): | |
330 | """Signal that this command request is finished. |
|
330 | """Signal that this command request is finished. | |
331 |
|
331 | |||
332 | When called, no more new commands may be issued. All outstanding |
|
332 | When called, no more new commands may be issued. All outstanding | |
333 | commands that have previously been issued are waited on before |
|
333 | commands that have previously been issued are waited on before | |
334 | returning. This not only includes waiting for the futures to resolve, |
|
334 | returning. This not only includes waiting for the futures to resolve, | |
335 | but also waiting for all response data to arrive. In other words, |
|
335 | but also waiting for all response data to arrive. In other words, | |
336 | calling this waits for all on-wire state for issued command requests |
|
336 | calling this waits for all on-wire state for issued command requests | |
337 | to finish. |
|
337 | to finish. | |
338 |
|
338 | |||
339 | When used as a context manager, this method is called when exiting the |
|
339 | When used as a context manager, this method is called when exiting the | |
340 | context manager. |
|
340 | context manager. | |
341 |
|
341 | |||
342 | This method may call ``sendcommands()`` if there are buffered commands. |
|
342 | This method may call ``sendcommands()`` if there are buffered commands. | |
343 | """ |
|
343 | """ | |
344 |
|
344 | |||
345 |
|
345 | |||
346 | class ipeerrequests(interfaceutil.Interface): |
|
346 | class ipeerrequests(interfaceutil.Interface): | |
347 | """Interface for executing commands on a peer.""" |
|
347 | """Interface for executing commands on a peer.""" | |
348 |
|
348 | |||
349 | limitedarguments = interfaceutil.Attribute( |
|
349 | limitedarguments = interfaceutil.Attribute( | |
350 | """True if the peer cannot receive large argument value for commands.""" |
|
350 | """True if the peer cannot receive large argument value for commands.""" | |
351 | ) |
|
351 | ) | |
352 |
|
352 | |||
353 | def commandexecutor(): |
|
353 | def commandexecutor(): | |
354 | """A context manager that resolves to an ipeercommandexecutor. |
|
354 | """A context manager that resolves to an ipeercommandexecutor. | |
355 |
|
355 | |||
356 | The object this resolves to can be used to issue command requests |
|
356 | The object this resolves to can be used to issue command requests | |
357 | to the peer. |
|
357 | to the peer. | |
358 |
|
358 | |||
359 | Callers should call its ``callcommand`` method to issue command |
|
359 | Callers should call its ``callcommand`` method to issue command | |
360 | requests. |
|
360 | requests. | |
361 |
|
361 | |||
362 | A new executor should be obtained for each distinct set of commands |
|
362 | A new executor should be obtained for each distinct set of commands | |
363 | (possibly just a single command) that the consumer wants to execute |
|
363 | (possibly just a single command) that the consumer wants to execute | |
364 | as part of a single operation or round trip. This is because some |
|
364 | as part of a single operation or round trip. This is because some | |
365 | peers are half-duplex and/or don't support persistent connections. |
|
365 | peers are half-duplex and/or don't support persistent connections. | |
366 | e.g. in the case of HTTP peers, commands sent to an executor represent |
|
366 | e.g. in the case of HTTP peers, commands sent to an executor represent | |
367 | a single HTTP request. While some peers may support multiple command |
|
367 | a single HTTP request. While some peers may support multiple command | |
368 | sends over the wire per executor, consumers need to code to the least |
|
368 | sends over the wire per executor, consumers need to code to the least | |
369 | capable peer. So it should be assumed that command executors buffer |
|
369 | capable peer. So it should be assumed that command executors buffer | |
370 | called commands until they are told to send them and that each |
|
370 | called commands until they are told to send them and that each | |
371 | command executor could result in a new connection or wire-level request |
|
371 | command executor could result in a new connection or wire-level request | |
372 | being issued. |
|
372 | being issued. | |
373 | """ |
|
373 | """ | |
374 |
|
374 | |||
375 |
|
375 | |||
376 | class ipeerbase(ipeerconnection, ipeercapabilities, ipeerrequests): |
|
376 | class ipeerbase(ipeerconnection, ipeercapabilities, ipeerrequests): | |
377 | """Unified interface for peer repositories. |
|
377 | """Unified interface for peer repositories. | |
378 |
|
378 | |||
379 | All peer instances must conform to this interface. |
|
379 | All peer instances must conform to this interface. | |
380 | """ |
|
380 | """ | |
381 |
|
381 | |||
382 |
|
382 | |||
383 | class ipeerv2(ipeerconnection, ipeercapabilities, ipeerrequests): |
|
383 | class ipeerv2(ipeerconnection, ipeercapabilities, ipeerrequests): | |
384 | """Unified peer interface for wire protocol version 2 peers.""" |
|
384 | """Unified peer interface for wire protocol version 2 peers.""" | |
385 |
|
385 | |||
386 | apidescriptor = interfaceutil.Attribute( |
|
386 | apidescriptor = interfaceutil.Attribute( | |
387 | """Data structure holding description of server API.""" |
|
387 | """Data structure holding description of server API.""" | |
388 | ) |
|
388 | ) | |
389 |
|
389 | |||
390 |
|
390 | |||
391 | @interfaceutil.implementer(ipeerbase) |
|
391 | @interfaceutil.implementer(ipeerbase) | |
392 | class peer: |
|
392 | class peer: | |
393 | """Base class for peer repositories.""" |
|
393 | """Base class for peer repositories.""" | |
394 |
|
394 | |||
395 | limitedarguments = False |
|
395 | limitedarguments = False | |
396 |
|
396 | |||
397 | def __init__(self, ui, path=None, remotehidden=False): |
|
397 | def __init__(self, ui, path=None, remotehidden=False): | |
398 | self.ui = ui |
|
398 | self.ui = ui | |
399 | self.path = path |
|
399 | self.path = path | |
400 |
|
400 | |||
401 | def capable(self, name): |
|
401 | def capable(self, name): | |
402 | caps = self.capabilities() |
|
402 | caps = self.capabilities() | |
403 | if name in caps: |
|
403 | if name in caps: | |
404 | return True |
|
404 | return True | |
405 |
|
405 | |||
406 | name = b'%s=' % name |
|
406 | name = b'%s=' % name | |
407 | for cap in caps: |
|
407 | for cap in caps: | |
408 | if cap.startswith(name): |
|
408 | if cap.startswith(name): | |
409 | return cap[len(name) :] |
|
409 | return cap[len(name) :] | |
410 |
|
410 | |||
411 | return False |
|
411 | return False | |
412 |
|
412 | |||
413 | def requirecap(self, name, purpose): |
|
413 | def requirecap(self, name, purpose): | |
414 | if self.capable(name): |
|
414 | if self.capable(name): | |
415 | return |
|
415 | return | |
416 |
|
416 | |||
417 | raise error.CapabilityError( |
|
417 | raise error.CapabilityError( | |
418 | _( |
|
418 | _( | |
419 | b'cannot %s; remote repository does not support the ' |
|
419 | b'cannot %s; remote repository does not support the ' | |
420 | b'\'%s\' capability' |
|
420 | b'\'%s\' capability' | |
421 | ) |
|
421 | ) | |
422 | % (purpose, name) |
|
422 | % (purpose, name) | |
423 | ) |
|
423 | ) | |
424 |
|
424 | |||
425 |
|
425 | |||
426 | class iverifyproblem(interfaceutil.Interface): |
|
426 | class iverifyproblem(interfaceutil.Interface): | |
427 | """Represents a problem with the integrity of the repository. |
|
427 | """Represents a problem with the integrity of the repository. | |
428 |
|
428 | |||
429 | Instances of this interface are emitted to describe an integrity issue |
|
429 | Instances of this interface are emitted to describe an integrity issue | |
430 | with a repository (e.g. corrupt storage, missing data, etc). |
|
430 | with a repository (e.g. corrupt storage, missing data, etc). | |
431 |
|
431 | |||
432 | Instances are essentially messages associated with severity. |
|
432 | Instances are essentially messages associated with severity. | |
433 | """ |
|
433 | """ | |
434 |
|
434 | |||
435 | warning = interfaceutil.Attribute( |
|
435 | warning = interfaceutil.Attribute( | |
436 | """Message indicating a non-fatal problem.""" |
|
436 | """Message indicating a non-fatal problem.""" | |
437 | ) |
|
437 | ) | |
438 |
|
438 | |||
439 | error = interfaceutil.Attribute("""Message indicating a fatal problem.""") |
|
439 | error = interfaceutil.Attribute("""Message indicating a fatal problem.""") | |
440 |
|
440 | |||
441 | node = interfaceutil.Attribute( |
|
441 | node = interfaceutil.Attribute( | |
442 | """Revision encountering the problem. |
|
442 | """Revision encountering the problem. | |
443 |
|
443 | |||
444 | ``None`` means the problem doesn't apply to a single revision. |
|
444 | ``None`` means the problem doesn't apply to a single revision. | |
445 | """ |
|
445 | """ | |
446 | ) |
|
446 | ) | |
447 |
|
447 | |||
448 |
|
448 | |||
449 | class irevisiondelta(interfaceutil.Interface): |
|
449 | class irevisiondelta(interfaceutil.Interface): | |
450 | """Represents a delta between one revision and another. |
|
450 | """Represents a delta between one revision and another. | |
451 |
|
451 | |||
452 | Instances convey enough information to allow a revision to be exchanged |
|
452 | Instances convey enough information to allow a revision to be exchanged | |
453 | with another repository. |
|
453 | with another repository. | |
454 |
|
454 | |||
455 | Instances represent the fulltext revision data or a delta against |
|
455 | Instances represent the fulltext revision data or a delta against | |
456 | another revision. Therefore the ``revision`` and ``delta`` attributes |
|
456 | another revision. Therefore the ``revision`` and ``delta`` attributes | |
457 | are mutually exclusive. |
|
457 | are mutually exclusive. | |
458 |
|
458 | |||
459 | Typically used for changegroup generation. |
|
459 | Typically used for changegroup generation. | |
460 | """ |
|
460 | """ | |
461 |
|
461 | |||
462 | node = interfaceutil.Attribute("""20 byte node of this revision.""") |
|
462 | node = interfaceutil.Attribute("""20 byte node of this revision.""") | |
463 |
|
463 | |||
464 | p1node = interfaceutil.Attribute( |
|
464 | p1node = interfaceutil.Attribute( | |
465 | """20 byte node of 1st parent of this revision.""" |
|
465 | """20 byte node of 1st parent of this revision.""" | |
466 | ) |
|
466 | ) | |
467 |
|
467 | |||
468 | p2node = interfaceutil.Attribute( |
|
468 | p2node = interfaceutil.Attribute( | |
469 | """20 byte node of 2nd parent of this revision.""" |
|
469 | """20 byte node of 2nd parent of this revision.""" | |
470 | ) |
|
470 | ) | |
471 |
|
471 | |||
472 | linknode = interfaceutil.Attribute( |
|
472 | linknode = interfaceutil.Attribute( | |
473 | """20 byte node of the changelog revision this node is linked to.""" |
|
473 | """20 byte node of the changelog revision this node is linked to.""" | |
474 | ) |
|
474 | ) | |
475 |
|
475 | |||
476 | flags = interfaceutil.Attribute( |
|
476 | flags = interfaceutil.Attribute( | |
477 | """2 bytes of integer flags that apply to this revision. |
|
477 | """2 bytes of integer flags that apply to this revision. | |
478 |
|
478 | |||
479 | This is a bitwise composition of the ``REVISION_FLAG_*`` constants. |
|
479 | This is a bitwise composition of the ``REVISION_FLAG_*`` constants. | |
480 | """ |
|
480 | """ | |
481 | ) |
|
481 | ) | |
482 |
|
482 | |||
483 | basenode = interfaceutil.Attribute( |
|
483 | basenode = interfaceutil.Attribute( | |
484 | """20 byte node of the revision this data is a delta against. |
|
484 | """20 byte node of the revision this data is a delta against. | |
485 |
|
485 | |||
486 | ``nullid`` indicates that the revision is a full revision and not |
|
486 | ``nullid`` indicates that the revision is a full revision and not | |
487 | a delta. |
|
487 | a delta. | |
488 | """ |
|
488 | """ | |
489 | ) |
|
489 | ) | |
490 |
|
490 | |||
491 | baserevisionsize = interfaceutil.Attribute( |
|
491 | baserevisionsize = interfaceutil.Attribute( | |
492 | """Size of base revision this delta is against. |
|
492 | """Size of base revision this delta is against. | |
493 |
|
493 | |||
494 | May be ``None`` if ``basenode`` is ``nullid``. |
|
494 | May be ``None`` if ``basenode`` is ``nullid``. | |
495 | """ |
|
495 | """ | |
496 | ) |
|
496 | ) | |
497 |
|
497 | |||
498 | revision = interfaceutil.Attribute( |
|
498 | revision = interfaceutil.Attribute( | |
499 | """Raw fulltext of revision data for this node.""" |
|
499 | """Raw fulltext of revision data for this node.""" | |
500 | ) |
|
500 | ) | |
501 |
|
501 | |||
502 | delta = interfaceutil.Attribute( |
|
502 | delta = interfaceutil.Attribute( | |
503 | """Delta between ``basenode`` and ``node``. |
|
503 | """Delta between ``basenode`` and ``node``. | |
504 |
|
504 | |||
505 | Stored in the bdiff delta format. |
|
505 | Stored in the bdiff delta format. | |
506 | """ |
|
506 | """ | |
507 | ) |
|
507 | ) | |
508 |
|
508 | |||
509 | sidedata = interfaceutil.Attribute( |
|
509 | sidedata = interfaceutil.Attribute( | |
510 | """Raw sidedata bytes for the given revision.""" |
|
510 | """Raw sidedata bytes for the given revision.""" | |
511 | ) |
|
511 | ) | |
512 |
|
512 | |||
513 | protocol_flags = interfaceutil.Attribute( |
|
513 | protocol_flags = interfaceutil.Attribute( | |
514 | """Single byte of integer flags that can influence the protocol. |
|
514 | """Single byte of integer flags that can influence the protocol. | |
515 |
|
515 | |||
516 | This is a bitwise composition of the ``storageutil.CG_FLAG*`` constants. |
|
516 | This is a bitwise composition of the ``storageutil.CG_FLAG*`` constants. | |
517 | """ |
|
517 | """ | |
518 | ) |
|
518 | ) | |
519 |
|
519 | |||
520 |
|
520 | |||
521 | class ifilerevisionssequence(interfaceutil.Interface): |
|
521 | class ifilerevisionssequence(interfaceutil.Interface): | |
522 | """Contains index data for all revisions of a file. |
|
522 | """Contains index data for all revisions of a file. | |
523 |
|
523 | |||
524 | Types implementing this behave like lists of tuples. The index |
|
524 | Types implementing this behave like lists of tuples. The index | |
525 | in the list corresponds to the revision number. The values contain |
|
525 | in the list corresponds to the revision number. The values contain | |
526 | index metadata. |
|
526 | index metadata. | |
527 |
|
527 | |||
528 | The *null* revision (revision number -1) is always the last item |
|
528 | The *null* revision (revision number -1) is always the last item | |
529 | in the index. |
|
529 | in the index. | |
530 | """ |
|
530 | """ | |
531 |
|
531 | |||
532 | def __len__(): |
|
532 | def __len__(): | |
533 | """The total number of revisions.""" |
|
533 | """The total number of revisions.""" | |
534 |
|
534 | |||
535 | def __getitem__(rev): |
|
535 | def __getitem__(rev): | |
536 | """Returns the object having a specific revision number. |
|
536 | """Returns the object having a specific revision number. | |
537 |
|
537 | |||
538 | Returns an 8-tuple with the following fields: |
|
538 | Returns an 8-tuple with the following fields: | |
539 |
|
539 | |||
540 | offset+flags |
|
540 | offset+flags | |
541 | Contains the offset and flags for the revision. 64-bit unsigned |
|
541 | Contains the offset and flags for the revision. 64-bit unsigned | |
542 | integer where first 6 bytes are the offset and the next 2 bytes |
|
542 | integer where first 6 bytes are the offset and the next 2 bytes | |
543 | are flags. The offset can be 0 if it is not used by the store. |
|
543 | are flags. The offset can be 0 if it is not used by the store. | |
544 | compressed size |
|
544 | compressed size | |
545 | Size of the revision data in the store. It can be 0 if it isn't |
|
545 | Size of the revision data in the store. It can be 0 if it isn't | |
546 | needed by the store. |
|
546 | needed by the store. | |
547 | uncompressed size |
|
547 | uncompressed size | |
548 | Fulltext size. It can be 0 if it isn't needed by the store. |
|
548 | Fulltext size. It can be 0 if it isn't needed by the store. | |
549 | base revision |
|
549 | base revision | |
550 | Revision number of revision the delta for storage is encoded |
|
550 | Revision number of revision the delta for storage is encoded | |
551 | against. -1 indicates not encoded against a base revision. |
|
551 | against. -1 indicates not encoded against a base revision. | |
552 | link revision |
|
552 | link revision | |
553 | Revision number of changelog revision this entry is related to. |
|
553 | Revision number of changelog revision this entry is related to. | |
554 | p1 revision |
|
554 | p1 revision | |
555 | Revision number of 1st parent. -1 if no 1st parent. |
|
555 | Revision number of 1st parent. -1 if no 1st parent. | |
556 | p2 revision |
|
556 | p2 revision | |
557 | Revision number of 2nd parent. -1 if no 1st parent. |
|
557 | Revision number of 2nd parent. -1 if no 1st parent. | |
558 | node |
|
558 | node | |
559 | Binary node value for this revision number. |
|
559 | Binary node value for this revision number. | |
560 |
|
560 | |||
561 | Negative values should index off the end of the sequence. ``-1`` |
|
561 | Negative values should index off the end of the sequence. ``-1`` | |
562 | should return the null revision. ``-2`` should return the most |
|
562 | should return the null revision. ``-2`` should return the most | |
563 | recent revision. |
|
563 | recent revision. | |
564 | """ |
|
564 | """ | |
565 |
|
565 | |||
566 | def __contains__(rev): |
|
566 | def __contains__(rev): | |
567 | """Whether a revision number exists.""" |
|
567 | """Whether a revision number exists.""" | |
568 |
|
568 | |||
569 | def insert(self, i, entry): |
|
569 | def insert(self, i, entry): | |
570 | """Add an item to the index at specific revision.""" |
|
570 | """Add an item to the index at specific revision.""" | |
571 |
|
571 | |||
572 |
|
572 | |||
573 | class ifileindex(interfaceutil.Interface): |
|
573 | class ifileindex(interfaceutil.Interface): | |
574 | """Storage interface for index data of a single file. |
|
574 | """Storage interface for index data of a single file. | |
575 |
|
575 | |||
576 | File storage data is divided into index metadata and data storage. |
|
576 | File storage data is divided into index metadata and data storage. | |
577 | This interface defines the index portion of the interface. |
|
577 | This interface defines the index portion of the interface. | |
578 |
|
578 | |||
579 | The index logically consists of: |
|
579 | The index logically consists of: | |
580 |
|
580 | |||
581 | * A mapping between revision numbers and nodes. |
|
581 | * A mapping between revision numbers and nodes. | |
582 | * DAG data (storing and querying the relationship between nodes). |
|
582 | * DAG data (storing and querying the relationship between nodes). | |
583 | * Metadata to facilitate storage. |
|
583 | * Metadata to facilitate storage. | |
584 | """ |
|
584 | """ | |
585 |
|
585 | |||
586 | nullid = interfaceutil.Attribute( |
|
586 | nullid = interfaceutil.Attribute( | |
587 | """node for the null revision for use as delta base.""" |
|
587 | """node for the null revision for use as delta base.""" | |
588 | ) |
|
588 | ) | |
589 |
|
589 | |||
590 | def __len__(): |
|
590 | def __len__(): | |
591 | """Obtain the number of revisions stored for this file.""" |
|
591 | """Obtain the number of revisions stored for this file.""" | |
592 |
|
592 | |||
593 | def __iter__(): |
|
593 | def __iter__(): | |
594 | """Iterate over revision numbers for this file.""" |
|
594 | """Iterate over revision numbers for this file.""" | |
595 |
|
595 | |||
596 | def hasnode(node): |
|
596 | def hasnode(node): | |
597 | """Returns a bool indicating if a node is known to this store. |
|
597 | """Returns a bool indicating if a node is known to this store. | |
598 |
|
598 | |||
599 | Implementations must only return True for full, binary node values: |
|
599 | Implementations must only return True for full, binary node values: | |
600 | hex nodes, revision numbers, and partial node matches must be |
|
600 | hex nodes, revision numbers, and partial node matches must be | |
601 | rejected. |
|
601 | rejected. | |
602 |
|
602 | |||
603 | The null node is never present. |
|
603 | The null node is never present. | |
604 | """ |
|
604 | """ | |
605 |
|
605 | |||
606 | def revs(start=0, stop=None): |
|
606 | def revs(start=0, stop=None): | |
607 | """Iterate over revision numbers for this file, with control.""" |
|
607 | """Iterate over revision numbers for this file, with control.""" | |
608 |
|
608 | |||
609 | def parents(node): |
|
609 | def parents(node): | |
610 | """Returns a 2-tuple of parent nodes for a revision. |
|
610 | """Returns a 2-tuple of parent nodes for a revision. | |
611 |
|
611 | |||
612 | Values will be ``nullid`` if the parent is empty. |
|
612 | Values will be ``nullid`` if the parent is empty. | |
613 | """ |
|
613 | """ | |
614 |
|
614 | |||
615 | def parentrevs(rev): |
|
615 | def parentrevs(rev): | |
616 | """Like parents() but operates on revision numbers.""" |
|
616 | """Like parents() but operates on revision numbers.""" | |
617 |
|
617 | |||
618 | def rev(node): |
|
618 | def rev(node): | |
619 | """Obtain the revision number given a node. |
|
619 | """Obtain the revision number given a node. | |
620 |
|
620 | |||
621 | Raises ``error.LookupError`` if the node is not known. |
|
621 | Raises ``error.LookupError`` if the node is not known. | |
622 | """ |
|
622 | """ | |
623 |
|
623 | |||
624 | def node(rev): |
|
624 | def node(rev): | |
625 | """Obtain the node value given a revision number. |
|
625 | """Obtain the node value given a revision number. | |
626 |
|
626 | |||
627 | Raises ``IndexError`` if the node is not known. |
|
627 | Raises ``IndexError`` if the node is not known. | |
628 | """ |
|
628 | """ | |
629 |
|
629 | |||
630 | def lookup(node): |
|
630 | def lookup(node): | |
631 | """Attempt to resolve a value to a node. |
|
631 | """Attempt to resolve a value to a node. | |
632 |
|
632 | |||
633 | Value can be a binary node, hex node, revision number, or a string |
|
633 | Value can be a binary node, hex node, revision number, or a string | |
634 | that can be converted to an integer. |
|
634 | that can be converted to an integer. | |
635 |
|
635 | |||
636 | Raises ``error.LookupError`` if a node could not be resolved. |
|
636 | Raises ``error.LookupError`` if a node could not be resolved. | |
637 | """ |
|
637 | """ | |
638 |
|
638 | |||
639 | def linkrev(rev): |
|
639 | def linkrev(rev): | |
640 | """Obtain the changeset revision number a revision is linked to.""" |
|
640 | """Obtain the changeset revision number a revision is linked to.""" | |
641 |
|
641 | |||
642 | def iscensored(rev): |
|
642 | def iscensored(rev): | |
643 | """Return whether a revision's content has been censored.""" |
|
643 | """Return whether a revision's content has been censored.""" | |
644 |
|
644 | |||
645 | def commonancestorsheads(node1, node2): |
|
645 | def commonancestorsheads(node1, node2): | |
646 | """Obtain an iterable of nodes containing heads of common ancestors. |
|
646 | """Obtain an iterable of nodes containing heads of common ancestors. | |
647 |
|
647 | |||
648 | See ``ancestor.commonancestorsheads()``. |
|
648 | See ``ancestor.commonancestorsheads()``. | |
649 | """ |
|
649 | """ | |
650 |
|
650 | |||
651 | def descendants(revs): |
|
651 | def descendants(revs): | |
652 | """Obtain descendant revision numbers for a set of revision numbers. |
|
652 | """Obtain descendant revision numbers for a set of revision numbers. | |
653 |
|
653 | |||
654 | If ``nullrev`` is in the set, this is equivalent to ``revs()``. |
|
654 | If ``nullrev`` is in the set, this is equivalent to ``revs()``. | |
655 | """ |
|
655 | """ | |
656 |
|
656 | |||
657 | def heads(start=None, stop=None): |
|
657 | def heads(start=None, stop=None): | |
658 | """Obtain a list of nodes that are DAG heads, with control. |
|
658 | """Obtain a list of nodes that are DAG heads, with control. | |
659 |
|
659 | |||
660 | The set of revisions examined can be limited by specifying |
|
660 | The set of revisions examined can be limited by specifying | |
661 | ``start`` and ``stop``. ``start`` is a node. ``stop`` is an |
|
661 | ``start`` and ``stop``. ``start`` is a node. ``stop`` is an | |
662 | iterable of nodes. DAG traversal starts at earlier revision |
|
662 | iterable of nodes. DAG traversal starts at earlier revision | |
663 | ``start`` and iterates forward until any node in ``stop`` is |
|
663 | ``start`` and iterates forward until any node in ``stop`` is | |
664 | encountered. |
|
664 | encountered. | |
665 | """ |
|
665 | """ | |
666 |
|
666 | |||
667 | def children(node): |
|
667 | def children(node): | |
668 | """Obtain nodes that are children of a node. |
|
668 | """Obtain nodes that are children of a node. | |
669 |
|
669 | |||
670 | Returns a list of nodes. |
|
670 | Returns a list of nodes. | |
671 | """ |
|
671 | """ | |
672 |
|
672 | |||
673 |
|
673 | |||
674 | class ifiledata(interfaceutil.Interface): |
|
674 | class ifiledata(interfaceutil.Interface): | |
675 | """Storage interface for data storage of a specific file. |
|
675 | """Storage interface for data storage of a specific file. | |
676 |
|
676 | |||
677 | This complements ``ifileindex`` and provides an interface for accessing |
|
677 | This complements ``ifileindex`` and provides an interface for accessing | |
678 | data for a tracked file. |
|
678 | data for a tracked file. | |
679 | """ |
|
679 | """ | |
680 |
|
680 | |||
681 | def size(rev): |
|
681 | def size(rev): | |
682 | """Obtain the fulltext size of file data. |
|
682 | """Obtain the fulltext size of file data. | |
683 |
|
683 | |||
684 | Any metadata is excluded from size measurements. |
|
684 | Any metadata is excluded from size measurements. | |
685 | """ |
|
685 | """ | |
686 |
|
686 | |||
687 | def revision(node): |
|
687 | def revision(node): | |
688 | """Obtain fulltext data for a node. |
|
688 | """Obtain fulltext data for a node. | |
689 |
|
689 | |||
690 | By default, any storage transformations are applied before the data |
|
690 | By default, any storage transformations are applied before the data | |
691 | is returned. If ``raw`` is True, non-raw storage transformations |
|
691 | is returned. If ``raw`` is True, non-raw storage transformations | |
692 | are not applied. |
|
692 | are not applied. | |
693 |
|
693 | |||
694 | The fulltext data may contain a header containing metadata. Most |
|
694 | The fulltext data may contain a header containing metadata. Most | |
695 | consumers should use ``read()`` to obtain the actual file data. |
|
695 | consumers should use ``read()`` to obtain the actual file data. | |
696 | """ |
|
696 | """ | |
697 |
|
697 | |||
698 | def rawdata(node): |
|
698 | def rawdata(node): | |
699 | """Obtain raw data for a node.""" |
|
699 | """Obtain raw data for a node.""" | |
700 |
|
700 | |||
701 | def read(node): |
|
701 | def read(node): | |
702 | """Resolve file fulltext data. |
|
702 | """Resolve file fulltext data. | |
703 |
|
703 | |||
704 | This is similar to ``revision()`` except any metadata in the data |
|
704 | This is similar to ``revision()`` except any metadata in the data | |
705 | headers is stripped. |
|
705 | headers is stripped. | |
706 | """ |
|
706 | """ | |
707 |
|
707 | |||
708 | def renamed(node): |
|
708 | def renamed(node): | |
709 | """Obtain copy metadata for a node. |
|
709 | """Obtain copy metadata for a node. | |
710 |
|
710 | |||
711 | Returns ``False`` if no copy metadata is stored or a 2-tuple of |
|
711 | Returns ``False`` if no copy metadata is stored or a 2-tuple of | |
712 | (path, node) from which this revision was copied. |
|
712 | (path, node) from which this revision was copied. | |
713 | """ |
|
713 | """ | |
714 |
|
714 | |||
715 | def cmp(node, fulltext): |
|
715 | def cmp(node, fulltext): | |
716 | """Compare fulltext to another revision. |
|
716 | """Compare fulltext to another revision. | |
717 |
|
717 | |||
718 | Returns True if the fulltext is different from what is stored. |
|
718 | Returns True if the fulltext is different from what is stored. | |
719 |
|
719 | |||
720 | This takes copy metadata into account. |
|
720 | This takes copy metadata into account. | |
721 |
|
721 | |||
722 | TODO better document the copy metadata and censoring logic. |
|
722 | TODO better document the copy metadata and censoring logic. | |
723 | """ |
|
723 | """ | |
724 |
|
724 | |||
725 | def emitrevisions( |
|
725 | def emitrevisions( | |
726 | nodes, |
|
726 | nodes, | |
727 | nodesorder=None, |
|
727 | nodesorder=None, | |
728 | revisiondata=False, |
|
728 | revisiondata=False, | |
729 | assumehaveparentrevisions=False, |
|
729 | assumehaveparentrevisions=False, | |
730 | deltamode=CG_DELTAMODE_STD, |
|
730 | deltamode=CG_DELTAMODE_STD, | |
731 | ): |
|
731 | ): | |
732 | """Produce ``irevisiondelta`` for revisions. |
|
732 | """Produce ``irevisiondelta`` for revisions. | |
733 |
|
733 | |||
734 | Given an iterable of nodes, emits objects conforming to the |
|
734 | Given an iterable of nodes, emits objects conforming to the | |
735 | ``irevisiondelta`` interface that describe revisions in storage. |
|
735 | ``irevisiondelta`` interface that describe revisions in storage. | |
736 |
|
736 | |||
737 | This method is a generator. |
|
737 | This method is a generator. | |
738 |
|
738 | |||
739 | The input nodes may be unordered. Implementations must ensure that a |
|
739 | The input nodes may be unordered. Implementations must ensure that a | |
740 | node's parents are emitted before the node itself. Transitively, this |
|
740 | node's parents are emitted before the node itself. Transitively, this | |
741 | means that a node may only be emitted once all its ancestors in |
|
741 | means that a node may only be emitted once all its ancestors in | |
742 | ``nodes`` have also been emitted. |
|
742 | ``nodes`` have also been emitted. | |
743 |
|
743 | |||
744 | By default, emits "index" data (the ``node``, ``p1node``, and |
|
744 | By default, emits "index" data (the ``node``, ``p1node``, and | |
745 | ``p2node`` attributes). If ``revisiondata`` is set, revision data |
|
745 | ``p2node`` attributes). If ``revisiondata`` is set, revision data | |
746 | will also be present on the emitted objects. |
|
746 | will also be present on the emitted objects. | |
747 |
|
747 | |||
748 | With default argument values, implementations can choose to emit |
|
748 | With default argument values, implementations can choose to emit | |
749 | either fulltext revision data or a delta. When emitting deltas, |
|
749 | either fulltext revision data or a delta. When emitting deltas, | |
750 | implementations must consider whether the delta's base revision |
|
750 | implementations must consider whether the delta's base revision | |
751 | fulltext is available to the receiver. |
|
751 | fulltext is available to the receiver. | |
752 |
|
752 | |||
753 | The base revision fulltext is guaranteed to be available if any of |
|
753 | The base revision fulltext is guaranteed to be available if any of | |
754 | the following are met: |
|
754 | the following are met: | |
755 |
|
755 | |||
756 | * Its fulltext revision was emitted by this method call. |
|
756 | * Its fulltext revision was emitted by this method call. | |
757 | * A delta for that revision was emitted by this method call. |
|
757 | * A delta for that revision was emitted by this method call. | |
758 | * ``assumehaveparentrevisions`` is True and the base revision is a |
|
758 | * ``assumehaveparentrevisions`` is True and the base revision is a | |
759 | parent of the node. |
|
759 | parent of the node. | |
760 |
|
760 | |||
761 | ``nodesorder`` can be used to control the order that revisions are |
|
761 | ``nodesorder`` can be used to control the order that revisions are | |
762 | emitted. By default, revisions can be reordered as long as they are |
|
762 | emitted. By default, revisions can be reordered as long as they are | |
763 | in DAG topological order (see above). If the value is ``nodes``, |
|
763 | in DAG topological order (see above). If the value is ``nodes``, | |
764 | the iteration order from ``nodes`` should be used. If the value is |
|
764 | the iteration order from ``nodes`` should be used. If the value is | |
765 | ``storage``, then the native order from the backing storage layer |
|
765 | ``storage``, then the native order from the backing storage layer | |
766 | is used. (Not all storage layers will have strong ordering and behavior |
|
766 | is used. (Not all storage layers will have strong ordering and behavior | |
767 | of this mode is storage-dependent.) ``nodes`` ordering can force |
|
767 | of this mode is storage-dependent.) ``nodes`` ordering can force | |
768 | revisions to be emitted before their ancestors, so consumers should |
|
768 | revisions to be emitted before their ancestors, so consumers should | |
769 | use it with care. |
|
769 | use it with care. | |
770 |
|
770 | |||
771 | The ``linknode`` attribute on the returned ``irevisiondelta`` may not |
|
771 | The ``linknode`` attribute on the returned ``irevisiondelta`` may not | |
772 | be set and it is the caller's responsibility to resolve it, if needed. |
|
772 | be set and it is the caller's responsibility to resolve it, if needed. | |
773 |
|
773 | |||
774 | If ``deltamode`` is CG_DELTAMODE_PREV and revision data is requested, |
|
774 | If ``deltamode`` is CG_DELTAMODE_PREV and revision data is requested, | |
775 | all revision data should be emitted as deltas against the revision |
|
775 | all revision data should be emitted as deltas against the revision | |
776 | emitted just prior. The initial revision should be a delta against its |
|
776 | emitted just prior. The initial revision should be a delta against its | |
777 | 1st parent. |
|
777 | 1st parent. | |
778 | """ |
|
778 | """ | |
779 |
|
779 | |||
780 |
|
780 | |||
781 | class ifilemutation(interfaceutil.Interface): |
|
781 | class ifilemutation(interfaceutil.Interface): | |
782 | """Storage interface for mutation events of a tracked file.""" |
|
782 | """Storage interface for mutation events of a tracked file.""" | |
783 |
|
783 | |||
784 | def add(filedata, meta, transaction, linkrev, p1, p2): |
|
784 | def add(filedata, meta, transaction, linkrev, p1, p2): | |
785 | """Add a new revision to the store. |
|
785 | """Add a new revision to the store. | |
786 |
|
786 | |||
787 | Takes file data, dictionary of metadata, a transaction, linkrev, |
|
787 | Takes file data, dictionary of metadata, a transaction, linkrev, | |
788 | and parent nodes. |
|
788 | and parent nodes. | |
789 |
|
789 | |||
790 | Returns the node that was added. |
|
790 | Returns the node that was added. | |
791 |
|
791 | |||
792 | May no-op if a revision matching the supplied data is already stored. |
|
792 | May no-op if a revision matching the supplied data is already stored. | |
793 | """ |
|
793 | """ | |
794 |
|
794 | |||
795 | def addrevision( |
|
795 | def addrevision( | |
796 | revisiondata, |
|
796 | revisiondata, | |
797 | transaction, |
|
797 | transaction, | |
798 | linkrev, |
|
798 | linkrev, | |
799 | p1, |
|
799 | p1, | |
800 | p2, |
|
800 | p2, | |
801 | node=None, |
|
801 | node=None, | |
802 | flags=0, |
|
802 | flags=0, | |
803 | cachedelta=None, |
|
803 | cachedelta=None, | |
804 | ): |
|
804 | ): | |
805 | """Add a new revision to the store and return its number. |
|
805 | """Add a new revision to the store and return its number. | |
806 |
|
806 | |||
807 | This is similar to ``add()`` except it operates at a lower level. |
|
807 | This is similar to ``add()`` except it operates at a lower level. | |
808 |
|
808 | |||
809 | The data passed in already contains a metadata header, if any. |
|
809 | The data passed in already contains a metadata header, if any. | |
810 |
|
810 | |||
811 | ``node`` and ``flags`` can be used to define the expected node and |
|
811 | ``node`` and ``flags`` can be used to define the expected node and | |
812 | the flags to use with storage. ``flags`` is a bitwise value composed |
|
812 | the flags to use with storage. ``flags`` is a bitwise value composed | |
813 | of the various ``REVISION_FLAG_*`` constants. |
|
813 | of the various ``REVISION_FLAG_*`` constants. | |
814 |
|
814 | |||
815 | ``add()`` is usually called when adding files from e.g. the working |
|
815 | ``add()`` is usually called when adding files from e.g. the working | |
816 | directory. ``addrevision()`` is often called by ``add()`` and for |
|
816 | directory. ``addrevision()`` is often called by ``add()`` and for | |
817 | scenarios where revision data has already been computed, such as when |
|
817 | scenarios where revision data has already been computed, such as when | |
818 | applying raw data from a peer repo. |
|
818 | applying raw data from a peer repo. | |
819 | """ |
|
819 | """ | |
820 |
|
820 | |||
821 | def addgroup( |
|
821 | def addgroup( | |
822 | deltas, |
|
822 | deltas, | |
823 | linkmapper, |
|
823 | linkmapper, | |
824 | transaction, |
|
824 | transaction, | |
825 | addrevisioncb=None, |
|
825 | addrevisioncb=None, | |
826 | duplicaterevisioncb=None, |
|
826 | duplicaterevisioncb=None, | |
827 | maybemissingparents=False, |
|
827 | maybemissingparents=False, | |
828 | ): |
|
828 | ): | |
829 | """Process a series of deltas for storage. |
|
829 | """Process a series of deltas for storage. | |
830 |
|
830 | |||
831 | ``deltas`` is an iterable of 7-tuples of |
|
831 | ``deltas`` is an iterable of 7-tuples of | |
832 | (node, p1, p2, linknode, deltabase, delta, flags) defining revisions |
|
832 | (node, p1, p2, linknode, deltabase, delta, flags) defining revisions | |
833 | to add. |
|
833 | to add. | |
834 |
|
834 | |||
835 | The ``delta`` field contains ``mpatch`` data to apply to a base |
|
835 | The ``delta`` field contains ``mpatch`` data to apply to a base | |
836 | revision, identified by ``deltabase``. The base node can be |
|
836 | revision, identified by ``deltabase``. The base node can be | |
837 | ``nullid``, in which case the header from the delta can be ignored |
|
837 | ``nullid``, in which case the header from the delta can be ignored | |
838 | and the delta used as the fulltext. |
|
838 | and the delta used as the fulltext. | |
839 |
|
839 | |||
840 | ``alwayscache`` instructs the lower layers to cache the content of the |
|
840 | ``alwayscache`` instructs the lower layers to cache the content of the | |
841 | newly added revision, even if it needs to be explicitly computed. |
|
841 | newly added revision, even if it needs to be explicitly computed. | |
842 | This used to be the default when ``addrevisioncb`` was provided up to |
|
842 | This used to be the default when ``addrevisioncb`` was provided up to | |
843 | Mercurial 5.8. |
|
843 | Mercurial 5.8. | |
844 |
|
844 | |||
845 | ``addrevisioncb`` should be called for each new rev as it is committed. |
|
845 | ``addrevisioncb`` should be called for each new rev as it is committed. | |
846 | ``duplicaterevisioncb`` should be called for all revs with a |
|
846 | ``duplicaterevisioncb`` should be called for all revs with a | |
847 | pre-existing node. |
|
847 | pre-existing node. | |
848 |
|
848 | |||
849 | ``maybemissingparents`` is a bool indicating whether the incoming |
|
849 | ``maybemissingparents`` is a bool indicating whether the incoming | |
850 | data may reference parents/ancestor revisions that aren't present. |
|
850 | data may reference parents/ancestor revisions that aren't present. | |
851 | This flag is set when receiving data into a "shallow" store that |
|
851 | This flag is set when receiving data into a "shallow" store that | |
852 | doesn't hold all history. |
|
852 | doesn't hold all history. | |
853 |
|
853 | |||
854 | Returns a list of nodes that were processed. A node will be in the list |
|
854 | Returns a list of nodes that were processed. A node will be in the list | |
855 | even if it existed in the store previously. |
|
855 | even if it existed in the store previously. | |
856 | """ |
|
856 | """ | |
857 |
|
857 | |||
858 | def censorrevision(tr, node, tombstone=b''): |
|
858 | def censorrevision(tr, node, tombstone=b''): | |
859 | """Remove the content of a single revision. |
|
859 | """Remove the content of a single revision. | |
860 |
|
860 | |||
861 | The specified ``node`` will have its content purged from storage. |
|
861 | The specified ``node`` will have its content purged from storage. | |
862 | Future attempts to access the revision data for this node will |
|
862 | Future attempts to access the revision data for this node will | |
863 | result in failure. |
|
863 | result in failure. | |
864 |
|
864 | |||
865 | A ``tombstone`` message can optionally be stored. This message may be |
|
865 | A ``tombstone`` message can optionally be stored. This message may be | |
866 | displayed to users when they attempt to access the missing revision |
|
866 | displayed to users when they attempt to access the missing revision | |
867 | data. |
|
867 | data. | |
868 |
|
868 | |||
869 | Storage backends may have stored deltas against the previous content |
|
869 | Storage backends may have stored deltas against the previous content | |
870 | in this revision. As part of censoring a revision, these storage |
|
870 | in this revision. As part of censoring a revision, these storage | |
871 | backends are expected to rewrite any internally stored deltas such |
|
871 | backends are expected to rewrite any internally stored deltas such | |
872 | that they no longer reference the deleted content. |
|
872 | that they no longer reference the deleted content. | |
873 | """ |
|
873 | """ | |
874 |
|
874 | |||
875 | def getstrippoint(minlink): |
|
875 | def getstrippoint(minlink): | |
876 | """Find the minimum revision that must be stripped to strip a linkrev. |
|
876 | """Find the minimum revision that must be stripped to strip a linkrev. | |
877 |
|
877 | |||
878 | Returns a 2-tuple containing the minimum revision number and a set |
|
878 | Returns a 2-tuple containing the minimum revision number and a set | |
879 | of all revisions numbers that would be broken by this strip. |
|
879 | of all revisions numbers that would be broken by this strip. | |
880 |
|
880 | |||
881 | TODO this is highly revlog centric and should be abstracted into |
|
881 | TODO this is highly revlog centric and should be abstracted into | |
882 | a higher-level deletion API. ``repair.strip()`` relies on this. |
|
882 | a higher-level deletion API. ``repair.strip()`` relies on this. | |
883 | """ |
|
883 | """ | |
884 |
|
884 | |||
885 | def strip(minlink, transaction): |
|
885 | def strip(minlink, transaction): | |
886 | """Remove storage of items starting at a linkrev. |
|
886 | """Remove storage of items starting at a linkrev. | |
887 |
|
887 | |||
888 | This uses ``getstrippoint()`` to determine the first node to remove. |
|
888 | This uses ``getstrippoint()`` to determine the first node to remove. | |
889 | Then it effectively truncates storage for all revisions after that. |
|
889 | Then it effectively truncates storage for all revisions after that. | |
890 |
|
890 | |||
891 | TODO this is highly revlog centric and should be abstracted into a |
|
891 | TODO this is highly revlog centric and should be abstracted into a | |
892 | higher-level deletion API. |
|
892 | higher-level deletion API. | |
893 | """ |
|
893 | """ | |
894 |
|
894 | |||
895 |
|
895 | |||
896 | class ifilestorage(ifileindex, ifiledata, ifilemutation): |
|
896 | class ifilestorage(ifileindex, ifiledata, ifilemutation): | |
897 | """Complete storage interface for a single tracked file.""" |
|
897 | """Complete storage interface for a single tracked file.""" | |
898 |
|
898 | |||
899 | def files(): |
|
899 | def files(): | |
900 | """Obtain paths that are backing storage for this file. |
|
900 | """Obtain paths that are backing storage for this file. | |
901 |
|
901 | |||
902 | TODO this is used heavily by verify code and there should probably |
|
902 | TODO this is used heavily by verify code and there should probably | |
903 | be a better API for that. |
|
903 | be a better API for that. | |
904 | """ |
|
904 | """ | |
905 |
|
905 | |||
906 | def storageinfo( |
|
906 | def storageinfo( | |
907 | exclusivefiles=False, |
|
907 | exclusivefiles=False, | |
908 | sharedfiles=False, |
|
908 | sharedfiles=False, | |
909 | revisionscount=False, |
|
909 | revisionscount=False, | |
910 | trackedsize=False, |
|
910 | trackedsize=False, | |
911 | storedsize=False, |
|
911 | storedsize=False, | |
912 | ): |
|
912 | ): | |
913 | """Obtain information about storage for this file's data. |
|
913 | """Obtain information about storage for this file's data. | |
914 |
|
914 | |||
915 | Returns a dict describing storage for this tracked path. The keys |
|
915 | Returns a dict describing storage for this tracked path. The keys | |
916 | in the dict map to arguments of the same. The arguments are bools |
|
916 | in the dict map to arguments of the same. The arguments are bools | |
917 | indicating whether to calculate and obtain that data. |
|
917 | indicating whether to calculate and obtain that data. | |
918 |
|
918 | |||
919 | exclusivefiles |
|
919 | exclusivefiles | |
920 | Iterable of (vfs, path) describing files that are exclusively |
|
920 | Iterable of (vfs, path) describing files that are exclusively | |
921 | used to back storage for this tracked path. |
|
921 | used to back storage for this tracked path. | |
922 |
|
922 | |||
923 | sharedfiles |
|
923 | sharedfiles | |
924 | Iterable of (vfs, path) describing files that are used to back |
|
924 | Iterable of (vfs, path) describing files that are used to back | |
925 | storage for this tracked path. Those files may also provide storage |
|
925 | storage for this tracked path. Those files may also provide storage | |
926 | for other stored entities. |
|
926 | for other stored entities. | |
927 |
|
927 | |||
928 | revisionscount |
|
928 | revisionscount | |
929 | Number of revisions available for retrieval. |
|
929 | Number of revisions available for retrieval. | |
930 |
|
930 | |||
931 | trackedsize |
|
931 | trackedsize | |
932 | Total size in bytes of all tracked revisions. This is a sum of the |
|
932 | Total size in bytes of all tracked revisions. This is a sum of the | |
933 | length of the fulltext of all revisions. |
|
933 | length of the fulltext of all revisions. | |
934 |
|
934 | |||
935 | storedsize |
|
935 | storedsize | |
936 | Total size in bytes used to store data for all tracked revisions. |
|
936 | Total size in bytes used to store data for all tracked revisions. | |
937 | This is commonly less than ``trackedsize`` due to internal usage |
|
937 | This is commonly less than ``trackedsize`` due to internal usage | |
938 | of deltas rather than fulltext revisions. |
|
938 | of deltas rather than fulltext revisions. | |
939 |
|
939 | |||
940 | Not all storage backends may support all queries are have a reasonable |
|
940 | Not all storage backends may support all queries are have a reasonable | |
941 | value to use. In that case, the value should be set to ``None`` and |
|
941 | value to use. In that case, the value should be set to ``None`` and | |
942 | callers are expected to handle this special value. |
|
942 | callers are expected to handle this special value. | |
943 | """ |
|
943 | """ | |
944 |
|
944 | |||
945 | def verifyintegrity(state): |
|
945 | def verifyintegrity(state): | |
946 | """Verifies the integrity of file storage. |
|
946 | """Verifies the integrity of file storage. | |
947 |
|
947 | |||
948 | ``state`` is a dict holding state of the verifier process. It can be |
|
948 | ``state`` is a dict holding state of the verifier process. It can be | |
949 | used to communicate data between invocations of multiple storage |
|
949 | used to communicate data between invocations of multiple storage | |
950 | primitives. |
|
950 | primitives. | |
951 |
|
951 | |||
952 | If individual revisions cannot have their revision content resolved, |
|
952 | If individual revisions cannot have their revision content resolved, | |
953 | the method is expected to set the ``skipread`` key to a set of nodes |
|
953 | the method is expected to set the ``skipread`` key to a set of nodes | |
954 | that encountered problems. If set, the method can also add the node(s) |
|
954 | that encountered problems. If set, the method can also add the node(s) | |
955 | to ``safe_renamed`` in order to indicate nodes that may perform the |
|
955 | to ``safe_renamed`` in order to indicate nodes that may perform the | |
956 | rename checks with currently accessible data. |
|
956 | rename checks with currently accessible data. | |
957 |
|
957 | |||
958 | The method yields objects conforming to the ``iverifyproblem`` |
|
958 | The method yields objects conforming to the ``iverifyproblem`` | |
959 | interface. |
|
959 | interface. | |
960 | """ |
|
960 | """ | |
961 |
|
961 | |||
962 |
|
962 | |||
963 | class idirs(interfaceutil.Interface): |
|
963 | class idirs(interfaceutil.Interface): | |
964 | """Interface representing a collection of directories from paths. |
|
964 | """Interface representing a collection of directories from paths. | |
965 |
|
965 | |||
966 | This interface is essentially a derived data structure representing |
|
966 | This interface is essentially a derived data structure representing | |
967 | directories from a collection of paths. |
|
967 | directories from a collection of paths. | |
968 | """ |
|
968 | """ | |
969 |
|
969 | |||
970 | def addpath(path): |
|
970 | def addpath(path): | |
971 | """Add a path to the collection. |
|
971 | """Add a path to the collection. | |
972 |
|
972 | |||
973 | All directories in the path will be added to the collection. |
|
973 | All directories in the path will be added to the collection. | |
974 | """ |
|
974 | """ | |
975 |
|
975 | |||
976 | def delpath(path): |
|
976 | def delpath(path): | |
977 | """Remove a path from the collection. |
|
977 | """Remove a path from the collection. | |
978 |
|
978 | |||
979 | If the removal was the last path in a particular directory, the |
|
979 | If the removal was the last path in a particular directory, the | |
980 | directory is removed from the collection. |
|
980 | directory is removed from the collection. | |
981 | """ |
|
981 | """ | |
982 |
|
982 | |||
983 | def __iter__(): |
|
983 | def __iter__(): | |
984 | """Iterate over the directories in this collection of paths.""" |
|
984 | """Iterate over the directories in this collection of paths.""" | |
985 |
|
985 | |||
986 | def __contains__(path): |
|
986 | def __contains__(path): | |
987 | """Whether a specific directory is in this collection.""" |
|
987 | """Whether a specific directory is in this collection.""" | |
988 |
|
988 | |||
989 |
|
989 | |||
990 | class imanifestdict(interfaceutil.Interface): |
|
990 | class imanifestdict(interfaceutil.Interface): | |
991 | """Interface representing a manifest data structure. |
|
991 | """Interface representing a manifest data structure. | |
992 |
|
992 | |||
993 | A manifest is effectively a dict mapping paths to entries. Each entry |
|
993 | A manifest is effectively a dict mapping paths to entries. Each entry | |
994 | consists of a binary node and extra flags affecting that entry. |
|
994 | consists of a binary node and extra flags affecting that entry. | |
995 | """ |
|
995 | """ | |
996 |
|
996 | |||
997 | def __getitem__(path): |
|
997 | def __getitem__(path): | |
998 | """Returns the binary node value for a path in the manifest. |
|
998 | """Returns the binary node value for a path in the manifest. | |
999 |
|
999 | |||
1000 | Raises ``KeyError`` if the path does not exist in the manifest. |
|
1000 | Raises ``KeyError`` if the path does not exist in the manifest. | |
1001 |
|
1001 | |||
1002 | Equivalent to ``self.find(path)[0]``. |
|
1002 | Equivalent to ``self.find(path)[0]``. | |
1003 | """ |
|
1003 | """ | |
1004 |
|
1004 | |||
1005 | def find(path): |
|
1005 | def find(path): | |
1006 | """Returns the entry for a path in the manifest. |
|
1006 | """Returns the entry for a path in the manifest. | |
1007 |
|
1007 | |||
1008 | Returns a 2-tuple of (node, flags). |
|
1008 | Returns a 2-tuple of (node, flags). | |
1009 |
|
1009 | |||
1010 | Raises ``KeyError`` if the path does not exist in the manifest. |
|
1010 | Raises ``KeyError`` if the path does not exist in the manifest. | |
1011 | """ |
|
1011 | """ | |
1012 |
|
1012 | |||
1013 | def __len__(): |
|
1013 | def __len__(): | |
1014 | """Return the number of entries in the manifest.""" |
|
1014 | """Return the number of entries in the manifest.""" | |
1015 |
|
1015 | |||
1016 | def __nonzero__(): |
|
1016 | def __nonzero__(): | |
1017 | """Returns True if the manifest has entries, False otherwise.""" |
|
1017 | """Returns True if the manifest has entries, False otherwise.""" | |
1018 |
|
1018 | |||
1019 | __bool__ = __nonzero__ |
|
1019 | __bool__ = __nonzero__ | |
1020 |
|
1020 | |||
1021 | def __setitem__(path, node): |
|
1021 | def __setitem__(path, node): | |
1022 | """Define the node value for a path in the manifest. |
|
1022 | """Define the node value for a path in the manifest. | |
1023 |
|
1023 | |||
1024 | If the path is already in the manifest, its flags will be copied to |
|
1024 | If the path is already in the manifest, its flags will be copied to | |
1025 | the new entry. |
|
1025 | the new entry. | |
1026 | """ |
|
1026 | """ | |
1027 |
|
1027 | |||
1028 | def __contains__(path): |
|
1028 | def __contains__(path): | |
1029 | """Whether a path exists in the manifest.""" |
|
1029 | """Whether a path exists in the manifest.""" | |
1030 |
|
1030 | |||
1031 | def __delitem__(path): |
|
1031 | def __delitem__(path): | |
1032 | """Remove a path from the manifest. |
|
1032 | """Remove a path from the manifest. | |
1033 |
|
1033 | |||
1034 | Raises ``KeyError`` if the path is not in the manifest. |
|
1034 | Raises ``KeyError`` if the path is not in the manifest. | |
1035 | """ |
|
1035 | """ | |
1036 |
|
1036 | |||
1037 | def __iter__(): |
|
1037 | def __iter__(): | |
1038 | """Iterate over paths in the manifest.""" |
|
1038 | """Iterate over paths in the manifest.""" | |
1039 |
|
1039 | |||
1040 | def iterkeys(): |
|
1040 | def iterkeys(): | |
1041 | """Iterate over paths in the manifest.""" |
|
1041 | """Iterate over paths in the manifest.""" | |
1042 |
|
1042 | |||
1043 | def keys(): |
|
1043 | def keys(): | |
1044 | """Obtain a list of paths in the manifest.""" |
|
1044 | """Obtain a list of paths in the manifest.""" | |
1045 |
|
1045 | |||
1046 | def filesnotin(other, match=None): |
|
1046 | def filesnotin(other, match=None): | |
1047 | """Obtain the set of paths in this manifest but not in another. |
|
1047 | """Obtain the set of paths in this manifest but not in another. | |
1048 |
|
1048 | |||
1049 | ``match`` is an optional matcher function to be applied to both |
|
1049 | ``match`` is an optional matcher function to be applied to both | |
1050 | manifests. |
|
1050 | manifests. | |
1051 |
|
1051 | |||
1052 | Returns a set of paths. |
|
1052 | Returns a set of paths. | |
1053 | """ |
|
1053 | """ | |
1054 |
|
1054 | |||
1055 | def dirs(): |
|
1055 | def dirs(): | |
1056 | """Returns an object implementing the ``idirs`` interface.""" |
|
1056 | """Returns an object implementing the ``idirs`` interface.""" | |
1057 |
|
1057 | |||
1058 | def hasdir(dir): |
|
1058 | def hasdir(dir): | |
1059 | """Returns a bool indicating if a directory is in this manifest.""" |
|
1059 | """Returns a bool indicating if a directory is in this manifest.""" | |
1060 |
|
1060 | |||
1061 | def walk(match): |
|
1061 | def walk(match): | |
1062 | """Generator of paths in manifest satisfying a matcher. |
|
1062 | """Generator of paths in manifest satisfying a matcher. | |
1063 |
|
1063 | |||
1064 | If the matcher has explicit files listed and they don't exist in |
|
1064 | If the matcher has explicit files listed and they don't exist in | |
1065 | the manifest, ``match.bad()`` is called for each missing file. |
|
1065 | the manifest, ``match.bad()`` is called for each missing file. | |
1066 | """ |
|
1066 | """ | |
1067 |
|
1067 | |||
1068 | def diff(other, match=None, clean=False): |
|
1068 | def diff(other, match=None, clean=False): | |
1069 | """Find differences between this manifest and another. |
|
1069 | """Find differences between this manifest and another. | |
1070 |
|
1070 | |||
1071 | This manifest is compared to ``other``. |
|
1071 | This manifest is compared to ``other``. | |
1072 |
|
1072 | |||
1073 | If ``match`` is provided, the two manifests are filtered against this |
|
1073 | If ``match`` is provided, the two manifests are filtered against this | |
1074 | matcher and only entries satisfying the matcher are compared. |
|
1074 | matcher and only entries satisfying the matcher are compared. | |
1075 |
|
1075 | |||
1076 | If ``clean`` is True, unchanged files are included in the returned |
|
1076 | If ``clean`` is True, unchanged files are included in the returned | |
1077 | object. |
|
1077 | object. | |
1078 |
|
1078 | |||
1079 | Returns a dict with paths as keys and values of 2-tuples of 2-tuples of |
|
1079 | Returns a dict with paths as keys and values of 2-tuples of 2-tuples of | |
1080 | the form ``((node1, flag1), (node2, flag2))`` where ``(node1, flag1)`` |
|
1080 | the form ``((node1, flag1), (node2, flag2))`` where ``(node1, flag1)`` | |
1081 | represents the node and flags for this manifest and ``(node2, flag2)`` |
|
1081 | represents the node and flags for this manifest and ``(node2, flag2)`` | |
1082 | are the same for the other manifest. |
|
1082 | are the same for the other manifest. | |
1083 | """ |
|
1083 | """ | |
1084 |
|
1084 | |||
1085 | def setflag(path, flag): |
|
1085 | def setflag(path, flag): | |
1086 | """Set the flag value for a given path. |
|
1086 | """Set the flag value for a given path. | |
1087 |
|
1087 | |||
1088 | Raises ``KeyError`` if the path is not already in the manifest. |
|
1088 | Raises ``KeyError`` if the path is not already in the manifest. | |
1089 | """ |
|
1089 | """ | |
1090 |
|
1090 | |||
1091 | def get(path, default=None): |
|
1091 | def get(path, default=None): | |
1092 | """Obtain the node value for a path or a default value if missing.""" |
|
1092 | """Obtain the node value for a path or a default value if missing.""" | |
1093 |
|
1093 | |||
1094 | def flags(path): |
|
1094 | def flags(path): | |
1095 | """Return the flags value for a path (default: empty bytestring).""" |
|
1095 | """Return the flags value for a path (default: empty bytestring).""" | |
1096 |
|
1096 | |||
1097 | def copy(): |
|
1097 | def copy(): | |
1098 | """Return a copy of this manifest.""" |
|
1098 | """Return a copy of this manifest.""" | |
1099 |
|
1099 | |||
1100 | def items(): |
|
1100 | def items(): | |
1101 | """Returns an iterable of (path, node) for items in this manifest.""" |
|
1101 | """Returns an iterable of (path, node) for items in this manifest.""" | |
1102 |
|
1102 | |||
1103 | def iteritems(): |
|
1103 | def iteritems(): | |
1104 | """Identical to items().""" |
|
1104 | """Identical to items().""" | |
1105 |
|
1105 | |||
1106 | def iterentries(): |
|
1106 | def iterentries(): | |
1107 | """Returns an iterable of (path, node, flags) for this manifest. |
|
1107 | """Returns an iterable of (path, node, flags) for this manifest. | |
1108 |
|
1108 | |||
1109 | Similar to ``iteritems()`` except items are a 3-tuple and include |
|
1109 | Similar to ``iteritems()`` except items are a 3-tuple and include | |
1110 | flags. |
|
1110 | flags. | |
1111 | """ |
|
1111 | """ | |
1112 |
|
1112 | |||
1113 | def text(): |
|
1113 | def text(): | |
1114 | """Obtain the raw data representation for this manifest. |
|
1114 | """Obtain the raw data representation for this manifest. | |
1115 |
|
1115 | |||
1116 | Result is used to create a manifest revision. |
|
1116 | Result is used to create a manifest revision. | |
1117 | """ |
|
1117 | """ | |
1118 |
|
1118 | |||
1119 | def fastdelta(base, changes): |
|
1119 | def fastdelta(base, changes): | |
1120 | """Obtain a delta between this manifest and another given changes. |
|
1120 | """Obtain a delta between this manifest and another given changes. | |
1121 |
|
1121 | |||
1122 | ``base`` in the raw data representation for another manifest. |
|
1122 | ``base`` in the raw data representation for another manifest. | |
1123 |
|
1123 | |||
1124 | ``changes`` is an iterable of ``(path, to_delete)``. |
|
1124 | ``changes`` is an iterable of ``(path, to_delete)``. | |
1125 |
|
1125 | |||
1126 | Returns a 2-tuple containing ``bytearray(self.text())`` and the |
|
1126 | Returns a 2-tuple containing ``bytearray(self.text())`` and the | |
1127 | delta between ``base`` and this manifest. |
|
1127 | delta between ``base`` and this manifest. | |
1128 |
|
1128 | |||
1129 | If this manifest implementation can't support ``fastdelta()``, |
|
1129 | If this manifest implementation can't support ``fastdelta()``, | |
1130 | raise ``mercurial.manifest.FastdeltaUnavailable``. |
|
1130 | raise ``mercurial.manifest.FastdeltaUnavailable``. | |
1131 | """ |
|
1131 | """ | |
1132 |
|
1132 | |||
1133 |
|
1133 | |||
1134 | class imanifestrevisionbase(interfaceutil.Interface): |
|
1134 | class imanifestrevisionbase(interfaceutil.Interface): | |
1135 | """Base interface representing a single revision of a manifest. |
|
1135 | """Base interface representing a single revision of a manifest. | |
1136 |
|
1136 | |||
1137 | Should not be used as a primary interface: should always be inherited |
|
1137 | Should not be used as a primary interface: should always be inherited | |
1138 | as part of a larger interface. |
|
1138 | as part of a larger interface. | |
1139 | """ |
|
1139 | """ | |
1140 |
|
1140 | |||
1141 | def copy(): |
|
1141 | def copy(): | |
1142 | """Obtain a copy of this manifest instance. |
|
1142 | """Obtain a copy of this manifest instance. | |
1143 |
|
1143 | |||
1144 | Returns an object conforming to the ``imanifestrevisionwritable`` |
|
1144 | Returns an object conforming to the ``imanifestrevisionwritable`` | |
1145 | interface. The instance will be associated with the same |
|
1145 | interface. The instance will be associated with the same | |
1146 | ``imanifestlog`` collection as this instance. |
|
1146 | ``imanifestlog`` collection as this instance. | |
1147 | """ |
|
1147 | """ | |
1148 |
|
1148 | |||
1149 | def read(): |
|
1149 | def read(): | |
1150 | """Obtain the parsed manifest data structure. |
|
1150 | """Obtain the parsed manifest data structure. | |
1151 |
|
1151 | |||
1152 | The returned object conforms to the ``imanifestdict`` interface. |
|
1152 | The returned object conforms to the ``imanifestdict`` interface. | |
1153 | """ |
|
1153 | """ | |
1154 |
|
1154 | |||
1155 |
|
1155 | |||
1156 | class imanifestrevisionstored(imanifestrevisionbase): |
|
1156 | class imanifestrevisionstored(imanifestrevisionbase): | |
1157 | """Interface representing a manifest revision committed to storage.""" |
|
1157 | """Interface representing a manifest revision committed to storage.""" | |
1158 |
|
1158 | |||
1159 | def node(): |
|
1159 | def node(): | |
1160 | """The binary node for this manifest.""" |
|
1160 | """The binary node for this manifest.""" | |
1161 |
|
1161 | |||
1162 | parents = interfaceutil.Attribute( |
|
1162 | parents = interfaceutil.Attribute( | |
1163 | """List of binary nodes that are parents for this manifest revision.""" |
|
1163 | """List of binary nodes that are parents for this manifest revision.""" | |
1164 | ) |
|
1164 | ) | |
1165 |
|
1165 | |||
1166 | def readdelta(shallow=False): |
|
1166 | def readdelta(shallow=False): | |
1167 | """Obtain the manifest data structure representing changes from parent. |
|
1167 | """Obtain the manifest data structure representing changes from parent. | |
1168 |
|
1168 | |||
1169 | This manifest is compared to its 1st parent. A new manifest representing |
|
1169 | This manifest is compared to its 1st parent. A new manifest representing | |
1170 | those differences is constructed. |
|
1170 | those differences is constructed. | |
1171 |
|
1171 | |||
1172 | The returned object conforms to the ``imanifestdict`` interface. |
|
1172 | The returned object conforms to the ``imanifestdict`` interface. | |
1173 | """ |
|
1173 | """ | |
1174 |
|
1174 | |||
1175 | def readfast(shallow=False): |
|
1175 | def readfast(shallow=False): | |
1176 | """Calls either ``read()`` or ``readdelta()``. |
|
1176 | """Calls either ``read()`` or ``readdelta()``. | |
1177 |
|
1177 | |||
1178 | The faster of the two options is called. |
|
1178 | The faster of the two options is called. | |
1179 | """ |
|
1179 | """ | |
1180 |
|
1180 | |||
1181 | def find(key): |
|
1181 | def find(key): | |
1182 | """Calls self.read().find(key)``. |
|
1182 | """Calls self.read().find(key)``. | |
1183 |
|
1183 | |||
1184 | Returns a 2-tuple of ``(node, flags)`` or raises ``KeyError``. |
|
1184 | Returns a 2-tuple of ``(node, flags)`` or raises ``KeyError``. | |
1185 | """ |
|
1185 | """ | |
1186 |
|
1186 | |||
1187 |
|
1187 | |||
1188 | class imanifestrevisionwritable(imanifestrevisionbase): |
|
1188 | class imanifestrevisionwritable(imanifestrevisionbase): | |
1189 | """Interface representing a manifest revision that can be committed.""" |
|
1189 | """Interface representing a manifest revision that can be committed.""" | |
1190 |
|
1190 | |||
1191 | def write(transaction, linkrev, p1node, p2node, added, removed, match=None): |
|
1191 | def write(transaction, linkrev, p1node, p2node, added, removed, match=None): | |
1192 | """Add this revision to storage. |
|
1192 | """Add this revision to storage. | |
1193 |
|
1193 | |||
1194 | Takes a transaction object, the changeset revision number it will |
|
1194 | Takes a transaction object, the changeset revision number it will | |
1195 | be associated with, its parent nodes, and lists of added and |
|
1195 | be associated with, its parent nodes, and lists of added and | |
1196 | removed paths. |
|
1196 | removed paths. | |
1197 |
|
1197 | |||
1198 | If match is provided, storage can choose not to inspect or write out |
|
1198 | If match is provided, storage can choose not to inspect or write out | |
1199 | items that do not match. Storage is still required to be able to provide |
|
1199 | items that do not match. Storage is still required to be able to provide | |
1200 | the full manifest in the future for any directories written (these |
|
1200 | the full manifest in the future for any directories written (these | |
1201 | manifests should not be "narrowed on disk"). |
|
1201 | manifests should not be "narrowed on disk"). | |
1202 |
|
1202 | |||
1203 | Returns the binary node of the created revision. |
|
1203 | Returns the binary node of the created revision. | |
1204 | """ |
|
1204 | """ | |
1205 |
|
1205 | |||
1206 |
|
1206 | |||
1207 | class imanifeststorage(interfaceutil.Interface): |
|
1207 | class imanifeststorage(interfaceutil.Interface): | |
1208 | """Storage interface for manifest data.""" |
|
1208 | """Storage interface for manifest data.""" | |
1209 |
|
1209 | |||
1210 | nodeconstants = interfaceutil.Attribute( |
|
1210 | nodeconstants = interfaceutil.Attribute( | |
1211 | """nodeconstants used by the current repository.""" |
|
1211 | """nodeconstants used by the current repository.""" | |
1212 | ) |
|
1212 | ) | |
1213 |
|
1213 | |||
1214 | tree = interfaceutil.Attribute( |
|
1214 | tree = interfaceutil.Attribute( | |
1215 | """The path to the directory this manifest tracks. |
|
1215 | """The path to the directory this manifest tracks. | |
1216 |
|
1216 | |||
1217 | The empty bytestring represents the root manifest. |
|
1217 | The empty bytestring represents the root manifest. | |
1218 | """ |
|
1218 | """ | |
1219 | ) |
|
1219 | ) | |
1220 |
|
1220 | |||
1221 | index = interfaceutil.Attribute( |
|
1221 | index = interfaceutil.Attribute( | |
1222 | """An ``ifilerevisionssequence`` instance.""" |
|
1222 | """An ``ifilerevisionssequence`` instance.""" | |
1223 | ) |
|
1223 | ) | |
1224 |
|
1224 | |||
1225 | opener = interfaceutil.Attribute( |
|
1225 | opener = interfaceutil.Attribute( | |
1226 | """VFS opener to use to access underlying files used for storage. |
|
1226 | """VFS opener to use to access underlying files used for storage. | |
1227 |
|
1227 | |||
1228 | TODO this is revlog specific and should not be exposed. |
|
1228 | TODO this is revlog specific and should not be exposed. | |
1229 | """ |
|
1229 | """ | |
1230 | ) |
|
1230 | ) | |
1231 |
|
1231 | |||
1232 | _generaldelta = interfaceutil.Attribute( |
|
1232 | _generaldelta = interfaceutil.Attribute( | |
1233 | """Whether generaldelta storage is being used. |
|
1233 | """Whether generaldelta storage is being used. | |
1234 |
|
1234 | |||
1235 | TODO this is revlog specific and should not be exposed. |
|
1235 | TODO this is revlog specific and should not be exposed. | |
1236 | """ |
|
1236 | """ | |
1237 | ) |
|
1237 | ) | |
1238 |
|
1238 | |||
1239 | fulltextcache = interfaceutil.Attribute( |
|
1239 | fulltextcache = interfaceutil.Attribute( | |
1240 | """Dict with cache of fulltexts. |
|
1240 | """Dict with cache of fulltexts. | |
1241 |
|
1241 | |||
1242 | TODO this doesn't feel appropriate for the storage interface. |
|
1242 | TODO this doesn't feel appropriate for the storage interface. | |
1243 | """ |
|
1243 | """ | |
1244 | ) |
|
1244 | ) | |
1245 |
|
1245 | |||
1246 | def __len__(): |
|
1246 | def __len__(): | |
1247 | """Obtain the number of revisions stored for this manifest.""" |
|
1247 | """Obtain the number of revisions stored for this manifest.""" | |
1248 |
|
1248 | |||
1249 | def __iter__(): |
|
1249 | def __iter__(): | |
1250 | """Iterate over revision numbers for this manifest.""" |
|
1250 | """Iterate over revision numbers for this manifest.""" | |
1251 |
|
1251 | |||
1252 | def rev(node): |
|
1252 | def rev(node): | |
1253 | """Obtain the revision number given a binary node. |
|
1253 | """Obtain the revision number given a binary node. | |
1254 |
|
1254 | |||
1255 | Raises ``error.LookupError`` if the node is not known. |
|
1255 | Raises ``error.LookupError`` if the node is not known. | |
1256 | """ |
|
1256 | """ | |
1257 |
|
1257 | |||
1258 | def node(rev): |
|
1258 | def node(rev): | |
1259 | """Obtain the node value given a revision number. |
|
1259 | """Obtain the node value given a revision number. | |
1260 |
|
1260 | |||
1261 | Raises ``error.LookupError`` if the revision is not known. |
|
1261 | Raises ``error.LookupError`` if the revision is not known. | |
1262 | """ |
|
1262 | """ | |
1263 |
|
1263 | |||
1264 | def lookup(value): |
|
1264 | def lookup(value): | |
1265 | """Attempt to resolve a value to a node. |
|
1265 | """Attempt to resolve a value to a node. | |
1266 |
|
1266 | |||
1267 | Value can be a binary node, hex node, revision number, or a bytes |
|
1267 | Value can be a binary node, hex node, revision number, or a bytes | |
1268 | that can be converted to an integer. |
|
1268 | that can be converted to an integer. | |
1269 |
|
1269 | |||
1270 | Raises ``error.LookupError`` if a ndoe could not be resolved. |
|
1270 | Raises ``error.LookupError`` if a ndoe could not be resolved. | |
1271 | """ |
|
1271 | """ | |
1272 |
|
1272 | |||
1273 | def parents(node): |
|
1273 | def parents(node): | |
1274 | """Returns a 2-tuple of parent nodes for a node. |
|
1274 | """Returns a 2-tuple of parent nodes for a node. | |
1275 |
|
1275 | |||
1276 | Values will be ``nullid`` if the parent is empty. |
|
1276 | Values will be ``nullid`` if the parent is empty. | |
1277 | """ |
|
1277 | """ | |
1278 |
|
1278 | |||
1279 | def parentrevs(rev): |
|
1279 | def parentrevs(rev): | |
1280 | """Like parents() but operates on revision numbers.""" |
|
1280 | """Like parents() but operates on revision numbers.""" | |
1281 |
|
1281 | |||
1282 | def linkrev(rev): |
|
1282 | def linkrev(rev): | |
1283 | """Obtain the changeset revision number a revision is linked to.""" |
|
1283 | """Obtain the changeset revision number a revision is linked to.""" | |
1284 |
|
1284 | |||
1285 | def revision(node): |
|
1285 | def revision(node): | |
1286 | """Obtain fulltext data for a node.""" |
|
1286 | """Obtain fulltext data for a node.""" | |
1287 |
|
1287 | |||
1288 |
def rawdata(node |
|
1288 | def rawdata(node): | |
1289 | """Obtain raw data for a node.""" |
|
1289 | """Obtain raw data for a node.""" | |
1290 |
|
1290 | |||
1291 | def revdiff(rev1, rev2): |
|
1291 | def revdiff(rev1, rev2): | |
1292 | """Obtain a delta between two revision numbers. |
|
1292 | """Obtain a delta between two revision numbers. | |
1293 |
|
1293 | |||
1294 | The returned data is the result of ``bdiff.bdiff()`` on the raw |
|
1294 | The returned data is the result of ``bdiff.bdiff()`` on the raw | |
1295 | revision data. |
|
1295 | revision data. | |
1296 | """ |
|
1296 | """ | |
1297 |
|
1297 | |||
1298 | def cmp(node, fulltext): |
|
1298 | def cmp(node, fulltext): | |
1299 | """Compare fulltext to another revision. |
|
1299 | """Compare fulltext to another revision. | |
1300 |
|
1300 | |||
1301 | Returns True if the fulltext is different from what is stored. |
|
1301 | Returns True if the fulltext is different from what is stored. | |
1302 | """ |
|
1302 | """ | |
1303 |
|
1303 | |||
1304 | def emitrevisions( |
|
1304 | def emitrevisions( | |
1305 | nodes, |
|
1305 | nodes, | |
1306 | nodesorder=None, |
|
1306 | nodesorder=None, | |
1307 | revisiondata=False, |
|
1307 | revisiondata=False, | |
1308 | assumehaveparentrevisions=False, |
|
1308 | assumehaveparentrevisions=False, | |
1309 | ): |
|
1309 | ): | |
1310 | """Produce ``irevisiondelta`` describing revisions. |
|
1310 | """Produce ``irevisiondelta`` describing revisions. | |
1311 |
|
1311 | |||
1312 | See the documentation for ``ifiledata`` for more. |
|
1312 | See the documentation for ``ifiledata`` for more. | |
1313 | """ |
|
1313 | """ | |
1314 |
|
1314 | |||
1315 | def addgroup( |
|
1315 | def addgroup( | |
1316 | deltas, |
|
1316 | deltas, | |
1317 | linkmapper, |
|
1317 | linkmapper, | |
1318 | transaction, |
|
1318 | transaction, | |
1319 | addrevisioncb=None, |
|
1319 | addrevisioncb=None, | |
1320 | duplicaterevisioncb=None, |
|
1320 | duplicaterevisioncb=None, | |
1321 | ): |
|
1321 | ): | |
1322 | """Process a series of deltas for storage. |
|
1322 | """Process a series of deltas for storage. | |
1323 |
|
1323 | |||
1324 | See the documentation in ``ifilemutation`` for more. |
|
1324 | See the documentation in ``ifilemutation`` for more. | |
1325 | """ |
|
1325 | """ | |
1326 |
|
1326 | |||
1327 | def rawsize(rev): |
|
1327 | def rawsize(rev): | |
1328 | """Obtain the size of tracked data. |
|
1328 | """Obtain the size of tracked data. | |
1329 |
|
1329 | |||
1330 | Is equivalent to ``len(m.rawdata(node))``. |
|
1330 | Is equivalent to ``len(m.rawdata(node))``. | |
1331 |
|
1331 | |||
1332 | TODO this method is only used by upgrade code and may be removed. |
|
1332 | TODO this method is only used by upgrade code and may be removed. | |
1333 | """ |
|
1333 | """ | |
1334 |
|
1334 | |||
1335 | def getstrippoint(minlink): |
|
1335 | def getstrippoint(minlink): | |
1336 | """Find minimum revision that must be stripped to strip a linkrev. |
|
1336 | """Find minimum revision that must be stripped to strip a linkrev. | |
1337 |
|
1337 | |||
1338 | See the documentation in ``ifilemutation`` for more. |
|
1338 | See the documentation in ``ifilemutation`` for more. | |
1339 | """ |
|
1339 | """ | |
1340 |
|
1340 | |||
1341 | def strip(minlink, transaction): |
|
1341 | def strip(minlink, transaction): | |
1342 | """Remove storage of items starting at a linkrev. |
|
1342 | """Remove storage of items starting at a linkrev. | |
1343 |
|
1343 | |||
1344 | See the documentation in ``ifilemutation`` for more. |
|
1344 | See the documentation in ``ifilemutation`` for more. | |
1345 | """ |
|
1345 | """ | |
1346 |
|
1346 | |||
1347 | def checksize(): |
|
1347 | def checksize(): | |
1348 | """Obtain the expected sizes of backing files. |
|
1348 | """Obtain the expected sizes of backing files. | |
1349 |
|
1349 | |||
1350 | TODO this is used by verify and it should not be part of the interface. |
|
1350 | TODO this is used by verify and it should not be part of the interface. | |
1351 | """ |
|
1351 | """ | |
1352 |
|
1352 | |||
1353 | def files(): |
|
1353 | def files(): | |
1354 | """Obtain paths that are backing storage for this manifest. |
|
1354 | """Obtain paths that are backing storage for this manifest. | |
1355 |
|
1355 | |||
1356 | TODO this is used by verify and there should probably be a better API |
|
1356 | TODO this is used by verify and there should probably be a better API | |
1357 | for this functionality. |
|
1357 | for this functionality. | |
1358 | """ |
|
1358 | """ | |
1359 |
|
1359 | |||
1360 | def deltaparent(rev): |
|
1360 | def deltaparent(rev): | |
1361 | """Obtain the revision that a revision is delta'd against. |
|
1361 | """Obtain the revision that a revision is delta'd against. | |
1362 |
|
1362 | |||
1363 | TODO delta encoding is an implementation detail of storage and should |
|
1363 | TODO delta encoding is an implementation detail of storage and should | |
1364 | not be exposed to the storage interface. |
|
1364 | not be exposed to the storage interface. | |
1365 | """ |
|
1365 | """ | |
1366 |
|
1366 | |||
1367 | def clone(tr, dest, **kwargs): |
|
1367 | def clone(tr, dest, **kwargs): | |
1368 | """Clone this instance to another.""" |
|
1368 | """Clone this instance to another.""" | |
1369 |
|
1369 | |||
1370 | def clearcaches(clear_persisted_data=False): |
|
1370 | def clearcaches(clear_persisted_data=False): | |
1371 | """Clear any caches associated with this instance.""" |
|
1371 | """Clear any caches associated with this instance.""" | |
1372 |
|
1372 | |||
1373 | def dirlog(d): |
|
1373 | def dirlog(d): | |
1374 | """Obtain a manifest storage instance for a tree.""" |
|
1374 | """Obtain a manifest storage instance for a tree.""" | |
1375 |
|
1375 | |||
1376 | def add( |
|
1376 | def add( | |
1377 | m, transaction, link, p1, p2, added, removed, readtree=None, match=None |
|
1377 | m, transaction, link, p1, p2, added, removed, readtree=None, match=None | |
1378 | ): |
|
1378 | ): | |
1379 | """Add a revision to storage. |
|
1379 | """Add a revision to storage. | |
1380 |
|
1380 | |||
1381 | ``m`` is an object conforming to ``imanifestdict``. |
|
1381 | ``m`` is an object conforming to ``imanifestdict``. | |
1382 |
|
1382 | |||
1383 | ``link`` is the linkrev revision number. |
|
1383 | ``link`` is the linkrev revision number. | |
1384 |
|
1384 | |||
1385 | ``p1`` and ``p2`` are the parent revision numbers. |
|
1385 | ``p1`` and ``p2`` are the parent revision numbers. | |
1386 |
|
1386 | |||
1387 | ``added`` and ``removed`` are iterables of added and removed paths, |
|
1387 | ``added`` and ``removed`` are iterables of added and removed paths, | |
1388 | respectively. |
|
1388 | respectively. | |
1389 |
|
1389 | |||
1390 | ``readtree`` is a function that can be used to read the child tree(s) |
|
1390 | ``readtree`` is a function that can be used to read the child tree(s) | |
1391 | when recursively writing the full tree structure when using |
|
1391 | when recursively writing the full tree structure when using | |
1392 | treemanifets. |
|
1392 | treemanifets. | |
1393 |
|
1393 | |||
1394 | ``match`` is a matcher that can be used to hint to storage that not all |
|
1394 | ``match`` is a matcher that can be used to hint to storage that not all | |
1395 | paths must be inspected; this is an optimization and can be safely |
|
1395 | paths must be inspected; this is an optimization and can be safely | |
1396 | ignored. Note that the storage must still be able to reproduce a full |
|
1396 | ignored. Note that the storage must still be able to reproduce a full | |
1397 | manifest including files that did not match. |
|
1397 | manifest including files that did not match. | |
1398 | """ |
|
1398 | """ | |
1399 |
|
1399 | |||
1400 | def storageinfo( |
|
1400 | def storageinfo( | |
1401 | exclusivefiles=False, |
|
1401 | exclusivefiles=False, | |
1402 | sharedfiles=False, |
|
1402 | sharedfiles=False, | |
1403 | revisionscount=False, |
|
1403 | revisionscount=False, | |
1404 | trackedsize=False, |
|
1404 | trackedsize=False, | |
1405 | storedsize=False, |
|
1405 | storedsize=False, | |
1406 | ): |
|
1406 | ): | |
1407 | """Obtain information about storage for this manifest's data. |
|
1407 | """Obtain information about storage for this manifest's data. | |
1408 |
|
1408 | |||
1409 | See ``ifilestorage.storageinfo()`` for a description of this method. |
|
1409 | See ``ifilestorage.storageinfo()`` for a description of this method. | |
1410 | This one behaves the same way, except for manifest data. |
|
1410 | This one behaves the same way, except for manifest data. | |
1411 | """ |
|
1411 | """ | |
1412 |
|
1412 | |||
1413 | def get_revlog(): |
|
1413 | def get_revlog(): | |
1414 | """return an actual revlog instance if any |
|
1414 | """return an actual revlog instance if any | |
1415 |
|
1415 | |||
1416 | This exist because a lot of code leverage the fact the underlying |
|
1416 | This exist because a lot of code leverage the fact the underlying | |
1417 | storage is a revlog for optimization, so giving simple way to access |
|
1417 | storage is a revlog for optimization, so giving simple way to access | |
1418 | the revlog instance helps such code. |
|
1418 | the revlog instance helps such code. | |
1419 | """ |
|
1419 | """ | |
1420 |
|
1420 | |||
1421 |
|
1421 | |||
1422 | class imanifestlog(interfaceutil.Interface): |
|
1422 | class imanifestlog(interfaceutil.Interface): | |
1423 | """Interface representing a collection of manifest snapshots. |
|
1423 | """Interface representing a collection of manifest snapshots. | |
1424 |
|
1424 | |||
1425 | Represents the root manifest in a repository. |
|
1425 | Represents the root manifest in a repository. | |
1426 |
|
1426 | |||
1427 | Also serves as a means to access nested tree manifests and to cache |
|
1427 | Also serves as a means to access nested tree manifests and to cache | |
1428 | tree manifests. |
|
1428 | tree manifests. | |
1429 | """ |
|
1429 | """ | |
1430 |
|
1430 | |||
1431 | nodeconstants = interfaceutil.Attribute( |
|
1431 | nodeconstants = interfaceutil.Attribute( | |
1432 | """nodeconstants used by the current repository.""" |
|
1432 | """nodeconstants used by the current repository.""" | |
1433 | ) |
|
1433 | ) | |
1434 |
|
1434 | |||
1435 | def __getitem__(node): |
|
1435 | def __getitem__(node): | |
1436 | """Obtain a manifest instance for a given binary node. |
|
1436 | """Obtain a manifest instance for a given binary node. | |
1437 |
|
1437 | |||
1438 | Equivalent to calling ``self.get('', node)``. |
|
1438 | Equivalent to calling ``self.get('', node)``. | |
1439 |
|
1439 | |||
1440 | The returned object conforms to the ``imanifestrevisionstored`` |
|
1440 | The returned object conforms to the ``imanifestrevisionstored`` | |
1441 | interface. |
|
1441 | interface. | |
1442 | """ |
|
1442 | """ | |
1443 |
|
1443 | |||
1444 | def get(tree, node, verify=True): |
|
1444 | def get(tree, node, verify=True): | |
1445 | """Retrieve the manifest instance for a given directory and binary node. |
|
1445 | """Retrieve the manifest instance for a given directory and binary node. | |
1446 |
|
1446 | |||
1447 | ``node`` always refers to the node of the root manifest (which will be |
|
1447 | ``node`` always refers to the node of the root manifest (which will be | |
1448 | the only manifest if flat manifests are being used). |
|
1448 | the only manifest if flat manifests are being used). | |
1449 |
|
1449 | |||
1450 | If ``tree`` is the empty string, the root manifest is returned. |
|
1450 | If ``tree`` is the empty string, the root manifest is returned. | |
1451 | Otherwise the manifest for the specified directory will be returned |
|
1451 | Otherwise the manifest for the specified directory will be returned | |
1452 | (requires tree manifests). |
|
1452 | (requires tree manifests). | |
1453 |
|
1453 | |||
1454 | If ``verify`` is True, ``LookupError`` is raised if the node is not |
|
1454 | If ``verify`` is True, ``LookupError`` is raised if the node is not | |
1455 | known. |
|
1455 | known. | |
1456 |
|
1456 | |||
1457 | The returned object conforms to the ``imanifestrevisionstored`` |
|
1457 | The returned object conforms to the ``imanifestrevisionstored`` | |
1458 | interface. |
|
1458 | interface. | |
1459 | """ |
|
1459 | """ | |
1460 |
|
1460 | |||
1461 | def getstorage(tree): |
|
1461 | def getstorage(tree): | |
1462 | """Retrieve an interface to storage for a particular tree. |
|
1462 | """Retrieve an interface to storage for a particular tree. | |
1463 |
|
1463 | |||
1464 | If ``tree`` is the empty bytestring, storage for the root manifest will |
|
1464 | If ``tree`` is the empty bytestring, storage for the root manifest will | |
1465 | be returned. Otherwise storage for a tree manifest is returned. |
|
1465 | be returned. Otherwise storage for a tree manifest is returned. | |
1466 |
|
1466 | |||
1467 | TODO formalize interface for returned object. |
|
1467 | TODO formalize interface for returned object. | |
1468 | """ |
|
1468 | """ | |
1469 |
|
1469 | |||
1470 | def clearcaches(): |
|
1470 | def clearcaches(): | |
1471 | """Clear caches associated with this collection.""" |
|
1471 | """Clear caches associated with this collection.""" | |
1472 |
|
1472 | |||
1473 | def rev(node): |
|
1473 | def rev(node): | |
1474 | """Obtain the revision number for a binary node. |
|
1474 | """Obtain the revision number for a binary node. | |
1475 |
|
1475 | |||
1476 | Raises ``error.LookupError`` if the node is not known. |
|
1476 | Raises ``error.LookupError`` if the node is not known. | |
1477 | """ |
|
1477 | """ | |
1478 |
|
1478 | |||
1479 | def update_caches(transaction): |
|
1479 | def update_caches(transaction): | |
1480 | """update whatever cache are relevant for the used storage.""" |
|
1480 | """update whatever cache are relevant for the used storage.""" | |
1481 |
|
1481 | |||
1482 |
|
1482 | |||
1483 | class ilocalrepositoryfilestorage(interfaceutil.Interface): |
|
1483 | class ilocalrepositoryfilestorage(interfaceutil.Interface): | |
1484 | """Local repository sub-interface providing access to tracked file storage. |
|
1484 | """Local repository sub-interface providing access to tracked file storage. | |
1485 |
|
1485 | |||
1486 | This interface defines how a repository accesses storage for a single |
|
1486 | This interface defines how a repository accesses storage for a single | |
1487 | tracked file path. |
|
1487 | tracked file path. | |
1488 | """ |
|
1488 | """ | |
1489 |
|
1489 | |||
1490 | def file(f): |
|
1490 | def file(f): | |
1491 | """Obtain a filelog for a tracked path. |
|
1491 | """Obtain a filelog for a tracked path. | |
1492 |
|
1492 | |||
1493 | The returned type conforms to the ``ifilestorage`` interface. |
|
1493 | The returned type conforms to the ``ifilestorage`` interface. | |
1494 | """ |
|
1494 | """ | |
1495 |
|
1495 | |||
1496 |
|
1496 | |||
1497 | class ilocalrepositorymain(interfaceutil.Interface): |
|
1497 | class ilocalrepositorymain(interfaceutil.Interface): | |
1498 | """Main interface for local repositories. |
|
1498 | """Main interface for local repositories. | |
1499 |
|
1499 | |||
1500 | This currently captures the reality of things - not how things should be. |
|
1500 | This currently captures the reality of things - not how things should be. | |
1501 | """ |
|
1501 | """ | |
1502 |
|
1502 | |||
1503 | nodeconstants = interfaceutil.Attribute( |
|
1503 | nodeconstants = interfaceutil.Attribute( | |
1504 | """Constant nodes matching the hash function used by the repository.""" |
|
1504 | """Constant nodes matching the hash function used by the repository.""" | |
1505 | ) |
|
1505 | ) | |
1506 | nullid = interfaceutil.Attribute( |
|
1506 | nullid = interfaceutil.Attribute( | |
1507 | """null revision for the hash function used by the repository.""" |
|
1507 | """null revision for the hash function used by the repository.""" | |
1508 | ) |
|
1508 | ) | |
1509 |
|
1509 | |||
1510 | supported = interfaceutil.Attribute( |
|
1510 | supported = interfaceutil.Attribute( | |
1511 | """Set of requirements that this repo is capable of opening.""" |
|
1511 | """Set of requirements that this repo is capable of opening.""" | |
1512 | ) |
|
1512 | ) | |
1513 |
|
1513 | |||
1514 | requirements = interfaceutil.Attribute( |
|
1514 | requirements = interfaceutil.Attribute( | |
1515 | """Set of requirements this repo uses.""" |
|
1515 | """Set of requirements this repo uses.""" | |
1516 | ) |
|
1516 | ) | |
1517 |
|
1517 | |||
1518 | features = interfaceutil.Attribute( |
|
1518 | features = interfaceutil.Attribute( | |
1519 | """Set of "features" this repository supports. |
|
1519 | """Set of "features" this repository supports. | |
1520 |
|
1520 | |||
1521 | A "feature" is a loosely-defined term. It can refer to a feature |
|
1521 | A "feature" is a loosely-defined term. It can refer to a feature | |
1522 | in the classical sense or can describe an implementation detail |
|
1522 | in the classical sense or can describe an implementation detail | |
1523 | of the repository. For example, a ``readonly`` feature may denote |
|
1523 | of the repository. For example, a ``readonly`` feature may denote | |
1524 | the repository as read-only. Or a ``revlogfilestore`` feature may |
|
1524 | the repository as read-only. Or a ``revlogfilestore`` feature may | |
1525 | denote that the repository is using revlogs for file storage. |
|
1525 | denote that the repository is using revlogs for file storage. | |
1526 |
|
1526 | |||
1527 | The intent of features is to provide a machine-queryable mechanism |
|
1527 | The intent of features is to provide a machine-queryable mechanism | |
1528 | for repo consumers to test for various repository characteristics. |
|
1528 | for repo consumers to test for various repository characteristics. | |
1529 |
|
1529 | |||
1530 | Features are similar to ``requirements``. The main difference is that |
|
1530 | Features are similar to ``requirements``. The main difference is that | |
1531 | requirements are stored on-disk and represent requirements to open the |
|
1531 | requirements are stored on-disk and represent requirements to open the | |
1532 | repository. Features are more run-time capabilities of the repository |
|
1532 | repository. Features are more run-time capabilities of the repository | |
1533 | and more granular capabilities (which may be derived from requirements). |
|
1533 | and more granular capabilities (which may be derived from requirements). | |
1534 | """ |
|
1534 | """ | |
1535 | ) |
|
1535 | ) | |
1536 |
|
1536 | |||
1537 | filtername = interfaceutil.Attribute( |
|
1537 | filtername = interfaceutil.Attribute( | |
1538 | """Name of the repoview that is active on this repo.""" |
|
1538 | """Name of the repoview that is active on this repo.""" | |
1539 | ) |
|
1539 | ) | |
1540 |
|
1540 | |||
1541 | vfs_map = interfaceutil.Attribute( |
|
1541 | vfs_map = interfaceutil.Attribute( | |
1542 | """a bytes-key β vfs mapping used by transaction and others""" |
|
1542 | """a bytes-key β vfs mapping used by transaction and others""" | |
1543 | ) |
|
1543 | ) | |
1544 |
|
1544 | |||
1545 | wvfs = interfaceutil.Attribute( |
|
1545 | wvfs = interfaceutil.Attribute( | |
1546 | """VFS used to access the working directory.""" |
|
1546 | """VFS used to access the working directory.""" | |
1547 | ) |
|
1547 | ) | |
1548 |
|
1548 | |||
1549 | vfs = interfaceutil.Attribute( |
|
1549 | vfs = interfaceutil.Attribute( | |
1550 | """VFS rooted at the .hg directory. |
|
1550 | """VFS rooted at the .hg directory. | |
1551 |
|
1551 | |||
1552 | Used to access repository data not in the store. |
|
1552 | Used to access repository data not in the store. | |
1553 | """ |
|
1553 | """ | |
1554 | ) |
|
1554 | ) | |
1555 |
|
1555 | |||
1556 | svfs = interfaceutil.Attribute( |
|
1556 | svfs = interfaceutil.Attribute( | |
1557 | """VFS rooted at the store. |
|
1557 | """VFS rooted at the store. | |
1558 |
|
1558 | |||
1559 | Used to access repository data in the store. Typically .hg/store. |
|
1559 | Used to access repository data in the store. Typically .hg/store. | |
1560 | But can point elsewhere if the store is shared. |
|
1560 | But can point elsewhere if the store is shared. | |
1561 | """ |
|
1561 | """ | |
1562 | ) |
|
1562 | ) | |
1563 |
|
1563 | |||
1564 | root = interfaceutil.Attribute( |
|
1564 | root = interfaceutil.Attribute( | |
1565 | """Path to the root of the working directory.""" |
|
1565 | """Path to the root of the working directory.""" | |
1566 | ) |
|
1566 | ) | |
1567 |
|
1567 | |||
1568 | path = interfaceutil.Attribute("""Path to the .hg directory.""") |
|
1568 | path = interfaceutil.Attribute("""Path to the .hg directory.""") | |
1569 |
|
1569 | |||
1570 | origroot = interfaceutil.Attribute( |
|
1570 | origroot = interfaceutil.Attribute( | |
1571 | """The filesystem path that was used to construct the repo.""" |
|
1571 | """The filesystem path that was used to construct the repo.""" | |
1572 | ) |
|
1572 | ) | |
1573 |
|
1573 | |||
1574 | auditor = interfaceutil.Attribute( |
|
1574 | auditor = interfaceutil.Attribute( | |
1575 | """A pathauditor for the working directory. |
|
1575 | """A pathauditor for the working directory. | |
1576 |
|
1576 | |||
1577 | This checks if a path refers to a nested repository. |
|
1577 | This checks if a path refers to a nested repository. | |
1578 |
|
1578 | |||
1579 | Operates on the filesystem. |
|
1579 | Operates on the filesystem. | |
1580 | """ |
|
1580 | """ | |
1581 | ) |
|
1581 | ) | |
1582 |
|
1582 | |||
1583 | nofsauditor = interfaceutil.Attribute( |
|
1583 | nofsauditor = interfaceutil.Attribute( | |
1584 | """A pathauditor for the working directory. |
|
1584 | """A pathauditor for the working directory. | |
1585 |
|
1585 | |||
1586 | This is like ``auditor`` except it doesn't do filesystem checks. |
|
1586 | This is like ``auditor`` except it doesn't do filesystem checks. | |
1587 | """ |
|
1587 | """ | |
1588 | ) |
|
1588 | ) | |
1589 |
|
1589 | |||
1590 | baseui = interfaceutil.Attribute( |
|
1590 | baseui = interfaceutil.Attribute( | |
1591 | """Original ui instance passed into constructor.""" |
|
1591 | """Original ui instance passed into constructor.""" | |
1592 | ) |
|
1592 | ) | |
1593 |
|
1593 | |||
1594 | ui = interfaceutil.Attribute("""Main ui instance for this instance.""") |
|
1594 | ui = interfaceutil.Attribute("""Main ui instance for this instance.""") | |
1595 |
|
1595 | |||
1596 | sharedpath = interfaceutil.Attribute( |
|
1596 | sharedpath = interfaceutil.Attribute( | |
1597 | """Path to the .hg directory of the repo this repo was shared from.""" |
|
1597 | """Path to the .hg directory of the repo this repo was shared from.""" | |
1598 | ) |
|
1598 | ) | |
1599 |
|
1599 | |||
1600 | store = interfaceutil.Attribute("""A store instance.""") |
|
1600 | store = interfaceutil.Attribute("""A store instance.""") | |
1601 |
|
1601 | |||
1602 | spath = interfaceutil.Attribute("""Path to the store.""") |
|
1602 | spath = interfaceutil.Attribute("""Path to the store.""") | |
1603 |
|
1603 | |||
1604 | sjoin = interfaceutil.Attribute("""Alias to self.store.join.""") |
|
1604 | sjoin = interfaceutil.Attribute("""Alias to self.store.join.""") | |
1605 |
|
1605 | |||
1606 | cachevfs = interfaceutil.Attribute( |
|
1606 | cachevfs = interfaceutil.Attribute( | |
1607 | """A VFS used to access the cache directory. |
|
1607 | """A VFS used to access the cache directory. | |
1608 |
|
1608 | |||
1609 | Typically .hg/cache. |
|
1609 | Typically .hg/cache. | |
1610 | """ |
|
1610 | """ | |
1611 | ) |
|
1611 | ) | |
1612 |
|
1612 | |||
1613 | wcachevfs = interfaceutil.Attribute( |
|
1613 | wcachevfs = interfaceutil.Attribute( | |
1614 | """A VFS used to access the cache directory dedicated to working copy |
|
1614 | """A VFS used to access the cache directory dedicated to working copy | |
1615 |
|
1615 | |||
1616 | Typically .hg/wcache. |
|
1616 | Typically .hg/wcache. | |
1617 | """ |
|
1617 | """ | |
1618 | ) |
|
1618 | ) | |
1619 |
|
1619 | |||
1620 | filteredrevcache = interfaceutil.Attribute( |
|
1620 | filteredrevcache = interfaceutil.Attribute( | |
1621 | """Holds sets of revisions to be filtered.""" |
|
1621 | """Holds sets of revisions to be filtered.""" | |
1622 | ) |
|
1622 | ) | |
1623 |
|
1623 | |||
1624 | names = interfaceutil.Attribute("""A ``namespaces`` instance.""") |
|
1624 | names = interfaceutil.Attribute("""A ``namespaces`` instance.""") | |
1625 |
|
1625 | |||
1626 | filecopiesmode = interfaceutil.Attribute( |
|
1626 | filecopiesmode = interfaceutil.Attribute( | |
1627 | """The way files copies should be dealt with in this repo.""" |
|
1627 | """The way files copies should be dealt with in this repo.""" | |
1628 | ) |
|
1628 | ) | |
1629 |
|
1629 | |||
1630 | def close(): |
|
1630 | def close(): | |
1631 | """Close the handle on this repository.""" |
|
1631 | """Close the handle on this repository.""" | |
1632 |
|
1632 | |||
1633 | def peer(path=None): |
|
1633 | def peer(path=None): | |
1634 | """Obtain an object conforming to the ``peer`` interface.""" |
|
1634 | """Obtain an object conforming to the ``peer`` interface.""" | |
1635 |
|
1635 | |||
1636 | def unfiltered(): |
|
1636 | def unfiltered(): | |
1637 | """Obtain an unfiltered/raw view of this repo.""" |
|
1637 | """Obtain an unfiltered/raw view of this repo.""" | |
1638 |
|
1638 | |||
1639 | def filtered(name, visibilityexceptions=None): |
|
1639 | def filtered(name, visibilityexceptions=None): | |
1640 | """Obtain a named view of this repository.""" |
|
1640 | """Obtain a named view of this repository.""" | |
1641 |
|
1641 | |||
1642 | obsstore = interfaceutil.Attribute("""A store of obsolescence data.""") |
|
1642 | obsstore = interfaceutil.Attribute("""A store of obsolescence data.""") | |
1643 |
|
1643 | |||
1644 | changelog = interfaceutil.Attribute("""A handle on the changelog revlog.""") |
|
1644 | changelog = interfaceutil.Attribute("""A handle on the changelog revlog.""") | |
1645 |
|
1645 | |||
1646 | manifestlog = interfaceutil.Attribute( |
|
1646 | manifestlog = interfaceutil.Attribute( | |
1647 | """An instance conforming to the ``imanifestlog`` interface. |
|
1647 | """An instance conforming to the ``imanifestlog`` interface. | |
1648 |
|
1648 | |||
1649 | Provides access to manifests for the repository. |
|
1649 | Provides access to manifests for the repository. | |
1650 | """ |
|
1650 | """ | |
1651 | ) |
|
1651 | ) | |
1652 |
|
1652 | |||
1653 | dirstate = interfaceutil.Attribute("""Working directory state.""") |
|
1653 | dirstate = interfaceutil.Attribute("""Working directory state.""") | |
1654 |
|
1654 | |||
1655 | narrowpats = interfaceutil.Attribute( |
|
1655 | narrowpats = interfaceutil.Attribute( | |
1656 | """Matcher patterns for this repository's narrowspec.""" |
|
1656 | """Matcher patterns for this repository's narrowspec.""" | |
1657 | ) |
|
1657 | ) | |
1658 |
|
1658 | |||
1659 | def narrowmatch(match=None, includeexact=False): |
|
1659 | def narrowmatch(match=None, includeexact=False): | |
1660 | """Obtain a matcher for the narrowspec.""" |
|
1660 | """Obtain a matcher for the narrowspec.""" | |
1661 |
|
1661 | |||
1662 | def setnarrowpats(newincludes, newexcludes): |
|
1662 | def setnarrowpats(newincludes, newexcludes): | |
1663 | """Define the narrowspec for this repository.""" |
|
1663 | """Define the narrowspec for this repository.""" | |
1664 |
|
1664 | |||
1665 | def __getitem__(changeid): |
|
1665 | def __getitem__(changeid): | |
1666 | """Try to resolve a changectx.""" |
|
1666 | """Try to resolve a changectx.""" | |
1667 |
|
1667 | |||
1668 | def __contains__(changeid): |
|
1668 | def __contains__(changeid): | |
1669 | """Whether a changeset exists.""" |
|
1669 | """Whether a changeset exists.""" | |
1670 |
|
1670 | |||
1671 | def __nonzero__(): |
|
1671 | def __nonzero__(): | |
1672 | """Always returns True.""" |
|
1672 | """Always returns True.""" | |
1673 | return True |
|
1673 | return True | |
1674 |
|
1674 | |||
1675 | __bool__ = __nonzero__ |
|
1675 | __bool__ = __nonzero__ | |
1676 |
|
1676 | |||
1677 | def __len__(): |
|
1677 | def __len__(): | |
1678 | """Returns the number of changesets in the repo.""" |
|
1678 | """Returns the number of changesets in the repo.""" | |
1679 |
|
1679 | |||
1680 | def __iter__(): |
|
1680 | def __iter__(): | |
1681 | """Iterate over revisions in the changelog.""" |
|
1681 | """Iterate over revisions in the changelog.""" | |
1682 |
|
1682 | |||
1683 | def revs(expr, *args): |
|
1683 | def revs(expr, *args): | |
1684 | """Evaluate a revset. |
|
1684 | """Evaluate a revset. | |
1685 |
|
1685 | |||
1686 | Emits revisions. |
|
1686 | Emits revisions. | |
1687 | """ |
|
1687 | """ | |
1688 |
|
1688 | |||
1689 | def set(expr, *args): |
|
1689 | def set(expr, *args): | |
1690 | """Evaluate a revset. |
|
1690 | """Evaluate a revset. | |
1691 |
|
1691 | |||
1692 | Emits changectx instances. |
|
1692 | Emits changectx instances. | |
1693 | """ |
|
1693 | """ | |
1694 |
|
1694 | |||
1695 | def anyrevs(specs, user=False, localalias=None): |
|
1695 | def anyrevs(specs, user=False, localalias=None): | |
1696 | """Find revisions matching one of the given revsets.""" |
|
1696 | """Find revisions matching one of the given revsets.""" | |
1697 |
|
1697 | |||
1698 | def url(): |
|
1698 | def url(): | |
1699 | """Returns a string representing the location of this repo.""" |
|
1699 | """Returns a string representing the location of this repo.""" | |
1700 |
|
1700 | |||
1701 | def hook(name, throw=False, **args): |
|
1701 | def hook(name, throw=False, **args): | |
1702 | """Call a hook.""" |
|
1702 | """Call a hook.""" | |
1703 |
|
1703 | |||
1704 | def tags(): |
|
1704 | def tags(): | |
1705 | """Return a mapping of tag to node.""" |
|
1705 | """Return a mapping of tag to node.""" | |
1706 |
|
1706 | |||
1707 | def tagtype(tagname): |
|
1707 | def tagtype(tagname): | |
1708 | """Return the type of a given tag.""" |
|
1708 | """Return the type of a given tag.""" | |
1709 |
|
1709 | |||
1710 | def tagslist(): |
|
1710 | def tagslist(): | |
1711 | """Return a list of tags ordered by revision.""" |
|
1711 | """Return a list of tags ordered by revision.""" | |
1712 |
|
1712 | |||
1713 | def nodetags(node): |
|
1713 | def nodetags(node): | |
1714 | """Return the tags associated with a node.""" |
|
1714 | """Return the tags associated with a node.""" | |
1715 |
|
1715 | |||
1716 | def nodebookmarks(node): |
|
1716 | def nodebookmarks(node): | |
1717 | """Return the list of bookmarks pointing to the specified node.""" |
|
1717 | """Return the list of bookmarks pointing to the specified node.""" | |
1718 |
|
1718 | |||
1719 | def branchmap(): |
|
1719 | def branchmap(): | |
1720 | """Return a mapping of branch to heads in that branch.""" |
|
1720 | """Return a mapping of branch to heads in that branch.""" | |
1721 |
|
1721 | |||
1722 | def revbranchcache(): |
|
1722 | def revbranchcache(): | |
1723 | pass |
|
1723 | pass | |
1724 |
|
1724 | |||
1725 | def register_changeset(rev, changelogrevision): |
|
1725 | def register_changeset(rev, changelogrevision): | |
1726 | """Extension point for caches for new nodes. |
|
1726 | """Extension point for caches for new nodes. | |
1727 |
|
1727 | |||
1728 | Multiple consumers are expected to need parts of the changelogrevision, |
|
1728 | Multiple consumers are expected to need parts of the changelogrevision, | |
1729 | so it is provided as optimization to avoid duplicate lookups. A simple |
|
1729 | so it is provided as optimization to avoid duplicate lookups. A simple | |
1730 | cache would be fragile when other revisions are accessed, too.""" |
|
1730 | cache would be fragile when other revisions are accessed, too.""" | |
1731 | pass |
|
1731 | pass | |
1732 |
|
1732 | |||
1733 | def branchtip(branchtip, ignoremissing=False): |
|
1733 | def branchtip(branchtip, ignoremissing=False): | |
1734 | """Return the tip node for a given branch.""" |
|
1734 | """Return the tip node for a given branch.""" | |
1735 |
|
1735 | |||
1736 | def lookup(key): |
|
1736 | def lookup(key): | |
1737 | """Resolve the node for a revision.""" |
|
1737 | """Resolve the node for a revision.""" | |
1738 |
|
1738 | |||
1739 | def lookupbranch(key): |
|
1739 | def lookupbranch(key): | |
1740 | """Look up the branch name of the given revision or branch name.""" |
|
1740 | """Look up the branch name of the given revision or branch name.""" | |
1741 |
|
1741 | |||
1742 | def known(nodes): |
|
1742 | def known(nodes): | |
1743 | """Determine whether a series of nodes is known. |
|
1743 | """Determine whether a series of nodes is known. | |
1744 |
|
1744 | |||
1745 | Returns a list of bools. |
|
1745 | Returns a list of bools. | |
1746 | """ |
|
1746 | """ | |
1747 |
|
1747 | |||
1748 | def local(): |
|
1748 | def local(): | |
1749 | """Whether the repository is local.""" |
|
1749 | """Whether the repository is local.""" | |
1750 | return True |
|
1750 | return True | |
1751 |
|
1751 | |||
1752 | def publishing(): |
|
1752 | def publishing(): | |
1753 | """Whether the repository is a publishing repository.""" |
|
1753 | """Whether the repository is a publishing repository.""" | |
1754 |
|
1754 | |||
1755 | def cancopy(): |
|
1755 | def cancopy(): | |
1756 | pass |
|
1756 | pass | |
1757 |
|
1757 | |||
1758 | def shared(): |
|
1758 | def shared(): | |
1759 | """The type of shared repository or None.""" |
|
1759 | """The type of shared repository or None.""" | |
1760 |
|
1760 | |||
1761 | def wjoin(f, *insidef): |
|
1761 | def wjoin(f, *insidef): | |
1762 | """Calls self.vfs.reljoin(self.root, f, *insidef)""" |
|
1762 | """Calls self.vfs.reljoin(self.root, f, *insidef)""" | |
1763 |
|
1763 | |||
1764 | def setparents(p1, p2): |
|
1764 | def setparents(p1, p2): | |
1765 | """Set the parent nodes of the working directory.""" |
|
1765 | """Set the parent nodes of the working directory.""" | |
1766 |
|
1766 | |||
1767 | def filectx(path, changeid=None, fileid=None): |
|
1767 | def filectx(path, changeid=None, fileid=None): | |
1768 | """Obtain a filectx for the given file revision.""" |
|
1768 | """Obtain a filectx for the given file revision.""" | |
1769 |
|
1769 | |||
1770 | def getcwd(): |
|
1770 | def getcwd(): | |
1771 | """Obtain the current working directory from the dirstate.""" |
|
1771 | """Obtain the current working directory from the dirstate.""" | |
1772 |
|
1772 | |||
1773 | def pathto(f, cwd=None): |
|
1773 | def pathto(f, cwd=None): | |
1774 | """Obtain the relative path to a file.""" |
|
1774 | """Obtain the relative path to a file.""" | |
1775 |
|
1775 | |||
1776 | def adddatafilter(name, fltr): |
|
1776 | def adddatafilter(name, fltr): | |
1777 | pass |
|
1777 | pass | |
1778 |
|
1778 | |||
1779 | def wread(filename): |
|
1779 | def wread(filename): | |
1780 | """Read a file from wvfs, using data filters.""" |
|
1780 | """Read a file from wvfs, using data filters.""" | |
1781 |
|
1781 | |||
1782 | def wwrite(filename, data, flags, backgroundclose=False, **kwargs): |
|
1782 | def wwrite(filename, data, flags, backgroundclose=False, **kwargs): | |
1783 | """Write data to a file in the wvfs, using data filters.""" |
|
1783 | """Write data to a file in the wvfs, using data filters.""" | |
1784 |
|
1784 | |||
1785 | def wwritedata(filename, data): |
|
1785 | def wwritedata(filename, data): | |
1786 | """Resolve data for writing to the wvfs, using data filters.""" |
|
1786 | """Resolve data for writing to the wvfs, using data filters.""" | |
1787 |
|
1787 | |||
1788 | def currenttransaction(): |
|
1788 | def currenttransaction(): | |
1789 | """Obtain the current transaction instance or None.""" |
|
1789 | """Obtain the current transaction instance or None.""" | |
1790 |
|
1790 | |||
1791 | def transaction(desc, report=None): |
|
1791 | def transaction(desc, report=None): | |
1792 | """Open a new transaction to write to the repository.""" |
|
1792 | """Open a new transaction to write to the repository.""" | |
1793 |
|
1793 | |||
1794 | def undofiles(): |
|
1794 | def undofiles(): | |
1795 | """Returns a list of (vfs, path) for files to undo transactions.""" |
|
1795 | """Returns a list of (vfs, path) for files to undo transactions.""" | |
1796 |
|
1796 | |||
1797 | def recover(): |
|
1797 | def recover(): | |
1798 | """Roll back an interrupted transaction.""" |
|
1798 | """Roll back an interrupted transaction.""" | |
1799 |
|
1799 | |||
1800 | def rollback(dryrun=False, force=False): |
|
1800 | def rollback(dryrun=False, force=False): | |
1801 | """Undo the last transaction. |
|
1801 | """Undo the last transaction. | |
1802 |
|
1802 | |||
1803 | DANGEROUS. |
|
1803 | DANGEROUS. | |
1804 | """ |
|
1804 | """ | |
1805 |
|
1805 | |||
1806 | def updatecaches(tr=None, full=False, caches=None): |
|
1806 | def updatecaches(tr=None, full=False, caches=None): | |
1807 | """Warm repo caches.""" |
|
1807 | """Warm repo caches.""" | |
1808 |
|
1808 | |||
1809 | def invalidatecaches(): |
|
1809 | def invalidatecaches(): | |
1810 | """Invalidate cached data due to the repository mutating.""" |
|
1810 | """Invalidate cached data due to the repository mutating.""" | |
1811 |
|
1811 | |||
1812 | def invalidatevolatilesets(): |
|
1812 | def invalidatevolatilesets(): | |
1813 | pass |
|
1813 | pass | |
1814 |
|
1814 | |||
1815 | def invalidatedirstate(): |
|
1815 | def invalidatedirstate(): | |
1816 | """Invalidate the dirstate.""" |
|
1816 | """Invalidate the dirstate.""" | |
1817 |
|
1817 | |||
1818 | def invalidate(clearfilecache=False): |
|
1818 | def invalidate(clearfilecache=False): | |
1819 | pass |
|
1819 | pass | |
1820 |
|
1820 | |||
1821 | def invalidateall(): |
|
1821 | def invalidateall(): | |
1822 | pass |
|
1822 | pass | |
1823 |
|
1823 | |||
1824 | def lock(wait=True): |
|
1824 | def lock(wait=True): | |
1825 | """Lock the repository store and return a lock instance.""" |
|
1825 | """Lock the repository store and return a lock instance.""" | |
1826 |
|
1826 | |||
1827 | def currentlock(): |
|
1827 | def currentlock(): | |
1828 | """Return the lock if it's held or None.""" |
|
1828 | """Return the lock if it's held or None.""" | |
1829 |
|
1829 | |||
1830 | def wlock(wait=True): |
|
1830 | def wlock(wait=True): | |
1831 | """Lock the non-store parts of the repository.""" |
|
1831 | """Lock the non-store parts of the repository.""" | |
1832 |
|
1832 | |||
1833 | def currentwlock(): |
|
1833 | def currentwlock(): | |
1834 | """Return the wlock if it's held or None.""" |
|
1834 | """Return the wlock if it's held or None.""" | |
1835 |
|
1835 | |||
1836 | def checkcommitpatterns(wctx, match, status, fail): |
|
1836 | def checkcommitpatterns(wctx, match, status, fail): | |
1837 | pass |
|
1837 | pass | |
1838 |
|
1838 | |||
1839 | def commit( |
|
1839 | def commit( | |
1840 | text=b'', |
|
1840 | text=b'', | |
1841 | user=None, |
|
1841 | user=None, | |
1842 | date=None, |
|
1842 | date=None, | |
1843 | match=None, |
|
1843 | match=None, | |
1844 | force=False, |
|
1844 | force=False, | |
1845 | editor=False, |
|
1845 | editor=False, | |
1846 | extra=None, |
|
1846 | extra=None, | |
1847 | ): |
|
1847 | ): | |
1848 | """Add a new revision to the repository.""" |
|
1848 | """Add a new revision to the repository.""" | |
1849 |
|
1849 | |||
1850 | def commitctx(ctx, error=False, origctx=None): |
|
1850 | def commitctx(ctx, error=False, origctx=None): | |
1851 | """Commit a commitctx instance to the repository.""" |
|
1851 | """Commit a commitctx instance to the repository.""" | |
1852 |
|
1852 | |||
1853 | def destroying(): |
|
1853 | def destroying(): | |
1854 | """Inform the repository that nodes are about to be destroyed.""" |
|
1854 | """Inform the repository that nodes are about to be destroyed.""" | |
1855 |
|
1855 | |||
1856 | def destroyed(): |
|
1856 | def destroyed(): | |
1857 | """Inform the repository that nodes have been destroyed.""" |
|
1857 | """Inform the repository that nodes have been destroyed.""" | |
1858 |
|
1858 | |||
1859 | def status( |
|
1859 | def status( | |
1860 | node1=b'.', |
|
1860 | node1=b'.', | |
1861 | node2=None, |
|
1861 | node2=None, | |
1862 | match=None, |
|
1862 | match=None, | |
1863 | ignored=False, |
|
1863 | ignored=False, | |
1864 | clean=False, |
|
1864 | clean=False, | |
1865 | unknown=False, |
|
1865 | unknown=False, | |
1866 | listsubrepos=False, |
|
1866 | listsubrepos=False, | |
1867 | ): |
|
1867 | ): | |
1868 | """Convenience method to call repo[x].status().""" |
|
1868 | """Convenience method to call repo[x].status().""" | |
1869 |
|
1869 | |||
1870 | def addpostdsstatus(ps): |
|
1870 | def addpostdsstatus(ps): | |
1871 | pass |
|
1871 | pass | |
1872 |
|
1872 | |||
1873 | def postdsstatus(): |
|
1873 | def postdsstatus(): | |
1874 | pass |
|
1874 | pass | |
1875 |
|
1875 | |||
1876 | def clearpostdsstatus(): |
|
1876 | def clearpostdsstatus(): | |
1877 | pass |
|
1877 | pass | |
1878 |
|
1878 | |||
1879 | def heads(start=None): |
|
1879 | def heads(start=None): | |
1880 | """Obtain list of nodes that are DAG heads.""" |
|
1880 | """Obtain list of nodes that are DAG heads.""" | |
1881 |
|
1881 | |||
1882 | def branchheads(branch=None, start=None, closed=False): |
|
1882 | def branchheads(branch=None, start=None, closed=False): | |
1883 | pass |
|
1883 | pass | |
1884 |
|
1884 | |||
1885 | def branches(nodes): |
|
1885 | def branches(nodes): | |
1886 | pass |
|
1886 | pass | |
1887 |
|
1887 | |||
1888 | def between(pairs): |
|
1888 | def between(pairs): | |
1889 | pass |
|
1889 | pass | |
1890 |
|
1890 | |||
1891 | def checkpush(pushop): |
|
1891 | def checkpush(pushop): | |
1892 | pass |
|
1892 | pass | |
1893 |
|
1893 | |||
1894 | prepushoutgoinghooks = interfaceutil.Attribute("""util.hooks instance.""") |
|
1894 | prepushoutgoinghooks = interfaceutil.Attribute("""util.hooks instance.""") | |
1895 |
|
1895 | |||
1896 | def pushkey(namespace, key, old, new): |
|
1896 | def pushkey(namespace, key, old, new): | |
1897 | pass |
|
1897 | pass | |
1898 |
|
1898 | |||
1899 | def listkeys(namespace): |
|
1899 | def listkeys(namespace): | |
1900 | pass |
|
1900 | pass | |
1901 |
|
1901 | |||
1902 | def debugwireargs(one, two, three=None, four=None, five=None): |
|
1902 | def debugwireargs(one, two, three=None, four=None, five=None): | |
1903 | pass |
|
1903 | pass | |
1904 |
|
1904 | |||
1905 | def savecommitmessage(text): |
|
1905 | def savecommitmessage(text): | |
1906 | pass |
|
1906 | pass | |
1907 |
|
1907 | |||
1908 | def register_sidedata_computer( |
|
1908 | def register_sidedata_computer( | |
1909 | kind, category, keys, computer, flags, replace=False |
|
1909 | kind, category, keys, computer, flags, replace=False | |
1910 | ): |
|
1910 | ): | |
1911 | pass |
|
1911 | pass | |
1912 |
|
1912 | |||
1913 | def register_wanted_sidedata(category): |
|
1913 | def register_wanted_sidedata(category): | |
1914 | pass |
|
1914 | pass | |
1915 |
|
1915 | |||
1916 |
|
1916 | |||
1917 | class completelocalrepository( |
|
1917 | class completelocalrepository( | |
1918 | ilocalrepositorymain, ilocalrepositoryfilestorage |
|
1918 | ilocalrepositorymain, ilocalrepositoryfilestorage | |
1919 | ): |
|
1919 | ): | |
1920 | """Complete interface for a local repository.""" |
|
1920 | """Complete interface for a local repository.""" | |
1921 |
|
1921 | |||
1922 |
|
1922 | |||
1923 | class iwireprotocolcommandcacher(interfaceutil.Interface): |
|
1923 | class iwireprotocolcommandcacher(interfaceutil.Interface): | |
1924 | """Represents a caching backend for wire protocol commands. |
|
1924 | """Represents a caching backend for wire protocol commands. | |
1925 |
|
1925 | |||
1926 | Wire protocol version 2 supports transparent caching of many commands. |
|
1926 | Wire protocol version 2 supports transparent caching of many commands. | |
1927 | To leverage this caching, servers can activate objects that cache |
|
1927 | To leverage this caching, servers can activate objects that cache | |
1928 | command responses. Objects handle both cache writing and reading. |
|
1928 | command responses. Objects handle both cache writing and reading. | |
1929 | This interface defines how that response caching mechanism works. |
|
1929 | This interface defines how that response caching mechanism works. | |
1930 |
|
1930 | |||
1931 | Wire protocol version 2 commands emit a series of objects that are |
|
1931 | Wire protocol version 2 commands emit a series of objects that are | |
1932 | serialized and sent to the client. The caching layer exists between |
|
1932 | serialized and sent to the client. The caching layer exists between | |
1933 | the invocation of the command function and the sending of its output |
|
1933 | the invocation of the command function and the sending of its output | |
1934 | objects to an output layer. |
|
1934 | objects to an output layer. | |
1935 |
|
1935 | |||
1936 | Instances of this interface represent a binding to a cache that |
|
1936 | Instances of this interface represent a binding to a cache that | |
1937 | can serve a response (in place of calling a command function) and/or |
|
1937 | can serve a response (in place of calling a command function) and/or | |
1938 | write responses to a cache for subsequent use. |
|
1938 | write responses to a cache for subsequent use. | |
1939 |
|
1939 | |||
1940 | When a command request arrives, the following happens with regards |
|
1940 | When a command request arrives, the following happens with regards | |
1941 | to this interface: |
|
1941 | to this interface: | |
1942 |
|
1942 | |||
1943 | 1. The server determines whether the command request is cacheable. |
|
1943 | 1. The server determines whether the command request is cacheable. | |
1944 | 2. If it is, an instance of this interface is spawned. |
|
1944 | 2. If it is, an instance of this interface is spawned. | |
1945 | 3. The cacher is activated in a context manager (``__enter__`` is called). |
|
1945 | 3. The cacher is activated in a context manager (``__enter__`` is called). | |
1946 | 4. A cache *key* for that request is derived. This will call the |
|
1946 | 4. A cache *key* for that request is derived. This will call the | |
1947 | instance's ``adjustcachekeystate()`` method so the derivation |
|
1947 | instance's ``adjustcachekeystate()`` method so the derivation | |
1948 | can be influenced. |
|
1948 | can be influenced. | |
1949 | 5. The cacher is informed of the derived cache key via a call to |
|
1949 | 5. The cacher is informed of the derived cache key via a call to | |
1950 | ``setcachekey()``. |
|
1950 | ``setcachekey()``. | |
1951 | 6. The cacher's ``lookup()`` method is called to test for presence of |
|
1951 | 6. The cacher's ``lookup()`` method is called to test for presence of | |
1952 | the derived key in the cache. |
|
1952 | the derived key in the cache. | |
1953 | 7. If ``lookup()`` returns a hit, that cached result is used in place |
|
1953 | 7. If ``lookup()`` returns a hit, that cached result is used in place | |
1954 | of invoking the command function. ``__exit__`` is called and the instance |
|
1954 | of invoking the command function. ``__exit__`` is called and the instance | |
1955 | is discarded. |
|
1955 | is discarded. | |
1956 | 8. The command function is invoked. |
|
1956 | 8. The command function is invoked. | |
1957 | 9. ``onobject()`` is called for each object emitted by the command |
|
1957 | 9. ``onobject()`` is called for each object emitted by the command | |
1958 | function. |
|
1958 | function. | |
1959 | 10. After the final object is seen, ``onfinished()`` is called. |
|
1959 | 10. After the final object is seen, ``onfinished()`` is called. | |
1960 | 11. ``__exit__`` is called to signal the end of use of the instance. |
|
1960 | 11. ``__exit__`` is called to signal the end of use of the instance. | |
1961 |
|
1961 | |||
1962 | Cache *key* derivation can be influenced by the instance. |
|
1962 | Cache *key* derivation can be influenced by the instance. | |
1963 |
|
1963 | |||
1964 | Cache keys are initially derived by a deterministic representation of |
|
1964 | Cache keys are initially derived by a deterministic representation of | |
1965 | the command request. This includes the command name, arguments, protocol |
|
1965 | the command request. This includes the command name, arguments, protocol | |
1966 | version, etc. This initial key derivation is performed by CBOR-encoding a |
|
1966 | version, etc. This initial key derivation is performed by CBOR-encoding a | |
1967 | data structure and feeding that output into a hasher. |
|
1967 | data structure and feeding that output into a hasher. | |
1968 |
|
1968 | |||
1969 | Instances of this interface can influence this initial key derivation |
|
1969 | Instances of this interface can influence this initial key derivation | |
1970 | via ``adjustcachekeystate()``. |
|
1970 | via ``adjustcachekeystate()``. | |
1971 |
|
1971 | |||
1972 | The instance is informed of the derived cache key via a call to |
|
1972 | The instance is informed of the derived cache key via a call to | |
1973 | ``setcachekey()``. The instance must store the key locally so it can |
|
1973 | ``setcachekey()``. The instance must store the key locally so it can | |
1974 | be consulted on subsequent operations that may require it. |
|
1974 | be consulted on subsequent operations that may require it. | |
1975 |
|
1975 | |||
1976 | When constructed, the instance has access to a callable that can be used |
|
1976 | When constructed, the instance has access to a callable that can be used | |
1977 | for encoding response objects. This callable receives as its single |
|
1977 | for encoding response objects. This callable receives as its single | |
1978 | argument an object emitted by a command function. It returns an iterable |
|
1978 | argument an object emitted by a command function. It returns an iterable | |
1979 | of bytes chunks representing the encoded object. Unless the cacher is |
|
1979 | of bytes chunks representing the encoded object. Unless the cacher is | |
1980 | caching native Python objects in memory or has a way of reconstructing |
|
1980 | caching native Python objects in memory or has a way of reconstructing | |
1981 | the original Python objects, implementations typically call this function |
|
1981 | the original Python objects, implementations typically call this function | |
1982 | to produce bytes from the output objects and then store those bytes in |
|
1982 | to produce bytes from the output objects and then store those bytes in | |
1983 | the cache. When it comes time to re-emit those bytes, they are wrapped |
|
1983 | the cache. When it comes time to re-emit those bytes, they are wrapped | |
1984 | in a ``wireprototypes.encodedresponse`` instance to tell the output |
|
1984 | in a ``wireprototypes.encodedresponse`` instance to tell the output | |
1985 | layer that they are pre-encoded. |
|
1985 | layer that they are pre-encoded. | |
1986 |
|
1986 | |||
1987 | When receiving the objects emitted by the command function, instances |
|
1987 | When receiving the objects emitted by the command function, instances | |
1988 | can choose what to do with those objects. The simplest thing to do is |
|
1988 | can choose what to do with those objects. The simplest thing to do is | |
1989 | re-emit the original objects. They will be forwarded to the output |
|
1989 | re-emit the original objects. They will be forwarded to the output | |
1990 | layer and will be processed as if the cacher did not exist. |
|
1990 | layer and will be processed as if the cacher did not exist. | |
1991 |
|
1991 | |||
1992 | Implementations could also choose to not emit objects - instead locally |
|
1992 | Implementations could also choose to not emit objects - instead locally | |
1993 | buffering objects or their encoded representation. They could then emit |
|
1993 | buffering objects or their encoded representation. They could then emit | |
1994 | a single "coalesced" object when ``onfinished()`` is called. In |
|
1994 | a single "coalesced" object when ``onfinished()`` is called. In | |
1995 | this way, the implementation would function as a filtering layer of |
|
1995 | this way, the implementation would function as a filtering layer of | |
1996 | sorts. |
|
1996 | sorts. | |
1997 |
|
1997 | |||
1998 | When caching objects, typically the encoded form of the object will |
|
1998 | When caching objects, typically the encoded form of the object will | |
1999 | be stored. Keep in mind that if the original object is forwarded to |
|
1999 | be stored. Keep in mind that if the original object is forwarded to | |
2000 | the output layer, it will need to be encoded there as well. For large |
|
2000 | the output layer, it will need to be encoded there as well. For large | |
2001 | output, this redundant encoding could add overhead. Implementations |
|
2001 | output, this redundant encoding could add overhead. Implementations | |
2002 | could wrap the encoded object data in ``wireprototypes.encodedresponse`` |
|
2002 | could wrap the encoded object data in ``wireprototypes.encodedresponse`` | |
2003 | instances to avoid this overhead. |
|
2003 | instances to avoid this overhead. | |
2004 | """ |
|
2004 | """ | |
2005 |
|
2005 | |||
2006 | def __enter__(): |
|
2006 | def __enter__(): | |
2007 | """Marks the instance as active. |
|
2007 | """Marks the instance as active. | |
2008 |
|
2008 | |||
2009 | Should return self. |
|
2009 | Should return self. | |
2010 | """ |
|
2010 | """ | |
2011 |
|
2011 | |||
2012 | def __exit__(exctype, excvalue, exctb): |
|
2012 | def __exit__(exctype, excvalue, exctb): | |
2013 | """Called when cacher is no longer used. |
|
2013 | """Called when cacher is no longer used. | |
2014 |
|
2014 | |||
2015 | This can be used by implementations to perform cleanup actions (e.g. |
|
2015 | This can be used by implementations to perform cleanup actions (e.g. | |
2016 | disconnecting network sockets, aborting a partially cached response. |
|
2016 | disconnecting network sockets, aborting a partially cached response. | |
2017 | """ |
|
2017 | """ | |
2018 |
|
2018 | |||
2019 | def adjustcachekeystate(state): |
|
2019 | def adjustcachekeystate(state): | |
2020 | """Influences cache key derivation by adjusting state to derive key. |
|
2020 | """Influences cache key derivation by adjusting state to derive key. | |
2021 |
|
2021 | |||
2022 | A dict defining the state used to derive the cache key is passed. |
|
2022 | A dict defining the state used to derive the cache key is passed. | |
2023 |
|
2023 | |||
2024 | Implementations can modify this dict to record additional state that |
|
2024 | Implementations can modify this dict to record additional state that | |
2025 | is wanted to influence key derivation. |
|
2025 | is wanted to influence key derivation. | |
2026 |
|
2026 | |||
2027 | Implementations are *highly* encouraged to not modify or delete |
|
2027 | Implementations are *highly* encouraged to not modify or delete | |
2028 | existing keys. |
|
2028 | existing keys. | |
2029 | """ |
|
2029 | """ | |
2030 |
|
2030 | |||
2031 | def setcachekey(key): |
|
2031 | def setcachekey(key): | |
2032 | """Record the derived cache key for this request. |
|
2032 | """Record the derived cache key for this request. | |
2033 |
|
2033 | |||
2034 | Instances may mutate the key for internal usage, as desired. e.g. |
|
2034 | Instances may mutate the key for internal usage, as desired. e.g. | |
2035 | instances may wish to prepend the repo name, introduce path |
|
2035 | instances may wish to prepend the repo name, introduce path | |
2036 | components for filesystem or URL addressing, etc. Behavior is up to |
|
2036 | components for filesystem or URL addressing, etc. Behavior is up to | |
2037 | the cache. |
|
2037 | the cache. | |
2038 |
|
2038 | |||
2039 | Returns a bool indicating if the request is cacheable by this |
|
2039 | Returns a bool indicating if the request is cacheable by this | |
2040 | instance. |
|
2040 | instance. | |
2041 | """ |
|
2041 | """ | |
2042 |
|
2042 | |||
2043 | def lookup(): |
|
2043 | def lookup(): | |
2044 | """Attempt to resolve an entry in the cache. |
|
2044 | """Attempt to resolve an entry in the cache. | |
2045 |
|
2045 | |||
2046 | The instance is instructed to look for the cache key that it was |
|
2046 | The instance is instructed to look for the cache key that it was | |
2047 | informed about via the call to ``setcachekey()``. |
|
2047 | informed about via the call to ``setcachekey()``. | |
2048 |
|
2048 | |||
2049 | If there's no cache hit or the cacher doesn't wish to use the cached |
|
2049 | If there's no cache hit or the cacher doesn't wish to use the cached | |
2050 | entry, ``None`` should be returned. |
|
2050 | entry, ``None`` should be returned. | |
2051 |
|
2051 | |||
2052 | Else, a dict defining the cached result should be returned. The |
|
2052 | Else, a dict defining the cached result should be returned. The | |
2053 | dict may have the following keys: |
|
2053 | dict may have the following keys: | |
2054 |
|
2054 | |||
2055 | objs |
|
2055 | objs | |
2056 | An iterable of objects that should be sent to the client. That |
|
2056 | An iterable of objects that should be sent to the client. That | |
2057 | iterable of objects is expected to be what the command function |
|
2057 | iterable of objects is expected to be what the command function | |
2058 | would return if invoked or an equivalent representation thereof. |
|
2058 | would return if invoked or an equivalent representation thereof. | |
2059 | """ |
|
2059 | """ | |
2060 |
|
2060 | |||
2061 | def onobject(obj): |
|
2061 | def onobject(obj): | |
2062 | """Called when a new object is emitted from the command function. |
|
2062 | """Called when a new object is emitted from the command function. | |
2063 |
|
2063 | |||
2064 | Receives as its argument the object that was emitted from the |
|
2064 | Receives as its argument the object that was emitted from the | |
2065 | command function. |
|
2065 | command function. | |
2066 |
|
2066 | |||
2067 | This method returns an iterator of objects to forward to the output |
|
2067 | This method returns an iterator of objects to forward to the output | |
2068 | layer. The easiest implementation is a generator that just |
|
2068 | layer. The easiest implementation is a generator that just | |
2069 | ``yield obj``. |
|
2069 | ``yield obj``. | |
2070 | """ |
|
2070 | """ | |
2071 |
|
2071 | |||
2072 | def onfinished(): |
|
2072 | def onfinished(): | |
2073 | """Called after all objects have been emitted from the command function. |
|
2073 | """Called after all objects have been emitted from the command function. | |
2074 |
|
2074 | |||
2075 | Implementations should return an iterator of objects to forward to |
|
2075 | Implementations should return an iterator of objects to forward to | |
2076 | the output layer. |
|
2076 | the output layer. | |
2077 |
|
2077 | |||
2078 | This method can be a generator. |
|
2078 | This method can be a generator. | |
2079 | """ |
|
2079 | """ |
@@ -1,2382 +1,2382 b'' | |||||
1 | # manifest.py - manifest revision class for mercurial |
|
1 | # manifest.py - manifest revision class for mercurial | |
2 | # |
|
2 | # | |
3 | # Copyright 2005-2007 Olivia Mackall <olivia@selenic.com> |
|
3 | # Copyright 2005-2007 Olivia Mackall <olivia@selenic.com> | |
4 | # |
|
4 | # | |
5 | # This software may be used and distributed according to the terms of the |
|
5 | # This software may be used and distributed according to the terms of the | |
6 | # GNU General Public License version 2 or any later version. |
|
6 | # GNU General Public License version 2 or any later version. | |
7 |
|
7 | |||
8 |
|
8 | |||
9 | import heapq |
|
9 | import heapq | |
10 | import itertools |
|
10 | import itertools | |
11 | import struct |
|
11 | import struct | |
12 | import weakref |
|
12 | import weakref | |
13 |
|
13 | |||
14 | from .i18n import _ |
|
14 | from .i18n import _ | |
15 | from .node import ( |
|
15 | from .node import ( | |
16 | bin, |
|
16 | bin, | |
17 | hex, |
|
17 | hex, | |
18 | nullrev, |
|
18 | nullrev, | |
19 | ) |
|
19 | ) | |
20 | from . import ( |
|
20 | from . import ( | |
21 | encoding, |
|
21 | encoding, | |
22 | error, |
|
22 | error, | |
23 | match as matchmod, |
|
23 | match as matchmod, | |
24 | mdiff, |
|
24 | mdiff, | |
25 | pathutil, |
|
25 | pathutil, | |
26 | policy, |
|
26 | policy, | |
27 | pycompat, |
|
27 | pycompat, | |
28 | revlog, |
|
28 | revlog, | |
29 | util, |
|
29 | util, | |
30 | ) |
|
30 | ) | |
31 | from .interfaces import ( |
|
31 | from .interfaces import ( | |
32 | repository, |
|
32 | repository, | |
33 | util as interfaceutil, |
|
33 | util as interfaceutil, | |
34 | ) |
|
34 | ) | |
35 | from .revlogutils import ( |
|
35 | from .revlogutils import ( | |
36 | constants as revlog_constants, |
|
36 | constants as revlog_constants, | |
37 | ) |
|
37 | ) | |
38 |
|
38 | |||
39 | parsers = policy.importmod('parsers') |
|
39 | parsers = policy.importmod('parsers') | |
40 | propertycache = util.propertycache |
|
40 | propertycache = util.propertycache | |
41 |
|
41 | |||
42 | # Allow tests to more easily test the alternate path in manifestdict.fastdelta() |
|
42 | # Allow tests to more easily test the alternate path in manifestdict.fastdelta() | |
43 | FASTDELTA_TEXTDIFF_THRESHOLD = 1000 |
|
43 | FASTDELTA_TEXTDIFF_THRESHOLD = 1000 | |
44 |
|
44 | |||
45 |
|
45 | |||
46 | def _parse(nodelen, data): |
|
46 | def _parse(nodelen, data): | |
47 | # This method does a little bit of excessive-looking |
|
47 | # This method does a little bit of excessive-looking | |
48 | # precondition checking. This is so that the behavior of this |
|
48 | # precondition checking. This is so that the behavior of this | |
49 | # class exactly matches its C counterpart to try and help |
|
49 | # class exactly matches its C counterpart to try and help | |
50 | # prevent surprise breakage for anyone that develops against |
|
50 | # prevent surprise breakage for anyone that develops against | |
51 | # the pure version. |
|
51 | # the pure version. | |
52 | if data and data[-1:] != b'\n': |
|
52 | if data and data[-1:] != b'\n': | |
53 | raise ValueError(b'Manifest did not end in a newline.') |
|
53 | raise ValueError(b'Manifest did not end in a newline.') | |
54 | prev = None |
|
54 | prev = None | |
55 | for l in data.splitlines(): |
|
55 | for l in data.splitlines(): | |
56 | if prev is not None and prev > l: |
|
56 | if prev is not None and prev > l: | |
57 | raise ValueError(b'Manifest lines not in sorted order.') |
|
57 | raise ValueError(b'Manifest lines not in sorted order.') | |
58 | prev = l |
|
58 | prev = l | |
59 | f, n = l.split(b'\0') |
|
59 | f, n = l.split(b'\0') | |
60 | nl = len(n) |
|
60 | nl = len(n) | |
61 | flags = n[-1:] |
|
61 | flags = n[-1:] | |
62 | if flags in _manifestflags: |
|
62 | if flags in _manifestflags: | |
63 | n = n[:-1] |
|
63 | n = n[:-1] | |
64 | nl -= 1 |
|
64 | nl -= 1 | |
65 | else: |
|
65 | else: | |
66 | flags = b'' |
|
66 | flags = b'' | |
67 | if nl != 2 * nodelen: |
|
67 | if nl != 2 * nodelen: | |
68 | raise ValueError(b'Invalid manifest line') |
|
68 | raise ValueError(b'Invalid manifest line') | |
69 |
|
69 | |||
70 | yield f, bin(n), flags |
|
70 | yield f, bin(n), flags | |
71 |
|
71 | |||
72 |
|
72 | |||
73 | def _text(it): |
|
73 | def _text(it): | |
74 | files = [] |
|
74 | files = [] | |
75 | lines = [] |
|
75 | lines = [] | |
76 | for f, n, fl in it: |
|
76 | for f, n, fl in it: | |
77 | files.append(f) |
|
77 | files.append(f) | |
78 | # if this is changed to support newlines in filenames, |
|
78 | # if this is changed to support newlines in filenames, | |
79 | # be sure to check the templates/ dir again (especially *-raw.tmpl) |
|
79 | # be sure to check the templates/ dir again (especially *-raw.tmpl) | |
80 | lines.append(b"%s\0%s%s\n" % (f, hex(n), fl)) |
|
80 | lines.append(b"%s\0%s%s\n" % (f, hex(n), fl)) | |
81 |
|
81 | |||
82 | _checkforbidden(files) |
|
82 | _checkforbidden(files) | |
83 | return b''.join(lines) |
|
83 | return b''.join(lines) | |
84 |
|
84 | |||
85 |
|
85 | |||
86 | class lazymanifestiter: |
|
86 | class lazymanifestiter: | |
87 | def __init__(self, lm): |
|
87 | def __init__(self, lm): | |
88 | self.pos = 0 |
|
88 | self.pos = 0 | |
89 | self.lm = lm |
|
89 | self.lm = lm | |
90 |
|
90 | |||
91 | def __iter__(self): |
|
91 | def __iter__(self): | |
92 | return self |
|
92 | return self | |
93 |
|
93 | |||
94 | def next(self): |
|
94 | def next(self): | |
95 | try: |
|
95 | try: | |
96 | data, pos = self.lm._get(self.pos) |
|
96 | data, pos = self.lm._get(self.pos) | |
97 | except IndexError: |
|
97 | except IndexError: | |
98 | raise StopIteration |
|
98 | raise StopIteration | |
99 | if pos == -1: |
|
99 | if pos == -1: | |
100 | self.pos += 1 |
|
100 | self.pos += 1 | |
101 | return data[0] |
|
101 | return data[0] | |
102 | self.pos += 1 |
|
102 | self.pos += 1 | |
103 | zeropos = data.find(b'\x00', pos) |
|
103 | zeropos = data.find(b'\x00', pos) | |
104 | return data[pos:zeropos] |
|
104 | return data[pos:zeropos] | |
105 |
|
105 | |||
106 | __next__ = next |
|
106 | __next__ = next | |
107 |
|
107 | |||
108 |
|
108 | |||
109 | class lazymanifestiterentries: |
|
109 | class lazymanifestiterentries: | |
110 | def __init__(self, lm): |
|
110 | def __init__(self, lm): | |
111 | self.lm = lm |
|
111 | self.lm = lm | |
112 | self.pos = 0 |
|
112 | self.pos = 0 | |
113 |
|
113 | |||
114 | def __iter__(self): |
|
114 | def __iter__(self): | |
115 | return self |
|
115 | return self | |
116 |
|
116 | |||
117 | def next(self): |
|
117 | def next(self): | |
118 | try: |
|
118 | try: | |
119 | data, pos = self.lm._get(self.pos) |
|
119 | data, pos = self.lm._get(self.pos) | |
120 | except IndexError: |
|
120 | except IndexError: | |
121 | raise StopIteration |
|
121 | raise StopIteration | |
122 | if pos == -1: |
|
122 | if pos == -1: | |
123 | self.pos += 1 |
|
123 | self.pos += 1 | |
124 | return data |
|
124 | return data | |
125 | zeropos = data.find(b'\x00', pos) |
|
125 | zeropos = data.find(b'\x00', pos) | |
126 | nlpos = data.find(b'\n', pos) |
|
126 | nlpos = data.find(b'\n', pos) | |
127 | if zeropos == -1 or nlpos == -1 or nlpos < zeropos: |
|
127 | if zeropos == -1 or nlpos == -1 or nlpos < zeropos: | |
128 | raise error.StorageError(b'Invalid manifest line') |
|
128 | raise error.StorageError(b'Invalid manifest line') | |
129 | flags = data[nlpos - 1 : nlpos] |
|
129 | flags = data[nlpos - 1 : nlpos] | |
130 | if flags in _manifestflags: |
|
130 | if flags in _manifestflags: | |
131 | hlen = nlpos - zeropos - 2 |
|
131 | hlen = nlpos - zeropos - 2 | |
132 | else: |
|
132 | else: | |
133 | hlen = nlpos - zeropos - 1 |
|
133 | hlen = nlpos - zeropos - 1 | |
134 | flags = b'' |
|
134 | flags = b'' | |
135 | if hlen != 2 * self.lm._nodelen: |
|
135 | if hlen != 2 * self.lm._nodelen: | |
136 | raise error.StorageError(b'Invalid manifest line') |
|
136 | raise error.StorageError(b'Invalid manifest line') | |
137 | hashval = unhexlify( |
|
137 | hashval = unhexlify( | |
138 | data, self.lm.extrainfo[self.pos], zeropos + 1, hlen |
|
138 | data, self.lm.extrainfo[self.pos], zeropos + 1, hlen | |
139 | ) |
|
139 | ) | |
140 | self.pos += 1 |
|
140 | self.pos += 1 | |
141 | return (data[pos:zeropos], hashval, flags) |
|
141 | return (data[pos:zeropos], hashval, flags) | |
142 |
|
142 | |||
143 | __next__ = next |
|
143 | __next__ = next | |
144 |
|
144 | |||
145 |
|
145 | |||
146 | def unhexlify(data, extra, pos, length): |
|
146 | def unhexlify(data, extra, pos, length): | |
147 | s = bin(data[pos : pos + length]) |
|
147 | s = bin(data[pos : pos + length]) | |
148 | if extra: |
|
148 | if extra: | |
149 | s += chr(extra & 0xFF) |
|
149 | s += chr(extra & 0xFF) | |
150 | return s |
|
150 | return s | |
151 |
|
151 | |||
152 |
|
152 | |||
153 | def _cmp(a, b): |
|
153 | def _cmp(a, b): | |
154 | return (a > b) - (a < b) |
|
154 | return (a > b) - (a < b) | |
155 |
|
155 | |||
156 |
|
156 | |||
157 | _manifestflags = {b'', b'l', b't', b'x'} |
|
157 | _manifestflags = {b'', b'l', b't', b'x'} | |
158 |
|
158 | |||
159 |
|
159 | |||
160 | class _lazymanifest: |
|
160 | class _lazymanifest: | |
161 | """A pure python manifest backed by a byte string. It is supplimented with |
|
161 | """A pure python manifest backed by a byte string. It is supplimented with | |
162 | internal lists as it is modified, until it is compacted back to a pure byte |
|
162 | internal lists as it is modified, until it is compacted back to a pure byte | |
163 | string. |
|
163 | string. | |
164 |
|
164 | |||
165 | ``data`` is the initial manifest data. |
|
165 | ``data`` is the initial manifest data. | |
166 |
|
166 | |||
167 | ``positions`` is a list of offsets, one per manifest entry. Positive |
|
167 | ``positions`` is a list of offsets, one per manifest entry. Positive | |
168 | values are offsets into ``data``, negative values are offsets into the |
|
168 | values are offsets into ``data``, negative values are offsets into the | |
169 | ``extradata`` list. When an entry is removed, its entry is dropped from |
|
169 | ``extradata`` list. When an entry is removed, its entry is dropped from | |
170 | ``positions``. The values are encoded such that when walking the list and |
|
170 | ``positions``. The values are encoded such that when walking the list and | |
171 | indexing into ``data`` or ``extradata`` as appropriate, the entries are |
|
171 | indexing into ``data`` or ``extradata`` as appropriate, the entries are | |
172 | sorted by filename. |
|
172 | sorted by filename. | |
173 |
|
173 | |||
174 | ``extradata`` is a list of (key, hash, flags) for entries that were added or |
|
174 | ``extradata`` is a list of (key, hash, flags) for entries that were added or | |
175 | modified since the manifest was created or compacted. |
|
175 | modified since the manifest was created or compacted. | |
176 | """ |
|
176 | """ | |
177 |
|
177 | |||
178 | def __init__( |
|
178 | def __init__( | |
179 | self, |
|
179 | self, | |
180 | nodelen, |
|
180 | nodelen, | |
181 | data, |
|
181 | data, | |
182 | positions=None, |
|
182 | positions=None, | |
183 | extrainfo=None, |
|
183 | extrainfo=None, | |
184 | extradata=None, |
|
184 | extradata=None, | |
185 | hasremovals=False, |
|
185 | hasremovals=False, | |
186 | ): |
|
186 | ): | |
187 | self._nodelen = nodelen |
|
187 | self._nodelen = nodelen | |
188 | if positions is None: |
|
188 | if positions is None: | |
189 | self.positions = self.findlines(data) |
|
189 | self.positions = self.findlines(data) | |
190 | self.extrainfo = [0] * len(self.positions) |
|
190 | self.extrainfo = [0] * len(self.positions) | |
191 | self.data = data |
|
191 | self.data = data | |
192 | self.extradata = [] |
|
192 | self.extradata = [] | |
193 | self.hasremovals = False |
|
193 | self.hasremovals = False | |
194 | else: |
|
194 | else: | |
195 | self.positions = positions[:] |
|
195 | self.positions = positions[:] | |
196 | self.extrainfo = extrainfo[:] |
|
196 | self.extrainfo = extrainfo[:] | |
197 | self.extradata = extradata[:] |
|
197 | self.extradata = extradata[:] | |
198 | self.data = data |
|
198 | self.data = data | |
199 | self.hasremovals = hasremovals |
|
199 | self.hasremovals = hasremovals | |
200 |
|
200 | |||
201 | def findlines(self, data): |
|
201 | def findlines(self, data): | |
202 | if not data: |
|
202 | if not data: | |
203 | return [] |
|
203 | return [] | |
204 | pos = data.find(b"\n") |
|
204 | pos = data.find(b"\n") | |
205 | if pos == -1 or data[-1:] != b'\n': |
|
205 | if pos == -1 or data[-1:] != b'\n': | |
206 | raise ValueError(b"Manifest did not end in a newline.") |
|
206 | raise ValueError(b"Manifest did not end in a newline.") | |
207 | positions = [0] |
|
207 | positions = [0] | |
208 | prev = data[: data.find(b'\x00')] |
|
208 | prev = data[: data.find(b'\x00')] | |
209 | while pos < len(data) - 1 and pos != -1: |
|
209 | while pos < len(data) - 1 and pos != -1: | |
210 | positions.append(pos + 1) |
|
210 | positions.append(pos + 1) | |
211 | nexts = data[pos + 1 : data.find(b'\x00', pos + 1)] |
|
211 | nexts = data[pos + 1 : data.find(b'\x00', pos + 1)] | |
212 | if nexts < prev: |
|
212 | if nexts < prev: | |
213 | raise ValueError(b"Manifest lines not in sorted order.") |
|
213 | raise ValueError(b"Manifest lines not in sorted order.") | |
214 | prev = nexts |
|
214 | prev = nexts | |
215 | pos = data.find(b"\n", pos + 1) |
|
215 | pos = data.find(b"\n", pos + 1) | |
216 | return positions |
|
216 | return positions | |
217 |
|
217 | |||
218 | def _get(self, index): |
|
218 | def _get(self, index): | |
219 | # get the position encoded in pos: |
|
219 | # get the position encoded in pos: | |
220 | # positive number is an index in 'data' |
|
220 | # positive number is an index in 'data' | |
221 | # negative number is in extrapieces |
|
221 | # negative number is in extrapieces | |
222 | pos = self.positions[index] |
|
222 | pos = self.positions[index] | |
223 | if pos >= 0: |
|
223 | if pos >= 0: | |
224 | return self.data, pos |
|
224 | return self.data, pos | |
225 | return self.extradata[-pos - 1], -1 |
|
225 | return self.extradata[-pos - 1], -1 | |
226 |
|
226 | |||
227 | def _getkey(self, pos): |
|
227 | def _getkey(self, pos): | |
228 | if pos >= 0: |
|
228 | if pos >= 0: | |
229 | return self.data[pos : self.data.find(b'\x00', pos + 1)] |
|
229 | return self.data[pos : self.data.find(b'\x00', pos + 1)] | |
230 | return self.extradata[-pos - 1][0] |
|
230 | return self.extradata[-pos - 1][0] | |
231 |
|
231 | |||
232 | def bsearch(self, key): |
|
232 | def bsearch(self, key): | |
233 | first = 0 |
|
233 | first = 0 | |
234 | last = len(self.positions) - 1 |
|
234 | last = len(self.positions) - 1 | |
235 |
|
235 | |||
236 | while first <= last: |
|
236 | while first <= last: | |
237 | midpoint = (first + last) // 2 |
|
237 | midpoint = (first + last) // 2 | |
238 | nextpos = self.positions[midpoint] |
|
238 | nextpos = self.positions[midpoint] | |
239 | candidate = self._getkey(nextpos) |
|
239 | candidate = self._getkey(nextpos) | |
240 | r = _cmp(key, candidate) |
|
240 | r = _cmp(key, candidate) | |
241 | if r == 0: |
|
241 | if r == 0: | |
242 | return midpoint |
|
242 | return midpoint | |
243 | else: |
|
243 | else: | |
244 | if r < 0: |
|
244 | if r < 0: | |
245 | last = midpoint - 1 |
|
245 | last = midpoint - 1 | |
246 | else: |
|
246 | else: | |
247 | first = midpoint + 1 |
|
247 | first = midpoint + 1 | |
248 | return -1 |
|
248 | return -1 | |
249 |
|
249 | |||
250 | def bsearch2(self, key): |
|
250 | def bsearch2(self, key): | |
251 | # same as the above, but will always return the position |
|
251 | # same as the above, but will always return the position | |
252 | # done for performance reasons |
|
252 | # done for performance reasons | |
253 | first = 0 |
|
253 | first = 0 | |
254 | last = len(self.positions) - 1 |
|
254 | last = len(self.positions) - 1 | |
255 |
|
255 | |||
256 | while first <= last: |
|
256 | while first <= last: | |
257 | midpoint = (first + last) // 2 |
|
257 | midpoint = (first + last) // 2 | |
258 | nextpos = self.positions[midpoint] |
|
258 | nextpos = self.positions[midpoint] | |
259 | candidate = self._getkey(nextpos) |
|
259 | candidate = self._getkey(nextpos) | |
260 | r = _cmp(key, candidate) |
|
260 | r = _cmp(key, candidate) | |
261 | if r == 0: |
|
261 | if r == 0: | |
262 | return (midpoint, True) |
|
262 | return (midpoint, True) | |
263 | else: |
|
263 | else: | |
264 | if r < 0: |
|
264 | if r < 0: | |
265 | last = midpoint - 1 |
|
265 | last = midpoint - 1 | |
266 | else: |
|
266 | else: | |
267 | first = midpoint + 1 |
|
267 | first = midpoint + 1 | |
268 | return (first, False) |
|
268 | return (first, False) | |
269 |
|
269 | |||
270 | def __contains__(self, key): |
|
270 | def __contains__(self, key): | |
271 | return self.bsearch(key) != -1 |
|
271 | return self.bsearch(key) != -1 | |
272 |
|
272 | |||
273 | def __getitem__(self, key): |
|
273 | def __getitem__(self, key): | |
274 | if not isinstance(key, bytes): |
|
274 | if not isinstance(key, bytes): | |
275 | raise TypeError(b"getitem: manifest keys must be a bytes.") |
|
275 | raise TypeError(b"getitem: manifest keys must be a bytes.") | |
276 | needle = self.bsearch(key) |
|
276 | needle = self.bsearch(key) | |
277 | if needle == -1: |
|
277 | if needle == -1: | |
278 | raise KeyError |
|
278 | raise KeyError | |
279 | data, pos = self._get(needle) |
|
279 | data, pos = self._get(needle) | |
280 | if pos == -1: |
|
280 | if pos == -1: | |
281 | return (data[1], data[2]) |
|
281 | return (data[1], data[2]) | |
282 | zeropos = data.find(b'\x00', pos) |
|
282 | zeropos = data.find(b'\x00', pos) | |
283 | nlpos = data.find(b'\n', zeropos) |
|
283 | nlpos = data.find(b'\n', zeropos) | |
284 | assert 0 <= needle <= len(self.positions) |
|
284 | assert 0 <= needle <= len(self.positions) | |
285 | assert len(self.extrainfo) == len(self.positions) |
|
285 | assert len(self.extrainfo) == len(self.positions) | |
286 | if zeropos == -1 or nlpos == -1 or nlpos < zeropos: |
|
286 | if zeropos == -1 or nlpos == -1 or nlpos < zeropos: | |
287 | raise error.StorageError(b'Invalid manifest line') |
|
287 | raise error.StorageError(b'Invalid manifest line') | |
288 | hlen = nlpos - zeropos - 1 |
|
288 | hlen = nlpos - zeropos - 1 | |
289 | flags = data[nlpos - 1 : nlpos] |
|
289 | flags = data[nlpos - 1 : nlpos] | |
290 | if flags in _manifestflags: |
|
290 | if flags in _manifestflags: | |
291 | hlen -= 1 |
|
291 | hlen -= 1 | |
292 | else: |
|
292 | else: | |
293 | flags = b'' |
|
293 | flags = b'' | |
294 | if hlen != 2 * self._nodelen: |
|
294 | if hlen != 2 * self._nodelen: | |
295 | raise error.StorageError(b'Invalid manifest line') |
|
295 | raise error.StorageError(b'Invalid manifest line') | |
296 | hashval = unhexlify(data, self.extrainfo[needle], zeropos + 1, hlen) |
|
296 | hashval = unhexlify(data, self.extrainfo[needle], zeropos + 1, hlen) | |
297 | return (hashval, flags) |
|
297 | return (hashval, flags) | |
298 |
|
298 | |||
299 | def __delitem__(self, key): |
|
299 | def __delitem__(self, key): | |
300 | needle, found = self.bsearch2(key) |
|
300 | needle, found = self.bsearch2(key) | |
301 | if not found: |
|
301 | if not found: | |
302 | raise KeyError |
|
302 | raise KeyError | |
303 | cur = self.positions[needle] |
|
303 | cur = self.positions[needle] | |
304 | self.positions = self.positions[:needle] + self.positions[needle + 1 :] |
|
304 | self.positions = self.positions[:needle] + self.positions[needle + 1 :] | |
305 | self.extrainfo = self.extrainfo[:needle] + self.extrainfo[needle + 1 :] |
|
305 | self.extrainfo = self.extrainfo[:needle] + self.extrainfo[needle + 1 :] | |
306 | if cur >= 0: |
|
306 | if cur >= 0: | |
307 | # This does NOT unsort the list as far as the search functions are |
|
307 | # This does NOT unsort the list as far as the search functions are | |
308 | # concerned, as they only examine lines mapped by self.positions. |
|
308 | # concerned, as they only examine lines mapped by self.positions. | |
309 | self.data = self.data[:cur] + b'\x00' + self.data[cur + 1 :] |
|
309 | self.data = self.data[:cur] + b'\x00' + self.data[cur + 1 :] | |
310 | self.hasremovals = True |
|
310 | self.hasremovals = True | |
311 |
|
311 | |||
312 | def __setitem__(self, key, value): |
|
312 | def __setitem__(self, key, value): | |
313 | if not isinstance(key, bytes): |
|
313 | if not isinstance(key, bytes): | |
314 | raise TypeError(b"setitem: manifest keys must be a byte string.") |
|
314 | raise TypeError(b"setitem: manifest keys must be a byte string.") | |
315 | if not isinstance(value, tuple) or len(value) != 2: |
|
315 | if not isinstance(value, tuple) or len(value) != 2: | |
316 | raise TypeError( |
|
316 | raise TypeError( | |
317 | b"Manifest values must be a tuple of (node, flags)." |
|
317 | b"Manifest values must be a tuple of (node, flags)." | |
318 | ) |
|
318 | ) | |
319 | hashval = value[0] |
|
319 | hashval = value[0] | |
320 | if not isinstance(hashval, bytes) or len(hashval) not in (20, 32): |
|
320 | if not isinstance(hashval, bytes) or len(hashval) not in (20, 32): | |
321 | raise TypeError(b"node must be a 20-byte or 32-byte byte string") |
|
321 | raise TypeError(b"node must be a 20-byte or 32-byte byte string") | |
322 | flags = value[1] |
|
322 | flags = value[1] | |
323 | if not isinstance(flags, bytes) or len(flags) > 1: |
|
323 | if not isinstance(flags, bytes) or len(flags) > 1: | |
324 | raise TypeError(b"flags must a 0 or 1 byte string, got %r", flags) |
|
324 | raise TypeError(b"flags must a 0 or 1 byte string, got %r", flags) | |
325 | needle, found = self.bsearch2(key) |
|
325 | needle, found = self.bsearch2(key) | |
326 | if found: |
|
326 | if found: | |
327 | # put the item |
|
327 | # put the item | |
328 | pos = self.positions[needle] |
|
328 | pos = self.positions[needle] | |
329 | if pos < 0: |
|
329 | if pos < 0: | |
330 | self.extradata[-pos - 1] = (key, hashval, value[1]) |
|
330 | self.extradata[-pos - 1] = (key, hashval, value[1]) | |
331 | else: |
|
331 | else: | |
332 | # just don't bother |
|
332 | # just don't bother | |
333 | self.extradata.append((key, hashval, value[1])) |
|
333 | self.extradata.append((key, hashval, value[1])) | |
334 | self.positions[needle] = -len(self.extradata) |
|
334 | self.positions[needle] = -len(self.extradata) | |
335 | else: |
|
335 | else: | |
336 | # not found, put it in with extra positions |
|
336 | # not found, put it in with extra positions | |
337 | self.extradata.append((key, hashval, value[1])) |
|
337 | self.extradata.append((key, hashval, value[1])) | |
338 | self.positions = ( |
|
338 | self.positions = ( | |
339 | self.positions[:needle] |
|
339 | self.positions[:needle] | |
340 | + [-len(self.extradata)] |
|
340 | + [-len(self.extradata)] | |
341 | + self.positions[needle:] |
|
341 | + self.positions[needle:] | |
342 | ) |
|
342 | ) | |
343 | self.extrainfo = ( |
|
343 | self.extrainfo = ( | |
344 | self.extrainfo[:needle] + [0] + self.extrainfo[needle:] |
|
344 | self.extrainfo[:needle] + [0] + self.extrainfo[needle:] | |
345 | ) |
|
345 | ) | |
346 |
|
346 | |||
347 | def copy(self): |
|
347 | def copy(self): | |
348 | # XXX call _compact like in C? |
|
348 | # XXX call _compact like in C? | |
349 | return _lazymanifest( |
|
349 | return _lazymanifest( | |
350 | self._nodelen, |
|
350 | self._nodelen, | |
351 | self.data, |
|
351 | self.data, | |
352 | self.positions, |
|
352 | self.positions, | |
353 | self.extrainfo, |
|
353 | self.extrainfo, | |
354 | self.extradata, |
|
354 | self.extradata, | |
355 | self.hasremovals, |
|
355 | self.hasremovals, | |
356 | ) |
|
356 | ) | |
357 |
|
357 | |||
358 | def _compact(self): |
|
358 | def _compact(self): | |
359 | # hopefully not called TOO often |
|
359 | # hopefully not called TOO often | |
360 | if len(self.extradata) == 0 and not self.hasremovals: |
|
360 | if len(self.extradata) == 0 and not self.hasremovals: | |
361 | return |
|
361 | return | |
362 | l = [] |
|
362 | l = [] | |
363 | i = 0 |
|
363 | i = 0 | |
364 | offset = 0 |
|
364 | offset = 0 | |
365 | self.extrainfo = [0] * len(self.positions) |
|
365 | self.extrainfo = [0] * len(self.positions) | |
366 | while i < len(self.positions): |
|
366 | while i < len(self.positions): | |
367 | if self.positions[i] >= 0: |
|
367 | if self.positions[i] >= 0: | |
368 | cur = self.positions[i] |
|
368 | cur = self.positions[i] | |
369 | last_cut = cur |
|
369 | last_cut = cur | |
370 |
|
370 | |||
371 | # Collect all contiguous entries in the buffer at the current |
|
371 | # Collect all contiguous entries in the buffer at the current | |
372 | # offset, breaking out only for added/modified items held in |
|
372 | # offset, breaking out only for added/modified items held in | |
373 | # extradata, or a deleted line prior to the next position. |
|
373 | # extradata, or a deleted line prior to the next position. | |
374 | while True: |
|
374 | while True: | |
375 | self.positions[i] = offset |
|
375 | self.positions[i] = offset | |
376 | i += 1 |
|
376 | i += 1 | |
377 | if i == len(self.positions) or self.positions[i] < 0: |
|
377 | if i == len(self.positions) or self.positions[i] < 0: | |
378 | break |
|
378 | break | |
379 |
|
379 | |||
380 | # A removed file has no positions[] entry, but does have an |
|
380 | # A removed file has no positions[] entry, but does have an | |
381 | # overwritten first byte. Break out and find the end of the |
|
381 | # overwritten first byte. Break out and find the end of the | |
382 | # current good entry/entries if there is a removed file |
|
382 | # current good entry/entries if there is a removed file | |
383 | # before the next position. |
|
383 | # before the next position. | |
384 | if ( |
|
384 | if ( | |
385 | self.hasremovals |
|
385 | self.hasremovals | |
386 | and self.data.find(b'\n\x00', cur, self.positions[i]) |
|
386 | and self.data.find(b'\n\x00', cur, self.positions[i]) | |
387 | != -1 |
|
387 | != -1 | |
388 | ): |
|
388 | ): | |
389 | break |
|
389 | break | |
390 |
|
390 | |||
391 | offset += self.positions[i] - cur |
|
391 | offset += self.positions[i] - cur | |
392 | cur = self.positions[i] |
|
392 | cur = self.positions[i] | |
393 | end_cut = self.data.find(b'\n', cur) |
|
393 | end_cut = self.data.find(b'\n', cur) | |
394 | if end_cut != -1: |
|
394 | if end_cut != -1: | |
395 | end_cut += 1 |
|
395 | end_cut += 1 | |
396 | offset += end_cut - cur |
|
396 | offset += end_cut - cur | |
397 | l.append(self.data[last_cut:end_cut]) |
|
397 | l.append(self.data[last_cut:end_cut]) | |
398 | else: |
|
398 | else: | |
399 | while i < len(self.positions) and self.positions[i] < 0: |
|
399 | while i < len(self.positions) and self.positions[i] < 0: | |
400 | cur = self.positions[i] |
|
400 | cur = self.positions[i] | |
401 | t = self.extradata[-cur - 1] |
|
401 | t = self.extradata[-cur - 1] | |
402 | l.append(self._pack(t)) |
|
402 | l.append(self._pack(t)) | |
403 | self.positions[i] = offset |
|
403 | self.positions[i] = offset | |
404 | # Hashes are either 20 bytes (old sha1s) or 32 |
|
404 | # Hashes are either 20 bytes (old sha1s) or 32 | |
405 | # bytes (new non-sha1). |
|
405 | # bytes (new non-sha1). | |
406 | hlen = 20 |
|
406 | hlen = 20 | |
407 | if len(t[1]) > 25: |
|
407 | if len(t[1]) > 25: | |
408 | hlen = 32 |
|
408 | hlen = 32 | |
409 | if len(t[1]) > hlen: |
|
409 | if len(t[1]) > hlen: | |
410 | self.extrainfo[i] = ord(t[1][hlen + 1]) |
|
410 | self.extrainfo[i] = ord(t[1][hlen + 1]) | |
411 | offset += len(l[-1]) |
|
411 | offset += len(l[-1]) | |
412 | i += 1 |
|
412 | i += 1 | |
413 | self.data = b''.join(l) |
|
413 | self.data = b''.join(l) | |
414 | self.hasremovals = False |
|
414 | self.hasremovals = False | |
415 | self.extradata = [] |
|
415 | self.extradata = [] | |
416 |
|
416 | |||
417 | def _pack(self, d): |
|
417 | def _pack(self, d): | |
418 | n = d[1] |
|
418 | n = d[1] | |
419 | assert len(n) in (20, 32) |
|
419 | assert len(n) in (20, 32) | |
420 | return d[0] + b'\x00' + hex(n) + d[2] + b'\n' |
|
420 | return d[0] + b'\x00' + hex(n) + d[2] + b'\n' | |
421 |
|
421 | |||
422 | def text(self): |
|
422 | def text(self): | |
423 | self._compact() |
|
423 | self._compact() | |
424 | return self.data |
|
424 | return self.data | |
425 |
|
425 | |||
426 | def diff(self, m2, clean=False): |
|
426 | def diff(self, m2, clean=False): | |
427 | '''Finds changes between the current manifest and m2.''' |
|
427 | '''Finds changes between the current manifest and m2.''' | |
428 | # XXX think whether efficiency matters here |
|
428 | # XXX think whether efficiency matters here | |
429 | diff = {} |
|
429 | diff = {} | |
430 |
|
430 | |||
431 | for fn, e1, flags in self.iterentries(): |
|
431 | for fn, e1, flags in self.iterentries(): | |
432 | if fn not in m2: |
|
432 | if fn not in m2: | |
433 | diff[fn] = (e1, flags), (None, b'') |
|
433 | diff[fn] = (e1, flags), (None, b'') | |
434 | else: |
|
434 | else: | |
435 | e2 = m2[fn] |
|
435 | e2 = m2[fn] | |
436 | if (e1, flags) != e2: |
|
436 | if (e1, flags) != e2: | |
437 | diff[fn] = (e1, flags), e2 |
|
437 | diff[fn] = (e1, flags), e2 | |
438 | elif clean: |
|
438 | elif clean: | |
439 | diff[fn] = None |
|
439 | diff[fn] = None | |
440 |
|
440 | |||
441 | for fn, e2, flags in m2.iterentries(): |
|
441 | for fn, e2, flags in m2.iterentries(): | |
442 | if fn not in self: |
|
442 | if fn not in self: | |
443 | diff[fn] = (None, b''), (e2, flags) |
|
443 | diff[fn] = (None, b''), (e2, flags) | |
444 |
|
444 | |||
445 | return diff |
|
445 | return diff | |
446 |
|
446 | |||
447 | def iterentries(self): |
|
447 | def iterentries(self): | |
448 | return lazymanifestiterentries(self) |
|
448 | return lazymanifestiterentries(self) | |
449 |
|
449 | |||
450 | def iterkeys(self): |
|
450 | def iterkeys(self): | |
451 | return lazymanifestiter(self) |
|
451 | return lazymanifestiter(self) | |
452 |
|
452 | |||
453 | def __iter__(self): |
|
453 | def __iter__(self): | |
454 | return lazymanifestiter(self) |
|
454 | return lazymanifestiter(self) | |
455 |
|
455 | |||
456 | def __len__(self): |
|
456 | def __len__(self): | |
457 | return len(self.positions) |
|
457 | return len(self.positions) | |
458 |
|
458 | |||
459 | def filtercopy(self, filterfn): |
|
459 | def filtercopy(self, filterfn): | |
460 | # XXX should be optimized |
|
460 | # XXX should be optimized | |
461 | c = _lazymanifest(self._nodelen, b'') |
|
461 | c = _lazymanifest(self._nodelen, b'') | |
462 | for f, n, fl in self.iterentries(): |
|
462 | for f, n, fl in self.iterentries(): | |
463 | if filterfn(f): |
|
463 | if filterfn(f): | |
464 | c[f] = n, fl |
|
464 | c[f] = n, fl | |
465 | return c |
|
465 | return c | |
466 |
|
466 | |||
467 |
|
467 | |||
468 | try: |
|
468 | try: | |
469 | _lazymanifest = parsers.lazymanifest |
|
469 | _lazymanifest = parsers.lazymanifest | |
470 | except AttributeError: |
|
470 | except AttributeError: | |
471 | pass |
|
471 | pass | |
472 |
|
472 | |||
473 |
|
473 | |||
474 | @interfaceutil.implementer(repository.imanifestdict) |
|
474 | @interfaceutil.implementer(repository.imanifestdict) | |
475 | class manifestdict: |
|
475 | class manifestdict: | |
476 | def __init__(self, nodelen, data=b''): |
|
476 | def __init__(self, nodelen, data=b''): | |
477 | self._nodelen = nodelen |
|
477 | self._nodelen = nodelen | |
478 | self._lm = _lazymanifest(nodelen, data) |
|
478 | self._lm = _lazymanifest(nodelen, data) | |
479 |
|
479 | |||
480 | def __getitem__(self, key): |
|
480 | def __getitem__(self, key): | |
481 | return self._lm[key][0] |
|
481 | return self._lm[key][0] | |
482 |
|
482 | |||
483 | def find(self, key): |
|
483 | def find(self, key): | |
484 | return self._lm[key] |
|
484 | return self._lm[key] | |
485 |
|
485 | |||
486 | def __len__(self): |
|
486 | def __len__(self): | |
487 | return len(self._lm) |
|
487 | return len(self._lm) | |
488 |
|
488 | |||
489 | def __nonzero__(self): |
|
489 | def __nonzero__(self): | |
490 | # nonzero is covered by the __len__ function, but implementing it here |
|
490 | # nonzero is covered by the __len__ function, but implementing it here | |
491 | # makes it easier for extensions to override. |
|
491 | # makes it easier for extensions to override. | |
492 | return len(self._lm) != 0 |
|
492 | return len(self._lm) != 0 | |
493 |
|
493 | |||
494 | __bool__ = __nonzero__ |
|
494 | __bool__ = __nonzero__ | |
495 |
|
495 | |||
496 | def __setitem__(self, key, node): |
|
496 | def __setitem__(self, key, node): | |
497 | self._lm[key] = node, self.flags(key) |
|
497 | self._lm[key] = node, self.flags(key) | |
498 |
|
498 | |||
499 | def __contains__(self, key): |
|
499 | def __contains__(self, key): | |
500 | if key is None: |
|
500 | if key is None: | |
501 | return False |
|
501 | return False | |
502 | return key in self._lm |
|
502 | return key in self._lm | |
503 |
|
503 | |||
504 | def __delitem__(self, key): |
|
504 | def __delitem__(self, key): | |
505 | del self._lm[key] |
|
505 | del self._lm[key] | |
506 |
|
506 | |||
507 | def __iter__(self): |
|
507 | def __iter__(self): | |
508 | return self._lm.__iter__() |
|
508 | return self._lm.__iter__() | |
509 |
|
509 | |||
510 | def iterkeys(self): |
|
510 | def iterkeys(self): | |
511 | return self._lm.iterkeys() |
|
511 | return self._lm.iterkeys() | |
512 |
|
512 | |||
513 | def keys(self): |
|
513 | def keys(self): | |
514 | return list(self.iterkeys()) |
|
514 | return list(self.iterkeys()) | |
515 |
|
515 | |||
516 | def filesnotin(self, m2, match=None): |
|
516 | def filesnotin(self, m2, match=None): | |
517 | '''Set of files in this manifest that are not in the other''' |
|
517 | '''Set of files in this manifest that are not in the other''' | |
518 | if match is not None: |
|
518 | if match is not None: | |
519 | match = matchmod.badmatch(match, lambda path, msg: None) |
|
519 | match = matchmod.badmatch(match, lambda path, msg: None) | |
520 | sm2 = set(m2.walk(match)) |
|
520 | sm2 = set(m2.walk(match)) | |
521 | return {f for f in self.walk(match) if f not in sm2} |
|
521 | return {f for f in self.walk(match) if f not in sm2} | |
522 | return {f for f in self if f not in m2} |
|
522 | return {f for f in self if f not in m2} | |
523 |
|
523 | |||
524 | @propertycache |
|
524 | @propertycache | |
525 | def _dirs(self): |
|
525 | def _dirs(self): | |
526 | return pathutil.dirs(self) |
|
526 | return pathutil.dirs(self) | |
527 |
|
527 | |||
528 | def dirs(self): |
|
528 | def dirs(self): | |
529 | return self._dirs |
|
529 | return self._dirs | |
530 |
|
530 | |||
531 | def hasdir(self, dir): |
|
531 | def hasdir(self, dir): | |
532 | return dir in self._dirs |
|
532 | return dir in self._dirs | |
533 |
|
533 | |||
534 | def _filesfastpath(self, match): |
|
534 | def _filesfastpath(self, match): | |
535 | """Checks whether we can correctly and quickly iterate over matcher |
|
535 | """Checks whether we can correctly and quickly iterate over matcher | |
536 | files instead of over manifest files.""" |
|
536 | files instead of over manifest files.""" | |
537 | files = match.files() |
|
537 | files = match.files() | |
538 | return len(files) < 100 and ( |
|
538 | return len(files) < 100 and ( | |
539 | match.isexact() |
|
539 | match.isexact() | |
540 | or (match.prefix() and all(fn in self for fn in files)) |
|
540 | or (match.prefix() and all(fn in self for fn in files)) | |
541 | ) |
|
541 | ) | |
542 |
|
542 | |||
543 | def walk(self, match): |
|
543 | def walk(self, match): | |
544 | """Generates matching file names. |
|
544 | """Generates matching file names. | |
545 |
|
545 | |||
546 | Equivalent to manifest.matches(match).iterkeys(), but without creating |
|
546 | Equivalent to manifest.matches(match).iterkeys(), but without creating | |
547 | an entirely new manifest. |
|
547 | an entirely new manifest. | |
548 |
|
548 | |||
549 | It also reports nonexistent files by marking them bad with match.bad(). |
|
549 | It also reports nonexistent files by marking them bad with match.bad(). | |
550 | """ |
|
550 | """ | |
551 | if match.always(): |
|
551 | if match.always(): | |
552 | for f in iter(self): |
|
552 | for f in iter(self): | |
553 | yield f |
|
553 | yield f | |
554 | return |
|
554 | return | |
555 |
|
555 | |||
556 | fset = set(match.files()) |
|
556 | fset = set(match.files()) | |
557 |
|
557 | |||
558 | # avoid the entire walk if we're only looking for specific files |
|
558 | # avoid the entire walk if we're only looking for specific files | |
559 | if self._filesfastpath(match): |
|
559 | if self._filesfastpath(match): | |
560 | for fn in sorted(fset): |
|
560 | for fn in sorted(fset): | |
561 | if fn in self: |
|
561 | if fn in self: | |
562 | yield fn |
|
562 | yield fn | |
563 | return |
|
563 | return | |
564 |
|
564 | |||
565 | for fn in self: |
|
565 | for fn in self: | |
566 | if fn in fset: |
|
566 | if fn in fset: | |
567 | # specified pattern is the exact name |
|
567 | # specified pattern is the exact name | |
568 | fset.remove(fn) |
|
568 | fset.remove(fn) | |
569 | if match(fn): |
|
569 | if match(fn): | |
570 | yield fn |
|
570 | yield fn | |
571 |
|
571 | |||
572 | # for dirstate.walk, files=[''] means "walk the whole tree". |
|
572 | # for dirstate.walk, files=[''] means "walk the whole tree". | |
573 | # follow that here, too |
|
573 | # follow that here, too | |
574 | fset.discard(b'') |
|
574 | fset.discard(b'') | |
575 |
|
575 | |||
576 | for fn in sorted(fset): |
|
576 | for fn in sorted(fset): | |
577 | if not self.hasdir(fn): |
|
577 | if not self.hasdir(fn): | |
578 | match.bad(fn, None) |
|
578 | match.bad(fn, None) | |
579 |
|
579 | |||
580 | def _matches(self, match): |
|
580 | def _matches(self, match): | |
581 | '''generate a new manifest filtered by the match argument''' |
|
581 | '''generate a new manifest filtered by the match argument''' | |
582 | if match.always(): |
|
582 | if match.always(): | |
583 | return self.copy() |
|
583 | return self.copy() | |
584 |
|
584 | |||
585 | if self._filesfastpath(match): |
|
585 | if self._filesfastpath(match): | |
586 | m = manifestdict(self._nodelen) |
|
586 | m = manifestdict(self._nodelen) | |
587 | lm = self._lm |
|
587 | lm = self._lm | |
588 | for fn in match.files(): |
|
588 | for fn in match.files(): | |
589 | if fn in lm: |
|
589 | if fn in lm: | |
590 | m._lm[fn] = lm[fn] |
|
590 | m._lm[fn] = lm[fn] | |
591 | return m |
|
591 | return m | |
592 |
|
592 | |||
593 | m = manifestdict(self._nodelen) |
|
593 | m = manifestdict(self._nodelen) | |
594 | m._lm = self._lm.filtercopy(match) |
|
594 | m._lm = self._lm.filtercopy(match) | |
595 | return m |
|
595 | return m | |
596 |
|
596 | |||
597 | def diff(self, m2, match=None, clean=False): |
|
597 | def diff(self, m2, match=None, clean=False): | |
598 | """Finds changes between the current manifest and m2. |
|
598 | """Finds changes between the current manifest and m2. | |
599 |
|
599 | |||
600 | Args: |
|
600 | Args: | |
601 | m2: the manifest to which this manifest should be compared. |
|
601 | m2: the manifest to which this manifest should be compared. | |
602 | clean: if true, include files unchanged between these manifests |
|
602 | clean: if true, include files unchanged between these manifests | |
603 | with a None value in the returned dictionary. |
|
603 | with a None value in the returned dictionary. | |
604 |
|
604 | |||
605 | The result is returned as a dict with filename as key and |
|
605 | The result is returned as a dict with filename as key and | |
606 | values of the form ((n1,fl1),(n2,fl2)), where n1/n2 is the |
|
606 | values of the form ((n1,fl1),(n2,fl2)), where n1/n2 is the | |
607 | nodeid in the current/other manifest and fl1/fl2 is the flag |
|
607 | nodeid in the current/other manifest and fl1/fl2 is the flag | |
608 | in the current/other manifest. Where the file does not exist, |
|
608 | in the current/other manifest. Where the file does not exist, | |
609 | the nodeid will be None and the flags will be the empty |
|
609 | the nodeid will be None and the flags will be the empty | |
610 | string. |
|
610 | string. | |
611 | """ |
|
611 | """ | |
612 | if match: |
|
612 | if match: | |
613 | m1 = self._matches(match) |
|
613 | m1 = self._matches(match) | |
614 | m2 = m2._matches(match) |
|
614 | m2 = m2._matches(match) | |
615 | return m1.diff(m2, clean=clean) |
|
615 | return m1.diff(m2, clean=clean) | |
616 | return self._lm.diff(m2._lm, clean) |
|
616 | return self._lm.diff(m2._lm, clean) | |
617 |
|
617 | |||
618 | def setflag(self, key, flag): |
|
618 | def setflag(self, key, flag): | |
619 | if flag not in _manifestflags: |
|
619 | if flag not in _manifestflags: | |
620 | raise TypeError(b"Invalid manifest flag set.") |
|
620 | raise TypeError(b"Invalid manifest flag set.") | |
621 | self._lm[key] = self[key], flag |
|
621 | self._lm[key] = self[key], flag | |
622 |
|
622 | |||
623 | def get(self, key, default=None): |
|
623 | def get(self, key, default=None): | |
624 | try: |
|
624 | try: | |
625 | return self._lm[key][0] |
|
625 | return self._lm[key][0] | |
626 | except KeyError: |
|
626 | except KeyError: | |
627 | return default |
|
627 | return default | |
628 |
|
628 | |||
629 | def flags(self, key): |
|
629 | def flags(self, key): | |
630 | try: |
|
630 | try: | |
631 | return self._lm[key][1] |
|
631 | return self._lm[key][1] | |
632 | except KeyError: |
|
632 | except KeyError: | |
633 | return b'' |
|
633 | return b'' | |
634 |
|
634 | |||
635 | def copy(self): |
|
635 | def copy(self): | |
636 | c = manifestdict(self._nodelen) |
|
636 | c = manifestdict(self._nodelen) | |
637 | c._lm = self._lm.copy() |
|
637 | c._lm = self._lm.copy() | |
638 | return c |
|
638 | return c | |
639 |
|
639 | |||
640 | def items(self): |
|
640 | def items(self): | |
641 | return (x[:2] for x in self._lm.iterentries()) |
|
641 | return (x[:2] for x in self._lm.iterentries()) | |
642 |
|
642 | |||
643 | def iteritems(self): |
|
643 | def iteritems(self): | |
644 | return (x[:2] for x in self._lm.iterentries()) |
|
644 | return (x[:2] for x in self._lm.iterentries()) | |
645 |
|
645 | |||
646 | def iterentries(self): |
|
646 | def iterentries(self): | |
647 | return self._lm.iterentries() |
|
647 | return self._lm.iterentries() | |
648 |
|
648 | |||
649 | def text(self): |
|
649 | def text(self): | |
650 | # most likely uses native version |
|
650 | # most likely uses native version | |
651 | return self._lm.text() |
|
651 | return self._lm.text() | |
652 |
|
652 | |||
653 | def fastdelta(self, base, changes): |
|
653 | def fastdelta(self, base, changes): | |
654 | """Given a base manifest text as a bytearray and a list of changes |
|
654 | """Given a base manifest text as a bytearray and a list of changes | |
655 | relative to that text, compute a delta that can be used by revlog. |
|
655 | relative to that text, compute a delta that can be used by revlog. | |
656 | """ |
|
656 | """ | |
657 | delta = [] |
|
657 | delta = [] | |
658 | dstart = None |
|
658 | dstart = None | |
659 | dend = None |
|
659 | dend = None | |
660 | dline = [b""] |
|
660 | dline = [b""] | |
661 | start = 0 |
|
661 | start = 0 | |
662 | # zero copy representation of base as a buffer |
|
662 | # zero copy representation of base as a buffer | |
663 | addbuf = util.buffer(base) |
|
663 | addbuf = util.buffer(base) | |
664 |
|
664 | |||
665 | changes = list(changes) |
|
665 | changes = list(changes) | |
666 | if len(changes) < FASTDELTA_TEXTDIFF_THRESHOLD: |
|
666 | if len(changes) < FASTDELTA_TEXTDIFF_THRESHOLD: | |
667 | # start with a readonly loop that finds the offset of |
|
667 | # start with a readonly loop that finds the offset of | |
668 | # each line and creates the deltas |
|
668 | # each line and creates the deltas | |
669 | for f, todelete in changes: |
|
669 | for f, todelete in changes: | |
670 | # bs will either be the index of the item or the insert point |
|
670 | # bs will either be the index of the item or the insert point | |
671 | start, end = _msearch(addbuf, f, start) |
|
671 | start, end = _msearch(addbuf, f, start) | |
672 | if not todelete: |
|
672 | if not todelete: | |
673 | h, fl = self._lm[f] |
|
673 | h, fl = self._lm[f] | |
674 | l = b"%s\0%s%s\n" % (f, hex(h), fl) |
|
674 | l = b"%s\0%s%s\n" % (f, hex(h), fl) | |
675 | else: |
|
675 | else: | |
676 | if start == end: |
|
676 | if start == end: | |
677 | # item we want to delete was not found, error out |
|
677 | # item we want to delete was not found, error out | |
678 | raise AssertionError( |
|
678 | raise AssertionError( | |
679 | _(b"failed to remove %s from manifest") % f |
|
679 | _(b"failed to remove %s from manifest") % f | |
680 | ) |
|
680 | ) | |
681 | l = b"" |
|
681 | l = b"" | |
682 | if dstart is not None and dstart <= start and dend >= start: |
|
682 | if dstart is not None and dstart <= start and dend >= start: | |
683 | if dend < end: |
|
683 | if dend < end: | |
684 | dend = end |
|
684 | dend = end | |
685 | if l: |
|
685 | if l: | |
686 | dline.append(l) |
|
686 | dline.append(l) | |
687 | else: |
|
687 | else: | |
688 | if dstart is not None: |
|
688 | if dstart is not None: | |
689 | delta.append([dstart, dend, b"".join(dline)]) |
|
689 | delta.append([dstart, dend, b"".join(dline)]) | |
690 | dstart = start |
|
690 | dstart = start | |
691 | dend = end |
|
691 | dend = end | |
692 | dline = [l] |
|
692 | dline = [l] | |
693 |
|
693 | |||
694 | if dstart is not None: |
|
694 | if dstart is not None: | |
695 | delta.append([dstart, dend, b"".join(dline)]) |
|
695 | delta.append([dstart, dend, b"".join(dline)]) | |
696 | # apply the delta to the base, and get a delta for addrevision |
|
696 | # apply the delta to the base, and get a delta for addrevision | |
697 | deltatext, arraytext = _addlistdelta(base, delta) |
|
697 | deltatext, arraytext = _addlistdelta(base, delta) | |
698 | else: |
|
698 | else: | |
699 | # For large changes, it's much cheaper to just build the text and |
|
699 | # For large changes, it's much cheaper to just build the text and | |
700 | # diff it. |
|
700 | # diff it. | |
701 | arraytext = bytearray(self.text()) |
|
701 | arraytext = bytearray(self.text()) | |
702 | deltatext = mdiff.textdiff( |
|
702 | deltatext = mdiff.textdiff( | |
703 | util.buffer(base), util.buffer(arraytext) |
|
703 | util.buffer(base), util.buffer(arraytext) | |
704 | ) |
|
704 | ) | |
705 |
|
705 | |||
706 | return arraytext, deltatext |
|
706 | return arraytext, deltatext | |
707 |
|
707 | |||
708 |
|
708 | |||
709 | def _msearch(m, s, lo=0, hi=None): |
|
709 | def _msearch(m, s, lo=0, hi=None): | |
710 | """return a tuple (start, end) that says where to find s within m. |
|
710 | """return a tuple (start, end) that says where to find s within m. | |
711 |
|
711 | |||
712 | If the string is found m[start:end] are the line containing |
|
712 | If the string is found m[start:end] are the line containing | |
713 | that string. If start == end the string was not found and |
|
713 | that string. If start == end the string was not found and | |
714 | they indicate the proper sorted insertion point. |
|
714 | they indicate the proper sorted insertion point. | |
715 |
|
715 | |||
716 | m should be a buffer, a memoryview or a byte string. |
|
716 | m should be a buffer, a memoryview or a byte string. | |
717 | s is a byte string""" |
|
717 | s is a byte string""" | |
718 |
|
718 | |||
719 | def advance(i, c): |
|
719 | def advance(i, c): | |
720 | while i < lenm and m[i : i + 1] != c: |
|
720 | while i < lenm and m[i : i + 1] != c: | |
721 | i += 1 |
|
721 | i += 1 | |
722 | return i |
|
722 | return i | |
723 |
|
723 | |||
724 | if not s: |
|
724 | if not s: | |
725 | return (lo, lo) |
|
725 | return (lo, lo) | |
726 | lenm = len(m) |
|
726 | lenm = len(m) | |
727 | if not hi: |
|
727 | if not hi: | |
728 | hi = lenm |
|
728 | hi = lenm | |
729 | while lo < hi: |
|
729 | while lo < hi: | |
730 | mid = (lo + hi) // 2 |
|
730 | mid = (lo + hi) // 2 | |
731 | start = mid |
|
731 | start = mid | |
732 | while start > 0 and m[start - 1 : start] != b'\n': |
|
732 | while start > 0 and m[start - 1 : start] != b'\n': | |
733 | start -= 1 |
|
733 | start -= 1 | |
734 | end = advance(start, b'\0') |
|
734 | end = advance(start, b'\0') | |
735 | if bytes(m[start:end]) < s: |
|
735 | if bytes(m[start:end]) < s: | |
736 | # we know that after the null there are 40 bytes of sha1 |
|
736 | # we know that after the null there are 40 bytes of sha1 | |
737 | # this translates to the bisect lo = mid + 1 |
|
737 | # this translates to the bisect lo = mid + 1 | |
738 | lo = advance(end + 40, b'\n') + 1 |
|
738 | lo = advance(end + 40, b'\n') + 1 | |
739 | else: |
|
739 | else: | |
740 | # this translates to the bisect hi = mid |
|
740 | # this translates to the bisect hi = mid | |
741 | hi = start |
|
741 | hi = start | |
742 | end = advance(lo, b'\0') |
|
742 | end = advance(lo, b'\0') | |
743 | found = m[lo:end] |
|
743 | found = m[lo:end] | |
744 | if s == found: |
|
744 | if s == found: | |
745 | # we know that after the null there are 40 bytes of sha1 |
|
745 | # we know that after the null there are 40 bytes of sha1 | |
746 | end = advance(end + 40, b'\n') |
|
746 | end = advance(end + 40, b'\n') | |
747 | return (lo, end + 1) |
|
747 | return (lo, end + 1) | |
748 | else: |
|
748 | else: | |
749 | return (lo, lo) |
|
749 | return (lo, lo) | |
750 |
|
750 | |||
751 |
|
751 | |||
752 | def _checkforbidden(l): |
|
752 | def _checkforbidden(l): | |
753 | """Check filenames for illegal characters.""" |
|
753 | """Check filenames for illegal characters.""" | |
754 | for f in l: |
|
754 | for f in l: | |
755 | if b'\n' in f or b'\r' in f: |
|
755 | if b'\n' in f or b'\r' in f: | |
756 | raise error.StorageError( |
|
756 | raise error.StorageError( | |
757 | _(b"'\\n' and '\\r' disallowed in filenames: %r") |
|
757 | _(b"'\\n' and '\\r' disallowed in filenames: %r") | |
758 | % pycompat.bytestr(f) |
|
758 | % pycompat.bytestr(f) | |
759 | ) |
|
759 | ) | |
760 |
|
760 | |||
761 |
|
761 | |||
762 | # apply the changes collected during the bisect loop to our addlist |
|
762 | # apply the changes collected during the bisect loop to our addlist | |
763 | # return a delta suitable for addrevision |
|
763 | # return a delta suitable for addrevision | |
764 | def _addlistdelta(addlist, x): |
|
764 | def _addlistdelta(addlist, x): | |
765 | # for large addlist arrays, building a new array is cheaper |
|
765 | # for large addlist arrays, building a new array is cheaper | |
766 | # than repeatedly modifying the existing one |
|
766 | # than repeatedly modifying the existing one | |
767 | currentposition = 0 |
|
767 | currentposition = 0 | |
768 | newaddlist = bytearray() |
|
768 | newaddlist = bytearray() | |
769 |
|
769 | |||
770 | for start, end, content in x: |
|
770 | for start, end, content in x: | |
771 | newaddlist += addlist[currentposition:start] |
|
771 | newaddlist += addlist[currentposition:start] | |
772 | if content: |
|
772 | if content: | |
773 | newaddlist += bytearray(content) |
|
773 | newaddlist += bytearray(content) | |
774 |
|
774 | |||
775 | currentposition = end |
|
775 | currentposition = end | |
776 |
|
776 | |||
777 | newaddlist += addlist[currentposition:] |
|
777 | newaddlist += addlist[currentposition:] | |
778 |
|
778 | |||
779 | deltatext = b"".join( |
|
779 | deltatext = b"".join( | |
780 | struct.pack(b">lll", start, end, len(content)) + content |
|
780 | struct.pack(b">lll", start, end, len(content)) + content | |
781 | for start, end, content in x |
|
781 | for start, end, content in x | |
782 | ) |
|
782 | ) | |
783 | return deltatext, newaddlist |
|
783 | return deltatext, newaddlist | |
784 |
|
784 | |||
785 |
|
785 | |||
786 | def _splittopdir(f): |
|
786 | def _splittopdir(f): | |
787 | if b'/' in f: |
|
787 | if b'/' in f: | |
788 | dir, subpath = f.split(b'/', 1) |
|
788 | dir, subpath = f.split(b'/', 1) | |
789 | return dir + b'/', subpath |
|
789 | return dir + b'/', subpath | |
790 | else: |
|
790 | else: | |
791 | return b'', f |
|
791 | return b'', f | |
792 |
|
792 | |||
793 |
|
793 | |||
794 | _noop = lambda s: None |
|
794 | _noop = lambda s: None | |
795 |
|
795 | |||
796 |
|
796 | |||
797 | @interfaceutil.implementer(repository.imanifestdict) |
|
797 | @interfaceutil.implementer(repository.imanifestdict) | |
798 | class treemanifest: |
|
798 | class treemanifest: | |
799 | def __init__(self, nodeconstants, dir=b'', text=b''): |
|
799 | def __init__(self, nodeconstants, dir=b'', text=b''): | |
800 | self._dir = dir |
|
800 | self._dir = dir | |
801 | self.nodeconstants = nodeconstants |
|
801 | self.nodeconstants = nodeconstants | |
802 | self._node = self.nodeconstants.nullid |
|
802 | self._node = self.nodeconstants.nullid | |
803 | self._nodelen = self.nodeconstants.nodelen |
|
803 | self._nodelen = self.nodeconstants.nodelen | |
804 | self._loadfunc = _noop |
|
804 | self._loadfunc = _noop | |
805 | self._copyfunc = _noop |
|
805 | self._copyfunc = _noop | |
806 | self._dirty = False |
|
806 | self._dirty = False | |
807 | self._dirs = {} |
|
807 | self._dirs = {} | |
808 | self._lazydirs = {} |
|
808 | self._lazydirs = {} | |
809 | # Using _lazymanifest here is a little slower than plain old dicts |
|
809 | # Using _lazymanifest here is a little slower than plain old dicts | |
810 | self._files = {} |
|
810 | self._files = {} | |
811 | self._flags = {} |
|
811 | self._flags = {} | |
812 | if text: |
|
812 | if text: | |
813 |
|
813 | |||
814 | def readsubtree(subdir, subm): |
|
814 | def readsubtree(subdir, subm): | |
815 | raise AssertionError( |
|
815 | raise AssertionError( | |
816 | b'treemanifest constructor only accepts flat manifests' |
|
816 | b'treemanifest constructor only accepts flat manifests' | |
817 | ) |
|
817 | ) | |
818 |
|
818 | |||
819 | self.parse(text, readsubtree) |
|
819 | self.parse(text, readsubtree) | |
820 | self._dirty = True # Mark flat manifest dirty after parsing |
|
820 | self._dirty = True # Mark flat manifest dirty after parsing | |
821 |
|
821 | |||
822 | def _subpath(self, path): |
|
822 | def _subpath(self, path): | |
823 | return self._dir + path |
|
823 | return self._dir + path | |
824 |
|
824 | |||
825 | def _loadalllazy(self): |
|
825 | def _loadalllazy(self): | |
826 | selfdirs = self._dirs |
|
826 | selfdirs = self._dirs | |
827 | subpath = self._subpath |
|
827 | subpath = self._subpath | |
828 | for d, (node, readsubtree, docopy) in self._lazydirs.items(): |
|
828 | for d, (node, readsubtree, docopy) in self._lazydirs.items(): | |
829 | if docopy: |
|
829 | if docopy: | |
830 | selfdirs[d] = readsubtree(subpath(d), node).copy() |
|
830 | selfdirs[d] = readsubtree(subpath(d), node).copy() | |
831 | else: |
|
831 | else: | |
832 | selfdirs[d] = readsubtree(subpath(d), node) |
|
832 | selfdirs[d] = readsubtree(subpath(d), node) | |
833 | self._lazydirs = {} |
|
833 | self._lazydirs = {} | |
834 |
|
834 | |||
835 | def _loadlazy(self, d): |
|
835 | def _loadlazy(self, d): | |
836 | v = self._lazydirs.get(d) |
|
836 | v = self._lazydirs.get(d) | |
837 | if v: |
|
837 | if v: | |
838 | node, readsubtree, docopy = v |
|
838 | node, readsubtree, docopy = v | |
839 | if docopy: |
|
839 | if docopy: | |
840 | self._dirs[d] = readsubtree(self._subpath(d), node).copy() |
|
840 | self._dirs[d] = readsubtree(self._subpath(d), node).copy() | |
841 | else: |
|
841 | else: | |
842 | self._dirs[d] = readsubtree(self._subpath(d), node) |
|
842 | self._dirs[d] = readsubtree(self._subpath(d), node) | |
843 | del self._lazydirs[d] |
|
843 | del self._lazydirs[d] | |
844 |
|
844 | |||
845 | def _loadchildrensetlazy(self, visit): |
|
845 | def _loadchildrensetlazy(self, visit): | |
846 | if not visit: |
|
846 | if not visit: | |
847 | return None |
|
847 | return None | |
848 | if visit == b'all' or visit == b'this': |
|
848 | if visit == b'all' or visit == b'this': | |
849 | self._loadalllazy() |
|
849 | self._loadalllazy() | |
850 | return None |
|
850 | return None | |
851 |
|
851 | |||
852 | loadlazy = self._loadlazy |
|
852 | loadlazy = self._loadlazy | |
853 | for k in visit: |
|
853 | for k in visit: | |
854 | loadlazy(k + b'/') |
|
854 | loadlazy(k + b'/') | |
855 | return visit |
|
855 | return visit | |
856 |
|
856 | |||
857 | def _loaddifflazy(self, t1, t2): |
|
857 | def _loaddifflazy(self, t1, t2): | |
858 | """load items in t1 and t2 if they're needed for diffing. |
|
858 | """load items in t1 and t2 if they're needed for diffing. | |
859 |
|
859 | |||
860 | The criteria currently is: |
|
860 | The criteria currently is: | |
861 | - if it's not present in _lazydirs in either t1 or t2, load it in the |
|
861 | - if it's not present in _lazydirs in either t1 or t2, load it in the | |
862 | other (it may already be loaded or it may not exist, doesn't matter) |
|
862 | other (it may already be loaded or it may not exist, doesn't matter) | |
863 | - if it's present in _lazydirs in both, compare the nodeid; if it |
|
863 | - if it's present in _lazydirs in both, compare the nodeid; if it | |
864 | differs, load it in both |
|
864 | differs, load it in both | |
865 | """ |
|
865 | """ | |
866 | toloadlazy = [] |
|
866 | toloadlazy = [] | |
867 | for d, v1 in t1._lazydirs.items(): |
|
867 | for d, v1 in t1._lazydirs.items(): | |
868 | v2 = t2._lazydirs.get(d) |
|
868 | v2 = t2._lazydirs.get(d) | |
869 | if not v2 or v2[0] != v1[0]: |
|
869 | if not v2 or v2[0] != v1[0]: | |
870 | toloadlazy.append(d) |
|
870 | toloadlazy.append(d) | |
871 | for d, v1 in t2._lazydirs.items(): |
|
871 | for d, v1 in t2._lazydirs.items(): | |
872 | if d not in t1._lazydirs: |
|
872 | if d not in t1._lazydirs: | |
873 | toloadlazy.append(d) |
|
873 | toloadlazy.append(d) | |
874 |
|
874 | |||
875 | for d in toloadlazy: |
|
875 | for d in toloadlazy: | |
876 | t1._loadlazy(d) |
|
876 | t1._loadlazy(d) | |
877 | t2._loadlazy(d) |
|
877 | t2._loadlazy(d) | |
878 |
|
878 | |||
879 | def __len__(self): |
|
879 | def __len__(self): | |
880 | self._load() |
|
880 | self._load() | |
881 | size = len(self._files) |
|
881 | size = len(self._files) | |
882 | self._loadalllazy() |
|
882 | self._loadalllazy() | |
883 | for m in self._dirs.values(): |
|
883 | for m in self._dirs.values(): | |
884 | size += m.__len__() |
|
884 | size += m.__len__() | |
885 | return size |
|
885 | return size | |
886 |
|
886 | |||
887 | def __nonzero__(self): |
|
887 | def __nonzero__(self): | |
888 | # Faster than "__len() != 0" since it avoids loading sub-manifests |
|
888 | # Faster than "__len() != 0" since it avoids loading sub-manifests | |
889 | return not self._isempty() |
|
889 | return not self._isempty() | |
890 |
|
890 | |||
891 | __bool__ = __nonzero__ |
|
891 | __bool__ = __nonzero__ | |
892 |
|
892 | |||
893 | def _isempty(self): |
|
893 | def _isempty(self): | |
894 | self._load() # for consistency; already loaded by all callers |
|
894 | self._load() # for consistency; already loaded by all callers | |
895 | # See if we can skip loading everything. |
|
895 | # See if we can skip loading everything. | |
896 | if self._files or ( |
|
896 | if self._files or ( | |
897 | self._dirs and any(not m._isempty() for m in self._dirs.values()) |
|
897 | self._dirs and any(not m._isempty() for m in self._dirs.values()) | |
898 | ): |
|
898 | ): | |
899 | return False |
|
899 | return False | |
900 | self._loadalllazy() |
|
900 | self._loadalllazy() | |
901 | return not self._dirs or all(m._isempty() for m in self._dirs.values()) |
|
901 | return not self._dirs or all(m._isempty() for m in self._dirs.values()) | |
902 |
|
902 | |||
903 | @encoding.strmethod |
|
903 | @encoding.strmethod | |
904 | def __repr__(self): |
|
904 | def __repr__(self): | |
905 | return ( |
|
905 | return ( | |
906 | b'<treemanifest dir=%s, node=%s, loaded=%r, dirty=%r at 0x%x>' |
|
906 | b'<treemanifest dir=%s, node=%s, loaded=%r, dirty=%r at 0x%x>' | |
907 | % ( |
|
907 | % ( | |
908 | self._dir, |
|
908 | self._dir, | |
909 | hex(self._node), |
|
909 | hex(self._node), | |
910 | bool(self._loadfunc is _noop), |
|
910 | bool(self._loadfunc is _noop), | |
911 | self._dirty, |
|
911 | self._dirty, | |
912 | id(self), |
|
912 | id(self), | |
913 | ) |
|
913 | ) | |
914 | ) |
|
914 | ) | |
915 |
|
915 | |||
916 | def dir(self): |
|
916 | def dir(self): | |
917 | """The directory that this tree manifest represents, including a |
|
917 | """The directory that this tree manifest represents, including a | |
918 | trailing '/'. Empty string for the repo root directory.""" |
|
918 | trailing '/'. Empty string for the repo root directory.""" | |
919 | return self._dir |
|
919 | return self._dir | |
920 |
|
920 | |||
921 | def node(self): |
|
921 | def node(self): | |
922 | """This node of this instance. nullid for unsaved instances. Should |
|
922 | """This node of this instance. nullid for unsaved instances. Should | |
923 | be updated when the instance is read or written from a revlog. |
|
923 | be updated when the instance is read or written from a revlog. | |
924 | """ |
|
924 | """ | |
925 | assert not self._dirty |
|
925 | assert not self._dirty | |
926 | return self._node |
|
926 | return self._node | |
927 |
|
927 | |||
928 | def setnode(self, node): |
|
928 | def setnode(self, node): | |
929 | self._node = node |
|
929 | self._node = node | |
930 | self._dirty = False |
|
930 | self._dirty = False | |
931 |
|
931 | |||
932 | def iterentries(self): |
|
932 | def iterentries(self): | |
933 | self._load() |
|
933 | self._load() | |
934 | self._loadalllazy() |
|
934 | self._loadalllazy() | |
935 | for p, n in sorted( |
|
935 | for p, n in sorted( | |
936 | itertools.chain(self._dirs.items(), self._files.items()) |
|
936 | itertools.chain(self._dirs.items(), self._files.items()) | |
937 | ): |
|
937 | ): | |
938 | if p in self._files: |
|
938 | if p in self._files: | |
939 | yield self._subpath(p), n, self._flags.get(p, b'') |
|
939 | yield self._subpath(p), n, self._flags.get(p, b'') | |
940 | else: |
|
940 | else: | |
941 | for x in n.iterentries(): |
|
941 | for x in n.iterentries(): | |
942 | yield x |
|
942 | yield x | |
943 |
|
943 | |||
944 | def items(self): |
|
944 | def items(self): | |
945 | self._load() |
|
945 | self._load() | |
946 | self._loadalllazy() |
|
946 | self._loadalllazy() | |
947 | for p, n in sorted( |
|
947 | for p, n in sorted( | |
948 | itertools.chain(self._dirs.items(), self._files.items()) |
|
948 | itertools.chain(self._dirs.items(), self._files.items()) | |
949 | ): |
|
949 | ): | |
950 | if p in self._files: |
|
950 | if p in self._files: | |
951 | yield self._subpath(p), n |
|
951 | yield self._subpath(p), n | |
952 | else: |
|
952 | else: | |
953 | for f, sn in n.items(): |
|
953 | for f, sn in n.items(): | |
954 | yield f, sn |
|
954 | yield f, sn | |
955 |
|
955 | |||
956 | iteritems = items |
|
956 | iteritems = items | |
957 |
|
957 | |||
958 | def iterkeys(self): |
|
958 | def iterkeys(self): | |
959 | self._load() |
|
959 | self._load() | |
960 | self._loadalllazy() |
|
960 | self._loadalllazy() | |
961 | for p in sorted(itertools.chain(self._dirs, self._files)): |
|
961 | for p in sorted(itertools.chain(self._dirs, self._files)): | |
962 | if p in self._files: |
|
962 | if p in self._files: | |
963 | yield self._subpath(p) |
|
963 | yield self._subpath(p) | |
964 | else: |
|
964 | else: | |
965 | for f in self._dirs[p]: |
|
965 | for f in self._dirs[p]: | |
966 | yield f |
|
966 | yield f | |
967 |
|
967 | |||
968 | def keys(self): |
|
968 | def keys(self): | |
969 | return list(self.iterkeys()) |
|
969 | return list(self.iterkeys()) | |
970 |
|
970 | |||
971 | def __iter__(self): |
|
971 | def __iter__(self): | |
972 | return self.iterkeys() |
|
972 | return self.iterkeys() | |
973 |
|
973 | |||
974 | def __contains__(self, f): |
|
974 | def __contains__(self, f): | |
975 | if f is None: |
|
975 | if f is None: | |
976 | return False |
|
976 | return False | |
977 | self._load() |
|
977 | self._load() | |
978 | dir, subpath = _splittopdir(f) |
|
978 | dir, subpath = _splittopdir(f) | |
979 | if dir: |
|
979 | if dir: | |
980 | self._loadlazy(dir) |
|
980 | self._loadlazy(dir) | |
981 |
|
981 | |||
982 | if dir not in self._dirs: |
|
982 | if dir not in self._dirs: | |
983 | return False |
|
983 | return False | |
984 |
|
984 | |||
985 | return self._dirs[dir].__contains__(subpath) |
|
985 | return self._dirs[dir].__contains__(subpath) | |
986 | else: |
|
986 | else: | |
987 | return f in self._files |
|
987 | return f in self._files | |
988 |
|
988 | |||
989 | def get(self, f, default=None): |
|
989 | def get(self, f, default=None): | |
990 | self._load() |
|
990 | self._load() | |
991 | dir, subpath = _splittopdir(f) |
|
991 | dir, subpath = _splittopdir(f) | |
992 | if dir: |
|
992 | if dir: | |
993 | self._loadlazy(dir) |
|
993 | self._loadlazy(dir) | |
994 |
|
994 | |||
995 | if dir not in self._dirs: |
|
995 | if dir not in self._dirs: | |
996 | return default |
|
996 | return default | |
997 | return self._dirs[dir].get(subpath, default) |
|
997 | return self._dirs[dir].get(subpath, default) | |
998 | else: |
|
998 | else: | |
999 | return self._files.get(f, default) |
|
999 | return self._files.get(f, default) | |
1000 |
|
1000 | |||
1001 | def __getitem__(self, f): |
|
1001 | def __getitem__(self, f): | |
1002 | self._load() |
|
1002 | self._load() | |
1003 | dir, subpath = _splittopdir(f) |
|
1003 | dir, subpath = _splittopdir(f) | |
1004 | if dir: |
|
1004 | if dir: | |
1005 | self._loadlazy(dir) |
|
1005 | self._loadlazy(dir) | |
1006 |
|
1006 | |||
1007 | return self._dirs[dir].__getitem__(subpath) |
|
1007 | return self._dirs[dir].__getitem__(subpath) | |
1008 | else: |
|
1008 | else: | |
1009 | return self._files[f] |
|
1009 | return self._files[f] | |
1010 |
|
1010 | |||
1011 | def flags(self, f): |
|
1011 | def flags(self, f): | |
1012 | self._load() |
|
1012 | self._load() | |
1013 | dir, subpath = _splittopdir(f) |
|
1013 | dir, subpath = _splittopdir(f) | |
1014 | if dir: |
|
1014 | if dir: | |
1015 | self._loadlazy(dir) |
|
1015 | self._loadlazy(dir) | |
1016 |
|
1016 | |||
1017 | if dir not in self._dirs: |
|
1017 | if dir not in self._dirs: | |
1018 | return b'' |
|
1018 | return b'' | |
1019 | return self._dirs[dir].flags(subpath) |
|
1019 | return self._dirs[dir].flags(subpath) | |
1020 | else: |
|
1020 | else: | |
1021 | if f in self._lazydirs or f in self._dirs: |
|
1021 | if f in self._lazydirs or f in self._dirs: | |
1022 | return b'' |
|
1022 | return b'' | |
1023 | return self._flags.get(f, b'') |
|
1023 | return self._flags.get(f, b'') | |
1024 |
|
1024 | |||
1025 | def find(self, f): |
|
1025 | def find(self, f): | |
1026 | self._load() |
|
1026 | self._load() | |
1027 | dir, subpath = _splittopdir(f) |
|
1027 | dir, subpath = _splittopdir(f) | |
1028 | if dir: |
|
1028 | if dir: | |
1029 | self._loadlazy(dir) |
|
1029 | self._loadlazy(dir) | |
1030 |
|
1030 | |||
1031 | return self._dirs[dir].find(subpath) |
|
1031 | return self._dirs[dir].find(subpath) | |
1032 | else: |
|
1032 | else: | |
1033 | return self._files[f], self._flags.get(f, b'') |
|
1033 | return self._files[f], self._flags.get(f, b'') | |
1034 |
|
1034 | |||
1035 | def __delitem__(self, f): |
|
1035 | def __delitem__(self, f): | |
1036 | self._load() |
|
1036 | self._load() | |
1037 | dir, subpath = _splittopdir(f) |
|
1037 | dir, subpath = _splittopdir(f) | |
1038 | if dir: |
|
1038 | if dir: | |
1039 | self._loadlazy(dir) |
|
1039 | self._loadlazy(dir) | |
1040 |
|
1040 | |||
1041 | self._dirs[dir].__delitem__(subpath) |
|
1041 | self._dirs[dir].__delitem__(subpath) | |
1042 | # If the directory is now empty, remove it |
|
1042 | # If the directory is now empty, remove it | |
1043 | if self._dirs[dir]._isempty(): |
|
1043 | if self._dirs[dir]._isempty(): | |
1044 | del self._dirs[dir] |
|
1044 | del self._dirs[dir] | |
1045 | else: |
|
1045 | else: | |
1046 | del self._files[f] |
|
1046 | del self._files[f] | |
1047 | if f in self._flags: |
|
1047 | if f in self._flags: | |
1048 | del self._flags[f] |
|
1048 | del self._flags[f] | |
1049 | self._dirty = True |
|
1049 | self._dirty = True | |
1050 |
|
1050 | |||
1051 | def __setitem__(self, f, n): |
|
1051 | def __setitem__(self, f, n): | |
1052 | assert n is not None |
|
1052 | assert n is not None | |
1053 | self._load() |
|
1053 | self._load() | |
1054 | dir, subpath = _splittopdir(f) |
|
1054 | dir, subpath = _splittopdir(f) | |
1055 | if dir: |
|
1055 | if dir: | |
1056 | self._loadlazy(dir) |
|
1056 | self._loadlazy(dir) | |
1057 | if dir not in self._dirs: |
|
1057 | if dir not in self._dirs: | |
1058 | self._dirs[dir] = treemanifest( |
|
1058 | self._dirs[dir] = treemanifest( | |
1059 | self.nodeconstants, self._subpath(dir) |
|
1059 | self.nodeconstants, self._subpath(dir) | |
1060 | ) |
|
1060 | ) | |
1061 | self._dirs[dir].__setitem__(subpath, n) |
|
1061 | self._dirs[dir].__setitem__(subpath, n) | |
1062 | else: |
|
1062 | else: | |
1063 | # manifest nodes are either 20 bytes or 32 bytes, |
|
1063 | # manifest nodes are either 20 bytes or 32 bytes, | |
1064 | # depending on the hash in use. Assert this as historically |
|
1064 | # depending on the hash in use. Assert this as historically | |
1065 | # sometimes extra bytes were added. |
|
1065 | # sometimes extra bytes were added. | |
1066 | assert len(n) in (20, 32) |
|
1066 | assert len(n) in (20, 32) | |
1067 | self._files[f] = n |
|
1067 | self._files[f] = n | |
1068 | self._dirty = True |
|
1068 | self._dirty = True | |
1069 |
|
1069 | |||
1070 | def _load(self): |
|
1070 | def _load(self): | |
1071 | if self._loadfunc is not _noop: |
|
1071 | if self._loadfunc is not _noop: | |
1072 | lf, self._loadfunc = self._loadfunc, _noop |
|
1072 | lf, self._loadfunc = self._loadfunc, _noop | |
1073 | lf(self) |
|
1073 | lf(self) | |
1074 | elif self._copyfunc is not _noop: |
|
1074 | elif self._copyfunc is not _noop: | |
1075 | cf, self._copyfunc = self._copyfunc, _noop |
|
1075 | cf, self._copyfunc = self._copyfunc, _noop | |
1076 | cf(self) |
|
1076 | cf(self) | |
1077 |
|
1077 | |||
1078 | def setflag(self, f, flags): |
|
1078 | def setflag(self, f, flags): | |
1079 | """Set the flags (symlink, executable) for path f.""" |
|
1079 | """Set the flags (symlink, executable) for path f.""" | |
1080 | if flags not in _manifestflags: |
|
1080 | if flags not in _manifestflags: | |
1081 | raise TypeError(b"Invalid manifest flag set.") |
|
1081 | raise TypeError(b"Invalid manifest flag set.") | |
1082 | self._load() |
|
1082 | self._load() | |
1083 | dir, subpath = _splittopdir(f) |
|
1083 | dir, subpath = _splittopdir(f) | |
1084 | if dir: |
|
1084 | if dir: | |
1085 | self._loadlazy(dir) |
|
1085 | self._loadlazy(dir) | |
1086 | if dir not in self._dirs: |
|
1086 | if dir not in self._dirs: | |
1087 | self._dirs[dir] = treemanifest( |
|
1087 | self._dirs[dir] = treemanifest( | |
1088 | self.nodeconstants, self._subpath(dir) |
|
1088 | self.nodeconstants, self._subpath(dir) | |
1089 | ) |
|
1089 | ) | |
1090 | self._dirs[dir].setflag(subpath, flags) |
|
1090 | self._dirs[dir].setflag(subpath, flags) | |
1091 | else: |
|
1091 | else: | |
1092 | self._flags[f] = flags |
|
1092 | self._flags[f] = flags | |
1093 | self._dirty = True |
|
1093 | self._dirty = True | |
1094 |
|
1094 | |||
1095 | def copy(self): |
|
1095 | def copy(self): | |
1096 | copy = treemanifest(self.nodeconstants, self._dir) |
|
1096 | copy = treemanifest(self.nodeconstants, self._dir) | |
1097 | copy._node = self._node |
|
1097 | copy._node = self._node | |
1098 | copy._dirty = self._dirty |
|
1098 | copy._dirty = self._dirty | |
1099 | if self._copyfunc is _noop: |
|
1099 | if self._copyfunc is _noop: | |
1100 |
|
1100 | |||
1101 | def _copyfunc(s): |
|
1101 | def _copyfunc(s): | |
1102 | self._load() |
|
1102 | self._load() | |
1103 | s._lazydirs = { |
|
1103 | s._lazydirs = { | |
1104 | d: (n, r, True) for d, (n, r, c) in self._lazydirs.items() |
|
1104 | d: (n, r, True) for d, (n, r, c) in self._lazydirs.items() | |
1105 | } |
|
1105 | } | |
1106 | sdirs = s._dirs |
|
1106 | sdirs = s._dirs | |
1107 | for d, v in self._dirs.items(): |
|
1107 | for d, v in self._dirs.items(): | |
1108 | sdirs[d] = v.copy() |
|
1108 | sdirs[d] = v.copy() | |
1109 | s._files = dict.copy(self._files) |
|
1109 | s._files = dict.copy(self._files) | |
1110 | s._flags = dict.copy(self._flags) |
|
1110 | s._flags = dict.copy(self._flags) | |
1111 |
|
1111 | |||
1112 | if self._loadfunc is _noop: |
|
1112 | if self._loadfunc is _noop: | |
1113 | _copyfunc(copy) |
|
1113 | _copyfunc(copy) | |
1114 | else: |
|
1114 | else: | |
1115 | copy._copyfunc = _copyfunc |
|
1115 | copy._copyfunc = _copyfunc | |
1116 | else: |
|
1116 | else: | |
1117 | copy._copyfunc = self._copyfunc |
|
1117 | copy._copyfunc = self._copyfunc | |
1118 | return copy |
|
1118 | return copy | |
1119 |
|
1119 | |||
1120 | def filesnotin(self, m2, match=None): |
|
1120 | def filesnotin(self, m2, match=None): | |
1121 | '''Set of files in this manifest that are not in the other''' |
|
1121 | '''Set of files in this manifest that are not in the other''' | |
1122 | if match and not match.always(): |
|
1122 | if match and not match.always(): | |
1123 | m1 = self._matches(match) |
|
1123 | m1 = self._matches(match) | |
1124 | m2 = m2._matches(match) |
|
1124 | m2 = m2._matches(match) | |
1125 | return m1.filesnotin(m2) |
|
1125 | return m1.filesnotin(m2) | |
1126 |
|
1126 | |||
1127 | files = set() |
|
1127 | files = set() | |
1128 |
|
1128 | |||
1129 | def _filesnotin(t1, t2): |
|
1129 | def _filesnotin(t1, t2): | |
1130 | if t1._node == t2._node and not t1._dirty and not t2._dirty: |
|
1130 | if t1._node == t2._node and not t1._dirty and not t2._dirty: | |
1131 | return |
|
1131 | return | |
1132 | t1._load() |
|
1132 | t1._load() | |
1133 | t2._load() |
|
1133 | t2._load() | |
1134 | self._loaddifflazy(t1, t2) |
|
1134 | self._loaddifflazy(t1, t2) | |
1135 | for d, m1 in t1._dirs.items(): |
|
1135 | for d, m1 in t1._dirs.items(): | |
1136 | if d in t2._dirs: |
|
1136 | if d in t2._dirs: | |
1137 | m2 = t2._dirs[d] |
|
1137 | m2 = t2._dirs[d] | |
1138 | _filesnotin(m1, m2) |
|
1138 | _filesnotin(m1, m2) | |
1139 | else: |
|
1139 | else: | |
1140 | files.update(m1.iterkeys()) |
|
1140 | files.update(m1.iterkeys()) | |
1141 |
|
1141 | |||
1142 | for fn in t1._files: |
|
1142 | for fn in t1._files: | |
1143 | if fn not in t2._files: |
|
1143 | if fn not in t2._files: | |
1144 | files.add(t1._subpath(fn)) |
|
1144 | files.add(t1._subpath(fn)) | |
1145 |
|
1145 | |||
1146 | _filesnotin(self, m2) |
|
1146 | _filesnotin(self, m2) | |
1147 | return files |
|
1147 | return files | |
1148 |
|
1148 | |||
1149 | @propertycache |
|
1149 | @propertycache | |
1150 | def _alldirs(self): |
|
1150 | def _alldirs(self): | |
1151 | return pathutil.dirs(self) |
|
1151 | return pathutil.dirs(self) | |
1152 |
|
1152 | |||
1153 | def dirs(self): |
|
1153 | def dirs(self): | |
1154 | return self._alldirs |
|
1154 | return self._alldirs | |
1155 |
|
1155 | |||
1156 | def hasdir(self, dir): |
|
1156 | def hasdir(self, dir): | |
1157 | self._load() |
|
1157 | self._load() | |
1158 | topdir, subdir = _splittopdir(dir) |
|
1158 | topdir, subdir = _splittopdir(dir) | |
1159 | if topdir: |
|
1159 | if topdir: | |
1160 | self._loadlazy(topdir) |
|
1160 | self._loadlazy(topdir) | |
1161 | if topdir in self._dirs: |
|
1161 | if topdir in self._dirs: | |
1162 | return self._dirs[topdir].hasdir(subdir) |
|
1162 | return self._dirs[topdir].hasdir(subdir) | |
1163 | return False |
|
1163 | return False | |
1164 | dirslash = dir + b'/' |
|
1164 | dirslash = dir + b'/' | |
1165 | return dirslash in self._dirs or dirslash in self._lazydirs |
|
1165 | return dirslash in self._dirs or dirslash in self._lazydirs | |
1166 |
|
1166 | |||
1167 | def walk(self, match): |
|
1167 | def walk(self, match): | |
1168 | """Generates matching file names. |
|
1168 | """Generates matching file names. | |
1169 |
|
1169 | |||
1170 | It also reports nonexistent files by marking them bad with match.bad(). |
|
1170 | It also reports nonexistent files by marking them bad with match.bad(). | |
1171 | """ |
|
1171 | """ | |
1172 | if match.always(): |
|
1172 | if match.always(): | |
1173 | for f in iter(self): |
|
1173 | for f in iter(self): | |
1174 | yield f |
|
1174 | yield f | |
1175 | return |
|
1175 | return | |
1176 |
|
1176 | |||
1177 | fset = set(match.files()) |
|
1177 | fset = set(match.files()) | |
1178 |
|
1178 | |||
1179 | for fn in self._walk(match): |
|
1179 | for fn in self._walk(match): | |
1180 | if fn in fset: |
|
1180 | if fn in fset: | |
1181 | # specified pattern is the exact name |
|
1181 | # specified pattern is the exact name | |
1182 | fset.remove(fn) |
|
1182 | fset.remove(fn) | |
1183 | yield fn |
|
1183 | yield fn | |
1184 |
|
1184 | |||
1185 | # for dirstate.walk, files=[''] means "walk the whole tree". |
|
1185 | # for dirstate.walk, files=[''] means "walk the whole tree". | |
1186 | # follow that here, too |
|
1186 | # follow that here, too | |
1187 | fset.discard(b'') |
|
1187 | fset.discard(b'') | |
1188 |
|
1188 | |||
1189 | for fn in sorted(fset): |
|
1189 | for fn in sorted(fset): | |
1190 | if not self.hasdir(fn): |
|
1190 | if not self.hasdir(fn): | |
1191 | match.bad(fn, None) |
|
1191 | match.bad(fn, None) | |
1192 |
|
1192 | |||
1193 | def _walk(self, match): |
|
1193 | def _walk(self, match): | |
1194 | '''Recursively generates matching file names for walk().''' |
|
1194 | '''Recursively generates matching file names for walk().''' | |
1195 | visit = match.visitchildrenset(self._dir[:-1]) |
|
1195 | visit = match.visitchildrenset(self._dir[:-1]) | |
1196 | if not visit: |
|
1196 | if not visit: | |
1197 | return |
|
1197 | return | |
1198 |
|
1198 | |||
1199 | # yield this dir's files and walk its submanifests |
|
1199 | # yield this dir's files and walk its submanifests | |
1200 | self._load() |
|
1200 | self._load() | |
1201 | visit = self._loadchildrensetlazy(visit) |
|
1201 | visit = self._loadchildrensetlazy(visit) | |
1202 | for p in sorted(list(self._dirs) + list(self._files)): |
|
1202 | for p in sorted(list(self._dirs) + list(self._files)): | |
1203 | if p in self._files: |
|
1203 | if p in self._files: | |
1204 | fullp = self._subpath(p) |
|
1204 | fullp = self._subpath(p) | |
1205 | if match(fullp): |
|
1205 | if match(fullp): | |
1206 | yield fullp |
|
1206 | yield fullp | |
1207 | else: |
|
1207 | else: | |
1208 | if not visit or p[:-1] in visit: |
|
1208 | if not visit or p[:-1] in visit: | |
1209 | for f in self._dirs[p]._walk(match): |
|
1209 | for f in self._dirs[p]._walk(match): | |
1210 | yield f |
|
1210 | yield f | |
1211 |
|
1211 | |||
1212 | def _matches(self, match): |
|
1212 | def _matches(self, match): | |
1213 | """recursively generate a new manifest filtered by the match argument.""" |
|
1213 | """recursively generate a new manifest filtered by the match argument.""" | |
1214 | if match.always(): |
|
1214 | if match.always(): | |
1215 | return self.copy() |
|
1215 | return self.copy() | |
1216 | return self._matches_inner(match) |
|
1216 | return self._matches_inner(match) | |
1217 |
|
1217 | |||
1218 | def _matches_inner(self, match): |
|
1218 | def _matches_inner(self, match): | |
1219 | if match.always(): |
|
1219 | if match.always(): | |
1220 | return self.copy() |
|
1220 | return self.copy() | |
1221 |
|
1221 | |||
1222 | visit = match.visitchildrenset(self._dir[:-1]) |
|
1222 | visit = match.visitchildrenset(self._dir[:-1]) | |
1223 | if visit == b'all': |
|
1223 | if visit == b'all': | |
1224 | return self.copy() |
|
1224 | return self.copy() | |
1225 | ret = treemanifest(self.nodeconstants, self._dir) |
|
1225 | ret = treemanifest(self.nodeconstants, self._dir) | |
1226 | if not visit: |
|
1226 | if not visit: | |
1227 | return ret |
|
1227 | return ret | |
1228 |
|
1228 | |||
1229 | self._load() |
|
1229 | self._load() | |
1230 | for fn in self._files: |
|
1230 | for fn in self._files: | |
1231 | # While visitchildrenset *usually* lists only subdirs, this is |
|
1231 | # While visitchildrenset *usually* lists only subdirs, this is | |
1232 | # actually up to the matcher and may have some files in the set(). |
|
1232 | # actually up to the matcher and may have some files in the set(). | |
1233 | # If visit == 'this', we should obviously look at the files in this |
|
1233 | # If visit == 'this', we should obviously look at the files in this | |
1234 | # directory; if visit is a set, and fn is in it, we should inspect |
|
1234 | # directory; if visit is a set, and fn is in it, we should inspect | |
1235 | # fn (but no need to inspect things not in the set). |
|
1235 | # fn (but no need to inspect things not in the set). | |
1236 | if visit != b'this' and fn not in visit: |
|
1236 | if visit != b'this' and fn not in visit: | |
1237 | continue |
|
1237 | continue | |
1238 | fullp = self._subpath(fn) |
|
1238 | fullp = self._subpath(fn) | |
1239 | # visitchildrenset isn't perfect, we still need to call the regular |
|
1239 | # visitchildrenset isn't perfect, we still need to call the regular | |
1240 | # matcher code to further filter results. |
|
1240 | # matcher code to further filter results. | |
1241 | if not match(fullp): |
|
1241 | if not match(fullp): | |
1242 | continue |
|
1242 | continue | |
1243 | ret._files[fn] = self._files[fn] |
|
1243 | ret._files[fn] = self._files[fn] | |
1244 | if fn in self._flags: |
|
1244 | if fn in self._flags: | |
1245 | ret._flags[fn] = self._flags[fn] |
|
1245 | ret._flags[fn] = self._flags[fn] | |
1246 |
|
1246 | |||
1247 | visit = self._loadchildrensetlazy(visit) |
|
1247 | visit = self._loadchildrensetlazy(visit) | |
1248 | for dir, subm in self._dirs.items(): |
|
1248 | for dir, subm in self._dirs.items(): | |
1249 | if visit and dir[:-1] not in visit: |
|
1249 | if visit and dir[:-1] not in visit: | |
1250 | continue |
|
1250 | continue | |
1251 | m = subm._matches_inner(match) |
|
1251 | m = subm._matches_inner(match) | |
1252 | if not m._isempty(): |
|
1252 | if not m._isempty(): | |
1253 | ret._dirs[dir] = m |
|
1253 | ret._dirs[dir] = m | |
1254 |
|
1254 | |||
1255 | if not ret._isempty(): |
|
1255 | if not ret._isempty(): | |
1256 | ret._dirty = True |
|
1256 | ret._dirty = True | |
1257 | return ret |
|
1257 | return ret | |
1258 |
|
1258 | |||
1259 | def fastdelta(self, base, changes): |
|
1259 | def fastdelta(self, base, changes): | |
1260 | raise FastdeltaUnavailable() |
|
1260 | raise FastdeltaUnavailable() | |
1261 |
|
1261 | |||
1262 | def diff(self, m2, match=None, clean=False): |
|
1262 | def diff(self, m2, match=None, clean=False): | |
1263 | """Finds changes between the current manifest and m2. |
|
1263 | """Finds changes between the current manifest and m2. | |
1264 |
|
1264 | |||
1265 | Args: |
|
1265 | Args: | |
1266 | m2: the manifest to which this manifest should be compared. |
|
1266 | m2: the manifest to which this manifest should be compared. | |
1267 | clean: if true, include files unchanged between these manifests |
|
1267 | clean: if true, include files unchanged between these manifests | |
1268 | with a None value in the returned dictionary. |
|
1268 | with a None value in the returned dictionary. | |
1269 |
|
1269 | |||
1270 | The result is returned as a dict with filename as key and |
|
1270 | The result is returned as a dict with filename as key and | |
1271 | values of the form ((n1,fl1),(n2,fl2)), where n1/n2 is the |
|
1271 | values of the form ((n1,fl1),(n2,fl2)), where n1/n2 is the | |
1272 | nodeid in the current/other manifest and fl1/fl2 is the flag |
|
1272 | nodeid in the current/other manifest and fl1/fl2 is the flag | |
1273 | in the current/other manifest. Where the file does not exist, |
|
1273 | in the current/other manifest. Where the file does not exist, | |
1274 | the nodeid will be None and the flags will be the empty |
|
1274 | the nodeid will be None and the flags will be the empty | |
1275 | string. |
|
1275 | string. | |
1276 | """ |
|
1276 | """ | |
1277 | if match and not match.always(): |
|
1277 | if match and not match.always(): | |
1278 | m1 = self._matches(match) |
|
1278 | m1 = self._matches(match) | |
1279 | m2 = m2._matches(match) |
|
1279 | m2 = m2._matches(match) | |
1280 | return m1.diff(m2, clean=clean) |
|
1280 | return m1.diff(m2, clean=clean) | |
1281 | result = {} |
|
1281 | result = {} | |
1282 | emptytree = treemanifest(self.nodeconstants) |
|
1282 | emptytree = treemanifest(self.nodeconstants) | |
1283 |
|
1283 | |||
1284 | def _iterativediff(t1, t2, stack): |
|
1284 | def _iterativediff(t1, t2, stack): | |
1285 | """compares two tree manifests and append new tree-manifests which |
|
1285 | """compares two tree manifests and append new tree-manifests which | |
1286 | needs to be compared to stack""" |
|
1286 | needs to be compared to stack""" | |
1287 | if t1._node == t2._node and not t1._dirty and not t2._dirty: |
|
1287 | if t1._node == t2._node and not t1._dirty and not t2._dirty: | |
1288 | return |
|
1288 | return | |
1289 | t1._load() |
|
1289 | t1._load() | |
1290 | t2._load() |
|
1290 | t2._load() | |
1291 | self._loaddifflazy(t1, t2) |
|
1291 | self._loaddifflazy(t1, t2) | |
1292 |
|
1292 | |||
1293 | for d, m1 in t1._dirs.items(): |
|
1293 | for d, m1 in t1._dirs.items(): | |
1294 | m2 = t2._dirs.get(d, emptytree) |
|
1294 | m2 = t2._dirs.get(d, emptytree) | |
1295 | stack.append((m1, m2)) |
|
1295 | stack.append((m1, m2)) | |
1296 |
|
1296 | |||
1297 | for d, m2 in t2._dirs.items(): |
|
1297 | for d, m2 in t2._dirs.items(): | |
1298 | if d not in t1._dirs: |
|
1298 | if d not in t1._dirs: | |
1299 | stack.append((emptytree, m2)) |
|
1299 | stack.append((emptytree, m2)) | |
1300 |
|
1300 | |||
1301 | for fn, n1 in t1._files.items(): |
|
1301 | for fn, n1 in t1._files.items(): | |
1302 | fl1 = t1._flags.get(fn, b'') |
|
1302 | fl1 = t1._flags.get(fn, b'') | |
1303 | n2 = t2._files.get(fn, None) |
|
1303 | n2 = t2._files.get(fn, None) | |
1304 | fl2 = t2._flags.get(fn, b'') |
|
1304 | fl2 = t2._flags.get(fn, b'') | |
1305 | if n1 != n2 or fl1 != fl2: |
|
1305 | if n1 != n2 or fl1 != fl2: | |
1306 | result[t1._subpath(fn)] = ((n1, fl1), (n2, fl2)) |
|
1306 | result[t1._subpath(fn)] = ((n1, fl1), (n2, fl2)) | |
1307 | elif clean: |
|
1307 | elif clean: | |
1308 | result[t1._subpath(fn)] = None |
|
1308 | result[t1._subpath(fn)] = None | |
1309 |
|
1309 | |||
1310 | for fn, n2 in t2._files.items(): |
|
1310 | for fn, n2 in t2._files.items(): | |
1311 | if fn not in t1._files: |
|
1311 | if fn not in t1._files: | |
1312 | fl2 = t2._flags.get(fn, b'') |
|
1312 | fl2 = t2._flags.get(fn, b'') | |
1313 | result[t2._subpath(fn)] = ((None, b''), (n2, fl2)) |
|
1313 | result[t2._subpath(fn)] = ((None, b''), (n2, fl2)) | |
1314 |
|
1314 | |||
1315 | stackls = [] |
|
1315 | stackls = [] | |
1316 | _iterativediff(self, m2, stackls) |
|
1316 | _iterativediff(self, m2, stackls) | |
1317 | while stackls: |
|
1317 | while stackls: | |
1318 | t1, t2 = stackls.pop() |
|
1318 | t1, t2 = stackls.pop() | |
1319 | # stackls is populated in the function call |
|
1319 | # stackls is populated in the function call | |
1320 | _iterativediff(t1, t2, stackls) |
|
1320 | _iterativediff(t1, t2, stackls) | |
1321 | return result |
|
1321 | return result | |
1322 |
|
1322 | |||
1323 | def unmodifiedsince(self, m2): |
|
1323 | def unmodifiedsince(self, m2): | |
1324 | return not self._dirty and not m2._dirty and self._node == m2._node |
|
1324 | return not self._dirty and not m2._dirty and self._node == m2._node | |
1325 |
|
1325 | |||
1326 | def parse(self, text, readsubtree): |
|
1326 | def parse(self, text, readsubtree): | |
1327 | selflazy = self._lazydirs |
|
1327 | selflazy = self._lazydirs | |
1328 | for f, n, fl in _parse(self._nodelen, text): |
|
1328 | for f, n, fl in _parse(self._nodelen, text): | |
1329 | if fl == b't': |
|
1329 | if fl == b't': | |
1330 | f = f + b'/' |
|
1330 | f = f + b'/' | |
1331 | # False below means "doesn't need to be copied" and can use the |
|
1331 | # False below means "doesn't need to be copied" and can use the | |
1332 | # cached value from readsubtree directly. |
|
1332 | # cached value from readsubtree directly. | |
1333 | selflazy[f] = (n, readsubtree, False) |
|
1333 | selflazy[f] = (n, readsubtree, False) | |
1334 | elif b'/' in f: |
|
1334 | elif b'/' in f: | |
1335 | # This is a flat manifest, so use __setitem__ and setflag rather |
|
1335 | # This is a flat manifest, so use __setitem__ and setflag rather | |
1336 | # than assigning directly to _files and _flags, so we can |
|
1336 | # than assigning directly to _files and _flags, so we can | |
1337 | # assign a path in a subdirectory, and to mark dirty (compared |
|
1337 | # assign a path in a subdirectory, and to mark dirty (compared | |
1338 | # to nullid). |
|
1338 | # to nullid). | |
1339 | self[f] = n |
|
1339 | self[f] = n | |
1340 | if fl: |
|
1340 | if fl: | |
1341 | self.setflag(f, fl) |
|
1341 | self.setflag(f, fl) | |
1342 | else: |
|
1342 | else: | |
1343 | # Assigning to _files and _flags avoids marking as dirty, |
|
1343 | # Assigning to _files and _flags avoids marking as dirty, | |
1344 | # and should be a little faster. |
|
1344 | # and should be a little faster. | |
1345 | self._files[f] = n |
|
1345 | self._files[f] = n | |
1346 | if fl: |
|
1346 | if fl: | |
1347 | self._flags[f] = fl |
|
1347 | self._flags[f] = fl | |
1348 |
|
1348 | |||
1349 | def text(self): |
|
1349 | def text(self): | |
1350 | """Get the full data of this manifest as a bytestring.""" |
|
1350 | """Get the full data of this manifest as a bytestring.""" | |
1351 | self._load() |
|
1351 | self._load() | |
1352 | return _text(self.iterentries()) |
|
1352 | return _text(self.iterentries()) | |
1353 |
|
1353 | |||
1354 | def dirtext(self): |
|
1354 | def dirtext(self): | |
1355 | """Get the full data of this directory as a bytestring. Make sure that |
|
1355 | """Get the full data of this directory as a bytestring. Make sure that | |
1356 | any submanifests have been written first, so their nodeids are correct. |
|
1356 | any submanifests have been written first, so their nodeids are correct. | |
1357 | """ |
|
1357 | """ | |
1358 | self._load() |
|
1358 | self._load() | |
1359 | flags = self.flags |
|
1359 | flags = self.flags | |
1360 | lazydirs = [(d[:-1], v[0], b't') for d, v in self._lazydirs.items()] |
|
1360 | lazydirs = [(d[:-1], v[0], b't') for d, v in self._lazydirs.items()] | |
1361 | dirs = [(d[:-1], self._dirs[d]._node, b't') for d in self._dirs] |
|
1361 | dirs = [(d[:-1], self._dirs[d]._node, b't') for d in self._dirs] | |
1362 | files = [(f, self._files[f], flags(f)) for f in self._files] |
|
1362 | files = [(f, self._files[f], flags(f)) for f in self._files] | |
1363 | return _text(sorted(dirs + files + lazydirs)) |
|
1363 | return _text(sorted(dirs + files + lazydirs)) | |
1364 |
|
1364 | |||
1365 | def read(self, gettext, readsubtree): |
|
1365 | def read(self, gettext, readsubtree): | |
1366 | def _load_for_read(s): |
|
1366 | def _load_for_read(s): | |
1367 | s.parse(gettext(), readsubtree) |
|
1367 | s.parse(gettext(), readsubtree) | |
1368 | s._dirty = False |
|
1368 | s._dirty = False | |
1369 |
|
1369 | |||
1370 | self._loadfunc = _load_for_read |
|
1370 | self._loadfunc = _load_for_read | |
1371 |
|
1371 | |||
1372 | def writesubtrees(self, m1, m2, writesubtree, match): |
|
1372 | def writesubtrees(self, m1, m2, writesubtree, match): | |
1373 | self._load() # for consistency; should never have any effect here |
|
1373 | self._load() # for consistency; should never have any effect here | |
1374 | m1._load() |
|
1374 | m1._load() | |
1375 | m2._load() |
|
1375 | m2._load() | |
1376 | emptytree = treemanifest(self.nodeconstants) |
|
1376 | emptytree = treemanifest(self.nodeconstants) | |
1377 |
|
1377 | |||
1378 | def getnode(m, d): |
|
1378 | def getnode(m, d): | |
1379 | ld = m._lazydirs.get(d) |
|
1379 | ld = m._lazydirs.get(d) | |
1380 | if ld: |
|
1380 | if ld: | |
1381 | return ld[0] |
|
1381 | return ld[0] | |
1382 | return m._dirs.get(d, emptytree)._node |
|
1382 | return m._dirs.get(d, emptytree)._node | |
1383 |
|
1383 | |||
1384 | # let's skip investigating things that `match` says we do not need. |
|
1384 | # let's skip investigating things that `match` says we do not need. | |
1385 | visit = match.visitchildrenset(self._dir[:-1]) |
|
1385 | visit = match.visitchildrenset(self._dir[:-1]) | |
1386 | visit = self._loadchildrensetlazy(visit) |
|
1386 | visit = self._loadchildrensetlazy(visit) | |
1387 | if visit == b'this' or visit == b'all': |
|
1387 | if visit == b'this' or visit == b'all': | |
1388 | visit = None |
|
1388 | visit = None | |
1389 | for d, subm in self._dirs.items(): |
|
1389 | for d, subm in self._dirs.items(): | |
1390 | if visit and d[:-1] not in visit: |
|
1390 | if visit and d[:-1] not in visit: | |
1391 | continue |
|
1391 | continue | |
1392 | subp1 = getnode(m1, d) |
|
1392 | subp1 = getnode(m1, d) | |
1393 | subp2 = getnode(m2, d) |
|
1393 | subp2 = getnode(m2, d) | |
1394 | if subp1 == self.nodeconstants.nullid: |
|
1394 | if subp1 == self.nodeconstants.nullid: | |
1395 | subp1, subp2 = subp2, subp1 |
|
1395 | subp1, subp2 = subp2, subp1 | |
1396 | writesubtree(subm, subp1, subp2, match) |
|
1396 | writesubtree(subm, subp1, subp2, match) | |
1397 |
|
1397 | |||
1398 | def walksubtrees(self, matcher=None): |
|
1398 | def walksubtrees(self, matcher=None): | |
1399 | """Returns an iterator of the subtrees of this manifest, including this |
|
1399 | """Returns an iterator of the subtrees of this manifest, including this | |
1400 | manifest itself. |
|
1400 | manifest itself. | |
1401 |
|
1401 | |||
1402 | If `matcher` is provided, it only returns subtrees that match. |
|
1402 | If `matcher` is provided, it only returns subtrees that match. | |
1403 | """ |
|
1403 | """ | |
1404 | if matcher and not matcher.visitdir(self._dir[:-1]): |
|
1404 | if matcher and not matcher.visitdir(self._dir[:-1]): | |
1405 | return |
|
1405 | return | |
1406 | if not matcher or matcher(self._dir[:-1]): |
|
1406 | if not matcher or matcher(self._dir[:-1]): | |
1407 | yield self |
|
1407 | yield self | |
1408 |
|
1408 | |||
1409 | self._load() |
|
1409 | self._load() | |
1410 | # OPT: use visitchildrenset to avoid loading everything. |
|
1410 | # OPT: use visitchildrenset to avoid loading everything. | |
1411 | self._loadalllazy() |
|
1411 | self._loadalllazy() | |
1412 | for d, subm in self._dirs.items(): |
|
1412 | for d, subm in self._dirs.items(): | |
1413 | for subtree in subm.walksubtrees(matcher=matcher): |
|
1413 | for subtree in subm.walksubtrees(matcher=matcher): | |
1414 | yield subtree |
|
1414 | yield subtree | |
1415 |
|
1415 | |||
1416 |
|
1416 | |||
1417 | class manifestfulltextcache(util.lrucachedict): |
|
1417 | class manifestfulltextcache(util.lrucachedict): | |
1418 | """File-backed LRU cache for the manifest cache |
|
1418 | """File-backed LRU cache for the manifest cache | |
1419 |
|
1419 | |||
1420 | File consists of entries, up to EOF: |
|
1420 | File consists of entries, up to EOF: | |
1421 |
|
1421 | |||
1422 | - 20 bytes node, 4 bytes length, <length> manifest data |
|
1422 | - 20 bytes node, 4 bytes length, <length> manifest data | |
1423 |
|
1423 | |||
1424 | These are written in reverse cache order (oldest to newest). |
|
1424 | These are written in reverse cache order (oldest to newest). | |
1425 |
|
1425 | |||
1426 | """ |
|
1426 | """ | |
1427 |
|
1427 | |||
1428 | _file = b'manifestfulltextcache' |
|
1428 | _file = b'manifestfulltextcache' | |
1429 |
|
1429 | |||
1430 | def __init__(self, max): |
|
1430 | def __init__(self, max): | |
1431 | super(manifestfulltextcache, self).__init__(max) |
|
1431 | super(manifestfulltextcache, self).__init__(max) | |
1432 | self._dirty = False |
|
1432 | self._dirty = False | |
1433 | self._read = False |
|
1433 | self._read = False | |
1434 | self._opener = None |
|
1434 | self._opener = None | |
1435 |
|
1435 | |||
1436 | def read(self): |
|
1436 | def read(self): | |
1437 | if self._read or self._opener is None: |
|
1437 | if self._read or self._opener is None: | |
1438 | return |
|
1438 | return | |
1439 |
|
1439 | |||
1440 | try: |
|
1440 | try: | |
1441 | with self._opener(self._file) as fp: |
|
1441 | with self._opener(self._file) as fp: | |
1442 | set = super(manifestfulltextcache, self).__setitem__ |
|
1442 | set = super(manifestfulltextcache, self).__setitem__ | |
1443 | # ignore trailing data, this is a cache, corruption is skipped |
|
1443 | # ignore trailing data, this is a cache, corruption is skipped | |
1444 | while True: |
|
1444 | while True: | |
1445 | # TODO do we need to do work here for sha1 portability? |
|
1445 | # TODO do we need to do work here for sha1 portability? | |
1446 | node = fp.read(20) |
|
1446 | node = fp.read(20) | |
1447 | if len(node) < 20: |
|
1447 | if len(node) < 20: | |
1448 | break |
|
1448 | break | |
1449 | try: |
|
1449 | try: | |
1450 | size = struct.unpack(b'>L', fp.read(4))[0] |
|
1450 | size = struct.unpack(b'>L', fp.read(4))[0] | |
1451 | except struct.error: |
|
1451 | except struct.error: | |
1452 | break |
|
1452 | break | |
1453 | value = bytearray(fp.read(size)) |
|
1453 | value = bytearray(fp.read(size)) | |
1454 | if len(value) != size: |
|
1454 | if len(value) != size: | |
1455 | break |
|
1455 | break | |
1456 | set(node, value) |
|
1456 | set(node, value) | |
1457 | except IOError: |
|
1457 | except IOError: | |
1458 | # the file is allowed to be missing |
|
1458 | # the file is allowed to be missing | |
1459 | pass |
|
1459 | pass | |
1460 |
|
1460 | |||
1461 | self._read = True |
|
1461 | self._read = True | |
1462 | self._dirty = False |
|
1462 | self._dirty = False | |
1463 |
|
1463 | |||
1464 | def write(self): |
|
1464 | def write(self): | |
1465 | if not self._dirty or self._opener is None: |
|
1465 | if not self._dirty or self._opener is None: | |
1466 | return |
|
1466 | return | |
1467 | # rotate backwards to the first used node |
|
1467 | # rotate backwards to the first used node | |
1468 | try: |
|
1468 | try: | |
1469 | with self._opener( |
|
1469 | with self._opener( | |
1470 | self._file, b'w', atomictemp=True, checkambig=True |
|
1470 | self._file, b'w', atomictemp=True, checkambig=True | |
1471 | ) as fp: |
|
1471 | ) as fp: | |
1472 | node = self._head.prev |
|
1472 | node = self._head.prev | |
1473 | while True: |
|
1473 | while True: | |
1474 | if node.key in self._cache: |
|
1474 | if node.key in self._cache: | |
1475 | fp.write(node.key) |
|
1475 | fp.write(node.key) | |
1476 | fp.write(struct.pack(b'>L', len(node.value))) |
|
1476 | fp.write(struct.pack(b'>L', len(node.value))) | |
1477 | fp.write(node.value) |
|
1477 | fp.write(node.value) | |
1478 | if node is self._head: |
|
1478 | if node is self._head: | |
1479 | break |
|
1479 | break | |
1480 | node = node.prev |
|
1480 | node = node.prev | |
1481 | except IOError: |
|
1481 | except IOError: | |
1482 | # We could not write the cache (eg: permission error) |
|
1482 | # We could not write the cache (eg: permission error) | |
1483 | # the content can be missing. |
|
1483 | # the content can be missing. | |
1484 | # |
|
1484 | # | |
1485 | # We could try harder and see if we could recreate a wcache |
|
1485 | # We could try harder and see if we could recreate a wcache | |
1486 | # directory were we coudl write too. |
|
1486 | # directory were we coudl write too. | |
1487 | # |
|
1487 | # | |
1488 | # XXX the error pass silently, having some way to issue an error |
|
1488 | # XXX the error pass silently, having some way to issue an error | |
1489 | # log `ui.log` would be nice. |
|
1489 | # log `ui.log` would be nice. | |
1490 | pass |
|
1490 | pass | |
1491 |
|
1491 | |||
1492 | def __len__(self): |
|
1492 | def __len__(self): | |
1493 | if not self._read: |
|
1493 | if not self._read: | |
1494 | self.read() |
|
1494 | self.read() | |
1495 | return super(manifestfulltextcache, self).__len__() |
|
1495 | return super(manifestfulltextcache, self).__len__() | |
1496 |
|
1496 | |||
1497 | def __contains__(self, k): |
|
1497 | def __contains__(self, k): | |
1498 | if not self._read: |
|
1498 | if not self._read: | |
1499 | self.read() |
|
1499 | self.read() | |
1500 | return super(manifestfulltextcache, self).__contains__(k) |
|
1500 | return super(manifestfulltextcache, self).__contains__(k) | |
1501 |
|
1501 | |||
1502 | def __iter__(self): |
|
1502 | def __iter__(self): | |
1503 | if not self._read: |
|
1503 | if not self._read: | |
1504 | self.read() |
|
1504 | self.read() | |
1505 | return super(manifestfulltextcache, self).__iter__() |
|
1505 | return super(manifestfulltextcache, self).__iter__() | |
1506 |
|
1506 | |||
1507 | def __getitem__(self, k): |
|
1507 | def __getitem__(self, k): | |
1508 | if not self._read: |
|
1508 | if not self._read: | |
1509 | self.read() |
|
1509 | self.read() | |
1510 | # the cache lru order can change on read |
|
1510 | # the cache lru order can change on read | |
1511 | setdirty = self._cache.get(k) is not self._head |
|
1511 | setdirty = self._cache.get(k) is not self._head | |
1512 | value = super(manifestfulltextcache, self).__getitem__(k) |
|
1512 | value = super(manifestfulltextcache, self).__getitem__(k) | |
1513 | if setdirty: |
|
1513 | if setdirty: | |
1514 | self._dirty = True |
|
1514 | self._dirty = True | |
1515 | return value |
|
1515 | return value | |
1516 |
|
1516 | |||
1517 | def __setitem__(self, k, v): |
|
1517 | def __setitem__(self, k, v): | |
1518 | if not self._read: |
|
1518 | if not self._read: | |
1519 | self.read() |
|
1519 | self.read() | |
1520 | super(manifestfulltextcache, self).__setitem__(k, v) |
|
1520 | super(manifestfulltextcache, self).__setitem__(k, v) | |
1521 | self._dirty = True |
|
1521 | self._dirty = True | |
1522 |
|
1522 | |||
1523 | def __delitem__(self, k): |
|
1523 | def __delitem__(self, k): | |
1524 | if not self._read: |
|
1524 | if not self._read: | |
1525 | self.read() |
|
1525 | self.read() | |
1526 | super(manifestfulltextcache, self).__delitem__(k) |
|
1526 | super(manifestfulltextcache, self).__delitem__(k) | |
1527 | self._dirty = True |
|
1527 | self._dirty = True | |
1528 |
|
1528 | |||
1529 | def get(self, k, default=None): |
|
1529 | def get(self, k, default=None): | |
1530 | if not self._read: |
|
1530 | if not self._read: | |
1531 | self.read() |
|
1531 | self.read() | |
1532 | return super(manifestfulltextcache, self).get(k, default=default) |
|
1532 | return super(manifestfulltextcache, self).get(k, default=default) | |
1533 |
|
1533 | |||
1534 | def clear(self, clear_persisted_data=False): |
|
1534 | def clear(self, clear_persisted_data=False): | |
1535 | super(manifestfulltextcache, self).clear() |
|
1535 | super(manifestfulltextcache, self).clear() | |
1536 | if clear_persisted_data: |
|
1536 | if clear_persisted_data: | |
1537 | self._dirty = True |
|
1537 | self._dirty = True | |
1538 | self.write() |
|
1538 | self.write() | |
1539 | self._read = False |
|
1539 | self._read = False | |
1540 |
|
1540 | |||
1541 |
|
1541 | |||
1542 | # and upper bound of what we expect from compression |
|
1542 | # and upper bound of what we expect from compression | |
1543 | # (real live value seems to be "3") |
|
1543 | # (real live value seems to be "3") | |
1544 | MAXCOMPRESSION = 3 |
|
1544 | MAXCOMPRESSION = 3 | |
1545 |
|
1545 | |||
1546 |
|
1546 | |||
1547 | class FastdeltaUnavailable(Exception): |
|
1547 | class FastdeltaUnavailable(Exception): | |
1548 | """Exception raised when fastdelta isn't usable on a manifest.""" |
|
1548 | """Exception raised when fastdelta isn't usable on a manifest.""" | |
1549 |
|
1549 | |||
1550 |
|
1550 | |||
1551 | @interfaceutil.implementer(repository.imanifeststorage) |
|
1551 | @interfaceutil.implementer(repository.imanifeststorage) | |
1552 | class manifestrevlog: |
|
1552 | class manifestrevlog: | |
1553 | """A revlog that stores manifest texts. This is responsible for caching the |
|
1553 | """A revlog that stores manifest texts. This is responsible for caching the | |
1554 | full-text manifest contents. |
|
1554 | full-text manifest contents. | |
1555 | """ |
|
1555 | """ | |
1556 |
|
1556 | |||
1557 | def __init__( |
|
1557 | def __init__( | |
1558 | self, |
|
1558 | self, | |
1559 | nodeconstants, |
|
1559 | nodeconstants, | |
1560 | opener, |
|
1560 | opener, | |
1561 | tree=b'', |
|
1561 | tree=b'', | |
1562 | dirlogcache=None, |
|
1562 | dirlogcache=None, | |
1563 | treemanifest=False, |
|
1563 | treemanifest=False, | |
1564 | ): |
|
1564 | ): | |
1565 | """Constructs a new manifest revlog |
|
1565 | """Constructs a new manifest revlog | |
1566 |
|
1566 | |||
1567 | `indexfile` - used by extensions to have two manifests at once, like |
|
1567 | `indexfile` - used by extensions to have two manifests at once, like | |
1568 | when transitioning between flatmanifeset and treemanifests. |
|
1568 | when transitioning between flatmanifeset and treemanifests. | |
1569 |
|
1569 | |||
1570 | `treemanifest` - used to indicate this is a tree manifest revlog. Opener |
|
1570 | `treemanifest` - used to indicate this is a tree manifest revlog. Opener | |
1571 | options can also be used to make this a tree manifest revlog. The opener |
|
1571 | options can also be used to make this a tree manifest revlog. The opener | |
1572 | option takes precedence, so if it is set to True, we ignore whatever |
|
1572 | option takes precedence, so if it is set to True, we ignore whatever | |
1573 | value is passed in to the constructor. |
|
1573 | value is passed in to the constructor. | |
1574 | """ |
|
1574 | """ | |
1575 | self.nodeconstants = nodeconstants |
|
1575 | self.nodeconstants = nodeconstants | |
1576 | # During normal operations, we expect to deal with not more than four |
|
1576 | # During normal operations, we expect to deal with not more than four | |
1577 | # revs at a time (such as during commit --amend). When rebasing large |
|
1577 | # revs at a time (such as during commit --amend). When rebasing large | |
1578 | # stacks of commits, the number can go up, hence the config knob below. |
|
1578 | # stacks of commits, the number can go up, hence the config knob below. | |
1579 | cachesize = 4 |
|
1579 | cachesize = 4 | |
1580 | optiontreemanifest = False |
|
1580 | optiontreemanifest = False | |
1581 | opts = getattr(opener, 'options', None) |
|
1581 | opts = getattr(opener, 'options', None) | |
1582 | if opts is not None: |
|
1582 | if opts is not None: | |
1583 | cachesize = opts.get(b'manifestcachesize', cachesize) |
|
1583 | cachesize = opts.get(b'manifestcachesize', cachesize) | |
1584 | optiontreemanifest = opts.get(b'treemanifest', False) |
|
1584 | optiontreemanifest = opts.get(b'treemanifest', False) | |
1585 |
|
1585 | |||
1586 | self._treeondisk = optiontreemanifest or treemanifest |
|
1586 | self._treeondisk = optiontreemanifest or treemanifest | |
1587 |
|
1587 | |||
1588 | self._fulltextcache = manifestfulltextcache(cachesize) |
|
1588 | self._fulltextcache = manifestfulltextcache(cachesize) | |
1589 |
|
1589 | |||
1590 | if tree: |
|
1590 | if tree: | |
1591 | assert self._treeondisk, (tree, b'opts is %r' % opts) |
|
1591 | assert self._treeondisk, (tree, b'opts is %r' % opts) | |
1592 |
|
1592 | |||
1593 | radix = b'00manifest' |
|
1593 | radix = b'00manifest' | |
1594 | if tree: |
|
1594 | if tree: | |
1595 | radix = b"meta/" + tree + radix |
|
1595 | radix = b"meta/" + tree + radix | |
1596 |
|
1596 | |||
1597 | self.tree = tree |
|
1597 | self.tree = tree | |
1598 |
|
1598 | |||
1599 | # The dirlogcache is kept on the root manifest log |
|
1599 | # The dirlogcache is kept on the root manifest log | |
1600 | if tree: |
|
1600 | if tree: | |
1601 | self._dirlogcache = dirlogcache |
|
1601 | self._dirlogcache = dirlogcache | |
1602 | else: |
|
1602 | else: | |
1603 | self._dirlogcache = {b'': self} |
|
1603 | self._dirlogcache = {b'': self} | |
1604 |
|
1604 | |||
1605 | self._revlog = revlog.revlog( |
|
1605 | self._revlog = revlog.revlog( | |
1606 | opener, |
|
1606 | opener, | |
1607 | target=(revlog_constants.KIND_MANIFESTLOG, self.tree), |
|
1607 | target=(revlog_constants.KIND_MANIFESTLOG, self.tree), | |
1608 | radix=radix, |
|
1608 | radix=radix, | |
1609 | # only root indexfile is cached |
|
1609 | # only root indexfile is cached | |
1610 | checkambig=not bool(tree), |
|
1610 | checkambig=not bool(tree), | |
1611 | mmaplargeindex=True, |
|
1611 | mmaplargeindex=True, | |
1612 | upperboundcomp=MAXCOMPRESSION, |
|
1612 | upperboundcomp=MAXCOMPRESSION, | |
1613 | persistentnodemap=opener.options.get(b'persistent-nodemap', False), |
|
1613 | persistentnodemap=opener.options.get(b'persistent-nodemap', False), | |
1614 | ) |
|
1614 | ) | |
1615 |
|
1615 | |||
1616 | self.index = self._revlog.index |
|
1616 | self.index = self._revlog.index | |
1617 | self._generaldelta = self._revlog._generaldelta |
|
1617 | self._generaldelta = self._revlog._generaldelta | |
1618 |
|
1618 | |||
1619 | def get_revlog(self): |
|
1619 | def get_revlog(self): | |
1620 | """return an actual revlog instance if any |
|
1620 | """return an actual revlog instance if any | |
1621 |
|
1621 | |||
1622 | This exist because a lot of code leverage the fact the underlying |
|
1622 | This exist because a lot of code leverage the fact the underlying | |
1623 | storage is a revlog for optimization, so giving simple way to access |
|
1623 | storage is a revlog for optimization, so giving simple way to access | |
1624 | the revlog instance helps such code. |
|
1624 | the revlog instance helps such code. | |
1625 | """ |
|
1625 | """ | |
1626 | return self._revlog |
|
1626 | return self._revlog | |
1627 |
|
1627 | |||
1628 | def _setupmanifestcachehooks(self, repo): |
|
1628 | def _setupmanifestcachehooks(self, repo): | |
1629 | """Persist the manifestfulltextcache on lock release""" |
|
1629 | """Persist the manifestfulltextcache on lock release""" | |
1630 | if not hasattr(repo, '_wlockref'): |
|
1630 | if not hasattr(repo, '_wlockref'): | |
1631 | return |
|
1631 | return | |
1632 |
|
1632 | |||
1633 | self._fulltextcache._opener = repo.wcachevfs |
|
1633 | self._fulltextcache._opener = repo.wcachevfs | |
1634 | if repo._currentlock(repo._wlockref) is None: |
|
1634 | if repo._currentlock(repo._wlockref) is None: | |
1635 | return |
|
1635 | return | |
1636 |
|
1636 | |||
1637 | reporef = weakref.ref(repo) |
|
1637 | reporef = weakref.ref(repo) | |
1638 | manifestrevlogref = weakref.ref(self) |
|
1638 | manifestrevlogref = weakref.ref(self) | |
1639 |
|
1639 | |||
1640 | def persistmanifestcache(success): |
|
1640 | def persistmanifestcache(success): | |
1641 | # Repo is in an unknown state, do not persist. |
|
1641 | # Repo is in an unknown state, do not persist. | |
1642 | if not success: |
|
1642 | if not success: | |
1643 | return |
|
1643 | return | |
1644 |
|
1644 | |||
1645 | repo = reporef() |
|
1645 | repo = reporef() | |
1646 | self = manifestrevlogref() |
|
1646 | self = manifestrevlogref() | |
1647 | if repo is None or self is None: |
|
1647 | if repo is None or self is None: | |
1648 | return |
|
1648 | return | |
1649 | if repo.manifestlog.getstorage(b'') is not self: |
|
1649 | if repo.manifestlog.getstorage(b'') is not self: | |
1650 | # there's a different manifest in play now, abort |
|
1650 | # there's a different manifest in play now, abort | |
1651 | return |
|
1651 | return | |
1652 | self._fulltextcache.write() |
|
1652 | self._fulltextcache.write() | |
1653 |
|
1653 | |||
1654 | repo._afterlock(persistmanifestcache) |
|
1654 | repo._afterlock(persistmanifestcache) | |
1655 |
|
1655 | |||
1656 | @property |
|
1656 | @property | |
1657 | def fulltextcache(self): |
|
1657 | def fulltextcache(self): | |
1658 | return self._fulltextcache |
|
1658 | return self._fulltextcache | |
1659 |
|
1659 | |||
1660 | def clearcaches(self, clear_persisted_data=False): |
|
1660 | def clearcaches(self, clear_persisted_data=False): | |
1661 | self._revlog.clearcaches() |
|
1661 | self._revlog.clearcaches() | |
1662 | self._fulltextcache.clear(clear_persisted_data=clear_persisted_data) |
|
1662 | self._fulltextcache.clear(clear_persisted_data=clear_persisted_data) | |
1663 | self._dirlogcache = {self.tree: self} |
|
1663 | self._dirlogcache = {self.tree: self} | |
1664 |
|
1664 | |||
1665 | def dirlog(self, d): |
|
1665 | def dirlog(self, d): | |
1666 | if d: |
|
1666 | if d: | |
1667 | assert self._treeondisk |
|
1667 | assert self._treeondisk | |
1668 | if d not in self._dirlogcache: |
|
1668 | if d not in self._dirlogcache: | |
1669 | mfrevlog = manifestrevlog( |
|
1669 | mfrevlog = manifestrevlog( | |
1670 | self.nodeconstants, |
|
1670 | self.nodeconstants, | |
1671 | self.opener, |
|
1671 | self.opener, | |
1672 | d, |
|
1672 | d, | |
1673 | self._dirlogcache, |
|
1673 | self._dirlogcache, | |
1674 | treemanifest=self._treeondisk, |
|
1674 | treemanifest=self._treeondisk, | |
1675 | ) |
|
1675 | ) | |
1676 | self._dirlogcache[d] = mfrevlog |
|
1676 | self._dirlogcache[d] = mfrevlog | |
1677 | return self._dirlogcache[d] |
|
1677 | return self._dirlogcache[d] | |
1678 |
|
1678 | |||
1679 | def add( |
|
1679 | def add( | |
1680 | self, |
|
1680 | self, | |
1681 | m, |
|
1681 | m, | |
1682 | transaction, |
|
1682 | transaction, | |
1683 | link, |
|
1683 | link, | |
1684 | p1, |
|
1684 | p1, | |
1685 | p2, |
|
1685 | p2, | |
1686 | added, |
|
1686 | added, | |
1687 | removed, |
|
1687 | removed, | |
1688 | readtree=None, |
|
1688 | readtree=None, | |
1689 | match=None, |
|
1689 | match=None, | |
1690 | ): |
|
1690 | ): | |
1691 | """add some manifest entry in to the manifest log |
|
1691 | """add some manifest entry in to the manifest log | |
1692 |
|
1692 | |||
1693 | input: |
|
1693 | input: | |
1694 |
|
1694 | |||
1695 | m: the manifest dict we want to store |
|
1695 | m: the manifest dict we want to store | |
1696 | transaction: the open transaction |
|
1696 | transaction: the open transaction | |
1697 | p1: manifest-node of p1 |
|
1697 | p1: manifest-node of p1 | |
1698 | p2: manifest-node of p2 |
|
1698 | p2: manifest-node of p2 | |
1699 | added: file added/changed compared to parent |
|
1699 | added: file added/changed compared to parent | |
1700 | removed: file removed compared to parent |
|
1700 | removed: file removed compared to parent | |
1701 |
|
1701 | |||
1702 | tree manifest input: |
|
1702 | tree manifest input: | |
1703 |
|
1703 | |||
1704 | readtree: a function to read a subtree |
|
1704 | readtree: a function to read a subtree | |
1705 | match: a filematcher for the subpart of the tree manifest |
|
1705 | match: a filematcher for the subpart of the tree manifest | |
1706 | """ |
|
1706 | """ | |
1707 | try: |
|
1707 | try: | |
1708 | if p1 not in self.fulltextcache: |
|
1708 | if p1 not in self.fulltextcache: | |
1709 | raise FastdeltaUnavailable() |
|
1709 | raise FastdeltaUnavailable() | |
1710 | # If our first parent is in the manifest cache, we can |
|
1710 | # If our first parent is in the manifest cache, we can | |
1711 | # compute a delta here using properties we know about the |
|
1711 | # compute a delta here using properties we know about the | |
1712 | # manifest up-front, which may save time later for the |
|
1712 | # manifest up-front, which may save time later for the | |
1713 | # revlog layer. |
|
1713 | # revlog layer. | |
1714 |
|
1714 | |||
1715 | _checkforbidden(added) |
|
1715 | _checkforbidden(added) | |
1716 | # combine the changed lists into one sorted iterator |
|
1716 | # combine the changed lists into one sorted iterator | |
1717 | work = heapq.merge( |
|
1717 | work = heapq.merge( | |
1718 | [(x, False) for x in sorted(added)], |
|
1718 | [(x, False) for x in sorted(added)], | |
1719 | [(x, True) for x in sorted(removed)], |
|
1719 | [(x, True) for x in sorted(removed)], | |
1720 | ) |
|
1720 | ) | |
1721 |
|
1721 | |||
1722 | arraytext, deltatext = m.fastdelta(self.fulltextcache[p1], work) |
|
1722 | arraytext, deltatext = m.fastdelta(self.fulltextcache[p1], work) | |
1723 | cachedelta = self._revlog.rev(p1), deltatext |
|
1723 | cachedelta = self._revlog.rev(p1), deltatext | |
1724 | text = util.buffer(arraytext) |
|
1724 | text = util.buffer(arraytext) | |
1725 | rev = self._revlog.addrevision( |
|
1725 | rev = self._revlog.addrevision( | |
1726 | text, transaction, link, p1, p2, cachedelta |
|
1726 | text, transaction, link, p1, p2, cachedelta | |
1727 | ) |
|
1727 | ) | |
1728 | n = self._revlog.node(rev) |
|
1728 | n = self._revlog.node(rev) | |
1729 | except FastdeltaUnavailable: |
|
1729 | except FastdeltaUnavailable: | |
1730 | # The first parent manifest isn't already loaded or the |
|
1730 | # The first parent manifest isn't already loaded or the | |
1731 | # manifest implementation doesn't support fastdelta, so |
|
1731 | # manifest implementation doesn't support fastdelta, so | |
1732 | # we'll just encode a fulltext of the manifest and pass |
|
1732 | # we'll just encode a fulltext of the manifest and pass | |
1733 | # that through to the revlog layer, and let it handle the |
|
1733 | # that through to the revlog layer, and let it handle the | |
1734 | # delta process. |
|
1734 | # delta process. | |
1735 | if self._treeondisk: |
|
1735 | if self._treeondisk: | |
1736 | assert readtree, b"readtree must be set for treemanifest writes" |
|
1736 | assert readtree, b"readtree must be set for treemanifest writes" | |
1737 | assert match, b"match must be specified for treemanifest writes" |
|
1737 | assert match, b"match must be specified for treemanifest writes" | |
1738 | m1 = readtree(self.tree, p1) |
|
1738 | m1 = readtree(self.tree, p1) | |
1739 | m2 = readtree(self.tree, p2) |
|
1739 | m2 = readtree(self.tree, p2) | |
1740 | n = self._addtree( |
|
1740 | n = self._addtree( | |
1741 | m, transaction, link, m1, m2, readtree, match=match |
|
1741 | m, transaction, link, m1, m2, readtree, match=match | |
1742 | ) |
|
1742 | ) | |
1743 | arraytext = None |
|
1743 | arraytext = None | |
1744 | else: |
|
1744 | else: | |
1745 | text = m.text() |
|
1745 | text = m.text() | |
1746 | rev = self._revlog.addrevision(text, transaction, link, p1, p2) |
|
1746 | rev = self._revlog.addrevision(text, transaction, link, p1, p2) | |
1747 | n = self._revlog.node(rev) |
|
1747 | n = self._revlog.node(rev) | |
1748 | arraytext = bytearray(text) |
|
1748 | arraytext = bytearray(text) | |
1749 |
|
1749 | |||
1750 | if arraytext is not None: |
|
1750 | if arraytext is not None: | |
1751 | self.fulltextcache[n] = arraytext |
|
1751 | self.fulltextcache[n] = arraytext | |
1752 |
|
1752 | |||
1753 | return n |
|
1753 | return n | |
1754 |
|
1754 | |||
1755 | def _addtree(self, m, transaction, link, m1, m2, readtree, match): |
|
1755 | def _addtree(self, m, transaction, link, m1, m2, readtree, match): | |
1756 | # If the manifest is unchanged compared to one parent, |
|
1756 | # If the manifest is unchanged compared to one parent, | |
1757 | # don't write a new revision |
|
1757 | # don't write a new revision | |
1758 | if self.tree != b'' and ( |
|
1758 | if self.tree != b'' and ( | |
1759 | m.unmodifiedsince(m1) or m.unmodifiedsince(m2) |
|
1759 | m.unmodifiedsince(m1) or m.unmodifiedsince(m2) | |
1760 | ): |
|
1760 | ): | |
1761 | return m.node() |
|
1761 | return m.node() | |
1762 |
|
1762 | |||
1763 | def writesubtree(subm, subp1, subp2, match): |
|
1763 | def writesubtree(subm, subp1, subp2, match): | |
1764 | sublog = self.dirlog(subm.dir()) |
|
1764 | sublog = self.dirlog(subm.dir()) | |
1765 | sublog.add( |
|
1765 | sublog.add( | |
1766 | subm, |
|
1766 | subm, | |
1767 | transaction, |
|
1767 | transaction, | |
1768 | link, |
|
1768 | link, | |
1769 | subp1, |
|
1769 | subp1, | |
1770 | subp2, |
|
1770 | subp2, | |
1771 | None, |
|
1771 | None, | |
1772 | None, |
|
1772 | None, | |
1773 | readtree=readtree, |
|
1773 | readtree=readtree, | |
1774 | match=match, |
|
1774 | match=match, | |
1775 | ) |
|
1775 | ) | |
1776 |
|
1776 | |||
1777 | m.writesubtrees(m1, m2, writesubtree, match) |
|
1777 | m.writesubtrees(m1, m2, writesubtree, match) | |
1778 | text = m.dirtext() |
|
1778 | text = m.dirtext() | |
1779 | n = None |
|
1779 | n = None | |
1780 | if self.tree != b'': |
|
1780 | if self.tree != b'': | |
1781 | # Double-check whether contents are unchanged to one parent |
|
1781 | # Double-check whether contents are unchanged to one parent | |
1782 | if text == m1.dirtext(): |
|
1782 | if text == m1.dirtext(): | |
1783 | n = m1.node() |
|
1783 | n = m1.node() | |
1784 | elif text == m2.dirtext(): |
|
1784 | elif text == m2.dirtext(): | |
1785 | n = m2.node() |
|
1785 | n = m2.node() | |
1786 |
|
1786 | |||
1787 | if not n: |
|
1787 | if not n: | |
1788 | rev = self._revlog.addrevision( |
|
1788 | rev = self._revlog.addrevision( | |
1789 | text, transaction, link, m1.node(), m2.node() |
|
1789 | text, transaction, link, m1.node(), m2.node() | |
1790 | ) |
|
1790 | ) | |
1791 | n = self._revlog.node(rev) |
|
1791 | n = self._revlog.node(rev) | |
1792 |
|
1792 | |||
1793 | # Save nodeid so parent manifest can calculate its nodeid |
|
1793 | # Save nodeid so parent manifest can calculate its nodeid | |
1794 | m.setnode(n) |
|
1794 | m.setnode(n) | |
1795 | return n |
|
1795 | return n | |
1796 |
|
1796 | |||
1797 | def __len__(self): |
|
1797 | def __len__(self): | |
1798 | return len(self._revlog) |
|
1798 | return len(self._revlog) | |
1799 |
|
1799 | |||
1800 | def __iter__(self): |
|
1800 | def __iter__(self): | |
1801 | return self._revlog.__iter__() |
|
1801 | return self._revlog.__iter__() | |
1802 |
|
1802 | |||
1803 | def rev(self, node): |
|
1803 | def rev(self, node): | |
1804 | return self._revlog.rev(node) |
|
1804 | return self._revlog.rev(node) | |
1805 |
|
1805 | |||
1806 | def node(self, rev): |
|
1806 | def node(self, rev): | |
1807 | return self._revlog.node(rev) |
|
1807 | return self._revlog.node(rev) | |
1808 |
|
1808 | |||
1809 | def lookup(self, value): |
|
1809 | def lookup(self, value): | |
1810 | return self._revlog.lookup(value) |
|
1810 | return self._revlog.lookup(value) | |
1811 |
|
1811 | |||
1812 | def parentrevs(self, rev): |
|
1812 | def parentrevs(self, rev): | |
1813 | return self._revlog.parentrevs(rev) |
|
1813 | return self._revlog.parentrevs(rev) | |
1814 |
|
1814 | |||
1815 | def parents(self, node): |
|
1815 | def parents(self, node): | |
1816 | return self._revlog.parents(node) |
|
1816 | return self._revlog.parents(node) | |
1817 |
|
1817 | |||
1818 | def linkrev(self, rev): |
|
1818 | def linkrev(self, rev): | |
1819 | return self._revlog.linkrev(rev) |
|
1819 | return self._revlog.linkrev(rev) | |
1820 |
|
1820 | |||
1821 | def checksize(self): |
|
1821 | def checksize(self): | |
1822 | return self._revlog.checksize() |
|
1822 | return self._revlog.checksize() | |
1823 |
|
1823 | |||
1824 | def revision(self, node): |
|
1824 | def revision(self, node): | |
1825 | return self._revlog.revision(node) |
|
1825 | return self._revlog.revision(node) | |
1826 |
|
1826 | |||
1827 |
def rawdata(self, node |
|
1827 | def rawdata(self, node): | |
1828 |
return self._revlog.rawdata(node |
|
1828 | return self._revlog.rawdata(node) | |
1829 |
|
1829 | |||
1830 | def revdiff(self, rev1, rev2): |
|
1830 | def revdiff(self, rev1, rev2): | |
1831 | return self._revlog.revdiff(rev1, rev2) |
|
1831 | return self._revlog.revdiff(rev1, rev2) | |
1832 |
|
1832 | |||
1833 | def cmp(self, node, text): |
|
1833 | def cmp(self, node, text): | |
1834 | return self._revlog.cmp(node, text) |
|
1834 | return self._revlog.cmp(node, text) | |
1835 |
|
1835 | |||
1836 | def deltaparent(self, rev): |
|
1836 | def deltaparent(self, rev): | |
1837 | return self._revlog.deltaparent(rev) |
|
1837 | return self._revlog.deltaparent(rev) | |
1838 |
|
1838 | |||
1839 | def emitrevisions( |
|
1839 | def emitrevisions( | |
1840 | self, |
|
1840 | self, | |
1841 | nodes, |
|
1841 | nodes, | |
1842 | nodesorder=None, |
|
1842 | nodesorder=None, | |
1843 | revisiondata=False, |
|
1843 | revisiondata=False, | |
1844 | assumehaveparentrevisions=False, |
|
1844 | assumehaveparentrevisions=False, | |
1845 | deltamode=repository.CG_DELTAMODE_STD, |
|
1845 | deltamode=repository.CG_DELTAMODE_STD, | |
1846 | sidedata_helpers=None, |
|
1846 | sidedata_helpers=None, | |
1847 | debug_info=None, |
|
1847 | debug_info=None, | |
1848 | ): |
|
1848 | ): | |
1849 | return self._revlog.emitrevisions( |
|
1849 | return self._revlog.emitrevisions( | |
1850 | nodes, |
|
1850 | nodes, | |
1851 | nodesorder=nodesorder, |
|
1851 | nodesorder=nodesorder, | |
1852 | revisiondata=revisiondata, |
|
1852 | revisiondata=revisiondata, | |
1853 | assumehaveparentrevisions=assumehaveparentrevisions, |
|
1853 | assumehaveparentrevisions=assumehaveparentrevisions, | |
1854 | deltamode=deltamode, |
|
1854 | deltamode=deltamode, | |
1855 | sidedata_helpers=sidedata_helpers, |
|
1855 | sidedata_helpers=sidedata_helpers, | |
1856 | debug_info=debug_info, |
|
1856 | debug_info=debug_info, | |
1857 | ) |
|
1857 | ) | |
1858 |
|
1858 | |||
1859 | def addgroup( |
|
1859 | def addgroup( | |
1860 | self, |
|
1860 | self, | |
1861 | deltas, |
|
1861 | deltas, | |
1862 | linkmapper, |
|
1862 | linkmapper, | |
1863 | transaction, |
|
1863 | transaction, | |
1864 | alwayscache=False, |
|
1864 | alwayscache=False, | |
1865 | addrevisioncb=None, |
|
1865 | addrevisioncb=None, | |
1866 | duplicaterevisioncb=None, |
|
1866 | duplicaterevisioncb=None, | |
1867 | debug_info=None, |
|
1867 | debug_info=None, | |
1868 | delta_base_reuse_policy=None, |
|
1868 | delta_base_reuse_policy=None, | |
1869 | ): |
|
1869 | ): | |
1870 | return self._revlog.addgroup( |
|
1870 | return self._revlog.addgroup( | |
1871 | deltas, |
|
1871 | deltas, | |
1872 | linkmapper, |
|
1872 | linkmapper, | |
1873 | transaction, |
|
1873 | transaction, | |
1874 | alwayscache=alwayscache, |
|
1874 | alwayscache=alwayscache, | |
1875 | addrevisioncb=addrevisioncb, |
|
1875 | addrevisioncb=addrevisioncb, | |
1876 | duplicaterevisioncb=duplicaterevisioncb, |
|
1876 | duplicaterevisioncb=duplicaterevisioncb, | |
1877 | debug_info=debug_info, |
|
1877 | debug_info=debug_info, | |
1878 | delta_base_reuse_policy=delta_base_reuse_policy, |
|
1878 | delta_base_reuse_policy=delta_base_reuse_policy, | |
1879 | ) |
|
1879 | ) | |
1880 |
|
1880 | |||
1881 | def rawsize(self, rev): |
|
1881 | def rawsize(self, rev): | |
1882 | return self._revlog.rawsize(rev) |
|
1882 | return self._revlog.rawsize(rev) | |
1883 |
|
1883 | |||
1884 | def getstrippoint(self, minlink): |
|
1884 | def getstrippoint(self, minlink): | |
1885 | return self._revlog.getstrippoint(minlink) |
|
1885 | return self._revlog.getstrippoint(minlink) | |
1886 |
|
1886 | |||
1887 | def strip(self, minlink, transaction): |
|
1887 | def strip(self, minlink, transaction): | |
1888 | return self._revlog.strip(minlink, transaction) |
|
1888 | return self._revlog.strip(minlink, transaction) | |
1889 |
|
1889 | |||
1890 | def files(self): |
|
1890 | def files(self): | |
1891 | return self._revlog.files() |
|
1891 | return self._revlog.files() | |
1892 |
|
1892 | |||
1893 | def clone(self, tr, destrevlog, **kwargs): |
|
1893 | def clone(self, tr, destrevlog, **kwargs): | |
1894 | if not isinstance(destrevlog, manifestrevlog): |
|
1894 | if not isinstance(destrevlog, manifestrevlog): | |
1895 | raise error.ProgrammingError(b'expected manifestrevlog to clone()') |
|
1895 | raise error.ProgrammingError(b'expected manifestrevlog to clone()') | |
1896 |
|
1896 | |||
1897 | return self._revlog.clone(tr, destrevlog._revlog, **kwargs) |
|
1897 | return self._revlog.clone(tr, destrevlog._revlog, **kwargs) | |
1898 |
|
1898 | |||
1899 | def storageinfo( |
|
1899 | def storageinfo( | |
1900 | self, |
|
1900 | self, | |
1901 | exclusivefiles=False, |
|
1901 | exclusivefiles=False, | |
1902 | sharedfiles=False, |
|
1902 | sharedfiles=False, | |
1903 | revisionscount=False, |
|
1903 | revisionscount=False, | |
1904 | trackedsize=False, |
|
1904 | trackedsize=False, | |
1905 | storedsize=False, |
|
1905 | storedsize=False, | |
1906 | ): |
|
1906 | ): | |
1907 | return self._revlog.storageinfo( |
|
1907 | return self._revlog.storageinfo( | |
1908 | exclusivefiles=exclusivefiles, |
|
1908 | exclusivefiles=exclusivefiles, | |
1909 | sharedfiles=sharedfiles, |
|
1909 | sharedfiles=sharedfiles, | |
1910 | revisionscount=revisionscount, |
|
1910 | revisionscount=revisionscount, | |
1911 | trackedsize=trackedsize, |
|
1911 | trackedsize=trackedsize, | |
1912 | storedsize=storedsize, |
|
1912 | storedsize=storedsize, | |
1913 | ) |
|
1913 | ) | |
1914 |
|
1914 | |||
1915 | @property |
|
1915 | @property | |
1916 | def opener(self): |
|
1916 | def opener(self): | |
1917 | return self._revlog.opener |
|
1917 | return self._revlog.opener | |
1918 |
|
1918 | |||
1919 | @opener.setter |
|
1919 | @opener.setter | |
1920 | def opener(self, value): |
|
1920 | def opener(self, value): | |
1921 | self._revlog.opener = value |
|
1921 | self._revlog.opener = value | |
1922 |
|
1922 | |||
1923 |
|
1923 | |||
1924 | @interfaceutil.implementer(repository.imanifestlog) |
|
1924 | @interfaceutil.implementer(repository.imanifestlog) | |
1925 | class manifestlog: |
|
1925 | class manifestlog: | |
1926 | """A collection class representing the collection of manifest snapshots |
|
1926 | """A collection class representing the collection of manifest snapshots | |
1927 | referenced by commits in the repository. |
|
1927 | referenced by commits in the repository. | |
1928 |
|
1928 | |||
1929 | In this situation, 'manifest' refers to the abstract concept of a snapshot |
|
1929 | In this situation, 'manifest' refers to the abstract concept of a snapshot | |
1930 | of the list of files in the given commit. Consumers of the output of this |
|
1930 | of the list of files in the given commit. Consumers of the output of this | |
1931 | class do not care about the implementation details of the actual manifests |
|
1931 | class do not care about the implementation details of the actual manifests | |
1932 | they receive (i.e. tree or flat or lazily loaded, etc).""" |
|
1932 | they receive (i.e. tree or flat or lazily loaded, etc).""" | |
1933 |
|
1933 | |||
1934 | def __init__(self, opener, repo, rootstore, narrowmatch): |
|
1934 | def __init__(self, opener, repo, rootstore, narrowmatch): | |
1935 | self.nodeconstants = repo.nodeconstants |
|
1935 | self.nodeconstants = repo.nodeconstants | |
1936 | usetreemanifest = False |
|
1936 | usetreemanifest = False | |
1937 | cachesize = 4 |
|
1937 | cachesize = 4 | |
1938 |
|
1938 | |||
1939 | opts = getattr(opener, 'options', None) |
|
1939 | opts = getattr(opener, 'options', None) | |
1940 | if opts is not None: |
|
1940 | if opts is not None: | |
1941 | usetreemanifest = opts.get(b'treemanifest', usetreemanifest) |
|
1941 | usetreemanifest = opts.get(b'treemanifest', usetreemanifest) | |
1942 | cachesize = opts.get(b'manifestcachesize', cachesize) |
|
1942 | cachesize = opts.get(b'manifestcachesize', cachesize) | |
1943 |
|
1943 | |||
1944 | self._treemanifests = usetreemanifest |
|
1944 | self._treemanifests = usetreemanifest | |
1945 |
|
1945 | |||
1946 | self._rootstore = rootstore |
|
1946 | self._rootstore = rootstore | |
1947 | self._rootstore._setupmanifestcachehooks(repo) |
|
1947 | self._rootstore._setupmanifestcachehooks(repo) | |
1948 | self._narrowmatch = narrowmatch |
|
1948 | self._narrowmatch = narrowmatch | |
1949 |
|
1949 | |||
1950 | # A cache of the manifestctx or treemanifestctx for each directory |
|
1950 | # A cache of the manifestctx or treemanifestctx for each directory | |
1951 | self._dirmancache = {} |
|
1951 | self._dirmancache = {} | |
1952 | self._dirmancache[b''] = util.lrucachedict(cachesize) |
|
1952 | self._dirmancache[b''] = util.lrucachedict(cachesize) | |
1953 |
|
1953 | |||
1954 | self._cachesize = cachesize |
|
1954 | self._cachesize = cachesize | |
1955 |
|
1955 | |||
1956 | def __getitem__(self, node): |
|
1956 | def __getitem__(self, node): | |
1957 | """Retrieves the manifest instance for the given node. Throws a |
|
1957 | """Retrieves the manifest instance for the given node. Throws a | |
1958 | LookupError if not found. |
|
1958 | LookupError if not found. | |
1959 | """ |
|
1959 | """ | |
1960 | return self.get(b'', node) |
|
1960 | return self.get(b'', node) | |
1961 |
|
1961 | |||
1962 | def get(self, tree, node, verify=True): |
|
1962 | def get(self, tree, node, verify=True): | |
1963 | """Retrieves the manifest instance for the given node. Throws a |
|
1963 | """Retrieves the manifest instance for the given node. Throws a | |
1964 | LookupError if not found. |
|
1964 | LookupError if not found. | |
1965 |
|
1965 | |||
1966 | `verify` - if True an exception will be thrown if the node is not in |
|
1966 | `verify` - if True an exception will be thrown if the node is not in | |
1967 | the revlog |
|
1967 | the revlog | |
1968 | """ |
|
1968 | """ | |
1969 | if node in self._dirmancache.get(tree, ()): |
|
1969 | if node in self._dirmancache.get(tree, ()): | |
1970 | return self._dirmancache[tree][node] |
|
1970 | return self._dirmancache[tree][node] | |
1971 |
|
1971 | |||
1972 | if not self._narrowmatch.always(): |
|
1972 | if not self._narrowmatch.always(): | |
1973 | if not self._narrowmatch.visitdir(tree[:-1]): |
|
1973 | if not self._narrowmatch.visitdir(tree[:-1]): | |
1974 | return excludeddirmanifestctx(self.nodeconstants, tree, node) |
|
1974 | return excludeddirmanifestctx(self.nodeconstants, tree, node) | |
1975 | if tree: |
|
1975 | if tree: | |
1976 | if self._rootstore._treeondisk: |
|
1976 | if self._rootstore._treeondisk: | |
1977 | if verify: |
|
1977 | if verify: | |
1978 | # Side-effect is LookupError is raised if node doesn't |
|
1978 | # Side-effect is LookupError is raised if node doesn't | |
1979 | # exist. |
|
1979 | # exist. | |
1980 | self.getstorage(tree).rev(node) |
|
1980 | self.getstorage(tree).rev(node) | |
1981 |
|
1981 | |||
1982 | m = treemanifestctx(self, tree, node) |
|
1982 | m = treemanifestctx(self, tree, node) | |
1983 | else: |
|
1983 | else: | |
1984 | raise error.Abort( |
|
1984 | raise error.Abort( | |
1985 | _( |
|
1985 | _( | |
1986 | b"cannot ask for manifest directory '%s' in a flat " |
|
1986 | b"cannot ask for manifest directory '%s' in a flat " | |
1987 | b"manifest" |
|
1987 | b"manifest" | |
1988 | ) |
|
1988 | ) | |
1989 | % tree |
|
1989 | % tree | |
1990 | ) |
|
1990 | ) | |
1991 | else: |
|
1991 | else: | |
1992 | if verify: |
|
1992 | if verify: | |
1993 | # Side-effect is LookupError is raised if node doesn't exist. |
|
1993 | # Side-effect is LookupError is raised if node doesn't exist. | |
1994 | self._rootstore.rev(node) |
|
1994 | self._rootstore.rev(node) | |
1995 |
|
1995 | |||
1996 | if self._treemanifests: |
|
1996 | if self._treemanifests: | |
1997 | m = treemanifestctx(self, b'', node) |
|
1997 | m = treemanifestctx(self, b'', node) | |
1998 | else: |
|
1998 | else: | |
1999 | m = manifestctx(self, node) |
|
1999 | m = manifestctx(self, node) | |
2000 |
|
2000 | |||
2001 | if node != self.nodeconstants.nullid: |
|
2001 | if node != self.nodeconstants.nullid: | |
2002 | mancache = self._dirmancache.get(tree) |
|
2002 | mancache = self._dirmancache.get(tree) | |
2003 | if not mancache: |
|
2003 | if not mancache: | |
2004 | mancache = util.lrucachedict(self._cachesize) |
|
2004 | mancache = util.lrucachedict(self._cachesize) | |
2005 | self._dirmancache[tree] = mancache |
|
2005 | self._dirmancache[tree] = mancache | |
2006 | mancache[node] = m |
|
2006 | mancache[node] = m | |
2007 | return m |
|
2007 | return m | |
2008 |
|
2008 | |||
2009 | def getstorage(self, tree): |
|
2009 | def getstorage(self, tree): | |
2010 | return self._rootstore.dirlog(tree) |
|
2010 | return self._rootstore.dirlog(tree) | |
2011 |
|
2011 | |||
2012 | def clearcaches(self, clear_persisted_data=False): |
|
2012 | def clearcaches(self, clear_persisted_data=False): | |
2013 | self._dirmancache.clear() |
|
2013 | self._dirmancache.clear() | |
2014 | self._rootstore.clearcaches(clear_persisted_data=clear_persisted_data) |
|
2014 | self._rootstore.clearcaches(clear_persisted_data=clear_persisted_data) | |
2015 |
|
2015 | |||
2016 | def rev(self, node): |
|
2016 | def rev(self, node): | |
2017 | return self._rootstore.rev(node) |
|
2017 | return self._rootstore.rev(node) | |
2018 |
|
2018 | |||
2019 | def update_caches(self, transaction): |
|
2019 | def update_caches(self, transaction): | |
2020 | return self._rootstore._revlog.update_caches(transaction=transaction) |
|
2020 | return self._rootstore._revlog.update_caches(transaction=transaction) | |
2021 |
|
2021 | |||
2022 |
|
2022 | |||
2023 | @interfaceutil.implementer(repository.imanifestrevisionwritable) |
|
2023 | @interfaceutil.implementer(repository.imanifestrevisionwritable) | |
2024 | class memmanifestctx: |
|
2024 | class memmanifestctx: | |
2025 | def __init__(self, manifestlog): |
|
2025 | def __init__(self, manifestlog): | |
2026 | self._manifestlog = manifestlog |
|
2026 | self._manifestlog = manifestlog | |
2027 | self._manifestdict = manifestdict(manifestlog.nodeconstants.nodelen) |
|
2027 | self._manifestdict = manifestdict(manifestlog.nodeconstants.nodelen) | |
2028 |
|
2028 | |||
2029 | def _storage(self): |
|
2029 | def _storage(self): | |
2030 | return self._manifestlog.getstorage(b'') |
|
2030 | return self._manifestlog.getstorage(b'') | |
2031 |
|
2031 | |||
2032 | def copy(self): |
|
2032 | def copy(self): | |
2033 | memmf = memmanifestctx(self._manifestlog) |
|
2033 | memmf = memmanifestctx(self._manifestlog) | |
2034 | memmf._manifestdict = self.read().copy() |
|
2034 | memmf._manifestdict = self.read().copy() | |
2035 | return memmf |
|
2035 | return memmf | |
2036 |
|
2036 | |||
2037 | def read(self): |
|
2037 | def read(self): | |
2038 | return self._manifestdict |
|
2038 | return self._manifestdict | |
2039 |
|
2039 | |||
2040 | def write(self, transaction, link, p1, p2, added, removed, match=None): |
|
2040 | def write(self, transaction, link, p1, p2, added, removed, match=None): | |
2041 | return self._storage().add( |
|
2041 | return self._storage().add( | |
2042 | self._manifestdict, |
|
2042 | self._manifestdict, | |
2043 | transaction, |
|
2043 | transaction, | |
2044 | link, |
|
2044 | link, | |
2045 | p1, |
|
2045 | p1, | |
2046 | p2, |
|
2046 | p2, | |
2047 | added, |
|
2047 | added, | |
2048 | removed, |
|
2048 | removed, | |
2049 | match=match, |
|
2049 | match=match, | |
2050 | ) |
|
2050 | ) | |
2051 |
|
2051 | |||
2052 |
|
2052 | |||
2053 | @interfaceutil.implementer(repository.imanifestrevisionstored) |
|
2053 | @interfaceutil.implementer(repository.imanifestrevisionstored) | |
2054 | class manifestctx: |
|
2054 | class manifestctx: | |
2055 | """A class representing a single revision of a manifest, including its |
|
2055 | """A class representing a single revision of a manifest, including its | |
2056 | contents, its parent revs, and its linkrev. |
|
2056 | contents, its parent revs, and its linkrev. | |
2057 | """ |
|
2057 | """ | |
2058 |
|
2058 | |||
2059 | def __init__(self, manifestlog, node): |
|
2059 | def __init__(self, manifestlog, node): | |
2060 | self._manifestlog = manifestlog |
|
2060 | self._manifestlog = manifestlog | |
2061 | self._data = None |
|
2061 | self._data = None | |
2062 |
|
2062 | |||
2063 | self._node = node |
|
2063 | self._node = node | |
2064 |
|
2064 | |||
2065 | # TODO: We eventually want p1, p2, and linkrev exposed on this class, |
|
2065 | # TODO: We eventually want p1, p2, and linkrev exposed on this class, | |
2066 | # but let's add it later when something needs it and we can load it |
|
2066 | # but let's add it later when something needs it and we can load it | |
2067 | # lazily. |
|
2067 | # lazily. | |
2068 | # self.p1, self.p2 = store.parents(node) |
|
2068 | # self.p1, self.p2 = store.parents(node) | |
2069 | # rev = store.rev(node) |
|
2069 | # rev = store.rev(node) | |
2070 | # self.linkrev = store.linkrev(rev) |
|
2070 | # self.linkrev = store.linkrev(rev) | |
2071 |
|
2071 | |||
2072 | def _storage(self): |
|
2072 | def _storage(self): | |
2073 | return self._manifestlog.getstorage(b'') |
|
2073 | return self._manifestlog.getstorage(b'') | |
2074 |
|
2074 | |||
2075 | def node(self): |
|
2075 | def node(self): | |
2076 | return self._node |
|
2076 | return self._node | |
2077 |
|
2077 | |||
2078 | def copy(self): |
|
2078 | def copy(self): | |
2079 | memmf = memmanifestctx(self._manifestlog) |
|
2079 | memmf = memmanifestctx(self._manifestlog) | |
2080 | memmf._manifestdict = self.read().copy() |
|
2080 | memmf._manifestdict = self.read().copy() | |
2081 | return memmf |
|
2081 | return memmf | |
2082 |
|
2082 | |||
2083 | @propertycache |
|
2083 | @propertycache | |
2084 | def parents(self): |
|
2084 | def parents(self): | |
2085 | return self._storage().parents(self._node) |
|
2085 | return self._storage().parents(self._node) | |
2086 |
|
2086 | |||
2087 | def read(self): |
|
2087 | def read(self): | |
2088 | if self._data is None: |
|
2088 | if self._data is None: | |
2089 | nc = self._manifestlog.nodeconstants |
|
2089 | nc = self._manifestlog.nodeconstants | |
2090 | if self._node == nc.nullid: |
|
2090 | if self._node == nc.nullid: | |
2091 | self._data = manifestdict(nc.nodelen) |
|
2091 | self._data = manifestdict(nc.nodelen) | |
2092 | else: |
|
2092 | else: | |
2093 | store = self._storage() |
|
2093 | store = self._storage() | |
2094 | if self._node in store.fulltextcache: |
|
2094 | if self._node in store.fulltextcache: | |
2095 | text = pycompat.bytestr(store.fulltextcache[self._node]) |
|
2095 | text = pycompat.bytestr(store.fulltextcache[self._node]) | |
2096 | else: |
|
2096 | else: | |
2097 | text = store.revision(self._node) |
|
2097 | text = store.revision(self._node) | |
2098 | arraytext = bytearray(text) |
|
2098 | arraytext = bytearray(text) | |
2099 | store.fulltextcache[self._node] = arraytext |
|
2099 | store.fulltextcache[self._node] = arraytext | |
2100 | self._data = manifestdict(nc.nodelen, text) |
|
2100 | self._data = manifestdict(nc.nodelen, text) | |
2101 | return self._data |
|
2101 | return self._data | |
2102 |
|
2102 | |||
2103 | def readfast(self, shallow=False): |
|
2103 | def readfast(self, shallow=False): | |
2104 | """Calls either readdelta or read, based on which would be less work. |
|
2104 | """Calls either readdelta or read, based on which would be less work. | |
2105 | readdelta is called if the delta is against the p1, and therefore can be |
|
2105 | readdelta is called if the delta is against the p1, and therefore can be | |
2106 | read quickly. |
|
2106 | read quickly. | |
2107 |
|
2107 | |||
2108 | If `shallow` is True, nothing changes since this is a flat manifest. |
|
2108 | If `shallow` is True, nothing changes since this is a flat manifest. | |
2109 | """ |
|
2109 | """ | |
2110 | store = self._storage() |
|
2110 | store = self._storage() | |
2111 | r = store.rev(self._node) |
|
2111 | r = store.rev(self._node) | |
2112 | deltaparent = store.deltaparent(r) |
|
2112 | deltaparent = store.deltaparent(r) | |
2113 | if deltaparent != nullrev and deltaparent in store.parentrevs(r): |
|
2113 | if deltaparent != nullrev and deltaparent in store.parentrevs(r): | |
2114 | return self.readdelta() |
|
2114 | return self.readdelta() | |
2115 | return self.read() |
|
2115 | return self.read() | |
2116 |
|
2116 | |||
2117 | def readdelta(self, shallow=False): |
|
2117 | def readdelta(self, shallow=False): | |
2118 | """Returns a manifest containing just the entries that are present |
|
2118 | """Returns a manifest containing just the entries that are present | |
2119 | in this manifest, but not in its p1 manifest. This is efficient to read |
|
2119 | in this manifest, but not in its p1 manifest. This is efficient to read | |
2120 | if the revlog delta is already p1. |
|
2120 | if the revlog delta is already p1. | |
2121 |
|
2121 | |||
2122 | Changing the value of `shallow` has no effect on flat manifests. |
|
2122 | Changing the value of `shallow` has no effect on flat manifests. | |
2123 | """ |
|
2123 | """ | |
2124 | store = self._storage() |
|
2124 | store = self._storage() | |
2125 | r = store.rev(self._node) |
|
2125 | r = store.rev(self._node) | |
2126 | d = mdiff.patchtext(store.revdiff(store.deltaparent(r), r)) |
|
2126 | d = mdiff.patchtext(store.revdiff(store.deltaparent(r), r)) | |
2127 | return manifestdict(store.nodeconstants.nodelen, d) |
|
2127 | return manifestdict(store.nodeconstants.nodelen, d) | |
2128 |
|
2128 | |||
2129 | def find(self, key): |
|
2129 | def find(self, key): | |
2130 | return self.read().find(key) |
|
2130 | return self.read().find(key) | |
2131 |
|
2131 | |||
2132 |
|
2132 | |||
2133 | @interfaceutil.implementer(repository.imanifestrevisionwritable) |
|
2133 | @interfaceutil.implementer(repository.imanifestrevisionwritable) | |
2134 | class memtreemanifestctx: |
|
2134 | class memtreemanifestctx: | |
2135 | def __init__(self, manifestlog, dir=b''): |
|
2135 | def __init__(self, manifestlog, dir=b''): | |
2136 | self._manifestlog = manifestlog |
|
2136 | self._manifestlog = manifestlog | |
2137 | self._dir = dir |
|
2137 | self._dir = dir | |
2138 | self._treemanifest = treemanifest(manifestlog.nodeconstants) |
|
2138 | self._treemanifest = treemanifest(manifestlog.nodeconstants) | |
2139 |
|
2139 | |||
2140 | def _storage(self): |
|
2140 | def _storage(self): | |
2141 | return self._manifestlog.getstorage(b'') |
|
2141 | return self._manifestlog.getstorage(b'') | |
2142 |
|
2142 | |||
2143 | def copy(self): |
|
2143 | def copy(self): | |
2144 | memmf = memtreemanifestctx(self._manifestlog, dir=self._dir) |
|
2144 | memmf = memtreemanifestctx(self._manifestlog, dir=self._dir) | |
2145 | memmf._treemanifest = self._treemanifest.copy() |
|
2145 | memmf._treemanifest = self._treemanifest.copy() | |
2146 | return memmf |
|
2146 | return memmf | |
2147 |
|
2147 | |||
2148 | def read(self): |
|
2148 | def read(self): | |
2149 | return self._treemanifest |
|
2149 | return self._treemanifest | |
2150 |
|
2150 | |||
2151 | def write(self, transaction, link, p1, p2, added, removed, match=None): |
|
2151 | def write(self, transaction, link, p1, p2, added, removed, match=None): | |
2152 | def readtree(dir, node): |
|
2152 | def readtree(dir, node): | |
2153 | return self._manifestlog.get(dir, node).read() |
|
2153 | return self._manifestlog.get(dir, node).read() | |
2154 |
|
2154 | |||
2155 | return self._storage().add( |
|
2155 | return self._storage().add( | |
2156 | self._treemanifest, |
|
2156 | self._treemanifest, | |
2157 | transaction, |
|
2157 | transaction, | |
2158 | link, |
|
2158 | link, | |
2159 | p1, |
|
2159 | p1, | |
2160 | p2, |
|
2160 | p2, | |
2161 | added, |
|
2161 | added, | |
2162 | removed, |
|
2162 | removed, | |
2163 | readtree=readtree, |
|
2163 | readtree=readtree, | |
2164 | match=match, |
|
2164 | match=match, | |
2165 | ) |
|
2165 | ) | |
2166 |
|
2166 | |||
2167 |
|
2167 | |||
2168 | @interfaceutil.implementer(repository.imanifestrevisionstored) |
|
2168 | @interfaceutil.implementer(repository.imanifestrevisionstored) | |
2169 | class treemanifestctx: |
|
2169 | class treemanifestctx: | |
2170 | def __init__(self, manifestlog, dir, node): |
|
2170 | def __init__(self, manifestlog, dir, node): | |
2171 | self._manifestlog = manifestlog |
|
2171 | self._manifestlog = manifestlog | |
2172 | self._dir = dir |
|
2172 | self._dir = dir | |
2173 | self._data = None |
|
2173 | self._data = None | |
2174 |
|
2174 | |||
2175 | self._node = node |
|
2175 | self._node = node | |
2176 |
|
2176 | |||
2177 | # TODO: Load p1/p2/linkrev lazily. They need to be lazily loaded so that |
|
2177 | # TODO: Load p1/p2/linkrev lazily. They need to be lazily loaded so that | |
2178 | # we can instantiate treemanifestctx objects for directories we don't |
|
2178 | # we can instantiate treemanifestctx objects for directories we don't | |
2179 | # have on disk. |
|
2179 | # have on disk. | |
2180 | # self.p1, self.p2 = store.parents(node) |
|
2180 | # self.p1, self.p2 = store.parents(node) | |
2181 | # rev = store.rev(node) |
|
2181 | # rev = store.rev(node) | |
2182 | # self.linkrev = store.linkrev(rev) |
|
2182 | # self.linkrev = store.linkrev(rev) | |
2183 |
|
2183 | |||
2184 | def _storage(self): |
|
2184 | def _storage(self): | |
2185 | narrowmatch = self._manifestlog._narrowmatch |
|
2185 | narrowmatch = self._manifestlog._narrowmatch | |
2186 | if not narrowmatch.always(): |
|
2186 | if not narrowmatch.always(): | |
2187 | if not narrowmatch.visitdir(self._dir[:-1]): |
|
2187 | if not narrowmatch.visitdir(self._dir[:-1]): | |
2188 | return excludedmanifestrevlog( |
|
2188 | return excludedmanifestrevlog( | |
2189 | self._manifestlog.nodeconstants, self._dir |
|
2189 | self._manifestlog.nodeconstants, self._dir | |
2190 | ) |
|
2190 | ) | |
2191 | return self._manifestlog.getstorage(self._dir) |
|
2191 | return self._manifestlog.getstorage(self._dir) | |
2192 |
|
2192 | |||
2193 | def read(self): |
|
2193 | def read(self): | |
2194 | if self._data is None: |
|
2194 | if self._data is None: | |
2195 | store = self._storage() |
|
2195 | store = self._storage() | |
2196 | if self._node == self._manifestlog.nodeconstants.nullid: |
|
2196 | if self._node == self._manifestlog.nodeconstants.nullid: | |
2197 | self._data = treemanifest(self._manifestlog.nodeconstants) |
|
2197 | self._data = treemanifest(self._manifestlog.nodeconstants) | |
2198 | # TODO accessing non-public API |
|
2198 | # TODO accessing non-public API | |
2199 | elif store._treeondisk: |
|
2199 | elif store._treeondisk: | |
2200 | m = treemanifest(self._manifestlog.nodeconstants, dir=self._dir) |
|
2200 | m = treemanifest(self._manifestlog.nodeconstants, dir=self._dir) | |
2201 |
|
2201 | |||
2202 | def gettext(): |
|
2202 | def gettext(): | |
2203 | return store.revision(self._node) |
|
2203 | return store.revision(self._node) | |
2204 |
|
2204 | |||
2205 | def readsubtree(dir, subm): |
|
2205 | def readsubtree(dir, subm): | |
2206 | # Set verify to False since we need to be able to create |
|
2206 | # Set verify to False since we need to be able to create | |
2207 | # subtrees for trees that don't exist on disk. |
|
2207 | # subtrees for trees that don't exist on disk. | |
2208 | return self._manifestlog.get(dir, subm, verify=False).read() |
|
2208 | return self._manifestlog.get(dir, subm, verify=False).read() | |
2209 |
|
2209 | |||
2210 | m.read(gettext, readsubtree) |
|
2210 | m.read(gettext, readsubtree) | |
2211 | m.setnode(self._node) |
|
2211 | m.setnode(self._node) | |
2212 | self._data = m |
|
2212 | self._data = m | |
2213 | else: |
|
2213 | else: | |
2214 | if self._node in store.fulltextcache: |
|
2214 | if self._node in store.fulltextcache: | |
2215 | text = pycompat.bytestr(store.fulltextcache[self._node]) |
|
2215 | text = pycompat.bytestr(store.fulltextcache[self._node]) | |
2216 | else: |
|
2216 | else: | |
2217 | text = store.revision(self._node) |
|
2217 | text = store.revision(self._node) | |
2218 | arraytext = bytearray(text) |
|
2218 | arraytext = bytearray(text) | |
2219 | store.fulltextcache[self._node] = arraytext |
|
2219 | store.fulltextcache[self._node] = arraytext | |
2220 | self._data = treemanifest( |
|
2220 | self._data = treemanifest( | |
2221 | self._manifestlog.nodeconstants, dir=self._dir, text=text |
|
2221 | self._manifestlog.nodeconstants, dir=self._dir, text=text | |
2222 | ) |
|
2222 | ) | |
2223 |
|
2223 | |||
2224 | return self._data |
|
2224 | return self._data | |
2225 |
|
2225 | |||
2226 | def node(self): |
|
2226 | def node(self): | |
2227 | return self._node |
|
2227 | return self._node | |
2228 |
|
2228 | |||
2229 | def copy(self): |
|
2229 | def copy(self): | |
2230 | memmf = memtreemanifestctx(self._manifestlog, dir=self._dir) |
|
2230 | memmf = memtreemanifestctx(self._manifestlog, dir=self._dir) | |
2231 | memmf._treemanifest = self.read().copy() |
|
2231 | memmf._treemanifest = self.read().copy() | |
2232 | return memmf |
|
2232 | return memmf | |
2233 |
|
2233 | |||
2234 | @propertycache |
|
2234 | @propertycache | |
2235 | def parents(self): |
|
2235 | def parents(self): | |
2236 | return self._storage().parents(self._node) |
|
2236 | return self._storage().parents(self._node) | |
2237 |
|
2237 | |||
2238 | def readdelta(self, shallow=False): |
|
2238 | def readdelta(self, shallow=False): | |
2239 | """Returns a manifest containing just the entries that are present |
|
2239 | """Returns a manifest containing just the entries that are present | |
2240 | in this manifest, but not in its p1 manifest. This is efficient to read |
|
2240 | in this manifest, but not in its p1 manifest. This is efficient to read | |
2241 | if the revlog delta is already p1. |
|
2241 | if the revlog delta is already p1. | |
2242 |
|
2242 | |||
2243 | If `shallow` is True, this will read the delta for this directory, |
|
2243 | If `shallow` is True, this will read the delta for this directory, | |
2244 | without recursively reading subdirectory manifests. Instead, any |
|
2244 | without recursively reading subdirectory manifests. Instead, any | |
2245 | subdirectory entry will be reported as it appears in the manifest, i.e. |
|
2245 | subdirectory entry will be reported as it appears in the manifest, i.e. | |
2246 | the subdirectory will be reported among files and distinguished only by |
|
2246 | the subdirectory will be reported among files and distinguished only by | |
2247 | its 't' flag. |
|
2247 | its 't' flag. | |
2248 | """ |
|
2248 | """ | |
2249 | store = self._storage() |
|
2249 | store = self._storage() | |
2250 | if shallow: |
|
2250 | if shallow: | |
2251 | r = store.rev(self._node) |
|
2251 | r = store.rev(self._node) | |
2252 | d = mdiff.patchtext(store.revdiff(store.deltaparent(r), r)) |
|
2252 | d = mdiff.patchtext(store.revdiff(store.deltaparent(r), r)) | |
2253 | return manifestdict(store.nodeconstants.nodelen, d) |
|
2253 | return manifestdict(store.nodeconstants.nodelen, d) | |
2254 | else: |
|
2254 | else: | |
2255 | # Need to perform a slow delta |
|
2255 | # Need to perform a slow delta | |
2256 | r0 = store.deltaparent(store.rev(self._node)) |
|
2256 | r0 = store.deltaparent(store.rev(self._node)) | |
2257 | m0 = self._manifestlog.get(self._dir, store.node(r0)).read() |
|
2257 | m0 = self._manifestlog.get(self._dir, store.node(r0)).read() | |
2258 | m1 = self.read() |
|
2258 | m1 = self.read() | |
2259 | md = treemanifest(self._manifestlog.nodeconstants, dir=self._dir) |
|
2259 | md = treemanifest(self._manifestlog.nodeconstants, dir=self._dir) | |
2260 | for f, ((n0, fl0), (n1, fl1)) in m0.diff(m1).items(): |
|
2260 | for f, ((n0, fl0), (n1, fl1)) in m0.diff(m1).items(): | |
2261 | if n1: |
|
2261 | if n1: | |
2262 | md[f] = n1 |
|
2262 | md[f] = n1 | |
2263 | if fl1: |
|
2263 | if fl1: | |
2264 | md.setflag(f, fl1) |
|
2264 | md.setflag(f, fl1) | |
2265 | return md |
|
2265 | return md | |
2266 |
|
2266 | |||
2267 | def readfast(self, shallow=False): |
|
2267 | def readfast(self, shallow=False): | |
2268 | """Calls either readdelta or read, based on which would be less work. |
|
2268 | """Calls either readdelta or read, based on which would be less work. | |
2269 | readdelta is called if the delta is against the p1, and therefore can be |
|
2269 | readdelta is called if the delta is against the p1, and therefore can be | |
2270 | read quickly. |
|
2270 | read quickly. | |
2271 |
|
2271 | |||
2272 | If `shallow` is True, it only returns the entries from this manifest, |
|
2272 | If `shallow` is True, it only returns the entries from this manifest, | |
2273 | and not any submanifests. |
|
2273 | and not any submanifests. | |
2274 | """ |
|
2274 | """ | |
2275 | store = self._storage() |
|
2275 | store = self._storage() | |
2276 | r = store.rev(self._node) |
|
2276 | r = store.rev(self._node) | |
2277 | deltaparent = store.deltaparent(r) |
|
2277 | deltaparent = store.deltaparent(r) | |
2278 | if deltaparent != nullrev and deltaparent in store.parentrevs(r): |
|
2278 | if deltaparent != nullrev and deltaparent in store.parentrevs(r): | |
2279 | return self.readdelta(shallow=shallow) |
|
2279 | return self.readdelta(shallow=shallow) | |
2280 |
|
2280 | |||
2281 | if shallow: |
|
2281 | if shallow: | |
2282 | return manifestdict( |
|
2282 | return manifestdict( | |
2283 | store.nodeconstants.nodelen, store.revision(self._node) |
|
2283 | store.nodeconstants.nodelen, store.revision(self._node) | |
2284 | ) |
|
2284 | ) | |
2285 | else: |
|
2285 | else: | |
2286 | return self.read() |
|
2286 | return self.read() | |
2287 |
|
2287 | |||
2288 | def find(self, key): |
|
2288 | def find(self, key): | |
2289 | return self.read().find(key) |
|
2289 | return self.read().find(key) | |
2290 |
|
2290 | |||
2291 |
|
2291 | |||
2292 | class excludeddir(treemanifest): |
|
2292 | class excludeddir(treemanifest): | |
2293 | """Stand-in for a directory that is excluded from the repository. |
|
2293 | """Stand-in for a directory that is excluded from the repository. | |
2294 |
|
2294 | |||
2295 | With narrowing active on a repository that uses treemanifests, |
|
2295 | With narrowing active on a repository that uses treemanifests, | |
2296 | some of the directory revlogs will be excluded from the resulting |
|
2296 | some of the directory revlogs will be excluded from the resulting | |
2297 | clone. This is a huge storage win for clients, but means we need |
|
2297 | clone. This is a huge storage win for clients, but means we need | |
2298 | some sort of pseudo-manifest to surface to internals so we can |
|
2298 | some sort of pseudo-manifest to surface to internals so we can | |
2299 | detect a merge conflict outside the narrowspec. That's what this |
|
2299 | detect a merge conflict outside the narrowspec. That's what this | |
2300 | class is: it stands in for a directory whose node is known, but |
|
2300 | class is: it stands in for a directory whose node is known, but | |
2301 | whose contents are unknown. |
|
2301 | whose contents are unknown. | |
2302 | """ |
|
2302 | """ | |
2303 |
|
2303 | |||
2304 | def __init__(self, nodeconstants, dir, node): |
|
2304 | def __init__(self, nodeconstants, dir, node): | |
2305 | super(excludeddir, self).__init__(nodeconstants, dir) |
|
2305 | super(excludeddir, self).__init__(nodeconstants, dir) | |
2306 | self._node = node |
|
2306 | self._node = node | |
2307 | # Add an empty file, which will be included by iterators and such, |
|
2307 | # Add an empty file, which will be included by iterators and such, | |
2308 | # appearing as the directory itself (i.e. something like "dir/") |
|
2308 | # appearing as the directory itself (i.e. something like "dir/") | |
2309 | self._files[b''] = node |
|
2309 | self._files[b''] = node | |
2310 | self._flags[b''] = b't' |
|
2310 | self._flags[b''] = b't' | |
2311 |
|
2311 | |||
2312 | # Manifests outside the narrowspec should never be modified, so avoid |
|
2312 | # Manifests outside the narrowspec should never be modified, so avoid | |
2313 | # copying. This makes a noticeable difference when there are very many |
|
2313 | # copying. This makes a noticeable difference when there are very many | |
2314 | # directories outside the narrowspec. Also, it makes sense for the copy to |
|
2314 | # directories outside the narrowspec. Also, it makes sense for the copy to | |
2315 | # be of the same type as the original, which would not happen with the |
|
2315 | # be of the same type as the original, which would not happen with the | |
2316 | # super type's copy(). |
|
2316 | # super type's copy(). | |
2317 | def copy(self): |
|
2317 | def copy(self): | |
2318 | return self |
|
2318 | return self | |
2319 |
|
2319 | |||
2320 |
|
2320 | |||
2321 | class excludeddirmanifestctx(treemanifestctx): |
|
2321 | class excludeddirmanifestctx(treemanifestctx): | |
2322 | """context wrapper for excludeddir - see that docstring for rationale""" |
|
2322 | """context wrapper for excludeddir - see that docstring for rationale""" | |
2323 |
|
2323 | |||
2324 | def __init__(self, nodeconstants, dir, node): |
|
2324 | def __init__(self, nodeconstants, dir, node): | |
2325 | self.nodeconstants = nodeconstants |
|
2325 | self.nodeconstants = nodeconstants | |
2326 | self._dir = dir |
|
2326 | self._dir = dir | |
2327 | self._node = node |
|
2327 | self._node = node | |
2328 |
|
2328 | |||
2329 | def read(self): |
|
2329 | def read(self): | |
2330 | return excludeddir(self.nodeconstants, self._dir, self._node) |
|
2330 | return excludeddir(self.nodeconstants, self._dir, self._node) | |
2331 |
|
2331 | |||
2332 | def readfast(self, shallow=False): |
|
2332 | def readfast(self, shallow=False): | |
2333 | # special version of readfast since we don't have underlying storage |
|
2333 | # special version of readfast since we don't have underlying storage | |
2334 | return self.read() |
|
2334 | return self.read() | |
2335 |
|
2335 | |||
2336 | def write(self, *args): |
|
2336 | def write(self, *args): | |
2337 | raise error.ProgrammingError( |
|
2337 | raise error.ProgrammingError( | |
2338 | b'attempt to write manifest from excluded dir %s' % self._dir |
|
2338 | b'attempt to write manifest from excluded dir %s' % self._dir | |
2339 | ) |
|
2339 | ) | |
2340 |
|
2340 | |||
2341 |
|
2341 | |||
2342 | class excludedmanifestrevlog(manifestrevlog): |
|
2342 | class excludedmanifestrevlog(manifestrevlog): | |
2343 | """Stand-in for excluded treemanifest revlogs. |
|
2343 | """Stand-in for excluded treemanifest revlogs. | |
2344 |
|
2344 | |||
2345 | When narrowing is active on a treemanifest repository, we'll have |
|
2345 | When narrowing is active on a treemanifest repository, we'll have | |
2346 | references to directories we can't see due to the revlog being |
|
2346 | references to directories we can't see due to the revlog being | |
2347 | skipped. This class exists to conform to the manifestrevlog |
|
2347 | skipped. This class exists to conform to the manifestrevlog | |
2348 | interface for those directories and proactively prevent writes to |
|
2348 | interface for those directories and proactively prevent writes to | |
2349 | outside the narrowspec. |
|
2349 | outside the narrowspec. | |
2350 | """ |
|
2350 | """ | |
2351 |
|
2351 | |||
2352 | def __init__(self, nodeconstants, dir): |
|
2352 | def __init__(self, nodeconstants, dir): | |
2353 | self.nodeconstants = nodeconstants |
|
2353 | self.nodeconstants = nodeconstants | |
2354 | self._dir = dir |
|
2354 | self._dir = dir | |
2355 |
|
2355 | |||
2356 | def __len__(self): |
|
2356 | def __len__(self): | |
2357 | raise error.ProgrammingError( |
|
2357 | raise error.ProgrammingError( | |
2358 | b'attempt to get length of excluded dir %s' % self._dir |
|
2358 | b'attempt to get length of excluded dir %s' % self._dir | |
2359 | ) |
|
2359 | ) | |
2360 |
|
2360 | |||
2361 | def rev(self, node): |
|
2361 | def rev(self, node): | |
2362 | raise error.ProgrammingError( |
|
2362 | raise error.ProgrammingError( | |
2363 | b'attempt to get rev from excluded dir %s' % self._dir |
|
2363 | b'attempt to get rev from excluded dir %s' % self._dir | |
2364 | ) |
|
2364 | ) | |
2365 |
|
2365 | |||
2366 | def linkrev(self, node): |
|
2366 | def linkrev(self, node): | |
2367 | raise error.ProgrammingError( |
|
2367 | raise error.ProgrammingError( | |
2368 | b'attempt to get linkrev from excluded dir %s' % self._dir |
|
2368 | b'attempt to get linkrev from excluded dir %s' % self._dir | |
2369 | ) |
|
2369 | ) | |
2370 |
|
2370 | |||
2371 | def node(self, rev): |
|
2371 | def node(self, rev): | |
2372 | raise error.ProgrammingError( |
|
2372 | raise error.ProgrammingError( | |
2373 | b'attempt to get node from excluded dir %s' % self._dir |
|
2373 | b'attempt to get node from excluded dir %s' % self._dir | |
2374 | ) |
|
2374 | ) | |
2375 |
|
2375 | |||
2376 | def add(self, *args, **kwargs): |
|
2376 | def add(self, *args, **kwargs): | |
2377 | # We should never write entries in dirlogs outside the narrow clone. |
|
2377 | # We should never write entries in dirlogs outside the narrow clone. | |
2378 | # However, the method still gets called from writesubtree() in |
|
2378 | # However, the method still gets called from writesubtree() in | |
2379 | # _addtree(), so we need to handle it. We should possibly make that |
|
2379 | # _addtree(), so we need to handle it. We should possibly make that | |
2380 | # avoid calling add() with a clean manifest (_dirty is always False |
|
2380 | # avoid calling add() with a clean manifest (_dirty is always False | |
2381 | # in excludeddir instances). |
|
2381 | # in excludeddir instances). | |
2382 | pass |
|
2382 | pass |
@@ -1,3549 +1,3546 b'' | |||||
1 | # revlog.py - storage back-end for mercurial |
|
1 | # revlog.py - storage back-end for mercurial | |
2 | # coding: utf8 |
|
2 | # coding: utf8 | |
3 | # |
|
3 | # | |
4 | # Copyright 2005-2007 Olivia Mackall <olivia@selenic.com> |
|
4 | # Copyright 2005-2007 Olivia Mackall <olivia@selenic.com> | |
5 | # |
|
5 | # | |
6 | # This software may be used and distributed according to the terms of the |
|
6 | # This software may be used and distributed according to the terms of the | |
7 | # GNU General Public License version 2 or any later version. |
|
7 | # GNU General Public License version 2 or any later version. | |
8 |
|
8 | |||
9 | """Storage back-end for Mercurial. |
|
9 | """Storage back-end for Mercurial. | |
10 |
|
10 | |||
11 | This provides efficient delta storage with O(1) retrieve and append |
|
11 | This provides efficient delta storage with O(1) retrieve and append | |
12 | and O(changes) merge between branches. |
|
12 | and O(changes) merge between branches. | |
13 | """ |
|
13 | """ | |
14 |
|
14 | |||
15 |
|
15 | |||
16 | import binascii |
|
16 | import binascii | |
17 | import collections |
|
17 | import collections | |
18 | import contextlib |
|
18 | import contextlib | |
19 | import io |
|
19 | import io | |
20 | import os |
|
20 | import os | |
21 | import struct |
|
21 | import struct | |
22 | import weakref |
|
22 | import weakref | |
23 | import zlib |
|
23 | import zlib | |
24 |
|
24 | |||
25 | # import stuff from node for others to import from revlog |
|
25 | # import stuff from node for others to import from revlog | |
26 | from .node import ( |
|
26 | from .node import ( | |
27 | bin, |
|
27 | bin, | |
28 | hex, |
|
28 | hex, | |
29 | nullrev, |
|
29 | nullrev, | |
30 | sha1nodeconstants, |
|
30 | sha1nodeconstants, | |
31 | short, |
|
31 | short, | |
32 | wdirrev, |
|
32 | wdirrev, | |
33 | ) |
|
33 | ) | |
34 | from .i18n import _ |
|
34 | from .i18n import _ | |
35 | from .revlogutils.constants import ( |
|
35 | from .revlogutils.constants import ( | |
36 | ALL_KINDS, |
|
36 | ALL_KINDS, | |
37 | CHANGELOGV2, |
|
37 | CHANGELOGV2, | |
38 | COMP_MODE_DEFAULT, |
|
38 | COMP_MODE_DEFAULT, | |
39 | COMP_MODE_INLINE, |
|
39 | COMP_MODE_INLINE, | |
40 | COMP_MODE_PLAIN, |
|
40 | COMP_MODE_PLAIN, | |
41 | DELTA_BASE_REUSE_NO, |
|
41 | DELTA_BASE_REUSE_NO, | |
42 | DELTA_BASE_REUSE_TRY, |
|
42 | DELTA_BASE_REUSE_TRY, | |
43 | ENTRY_RANK, |
|
43 | ENTRY_RANK, | |
44 | FEATURES_BY_VERSION, |
|
44 | FEATURES_BY_VERSION, | |
45 | FLAG_GENERALDELTA, |
|
45 | FLAG_GENERALDELTA, | |
46 | FLAG_INLINE_DATA, |
|
46 | FLAG_INLINE_DATA, | |
47 | INDEX_HEADER, |
|
47 | INDEX_HEADER, | |
48 | KIND_CHANGELOG, |
|
48 | KIND_CHANGELOG, | |
49 | KIND_FILELOG, |
|
49 | KIND_FILELOG, | |
50 | RANK_UNKNOWN, |
|
50 | RANK_UNKNOWN, | |
51 | REVLOGV0, |
|
51 | REVLOGV0, | |
52 | REVLOGV1, |
|
52 | REVLOGV1, | |
53 | REVLOGV1_FLAGS, |
|
53 | REVLOGV1_FLAGS, | |
54 | REVLOGV2, |
|
54 | REVLOGV2, | |
55 | REVLOGV2_FLAGS, |
|
55 | REVLOGV2_FLAGS, | |
56 | REVLOG_DEFAULT_FLAGS, |
|
56 | REVLOG_DEFAULT_FLAGS, | |
57 | REVLOG_DEFAULT_FORMAT, |
|
57 | REVLOG_DEFAULT_FORMAT, | |
58 | REVLOG_DEFAULT_VERSION, |
|
58 | REVLOG_DEFAULT_VERSION, | |
59 | SUPPORTED_FLAGS, |
|
59 | SUPPORTED_FLAGS, | |
60 | ) |
|
60 | ) | |
61 | from .revlogutils.flagutil import ( |
|
61 | from .revlogutils.flagutil import ( | |
62 | REVIDX_DEFAULT_FLAGS, |
|
62 | REVIDX_DEFAULT_FLAGS, | |
63 | REVIDX_ELLIPSIS, |
|
63 | REVIDX_ELLIPSIS, | |
64 | REVIDX_EXTSTORED, |
|
64 | REVIDX_EXTSTORED, | |
65 | REVIDX_FLAGS_ORDER, |
|
65 | REVIDX_FLAGS_ORDER, | |
66 | REVIDX_HASCOPIESINFO, |
|
66 | REVIDX_HASCOPIESINFO, | |
67 | REVIDX_ISCENSORED, |
|
67 | REVIDX_ISCENSORED, | |
68 | REVIDX_RAWTEXT_CHANGING_FLAGS, |
|
68 | REVIDX_RAWTEXT_CHANGING_FLAGS, | |
69 | ) |
|
69 | ) | |
70 | from .thirdparty import attr |
|
70 | from .thirdparty import attr | |
71 | from . import ( |
|
71 | from . import ( | |
72 | ancestor, |
|
72 | ancestor, | |
73 | dagop, |
|
73 | dagop, | |
74 | error, |
|
74 | error, | |
75 | mdiff, |
|
75 | mdiff, | |
76 | policy, |
|
76 | policy, | |
77 | pycompat, |
|
77 | pycompat, | |
78 | revlogutils, |
|
78 | revlogutils, | |
79 | templatefilters, |
|
79 | templatefilters, | |
80 | util, |
|
80 | util, | |
81 | ) |
|
81 | ) | |
82 | from .interfaces import ( |
|
82 | from .interfaces import ( | |
83 | repository, |
|
83 | repository, | |
84 | util as interfaceutil, |
|
84 | util as interfaceutil, | |
85 | ) |
|
85 | ) | |
86 | from .revlogutils import ( |
|
86 | from .revlogutils import ( | |
87 | deltas as deltautil, |
|
87 | deltas as deltautil, | |
88 | docket as docketutil, |
|
88 | docket as docketutil, | |
89 | flagutil, |
|
89 | flagutil, | |
90 | nodemap as nodemaputil, |
|
90 | nodemap as nodemaputil, | |
91 | randomaccessfile, |
|
91 | randomaccessfile, | |
92 | revlogv0, |
|
92 | revlogv0, | |
93 | rewrite, |
|
93 | rewrite, | |
94 | sidedata as sidedatautil, |
|
94 | sidedata as sidedatautil, | |
95 | ) |
|
95 | ) | |
96 | from .utils import ( |
|
96 | from .utils import ( | |
97 | storageutil, |
|
97 | storageutil, | |
98 | stringutil, |
|
98 | stringutil, | |
99 | ) |
|
99 | ) | |
100 |
|
100 | |||
101 | # blanked usage of all the name to prevent pyflakes constraints |
|
101 | # blanked usage of all the name to prevent pyflakes constraints | |
102 | # We need these name available in the module for extensions. |
|
102 | # We need these name available in the module for extensions. | |
103 |
|
103 | |||
104 | REVLOGV0 |
|
104 | REVLOGV0 | |
105 | REVLOGV1 |
|
105 | REVLOGV1 | |
106 | REVLOGV2 |
|
106 | REVLOGV2 | |
107 | CHANGELOGV2 |
|
107 | CHANGELOGV2 | |
108 | FLAG_INLINE_DATA |
|
108 | FLAG_INLINE_DATA | |
109 | FLAG_GENERALDELTA |
|
109 | FLAG_GENERALDELTA | |
110 | REVLOG_DEFAULT_FLAGS |
|
110 | REVLOG_DEFAULT_FLAGS | |
111 | REVLOG_DEFAULT_FORMAT |
|
111 | REVLOG_DEFAULT_FORMAT | |
112 | REVLOG_DEFAULT_VERSION |
|
112 | REVLOG_DEFAULT_VERSION | |
113 | REVLOGV1_FLAGS |
|
113 | REVLOGV1_FLAGS | |
114 | REVLOGV2_FLAGS |
|
114 | REVLOGV2_FLAGS | |
115 | REVIDX_ISCENSORED |
|
115 | REVIDX_ISCENSORED | |
116 | REVIDX_ELLIPSIS |
|
116 | REVIDX_ELLIPSIS | |
117 | REVIDX_HASCOPIESINFO |
|
117 | REVIDX_HASCOPIESINFO | |
118 | REVIDX_EXTSTORED |
|
118 | REVIDX_EXTSTORED | |
119 | REVIDX_DEFAULT_FLAGS |
|
119 | REVIDX_DEFAULT_FLAGS | |
120 | REVIDX_FLAGS_ORDER |
|
120 | REVIDX_FLAGS_ORDER | |
121 | REVIDX_RAWTEXT_CHANGING_FLAGS |
|
121 | REVIDX_RAWTEXT_CHANGING_FLAGS | |
122 |
|
122 | |||
123 | parsers = policy.importmod('parsers') |
|
123 | parsers = policy.importmod('parsers') | |
124 | rustancestor = policy.importrust('ancestor') |
|
124 | rustancestor = policy.importrust('ancestor') | |
125 | rustdagop = policy.importrust('dagop') |
|
125 | rustdagop = policy.importrust('dagop') | |
126 | rustrevlog = policy.importrust('revlog') |
|
126 | rustrevlog = policy.importrust('revlog') | |
127 |
|
127 | |||
128 | # Aliased for performance. |
|
128 | # Aliased for performance. | |
129 | _zlibdecompress = zlib.decompress |
|
129 | _zlibdecompress = zlib.decompress | |
130 |
|
130 | |||
131 | # max size of inline data embedded into a revlog |
|
131 | # max size of inline data embedded into a revlog | |
132 | _maxinline = 131072 |
|
132 | _maxinline = 131072 | |
133 |
|
133 | |||
134 | # Flag processors for REVIDX_ELLIPSIS. |
|
134 | # Flag processors for REVIDX_ELLIPSIS. | |
135 | def ellipsisreadprocessor(rl, text): |
|
135 | def ellipsisreadprocessor(rl, text): | |
136 | return text, False |
|
136 | return text, False | |
137 |
|
137 | |||
138 |
|
138 | |||
139 | def ellipsiswriteprocessor(rl, text): |
|
139 | def ellipsiswriteprocessor(rl, text): | |
140 | return text, False |
|
140 | return text, False | |
141 |
|
141 | |||
142 |
|
142 | |||
143 | def ellipsisrawprocessor(rl, text): |
|
143 | def ellipsisrawprocessor(rl, text): | |
144 | return False |
|
144 | return False | |
145 |
|
145 | |||
146 |
|
146 | |||
147 | ellipsisprocessor = ( |
|
147 | ellipsisprocessor = ( | |
148 | ellipsisreadprocessor, |
|
148 | ellipsisreadprocessor, | |
149 | ellipsiswriteprocessor, |
|
149 | ellipsiswriteprocessor, | |
150 | ellipsisrawprocessor, |
|
150 | ellipsisrawprocessor, | |
151 | ) |
|
151 | ) | |
152 |
|
152 | |||
153 |
|
153 | |||
154 | def _verify_revision(rl, skipflags, state, node): |
|
154 | def _verify_revision(rl, skipflags, state, node): | |
155 | """Verify the integrity of the given revlog ``node`` while providing a hook |
|
155 | """Verify the integrity of the given revlog ``node`` while providing a hook | |
156 | point for extensions to influence the operation.""" |
|
156 | point for extensions to influence the operation.""" | |
157 | if skipflags: |
|
157 | if skipflags: | |
158 | state[b'skipread'].add(node) |
|
158 | state[b'skipread'].add(node) | |
159 | else: |
|
159 | else: | |
160 | # Side-effect: read content and verify hash. |
|
160 | # Side-effect: read content and verify hash. | |
161 | rl.revision(node) |
|
161 | rl.revision(node) | |
162 |
|
162 | |||
163 |
|
163 | |||
164 | # True if a fast implementation for persistent-nodemap is available |
|
164 | # True if a fast implementation for persistent-nodemap is available | |
165 | # |
|
165 | # | |
166 | # We also consider we have a "fast" implementation in "pure" python because |
|
166 | # We also consider we have a "fast" implementation in "pure" python because | |
167 | # people using pure don't really have performance consideration (and a |
|
167 | # people using pure don't really have performance consideration (and a | |
168 | # wheelbarrow of other slowness source) |
|
168 | # wheelbarrow of other slowness source) | |
169 | HAS_FAST_PERSISTENT_NODEMAP = rustrevlog is not None or hasattr( |
|
169 | HAS_FAST_PERSISTENT_NODEMAP = rustrevlog is not None or hasattr( | |
170 | parsers, 'BaseIndexObject' |
|
170 | parsers, 'BaseIndexObject' | |
171 | ) |
|
171 | ) | |
172 |
|
172 | |||
173 |
|
173 | |||
174 | @interfaceutil.implementer(repository.irevisiondelta) |
|
174 | @interfaceutil.implementer(repository.irevisiondelta) | |
175 | @attr.s(slots=True) |
|
175 | @attr.s(slots=True) | |
176 | class revlogrevisiondelta: |
|
176 | class revlogrevisiondelta: | |
177 | node = attr.ib() |
|
177 | node = attr.ib() | |
178 | p1node = attr.ib() |
|
178 | p1node = attr.ib() | |
179 | p2node = attr.ib() |
|
179 | p2node = attr.ib() | |
180 | basenode = attr.ib() |
|
180 | basenode = attr.ib() | |
181 | flags = attr.ib() |
|
181 | flags = attr.ib() | |
182 | baserevisionsize = attr.ib() |
|
182 | baserevisionsize = attr.ib() | |
183 | revision = attr.ib() |
|
183 | revision = attr.ib() | |
184 | delta = attr.ib() |
|
184 | delta = attr.ib() | |
185 | sidedata = attr.ib() |
|
185 | sidedata = attr.ib() | |
186 | protocol_flags = attr.ib() |
|
186 | protocol_flags = attr.ib() | |
187 | linknode = attr.ib(default=None) |
|
187 | linknode = attr.ib(default=None) | |
188 |
|
188 | |||
189 |
|
189 | |||
190 | @interfaceutil.implementer(repository.iverifyproblem) |
|
190 | @interfaceutil.implementer(repository.iverifyproblem) | |
191 | @attr.s(frozen=True) |
|
191 | @attr.s(frozen=True) | |
192 | class revlogproblem: |
|
192 | class revlogproblem: | |
193 | warning = attr.ib(default=None) |
|
193 | warning = attr.ib(default=None) | |
194 | error = attr.ib(default=None) |
|
194 | error = attr.ib(default=None) | |
195 | node = attr.ib(default=None) |
|
195 | node = attr.ib(default=None) | |
196 |
|
196 | |||
197 |
|
197 | |||
198 | def parse_index_v1(data, inline): |
|
198 | def parse_index_v1(data, inline): | |
199 | # call the C implementation to parse the index data |
|
199 | # call the C implementation to parse the index data | |
200 | index, cache = parsers.parse_index2(data, inline) |
|
200 | index, cache = parsers.parse_index2(data, inline) | |
201 | return index, cache |
|
201 | return index, cache | |
202 |
|
202 | |||
203 |
|
203 | |||
204 | def parse_index_v2(data, inline): |
|
204 | def parse_index_v2(data, inline): | |
205 | # call the C implementation to parse the index data |
|
205 | # call the C implementation to parse the index data | |
206 | index, cache = parsers.parse_index2(data, inline, format=REVLOGV2) |
|
206 | index, cache = parsers.parse_index2(data, inline, format=REVLOGV2) | |
207 | return index, cache |
|
207 | return index, cache | |
208 |
|
208 | |||
209 |
|
209 | |||
210 | def parse_index_cl_v2(data, inline): |
|
210 | def parse_index_cl_v2(data, inline): | |
211 | # call the C implementation to parse the index data |
|
211 | # call the C implementation to parse the index data | |
212 | index, cache = parsers.parse_index2(data, inline, format=CHANGELOGV2) |
|
212 | index, cache = parsers.parse_index2(data, inline, format=CHANGELOGV2) | |
213 | return index, cache |
|
213 | return index, cache | |
214 |
|
214 | |||
215 |
|
215 | |||
216 | if hasattr(parsers, 'parse_index_devel_nodemap'): |
|
216 | if hasattr(parsers, 'parse_index_devel_nodemap'): | |
217 |
|
217 | |||
218 | def parse_index_v1_nodemap(data, inline): |
|
218 | def parse_index_v1_nodemap(data, inline): | |
219 | index, cache = parsers.parse_index_devel_nodemap(data, inline) |
|
219 | index, cache = parsers.parse_index_devel_nodemap(data, inline) | |
220 | return index, cache |
|
220 | return index, cache | |
221 |
|
221 | |||
222 |
|
222 | |||
223 | else: |
|
223 | else: | |
224 | parse_index_v1_nodemap = None |
|
224 | parse_index_v1_nodemap = None | |
225 |
|
225 | |||
226 |
|
226 | |||
227 | def parse_index_v1_mixed(data, inline): |
|
227 | def parse_index_v1_mixed(data, inline): | |
228 | index, cache = parse_index_v1(data, inline) |
|
228 | index, cache = parse_index_v1(data, inline) | |
229 | return rustrevlog.MixedIndex(index), cache |
|
229 | return rustrevlog.MixedIndex(index), cache | |
230 |
|
230 | |||
231 |
|
231 | |||
232 | # corresponds to uncompressed length of indexformatng (2 gigs, 4-byte |
|
232 | # corresponds to uncompressed length of indexformatng (2 gigs, 4-byte | |
233 | # signed integer) |
|
233 | # signed integer) | |
234 | _maxentrysize = 0x7FFFFFFF |
|
234 | _maxentrysize = 0x7FFFFFFF | |
235 |
|
235 | |||
236 | FILE_TOO_SHORT_MSG = _( |
|
236 | FILE_TOO_SHORT_MSG = _( | |
237 | b'cannot read from revlog %s;' |
|
237 | b'cannot read from revlog %s;' | |
238 | b' expected %d bytes from offset %d, data size is %d' |
|
238 | b' expected %d bytes from offset %d, data size is %d' | |
239 | ) |
|
239 | ) | |
240 |
|
240 | |||
241 | hexdigits = b'0123456789abcdefABCDEF' |
|
241 | hexdigits = b'0123456789abcdefABCDEF' | |
242 |
|
242 | |||
243 |
|
243 | |||
244 | class revlog: |
|
244 | class revlog: | |
245 | """ |
|
245 | """ | |
246 | the underlying revision storage object |
|
246 | the underlying revision storage object | |
247 |
|
247 | |||
248 | A revlog consists of two parts, an index and the revision data. |
|
248 | A revlog consists of two parts, an index and the revision data. | |
249 |
|
249 | |||
250 | The index is a file with a fixed record size containing |
|
250 | The index is a file with a fixed record size containing | |
251 | information on each revision, including its nodeid (hash), the |
|
251 | information on each revision, including its nodeid (hash), the | |
252 | nodeids of its parents, the position and offset of its data within |
|
252 | nodeids of its parents, the position and offset of its data within | |
253 | the data file, and the revision it's based on. Finally, each entry |
|
253 | the data file, and the revision it's based on. Finally, each entry | |
254 | contains a linkrev entry that can serve as a pointer to external |
|
254 | contains a linkrev entry that can serve as a pointer to external | |
255 | data. |
|
255 | data. | |
256 |
|
256 | |||
257 | The revision data itself is a linear collection of data chunks. |
|
257 | The revision data itself is a linear collection of data chunks. | |
258 | Each chunk represents a revision and is usually represented as a |
|
258 | Each chunk represents a revision and is usually represented as a | |
259 | delta against the previous chunk. To bound lookup time, runs of |
|
259 | delta against the previous chunk. To bound lookup time, runs of | |
260 | deltas are limited to about 2 times the length of the original |
|
260 | deltas are limited to about 2 times the length of the original | |
261 | version data. This makes retrieval of a version proportional to |
|
261 | version data. This makes retrieval of a version proportional to | |
262 | its size, or O(1) relative to the number of revisions. |
|
262 | its size, or O(1) relative to the number of revisions. | |
263 |
|
263 | |||
264 | Both pieces of the revlog are written to in an append-only |
|
264 | Both pieces of the revlog are written to in an append-only | |
265 | fashion, which means we never need to rewrite a file to insert or |
|
265 | fashion, which means we never need to rewrite a file to insert or | |
266 | remove data, and can use some simple techniques to avoid the need |
|
266 | remove data, and can use some simple techniques to avoid the need | |
267 | for locking while reading. |
|
267 | for locking while reading. | |
268 |
|
268 | |||
269 | If checkambig, indexfile is opened with checkambig=True at |
|
269 | If checkambig, indexfile is opened with checkambig=True at | |
270 | writing, to avoid file stat ambiguity. |
|
270 | writing, to avoid file stat ambiguity. | |
271 |
|
271 | |||
272 | If mmaplargeindex is True, and an mmapindexthreshold is set, the |
|
272 | If mmaplargeindex is True, and an mmapindexthreshold is set, the | |
273 | index will be mmapped rather than read if it is larger than the |
|
273 | index will be mmapped rather than read if it is larger than the | |
274 | configured threshold. |
|
274 | configured threshold. | |
275 |
|
275 | |||
276 | If censorable is True, the revlog can have censored revisions. |
|
276 | If censorable is True, the revlog can have censored revisions. | |
277 |
|
277 | |||
278 | If `upperboundcomp` is not None, this is the expected maximal gain from |
|
278 | If `upperboundcomp` is not None, this is the expected maximal gain from | |
279 | compression for the data content. |
|
279 | compression for the data content. | |
280 |
|
280 | |||
281 | `concurrencychecker` is an optional function that receives 3 arguments: a |
|
281 | `concurrencychecker` is an optional function that receives 3 arguments: a | |
282 | file handle, a filename, and an expected position. It should check whether |
|
282 | file handle, a filename, and an expected position. It should check whether | |
283 | the current position in the file handle is valid, and log/warn/fail (by |
|
283 | the current position in the file handle is valid, and log/warn/fail (by | |
284 | raising). |
|
284 | raising). | |
285 |
|
285 | |||
286 | See mercurial/revlogutils/contants.py for details about the content of an |
|
286 | See mercurial/revlogutils/contants.py for details about the content of an | |
287 | index entry. |
|
287 | index entry. | |
288 | """ |
|
288 | """ | |
289 |
|
289 | |||
290 | _flagserrorclass = error.RevlogError |
|
290 | _flagserrorclass = error.RevlogError | |
291 |
|
291 | |||
292 | @staticmethod |
|
292 | @staticmethod | |
293 | def is_inline_index(header_bytes): |
|
293 | def is_inline_index(header_bytes): | |
294 | """Determine if a revlog is inline from the initial bytes of the index""" |
|
294 | """Determine if a revlog is inline from the initial bytes of the index""" | |
295 | header = INDEX_HEADER.unpack(header_bytes)[0] |
|
295 | header = INDEX_HEADER.unpack(header_bytes)[0] | |
296 |
|
296 | |||
297 | _format_flags = header & ~0xFFFF |
|
297 | _format_flags = header & ~0xFFFF | |
298 | _format_version = header & 0xFFFF |
|
298 | _format_version = header & 0xFFFF | |
299 |
|
299 | |||
300 | features = FEATURES_BY_VERSION[_format_version] |
|
300 | features = FEATURES_BY_VERSION[_format_version] | |
301 | return features[b'inline'](_format_flags) |
|
301 | return features[b'inline'](_format_flags) | |
302 |
|
302 | |||
303 | def __init__( |
|
303 | def __init__( | |
304 | self, |
|
304 | self, | |
305 | opener, |
|
305 | opener, | |
306 | target, |
|
306 | target, | |
307 | radix, |
|
307 | radix, | |
308 | postfix=None, # only exist for `tmpcensored` now |
|
308 | postfix=None, # only exist for `tmpcensored` now | |
309 | checkambig=False, |
|
309 | checkambig=False, | |
310 | mmaplargeindex=False, |
|
310 | mmaplargeindex=False, | |
311 | censorable=False, |
|
311 | censorable=False, | |
312 | upperboundcomp=None, |
|
312 | upperboundcomp=None, | |
313 | persistentnodemap=False, |
|
313 | persistentnodemap=False, | |
314 | concurrencychecker=None, |
|
314 | concurrencychecker=None, | |
315 | trypending=False, |
|
315 | trypending=False, | |
316 | try_split=False, |
|
316 | try_split=False, | |
317 | canonical_parent_order=True, |
|
317 | canonical_parent_order=True, | |
318 | ): |
|
318 | ): | |
319 | """ |
|
319 | """ | |
320 | create a revlog object |
|
320 | create a revlog object | |
321 |
|
321 | |||
322 | opener is a function that abstracts the file opening operation |
|
322 | opener is a function that abstracts the file opening operation | |
323 | and can be used to implement COW semantics or the like. |
|
323 | and can be used to implement COW semantics or the like. | |
324 |
|
324 | |||
325 | `target`: a (KIND, ID) tuple that identify the content stored in |
|
325 | `target`: a (KIND, ID) tuple that identify the content stored in | |
326 | this revlog. It help the rest of the code to understand what the revlog |
|
326 | this revlog. It help the rest of the code to understand what the revlog | |
327 | is about without having to resort to heuristic and index filename |
|
327 | is about without having to resort to heuristic and index filename | |
328 | analysis. Note: that this must be reliably be set by normal code, but |
|
328 | analysis. Note: that this must be reliably be set by normal code, but | |
329 | that test, debug, or performance measurement code might not set this to |
|
329 | that test, debug, or performance measurement code might not set this to | |
330 | accurate value. |
|
330 | accurate value. | |
331 | """ |
|
331 | """ | |
332 | self.upperboundcomp = upperboundcomp |
|
332 | self.upperboundcomp = upperboundcomp | |
333 |
|
333 | |||
334 | self.radix = radix |
|
334 | self.radix = radix | |
335 |
|
335 | |||
336 | self._docket_file = None |
|
336 | self._docket_file = None | |
337 | self._indexfile = None |
|
337 | self._indexfile = None | |
338 | self._datafile = None |
|
338 | self._datafile = None | |
339 | self._sidedatafile = None |
|
339 | self._sidedatafile = None | |
340 | self._nodemap_file = None |
|
340 | self._nodemap_file = None | |
341 | self.postfix = postfix |
|
341 | self.postfix = postfix | |
342 | self._trypending = trypending |
|
342 | self._trypending = trypending | |
343 | self._try_split = try_split |
|
343 | self._try_split = try_split | |
344 | self.opener = opener |
|
344 | self.opener = opener | |
345 | if persistentnodemap: |
|
345 | if persistentnodemap: | |
346 | self._nodemap_file = nodemaputil.get_nodemap_file(self) |
|
346 | self._nodemap_file = nodemaputil.get_nodemap_file(self) | |
347 |
|
347 | |||
348 | assert target[0] in ALL_KINDS |
|
348 | assert target[0] in ALL_KINDS | |
349 | assert len(target) == 2 |
|
349 | assert len(target) == 2 | |
350 | self.target = target |
|
350 | self.target = target | |
351 | # When True, indexfile is opened with checkambig=True at writing, to |
|
351 | # When True, indexfile is opened with checkambig=True at writing, to | |
352 | # avoid file stat ambiguity. |
|
352 | # avoid file stat ambiguity. | |
353 | self._checkambig = checkambig |
|
353 | self._checkambig = checkambig | |
354 | self._mmaplargeindex = mmaplargeindex |
|
354 | self._mmaplargeindex = mmaplargeindex | |
355 | self._censorable = censorable |
|
355 | self._censorable = censorable | |
356 | # 3-tuple of (node, rev, text) for a raw revision. |
|
356 | # 3-tuple of (node, rev, text) for a raw revision. | |
357 | self._revisioncache = None |
|
357 | self._revisioncache = None | |
358 | # Maps rev to chain base rev. |
|
358 | # Maps rev to chain base rev. | |
359 | self._chainbasecache = util.lrucachedict(100) |
|
359 | self._chainbasecache = util.lrucachedict(100) | |
360 | # 2-tuple of (offset, data) of raw data from the revlog at an offset. |
|
360 | # 2-tuple of (offset, data) of raw data from the revlog at an offset. | |
361 | self._chunkcache = (0, b'') |
|
361 | self._chunkcache = (0, b'') | |
362 | # How much data to read and cache into the raw revlog data cache. |
|
362 | # How much data to read and cache into the raw revlog data cache. | |
363 | self._chunkcachesize = 65536 |
|
363 | self._chunkcachesize = 65536 | |
364 | self._maxchainlen = None |
|
364 | self._maxchainlen = None | |
365 | self._deltabothparents = True |
|
365 | self._deltabothparents = True | |
366 | self._candidate_group_chunk_size = 0 |
|
366 | self._candidate_group_chunk_size = 0 | |
367 | self._debug_delta = False |
|
367 | self._debug_delta = False | |
368 | self.index = None |
|
368 | self.index = None | |
369 | self._docket = None |
|
369 | self._docket = None | |
370 | self._nodemap_docket = None |
|
370 | self._nodemap_docket = None | |
371 | # Mapping of partial identifiers to full nodes. |
|
371 | # Mapping of partial identifiers to full nodes. | |
372 | self._pcache = {} |
|
372 | self._pcache = {} | |
373 | # Mapping of revision integer to full node. |
|
373 | # Mapping of revision integer to full node. | |
374 | self._compengine = b'zlib' |
|
374 | self._compengine = b'zlib' | |
375 | self._compengineopts = {} |
|
375 | self._compengineopts = {} | |
376 | self._maxdeltachainspan = -1 |
|
376 | self._maxdeltachainspan = -1 | |
377 | self._withsparseread = False |
|
377 | self._withsparseread = False | |
378 | self._sparserevlog = False |
|
378 | self._sparserevlog = False | |
379 | self.hassidedata = False |
|
379 | self.hassidedata = False | |
380 | self._srdensitythreshold = 0.50 |
|
380 | self._srdensitythreshold = 0.50 | |
381 | self._srmingapsize = 262144 |
|
381 | self._srmingapsize = 262144 | |
382 |
|
382 | |||
383 | # other optionnals features |
|
383 | # other optionnals features | |
384 |
|
384 | |||
385 | # might remove rank configuration once the computation has no impact |
|
385 | # might remove rank configuration once the computation has no impact | |
386 | self._compute_rank = False |
|
386 | self._compute_rank = False | |
387 |
|
387 | |||
388 | # Make copy of flag processors so each revlog instance can support |
|
388 | # Make copy of flag processors so each revlog instance can support | |
389 | # custom flags. |
|
389 | # custom flags. | |
390 | self._flagprocessors = dict(flagutil.flagprocessors) |
|
390 | self._flagprocessors = dict(flagutil.flagprocessors) | |
391 |
|
391 | |||
392 | # 3-tuple of file handles being used for active writing. |
|
392 | # 3-tuple of file handles being used for active writing. | |
393 | self._writinghandles = None |
|
393 | self._writinghandles = None | |
394 | # prevent nesting of addgroup |
|
394 | # prevent nesting of addgroup | |
395 | self._adding_group = None |
|
395 | self._adding_group = None | |
396 |
|
396 | |||
397 | self._loadindex() |
|
397 | self._loadindex() | |
398 |
|
398 | |||
399 | self._concurrencychecker = concurrencychecker |
|
399 | self._concurrencychecker = concurrencychecker | |
400 |
|
400 | |||
401 | # parent order is supposed to be semantically irrelevant, so we |
|
401 | # parent order is supposed to be semantically irrelevant, so we | |
402 | # normally resort parents to ensure that the first parent is non-null, |
|
402 | # normally resort parents to ensure that the first parent is non-null, | |
403 | # if there is a non-null parent at all. |
|
403 | # if there is a non-null parent at all. | |
404 | # filelog abuses the parent order as flag to mark some instances of |
|
404 | # filelog abuses the parent order as flag to mark some instances of | |
405 | # meta-encoded files, so allow it to disable this behavior. |
|
405 | # meta-encoded files, so allow it to disable this behavior. | |
406 | self.canonical_parent_order = canonical_parent_order |
|
406 | self.canonical_parent_order = canonical_parent_order | |
407 |
|
407 | |||
408 | def _init_opts(self): |
|
408 | def _init_opts(self): | |
409 | """process options (from above/config) to setup associated default revlog mode |
|
409 | """process options (from above/config) to setup associated default revlog mode | |
410 |
|
410 | |||
411 | These values might be affected when actually reading on disk information. |
|
411 | These values might be affected when actually reading on disk information. | |
412 |
|
412 | |||
413 | The relevant values are returned for use in _loadindex(). |
|
413 | The relevant values are returned for use in _loadindex(). | |
414 |
|
414 | |||
415 | * newversionflags: |
|
415 | * newversionflags: | |
416 | version header to use if we need to create a new revlog |
|
416 | version header to use if we need to create a new revlog | |
417 |
|
417 | |||
418 | * mmapindexthreshold: |
|
418 | * mmapindexthreshold: | |
419 | minimal index size for start to use mmap |
|
419 | minimal index size for start to use mmap | |
420 |
|
420 | |||
421 | * force_nodemap: |
|
421 | * force_nodemap: | |
422 | force the usage of a "development" version of the nodemap code |
|
422 | force the usage of a "development" version of the nodemap code | |
423 | """ |
|
423 | """ | |
424 | mmapindexthreshold = None |
|
424 | mmapindexthreshold = None | |
425 | opts = self.opener.options |
|
425 | opts = self.opener.options | |
426 |
|
426 | |||
427 | if b'changelogv2' in opts and self.revlog_kind == KIND_CHANGELOG: |
|
427 | if b'changelogv2' in opts and self.revlog_kind == KIND_CHANGELOG: | |
428 | new_header = CHANGELOGV2 |
|
428 | new_header = CHANGELOGV2 | |
429 | self._compute_rank = opts.get(b'changelogv2.compute-rank', True) |
|
429 | self._compute_rank = opts.get(b'changelogv2.compute-rank', True) | |
430 | elif b'revlogv2' in opts: |
|
430 | elif b'revlogv2' in opts: | |
431 | new_header = REVLOGV2 |
|
431 | new_header = REVLOGV2 | |
432 | elif b'revlogv1' in opts: |
|
432 | elif b'revlogv1' in opts: | |
433 | new_header = REVLOGV1 | FLAG_INLINE_DATA |
|
433 | new_header = REVLOGV1 | FLAG_INLINE_DATA | |
434 | if b'generaldelta' in opts: |
|
434 | if b'generaldelta' in opts: | |
435 | new_header |= FLAG_GENERALDELTA |
|
435 | new_header |= FLAG_GENERALDELTA | |
436 | elif b'revlogv0' in self.opener.options: |
|
436 | elif b'revlogv0' in self.opener.options: | |
437 | new_header = REVLOGV0 |
|
437 | new_header = REVLOGV0 | |
438 | else: |
|
438 | else: | |
439 | new_header = REVLOG_DEFAULT_VERSION |
|
439 | new_header = REVLOG_DEFAULT_VERSION | |
440 |
|
440 | |||
441 | if b'chunkcachesize' in opts: |
|
441 | if b'chunkcachesize' in opts: | |
442 | self._chunkcachesize = opts[b'chunkcachesize'] |
|
442 | self._chunkcachesize = opts[b'chunkcachesize'] | |
443 | if b'maxchainlen' in opts: |
|
443 | if b'maxchainlen' in opts: | |
444 | self._maxchainlen = opts[b'maxchainlen'] |
|
444 | self._maxchainlen = opts[b'maxchainlen'] | |
445 | if b'deltabothparents' in opts: |
|
445 | if b'deltabothparents' in opts: | |
446 | self._deltabothparents = opts[b'deltabothparents'] |
|
446 | self._deltabothparents = opts[b'deltabothparents'] | |
447 | dps_cgds = opts.get(b'delta-parent-search.candidate-group-chunk-size') |
|
447 | dps_cgds = opts.get(b'delta-parent-search.candidate-group-chunk-size') | |
448 | if dps_cgds: |
|
448 | if dps_cgds: | |
449 | self._candidate_group_chunk_size = dps_cgds |
|
449 | self._candidate_group_chunk_size = dps_cgds | |
450 | self._lazydelta = bool(opts.get(b'lazydelta', True)) |
|
450 | self._lazydelta = bool(opts.get(b'lazydelta', True)) | |
451 | self._lazydeltabase = False |
|
451 | self._lazydeltabase = False | |
452 | if self._lazydelta: |
|
452 | if self._lazydelta: | |
453 | self._lazydeltabase = bool(opts.get(b'lazydeltabase', False)) |
|
453 | self._lazydeltabase = bool(opts.get(b'lazydeltabase', False)) | |
454 | if b'debug-delta' in opts: |
|
454 | if b'debug-delta' in opts: | |
455 | self._debug_delta = opts[b'debug-delta'] |
|
455 | self._debug_delta = opts[b'debug-delta'] | |
456 | if b'compengine' in opts: |
|
456 | if b'compengine' in opts: | |
457 | self._compengine = opts[b'compengine'] |
|
457 | self._compengine = opts[b'compengine'] | |
458 | if b'zlib.level' in opts: |
|
458 | if b'zlib.level' in opts: | |
459 | self._compengineopts[b'zlib.level'] = opts[b'zlib.level'] |
|
459 | self._compengineopts[b'zlib.level'] = opts[b'zlib.level'] | |
460 | if b'zstd.level' in opts: |
|
460 | if b'zstd.level' in opts: | |
461 | self._compengineopts[b'zstd.level'] = opts[b'zstd.level'] |
|
461 | self._compengineopts[b'zstd.level'] = opts[b'zstd.level'] | |
462 | if b'maxdeltachainspan' in opts: |
|
462 | if b'maxdeltachainspan' in opts: | |
463 | self._maxdeltachainspan = opts[b'maxdeltachainspan'] |
|
463 | self._maxdeltachainspan = opts[b'maxdeltachainspan'] | |
464 | if self._mmaplargeindex and b'mmapindexthreshold' in opts: |
|
464 | if self._mmaplargeindex and b'mmapindexthreshold' in opts: | |
465 | mmapindexthreshold = opts[b'mmapindexthreshold'] |
|
465 | mmapindexthreshold = opts[b'mmapindexthreshold'] | |
466 | self._sparserevlog = bool(opts.get(b'sparse-revlog', False)) |
|
466 | self._sparserevlog = bool(opts.get(b'sparse-revlog', False)) | |
467 | withsparseread = bool(opts.get(b'with-sparse-read', False)) |
|
467 | withsparseread = bool(opts.get(b'with-sparse-read', False)) | |
468 | # sparse-revlog forces sparse-read |
|
468 | # sparse-revlog forces sparse-read | |
469 | self._withsparseread = self._sparserevlog or withsparseread |
|
469 | self._withsparseread = self._sparserevlog or withsparseread | |
470 | if b'sparse-read-density-threshold' in opts: |
|
470 | if b'sparse-read-density-threshold' in opts: | |
471 | self._srdensitythreshold = opts[b'sparse-read-density-threshold'] |
|
471 | self._srdensitythreshold = opts[b'sparse-read-density-threshold'] | |
472 | if b'sparse-read-min-gap-size' in opts: |
|
472 | if b'sparse-read-min-gap-size' in opts: | |
473 | self._srmingapsize = opts[b'sparse-read-min-gap-size'] |
|
473 | self._srmingapsize = opts[b'sparse-read-min-gap-size'] | |
474 | if opts.get(b'enableellipsis'): |
|
474 | if opts.get(b'enableellipsis'): | |
475 | self._flagprocessors[REVIDX_ELLIPSIS] = ellipsisprocessor |
|
475 | self._flagprocessors[REVIDX_ELLIPSIS] = ellipsisprocessor | |
476 |
|
476 | |||
477 | # revlog v0 doesn't have flag processors |
|
477 | # revlog v0 doesn't have flag processors | |
478 | for flag, processor in opts.get(b'flagprocessors', {}).items(): |
|
478 | for flag, processor in opts.get(b'flagprocessors', {}).items(): | |
479 | flagutil.insertflagprocessor(flag, processor, self._flagprocessors) |
|
479 | flagutil.insertflagprocessor(flag, processor, self._flagprocessors) | |
480 |
|
480 | |||
481 | if self._chunkcachesize <= 0: |
|
481 | if self._chunkcachesize <= 0: | |
482 | raise error.RevlogError( |
|
482 | raise error.RevlogError( | |
483 | _(b'revlog chunk cache size %r is not greater than 0') |
|
483 | _(b'revlog chunk cache size %r is not greater than 0') | |
484 | % self._chunkcachesize |
|
484 | % self._chunkcachesize | |
485 | ) |
|
485 | ) | |
486 | elif self._chunkcachesize & (self._chunkcachesize - 1): |
|
486 | elif self._chunkcachesize & (self._chunkcachesize - 1): | |
487 | raise error.RevlogError( |
|
487 | raise error.RevlogError( | |
488 | _(b'revlog chunk cache size %r is not a power of 2') |
|
488 | _(b'revlog chunk cache size %r is not a power of 2') | |
489 | % self._chunkcachesize |
|
489 | % self._chunkcachesize | |
490 | ) |
|
490 | ) | |
491 | force_nodemap = opts.get(b'devel-force-nodemap', False) |
|
491 | force_nodemap = opts.get(b'devel-force-nodemap', False) | |
492 | return new_header, mmapindexthreshold, force_nodemap |
|
492 | return new_header, mmapindexthreshold, force_nodemap | |
493 |
|
493 | |||
494 | def _get_data(self, filepath, mmap_threshold, size=None): |
|
494 | def _get_data(self, filepath, mmap_threshold, size=None): | |
495 | """return a file content with or without mmap |
|
495 | """return a file content with or without mmap | |
496 |
|
496 | |||
497 | If the file is missing return the empty string""" |
|
497 | If the file is missing return the empty string""" | |
498 | try: |
|
498 | try: | |
499 | with self.opener(filepath) as fp: |
|
499 | with self.opener(filepath) as fp: | |
500 | if mmap_threshold is not None: |
|
500 | if mmap_threshold is not None: | |
501 | file_size = self.opener.fstat(fp).st_size |
|
501 | file_size = self.opener.fstat(fp).st_size | |
502 | if file_size >= mmap_threshold: |
|
502 | if file_size >= mmap_threshold: | |
503 | if size is not None: |
|
503 | if size is not None: | |
504 | # avoid potentiel mmap crash |
|
504 | # avoid potentiel mmap crash | |
505 | size = min(file_size, size) |
|
505 | size = min(file_size, size) | |
506 | # TODO: should .close() to release resources without |
|
506 | # TODO: should .close() to release resources without | |
507 | # relying on Python GC |
|
507 | # relying on Python GC | |
508 | if size is None: |
|
508 | if size is None: | |
509 | return util.buffer(util.mmapread(fp)) |
|
509 | return util.buffer(util.mmapread(fp)) | |
510 | else: |
|
510 | else: | |
511 | return util.buffer(util.mmapread(fp, size)) |
|
511 | return util.buffer(util.mmapread(fp, size)) | |
512 | if size is None: |
|
512 | if size is None: | |
513 | return fp.read() |
|
513 | return fp.read() | |
514 | else: |
|
514 | else: | |
515 | return fp.read(size) |
|
515 | return fp.read(size) | |
516 | except FileNotFoundError: |
|
516 | except FileNotFoundError: | |
517 | return b'' |
|
517 | return b'' | |
518 |
|
518 | |||
519 | def get_streams(self, max_linkrev, force_inline=False): |
|
519 | def get_streams(self, max_linkrev, force_inline=False): | |
520 | """return a list of streams that represent this revlog |
|
520 | """return a list of streams that represent this revlog | |
521 |
|
521 | |||
522 | This is used by stream-clone to do bytes to bytes copies of a repository. |
|
522 | This is used by stream-clone to do bytes to bytes copies of a repository. | |
523 |
|
523 | |||
524 | This streams data for all revisions that refer to a changelog revision up |
|
524 | This streams data for all revisions that refer to a changelog revision up | |
525 | to `max_linkrev`. |
|
525 | to `max_linkrev`. | |
526 |
|
526 | |||
527 | If `force_inline` is set, it enforces that the stream will represent an inline revlog. |
|
527 | If `force_inline` is set, it enforces that the stream will represent an inline revlog. | |
528 |
|
528 | |||
529 | It returns is a list of three-tuple: |
|
529 | It returns is a list of three-tuple: | |
530 |
|
530 | |||
531 | [ |
|
531 | [ | |
532 | (filename, bytes_stream, stream_size), |
|
532 | (filename, bytes_stream, stream_size), | |
533 | β¦ |
|
533 | β¦ | |
534 | ] |
|
534 | ] | |
535 | """ |
|
535 | """ | |
536 | n = len(self) |
|
536 | n = len(self) | |
537 | index = self.index |
|
537 | index = self.index | |
538 | while n > 0: |
|
538 | while n > 0: | |
539 | linkrev = index[n - 1][4] |
|
539 | linkrev = index[n - 1][4] | |
540 | if linkrev < max_linkrev: |
|
540 | if linkrev < max_linkrev: | |
541 | break |
|
541 | break | |
542 | # note: this loop will rarely go through multiple iterations, since |
|
542 | # note: this loop will rarely go through multiple iterations, since | |
543 | # it only traverses commits created during the current streaming |
|
543 | # it only traverses commits created during the current streaming | |
544 | # pull operation. |
|
544 | # pull operation. | |
545 | # |
|
545 | # | |
546 | # If this become a problem, using a binary search should cap the |
|
546 | # If this become a problem, using a binary search should cap the | |
547 | # runtime of this. |
|
547 | # runtime of this. | |
548 | n = n - 1 |
|
548 | n = n - 1 | |
549 | if n == 0: |
|
549 | if n == 0: | |
550 | # no data to send |
|
550 | # no data to send | |
551 | return [] |
|
551 | return [] | |
552 | index_size = n * index.entry_size |
|
552 | index_size = n * index.entry_size | |
553 | data_size = self.end(n - 1) |
|
553 | data_size = self.end(n - 1) | |
554 |
|
554 | |||
555 | # XXX we might have been split (or stripped) since the object |
|
555 | # XXX we might have been split (or stripped) since the object | |
556 | # initialization, We need to close this race too, but having a way to |
|
556 | # initialization, We need to close this race too, but having a way to | |
557 | # pre-open the file we feed to the revlog and never closing them before |
|
557 | # pre-open the file we feed to the revlog and never closing them before | |
558 | # we are done streaming. |
|
558 | # we are done streaming. | |
559 |
|
559 | |||
560 | if self._inline: |
|
560 | if self._inline: | |
561 |
|
561 | |||
562 | def get_stream(): |
|
562 | def get_stream(): | |
563 | with self._indexfp() as fp: |
|
563 | with self._indexfp() as fp: | |
564 | yield None |
|
564 | yield None | |
565 | size = index_size + data_size |
|
565 | size = index_size + data_size | |
566 | if size <= 65536: |
|
566 | if size <= 65536: | |
567 | yield fp.read(size) |
|
567 | yield fp.read(size) | |
568 | else: |
|
568 | else: | |
569 | yield from util.filechunkiter(fp, limit=size) |
|
569 | yield from util.filechunkiter(fp, limit=size) | |
570 |
|
570 | |||
571 | inline_stream = get_stream() |
|
571 | inline_stream = get_stream() | |
572 | next(inline_stream) |
|
572 | next(inline_stream) | |
573 | return [ |
|
573 | return [ | |
574 | (self._indexfile, inline_stream, index_size + data_size), |
|
574 | (self._indexfile, inline_stream, index_size + data_size), | |
575 | ] |
|
575 | ] | |
576 | elif force_inline: |
|
576 | elif force_inline: | |
577 |
|
577 | |||
578 | def get_stream(): |
|
578 | def get_stream(): | |
579 | with self.reading(): |
|
579 | with self.reading(): | |
580 | yield None |
|
580 | yield None | |
581 |
|
581 | |||
582 | for rev in range(n): |
|
582 | for rev in range(n): | |
583 | idx = self.index.entry_binary(rev) |
|
583 | idx = self.index.entry_binary(rev) | |
584 | if rev == 0 and self._docket is None: |
|
584 | if rev == 0 and self._docket is None: | |
585 | # re-inject the inline flag |
|
585 | # re-inject the inline flag | |
586 | header = self._format_flags |
|
586 | header = self._format_flags | |
587 | header |= self._format_version |
|
587 | header |= self._format_version | |
588 | header |= FLAG_INLINE_DATA |
|
588 | header |= FLAG_INLINE_DATA | |
589 | header = self.index.pack_header(header) |
|
589 | header = self.index.pack_header(header) | |
590 | idx = header + idx |
|
590 | idx = header + idx | |
591 | yield idx |
|
591 | yield idx | |
592 | yield self._getsegmentforrevs(rev, rev)[1] |
|
592 | yield self._getsegmentforrevs(rev, rev)[1] | |
593 |
|
593 | |||
594 | inline_stream = get_stream() |
|
594 | inline_stream = get_stream() | |
595 | next(inline_stream) |
|
595 | next(inline_stream) | |
596 | return [ |
|
596 | return [ | |
597 | (self._indexfile, inline_stream, index_size + data_size), |
|
597 | (self._indexfile, inline_stream, index_size + data_size), | |
598 | ] |
|
598 | ] | |
599 | else: |
|
599 | else: | |
600 |
|
600 | |||
601 | def get_index_stream(): |
|
601 | def get_index_stream(): | |
602 | with self._indexfp() as fp: |
|
602 | with self._indexfp() as fp: | |
603 | yield None |
|
603 | yield None | |
604 | if index_size <= 65536: |
|
604 | if index_size <= 65536: | |
605 | yield fp.read(index_size) |
|
605 | yield fp.read(index_size) | |
606 | else: |
|
606 | else: | |
607 | yield from util.filechunkiter(fp, limit=index_size) |
|
607 | yield from util.filechunkiter(fp, limit=index_size) | |
608 |
|
608 | |||
609 | def get_data_stream(): |
|
609 | def get_data_stream(): | |
610 | with self._datafp() as fp: |
|
610 | with self._datafp() as fp: | |
611 | yield None |
|
611 | yield None | |
612 | if data_size <= 65536: |
|
612 | if data_size <= 65536: | |
613 | yield fp.read(data_size) |
|
613 | yield fp.read(data_size) | |
614 | else: |
|
614 | else: | |
615 | yield from util.filechunkiter(fp, limit=data_size) |
|
615 | yield from util.filechunkiter(fp, limit=data_size) | |
616 |
|
616 | |||
617 | index_stream = get_index_stream() |
|
617 | index_stream = get_index_stream() | |
618 | next(index_stream) |
|
618 | next(index_stream) | |
619 | data_stream = get_data_stream() |
|
619 | data_stream = get_data_stream() | |
620 | next(data_stream) |
|
620 | next(data_stream) | |
621 | return [ |
|
621 | return [ | |
622 | (self._datafile, data_stream, data_size), |
|
622 | (self._datafile, data_stream, data_size), | |
623 | (self._indexfile, index_stream, index_size), |
|
623 | (self._indexfile, index_stream, index_size), | |
624 | ] |
|
624 | ] | |
625 |
|
625 | |||
626 | def _loadindex(self, docket=None): |
|
626 | def _loadindex(self, docket=None): | |
627 |
|
627 | |||
628 | new_header, mmapindexthreshold, force_nodemap = self._init_opts() |
|
628 | new_header, mmapindexthreshold, force_nodemap = self._init_opts() | |
629 |
|
629 | |||
630 | if self.postfix is not None: |
|
630 | if self.postfix is not None: | |
631 | entry_point = b'%s.i.%s' % (self.radix, self.postfix) |
|
631 | entry_point = b'%s.i.%s' % (self.radix, self.postfix) | |
632 | elif self._trypending and self.opener.exists(b'%s.i.a' % self.radix): |
|
632 | elif self._trypending and self.opener.exists(b'%s.i.a' % self.radix): | |
633 | entry_point = b'%s.i.a' % self.radix |
|
633 | entry_point = b'%s.i.a' % self.radix | |
634 | elif self._try_split and self.opener.exists(self._split_index_file): |
|
634 | elif self._try_split and self.opener.exists(self._split_index_file): | |
635 | entry_point = self._split_index_file |
|
635 | entry_point = self._split_index_file | |
636 | else: |
|
636 | else: | |
637 | entry_point = b'%s.i' % self.radix |
|
637 | entry_point = b'%s.i' % self.radix | |
638 |
|
638 | |||
639 | if docket is not None: |
|
639 | if docket is not None: | |
640 | self._docket = docket |
|
640 | self._docket = docket | |
641 | self._docket_file = entry_point |
|
641 | self._docket_file = entry_point | |
642 | else: |
|
642 | else: | |
643 | self._initempty = True |
|
643 | self._initempty = True | |
644 | entry_data = self._get_data(entry_point, mmapindexthreshold) |
|
644 | entry_data = self._get_data(entry_point, mmapindexthreshold) | |
645 | if len(entry_data) > 0: |
|
645 | if len(entry_data) > 0: | |
646 | header = INDEX_HEADER.unpack(entry_data[:4])[0] |
|
646 | header = INDEX_HEADER.unpack(entry_data[:4])[0] | |
647 | self._initempty = False |
|
647 | self._initempty = False | |
648 | else: |
|
648 | else: | |
649 | header = new_header |
|
649 | header = new_header | |
650 |
|
650 | |||
651 | self._format_flags = header & ~0xFFFF |
|
651 | self._format_flags = header & ~0xFFFF | |
652 | self._format_version = header & 0xFFFF |
|
652 | self._format_version = header & 0xFFFF | |
653 |
|
653 | |||
654 | supported_flags = SUPPORTED_FLAGS.get(self._format_version) |
|
654 | supported_flags = SUPPORTED_FLAGS.get(self._format_version) | |
655 | if supported_flags is None: |
|
655 | if supported_flags is None: | |
656 | msg = _(b'unknown version (%d) in revlog %s') |
|
656 | msg = _(b'unknown version (%d) in revlog %s') | |
657 | msg %= (self._format_version, self.display_id) |
|
657 | msg %= (self._format_version, self.display_id) | |
658 | raise error.RevlogError(msg) |
|
658 | raise error.RevlogError(msg) | |
659 | elif self._format_flags & ~supported_flags: |
|
659 | elif self._format_flags & ~supported_flags: | |
660 | msg = _(b'unknown flags (%#04x) in version %d revlog %s') |
|
660 | msg = _(b'unknown flags (%#04x) in version %d revlog %s') | |
661 | display_flag = self._format_flags >> 16 |
|
661 | display_flag = self._format_flags >> 16 | |
662 | msg %= (display_flag, self._format_version, self.display_id) |
|
662 | msg %= (display_flag, self._format_version, self.display_id) | |
663 | raise error.RevlogError(msg) |
|
663 | raise error.RevlogError(msg) | |
664 |
|
664 | |||
665 | features = FEATURES_BY_VERSION[self._format_version] |
|
665 | features = FEATURES_BY_VERSION[self._format_version] | |
666 | self._inline = features[b'inline'](self._format_flags) |
|
666 | self._inline = features[b'inline'](self._format_flags) | |
667 | self._generaldelta = features[b'generaldelta'](self._format_flags) |
|
667 | self._generaldelta = features[b'generaldelta'](self._format_flags) | |
668 | self.hassidedata = features[b'sidedata'] |
|
668 | self.hassidedata = features[b'sidedata'] | |
669 |
|
669 | |||
670 | if not features[b'docket']: |
|
670 | if not features[b'docket']: | |
671 | self._indexfile = entry_point |
|
671 | self._indexfile = entry_point | |
672 | index_data = entry_data |
|
672 | index_data = entry_data | |
673 | else: |
|
673 | else: | |
674 | self._docket_file = entry_point |
|
674 | self._docket_file = entry_point | |
675 | if self._initempty: |
|
675 | if self._initempty: | |
676 | self._docket = docketutil.default_docket(self, header) |
|
676 | self._docket = docketutil.default_docket(self, header) | |
677 | else: |
|
677 | else: | |
678 | self._docket = docketutil.parse_docket( |
|
678 | self._docket = docketutil.parse_docket( | |
679 | self, entry_data, use_pending=self._trypending |
|
679 | self, entry_data, use_pending=self._trypending | |
680 | ) |
|
680 | ) | |
681 |
|
681 | |||
682 | if self._docket is not None: |
|
682 | if self._docket is not None: | |
683 | self._indexfile = self._docket.index_filepath() |
|
683 | self._indexfile = self._docket.index_filepath() | |
684 | index_data = b'' |
|
684 | index_data = b'' | |
685 | index_size = self._docket.index_end |
|
685 | index_size = self._docket.index_end | |
686 | if index_size > 0: |
|
686 | if index_size > 0: | |
687 | index_data = self._get_data( |
|
687 | index_data = self._get_data( | |
688 | self._indexfile, mmapindexthreshold, size=index_size |
|
688 | self._indexfile, mmapindexthreshold, size=index_size | |
689 | ) |
|
689 | ) | |
690 | if len(index_data) < index_size: |
|
690 | if len(index_data) < index_size: | |
691 | msg = _(b'too few index data for %s: got %d, expected %d') |
|
691 | msg = _(b'too few index data for %s: got %d, expected %d') | |
692 | msg %= (self.display_id, len(index_data), index_size) |
|
692 | msg %= (self.display_id, len(index_data), index_size) | |
693 | raise error.RevlogError(msg) |
|
693 | raise error.RevlogError(msg) | |
694 |
|
694 | |||
695 | self._inline = False |
|
695 | self._inline = False | |
696 | # generaldelta implied by version 2 revlogs. |
|
696 | # generaldelta implied by version 2 revlogs. | |
697 | self._generaldelta = True |
|
697 | self._generaldelta = True | |
698 | # the logic for persistent nodemap will be dealt with within the |
|
698 | # the logic for persistent nodemap will be dealt with within the | |
699 | # main docket, so disable it for now. |
|
699 | # main docket, so disable it for now. | |
700 | self._nodemap_file = None |
|
700 | self._nodemap_file = None | |
701 |
|
701 | |||
702 | if self._docket is not None: |
|
702 | if self._docket is not None: | |
703 | self._datafile = self._docket.data_filepath() |
|
703 | self._datafile = self._docket.data_filepath() | |
704 | self._sidedatafile = self._docket.sidedata_filepath() |
|
704 | self._sidedatafile = self._docket.sidedata_filepath() | |
705 | elif self.postfix is None: |
|
705 | elif self.postfix is None: | |
706 | self._datafile = b'%s.d' % self.radix |
|
706 | self._datafile = b'%s.d' % self.radix | |
707 | else: |
|
707 | else: | |
708 | self._datafile = b'%s.d.%s' % (self.radix, self.postfix) |
|
708 | self._datafile = b'%s.d.%s' % (self.radix, self.postfix) | |
709 |
|
709 | |||
710 | self.nodeconstants = sha1nodeconstants |
|
710 | self.nodeconstants = sha1nodeconstants | |
711 | self.nullid = self.nodeconstants.nullid |
|
711 | self.nullid = self.nodeconstants.nullid | |
712 |
|
712 | |||
713 | # sparse-revlog can't be on without general-delta (issue6056) |
|
713 | # sparse-revlog can't be on without general-delta (issue6056) | |
714 | if not self._generaldelta: |
|
714 | if not self._generaldelta: | |
715 | self._sparserevlog = False |
|
715 | self._sparserevlog = False | |
716 |
|
716 | |||
717 | self._storedeltachains = True |
|
717 | self._storedeltachains = True | |
718 |
|
718 | |||
719 | devel_nodemap = ( |
|
719 | devel_nodemap = ( | |
720 | self._nodemap_file |
|
720 | self._nodemap_file | |
721 | and force_nodemap |
|
721 | and force_nodemap | |
722 | and parse_index_v1_nodemap is not None |
|
722 | and parse_index_v1_nodemap is not None | |
723 | ) |
|
723 | ) | |
724 |
|
724 | |||
725 | use_rust_index = False |
|
725 | use_rust_index = False | |
726 | if rustrevlog is not None: |
|
726 | if rustrevlog is not None: | |
727 | if self._nodemap_file is not None: |
|
727 | if self._nodemap_file is not None: | |
728 | use_rust_index = True |
|
728 | use_rust_index = True | |
729 | else: |
|
729 | else: | |
730 | use_rust_index = self.opener.options.get(b'rust.index') |
|
730 | use_rust_index = self.opener.options.get(b'rust.index') | |
731 |
|
731 | |||
732 | self._parse_index = parse_index_v1 |
|
732 | self._parse_index = parse_index_v1 | |
733 | if self._format_version == REVLOGV0: |
|
733 | if self._format_version == REVLOGV0: | |
734 | self._parse_index = revlogv0.parse_index_v0 |
|
734 | self._parse_index = revlogv0.parse_index_v0 | |
735 | elif self._format_version == REVLOGV2: |
|
735 | elif self._format_version == REVLOGV2: | |
736 | self._parse_index = parse_index_v2 |
|
736 | self._parse_index = parse_index_v2 | |
737 | elif self._format_version == CHANGELOGV2: |
|
737 | elif self._format_version == CHANGELOGV2: | |
738 | self._parse_index = parse_index_cl_v2 |
|
738 | self._parse_index = parse_index_cl_v2 | |
739 | elif devel_nodemap: |
|
739 | elif devel_nodemap: | |
740 | self._parse_index = parse_index_v1_nodemap |
|
740 | self._parse_index = parse_index_v1_nodemap | |
741 | elif use_rust_index: |
|
741 | elif use_rust_index: | |
742 | self._parse_index = parse_index_v1_mixed |
|
742 | self._parse_index = parse_index_v1_mixed | |
743 | try: |
|
743 | try: | |
744 | d = self._parse_index(index_data, self._inline) |
|
744 | d = self._parse_index(index_data, self._inline) | |
745 | index, chunkcache = d |
|
745 | index, chunkcache = d | |
746 | use_nodemap = ( |
|
746 | use_nodemap = ( | |
747 | not self._inline |
|
747 | not self._inline | |
748 | and self._nodemap_file is not None |
|
748 | and self._nodemap_file is not None | |
749 | and hasattr(index, 'update_nodemap_data') |
|
749 | and hasattr(index, 'update_nodemap_data') | |
750 | ) |
|
750 | ) | |
751 | if use_nodemap: |
|
751 | if use_nodemap: | |
752 | nodemap_data = nodemaputil.persisted_data(self) |
|
752 | nodemap_data = nodemaputil.persisted_data(self) | |
753 | if nodemap_data is not None: |
|
753 | if nodemap_data is not None: | |
754 | docket = nodemap_data[0] |
|
754 | docket = nodemap_data[0] | |
755 | if ( |
|
755 | if ( | |
756 | len(d[0]) > docket.tip_rev |
|
756 | len(d[0]) > docket.tip_rev | |
757 | and d[0][docket.tip_rev][7] == docket.tip_node |
|
757 | and d[0][docket.tip_rev][7] == docket.tip_node | |
758 | ): |
|
758 | ): | |
759 | # no changelog tampering |
|
759 | # no changelog tampering | |
760 | self._nodemap_docket = docket |
|
760 | self._nodemap_docket = docket | |
761 | index.update_nodemap_data(*nodemap_data) |
|
761 | index.update_nodemap_data(*nodemap_data) | |
762 | except (ValueError, IndexError): |
|
762 | except (ValueError, IndexError): | |
763 | raise error.RevlogError( |
|
763 | raise error.RevlogError( | |
764 | _(b"index %s is corrupted") % self.display_id |
|
764 | _(b"index %s is corrupted") % self.display_id | |
765 | ) |
|
765 | ) | |
766 | self.index = index |
|
766 | self.index = index | |
767 | self._segmentfile = randomaccessfile.randomaccessfile( |
|
767 | self._segmentfile = randomaccessfile.randomaccessfile( | |
768 | self.opener, |
|
768 | self.opener, | |
769 | (self._indexfile if self._inline else self._datafile), |
|
769 | (self._indexfile if self._inline else self._datafile), | |
770 | self._chunkcachesize, |
|
770 | self._chunkcachesize, | |
771 | chunkcache, |
|
771 | chunkcache, | |
772 | ) |
|
772 | ) | |
773 | self._segmentfile_sidedata = randomaccessfile.randomaccessfile( |
|
773 | self._segmentfile_sidedata = randomaccessfile.randomaccessfile( | |
774 | self.opener, |
|
774 | self.opener, | |
775 | self._sidedatafile, |
|
775 | self._sidedatafile, | |
776 | self._chunkcachesize, |
|
776 | self._chunkcachesize, | |
777 | ) |
|
777 | ) | |
778 | # revnum -> (chain-length, sum-delta-length) |
|
778 | # revnum -> (chain-length, sum-delta-length) | |
779 | self._chaininfocache = util.lrucachedict(500) |
|
779 | self._chaininfocache = util.lrucachedict(500) | |
780 | # revlog header -> revlog compressor |
|
780 | # revlog header -> revlog compressor | |
781 | self._decompressors = {} |
|
781 | self._decompressors = {} | |
782 |
|
782 | |||
783 | def get_revlog(self): |
|
783 | def get_revlog(self): | |
784 | """simple function to mirror API of other not-really-revlog API""" |
|
784 | """simple function to mirror API of other not-really-revlog API""" | |
785 | return self |
|
785 | return self | |
786 |
|
786 | |||
787 | @util.propertycache |
|
787 | @util.propertycache | |
788 | def revlog_kind(self): |
|
788 | def revlog_kind(self): | |
789 | return self.target[0] |
|
789 | return self.target[0] | |
790 |
|
790 | |||
791 | @util.propertycache |
|
791 | @util.propertycache | |
792 | def display_id(self): |
|
792 | def display_id(self): | |
793 | """The public facing "ID" of the revlog that we use in message""" |
|
793 | """The public facing "ID" of the revlog that we use in message""" | |
794 | if self.revlog_kind == KIND_FILELOG: |
|
794 | if self.revlog_kind == KIND_FILELOG: | |
795 | # Reference the file without the "data/" prefix, so it is familiar |
|
795 | # Reference the file without the "data/" prefix, so it is familiar | |
796 | # to the user. |
|
796 | # to the user. | |
797 | return self.target[1] |
|
797 | return self.target[1] | |
798 | else: |
|
798 | else: | |
799 | return self.radix |
|
799 | return self.radix | |
800 |
|
800 | |||
801 | def _get_decompressor(self, t): |
|
801 | def _get_decompressor(self, t): | |
802 | try: |
|
802 | try: | |
803 | compressor = self._decompressors[t] |
|
803 | compressor = self._decompressors[t] | |
804 | except KeyError: |
|
804 | except KeyError: | |
805 | try: |
|
805 | try: | |
806 | engine = util.compengines.forrevlogheader(t) |
|
806 | engine = util.compengines.forrevlogheader(t) | |
807 | compressor = engine.revlogcompressor(self._compengineopts) |
|
807 | compressor = engine.revlogcompressor(self._compengineopts) | |
808 | self._decompressors[t] = compressor |
|
808 | self._decompressors[t] = compressor | |
809 | except KeyError: |
|
809 | except KeyError: | |
810 | raise error.RevlogError( |
|
810 | raise error.RevlogError( | |
811 | _(b'unknown compression type %s') % binascii.hexlify(t) |
|
811 | _(b'unknown compression type %s') % binascii.hexlify(t) | |
812 | ) |
|
812 | ) | |
813 | return compressor |
|
813 | return compressor | |
814 |
|
814 | |||
815 | @util.propertycache |
|
815 | @util.propertycache | |
816 | def _compressor(self): |
|
816 | def _compressor(self): | |
817 | engine = util.compengines[self._compengine] |
|
817 | engine = util.compengines[self._compengine] | |
818 | return engine.revlogcompressor(self._compengineopts) |
|
818 | return engine.revlogcompressor(self._compengineopts) | |
819 |
|
819 | |||
820 | @util.propertycache |
|
820 | @util.propertycache | |
821 | def _decompressor(self): |
|
821 | def _decompressor(self): | |
822 | """the default decompressor""" |
|
822 | """the default decompressor""" | |
823 | if self._docket is None: |
|
823 | if self._docket is None: | |
824 | return None |
|
824 | return None | |
825 | t = self._docket.default_compression_header |
|
825 | t = self._docket.default_compression_header | |
826 | c = self._get_decompressor(t) |
|
826 | c = self._get_decompressor(t) | |
827 | return c.decompress |
|
827 | return c.decompress | |
828 |
|
828 | |||
829 | def _indexfp(self): |
|
829 | def _indexfp(self): | |
830 | """file object for the revlog's index file""" |
|
830 | """file object for the revlog's index file""" | |
831 | return self.opener(self._indexfile, mode=b"r") |
|
831 | return self.opener(self._indexfile, mode=b"r") | |
832 |
|
832 | |||
833 | def __index_write_fp(self): |
|
833 | def __index_write_fp(self): | |
834 | # You should not use this directly and use `_writing` instead |
|
834 | # You should not use this directly and use `_writing` instead | |
835 | try: |
|
835 | try: | |
836 | f = self.opener( |
|
836 | f = self.opener( | |
837 | self._indexfile, mode=b"r+", checkambig=self._checkambig |
|
837 | self._indexfile, mode=b"r+", checkambig=self._checkambig | |
838 | ) |
|
838 | ) | |
839 | if self._docket is None: |
|
839 | if self._docket is None: | |
840 | f.seek(0, os.SEEK_END) |
|
840 | f.seek(0, os.SEEK_END) | |
841 | else: |
|
841 | else: | |
842 | f.seek(self._docket.index_end, os.SEEK_SET) |
|
842 | f.seek(self._docket.index_end, os.SEEK_SET) | |
843 | return f |
|
843 | return f | |
844 | except FileNotFoundError: |
|
844 | except FileNotFoundError: | |
845 | return self.opener( |
|
845 | return self.opener( | |
846 | self._indexfile, mode=b"w+", checkambig=self._checkambig |
|
846 | self._indexfile, mode=b"w+", checkambig=self._checkambig | |
847 | ) |
|
847 | ) | |
848 |
|
848 | |||
849 | def __index_new_fp(self): |
|
849 | def __index_new_fp(self): | |
850 | # You should not use this unless you are upgrading from inline revlog |
|
850 | # You should not use this unless you are upgrading from inline revlog | |
851 | return self.opener( |
|
851 | return self.opener( | |
852 | self._indexfile, |
|
852 | self._indexfile, | |
853 | mode=b"w", |
|
853 | mode=b"w", | |
854 | checkambig=self._checkambig, |
|
854 | checkambig=self._checkambig, | |
855 | atomictemp=True, |
|
855 | atomictemp=True, | |
856 | ) |
|
856 | ) | |
857 |
|
857 | |||
858 | def _datafp(self, mode=b'r'): |
|
858 | def _datafp(self, mode=b'r'): | |
859 | """file object for the revlog's data file""" |
|
859 | """file object for the revlog's data file""" | |
860 | return self.opener(self._datafile, mode=mode) |
|
860 | return self.opener(self._datafile, mode=mode) | |
861 |
|
861 | |||
862 | @contextlib.contextmanager |
|
862 | @contextlib.contextmanager | |
863 | def _sidedatareadfp(self): |
|
863 | def _sidedatareadfp(self): | |
864 | """file object suitable to read sidedata""" |
|
864 | """file object suitable to read sidedata""" | |
865 | if self._writinghandles: |
|
865 | if self._writinghandles: | |
866 | yield self._writinghandles[2] |
|
866 | yield self._writinghandles[2] | |
867 | else: |
|
867 | else: | |
868 | with self.opener(self._sidedatafile) as fp: |
|
868 | with self.opener(self._sidedatafile) as fp: | |
869 | yield fp |
|
869 | yield fp | |
870 |
|
870 | |||
871 | def tiprev(self): |
|
871 | def tiprev(self): | |
872 | return len(self.index) - 1 |
|
872 | return len(self.index) - 1 | |
873 |
|
873 | |||
874 | def tip(self): |
|
874 | def tip(self): | |
875 | return self.node(self.tiprev()) |
|
875 | return self.node(self.tiprev()) | |
876 |
|
876 | |||
877 | def __contains__(self, rev): |
|
877 | def __contains__(self, rev): | |
878 | return 0 <= rev < len(self) |
|
878 | return 0 <= rev < len(self) | |
879 |
|
879 | |||
880 | def __len__(self): |
|
880 | def __len__(self): | |
881 | return len(self.index) |
|
881 | return len(self.index) | |
882 |
|
882 | |||
883 | def __iter__(self): |
|
883 | def __iter__(self): | |
884 | return iter(range(len(self))) |
|
884 | return iter(range(len(self))) | |
885 |
|
885 | |||
886 | def revs(self, start=0, stop=None): |
|
886 | def revs(self, start=0, stop=None): | |
887 | """iterate over all rev in this revlog (from start to stop)""" |
|
887 | """iterate over all rev in this revlog (from start to stop)""" | |
888 | return storageutil.iterrevs(len(self), start=start, stop=stop) |
|
888 | return storageutil.iterrevs(len(self), start=start, stop=stop) | |
889 |
|
889 | |||
890 | def hasnode(self, node): |
|
890 | def hasnode(self, node): | |
891 | try: |
|
891 | try: | |
892 | self.rev(node) |
|
892 | self.rev(node) | |
893 | return True |
|
893 | return True | |
894 | except KeyError: |
|
894 | except KeyError: | |
895 | return False |
|
895 | return False | |
896 |
|
896 | |||
897 | def _candelta(self, baserev, rev): |
|
897 | def _candelta(self, baserev, rev): | |
898 | """whether two revisions (baserev, rev) can be delta-ed or not""" |
|
898 | """whether two revisions (baserev, rev) can be delta-ed or not""" | |
899 | # Disable delta if either rev requires a content-changing flag |
|
899 | # Disable delta if either rev requires a content-changing flag | |
900 | # processor (ex. LFS). This is because such flag processor can alter |
|
900 | # processor (ex. LFS). This is because such flag processor can alter | |
901 | # the rawtext content that the delta will be based on, and two clients |
|
901 | # the rawtext content that the delta will be based on, and two clients | |
902 | # could have a same revlog node with different flags (i.e. different |
|
902 | # could have a same revlog node with different flags (i.e. different | |
903 | # rawtext contents) and the delta could be incompatible. |
|
903 | # rawtext contents) and the delta could be incompatible. | |
904 | if (self.flags(baserev) & REVIDX_RAWTEXT_CHANGING_FLAGS) or ( |
|
904 | if (self.flags(baserev) & REVIDX_RAWTEXT_CHANGING_FLAGS) or ( | |
905 | self.flags(rev) & REVIDX_RAWTEXT_CHANGING_FLAGS |
|
905 | self.flags(rev) & REVIDX_RAWTEXT_CHANGING_FLAGS | |
906 | ): |
|
906 | ): | |
907 | return False |
|
907 | return False | |
908 | return True |
|
908 | return True | |
909 |
|
909 | |||
910 | def update_caches(self, transaction): |
|
910 | def update_caches(self, transaction): | |
911 | """update on disk cache |
|
911 | """update on disk cache | |
912 |
|
912 | |||
913 | If a transaction is passed, the update may be delayed to transaction |
|
913 | If a transaction is passed, the update may be delayed to transaction | |
914 | commit.""" |
|
914 | commit.""" | |
915 | if self._nodemap_file is not None: |
|
915 | if self._nodemap_file is not None: | |
916 | if transaction is None: |
|
916 | if transaction is None: | |
917 | nodemaputil.update_persistent_nodemap(self) |
|
917 | nodemaputil.update_persistent_nodemap(self) | |
918 | else: |
|
918 | else: | |
919 | nodemaputil.setup_persistent_nodemap(transaction, self) |
|
919 | nodemaputil.setup_persistent_nodemap(transaction, self) | |
920 |
|
920 | |||
921 | def clearcaches(self): |
|
921 | def clearcaches(self): | |
922 | """Clear in-memory caches""" |
|
922 | """Clear in-memory caches""" | |
923 | self._revisioncache = None |
|
923 | self._revisioncache = None | |
924 | self._chainbasecache.clear() |
|
924 | self._chainbasecache.clear() | |
925 | self._segmentfile.clear_cache() |
|
925 | self._segmentfile.clear_cache() | |
926 | self._segmentfile_sidedata.clear_cache() |
|
926 | self._segmentfile_sidedata.clear_cache() | |
927 | self._pcache = {} |
|
927 | self._pcache = {} | |
928 | self._nodemap_docket = None |
|
928 | self._nodemap_docket = None | |
929 | self.index.clearcaches() |
|
929 | self.index.clearcaches() | |
930 | # The python code is the one responsible for validating the docket, we |
|
930 | # The python code is the one responsible for validating the docket, we | |
931 | # end up having to refresh it here. |
|
931 | # end up having to refresh it here. | |
932 | use_nodemap = ( |
|
932 | use_nodemap = ( | |
933 | not self._inline |
|
933 | not self._inline | |
934 | and self._nodemap_file is not None |
|
934 | and self._nodemap_file is not None | |
935 | and hasattr(self.index, 'update_nodemap_data') |
|
935 | and hasattr(self.index, 'update_nodemap_data') | |
936 | ) |
|
936 | ) | |
937 | if use_nodemap: |
|
937 | if use_nodemap: | |
938 | nodemap_data = nodemaputil.persisted_data(self) |
|
938 | nodemap_data = nodemaputil.persisted_data(self) | |
939 | if nodemap_data is not None: |
|
939 | if nodemap_data is not None: | |
940 | self._nodemap_docket = nodemap_data[0] |
|
940 | self._nodemap_docket = nodemap_data[0] | |
941 | self.index.update_nodemap_data(*nodemap_data) |
|
941 | self.index.update_nodemap_data(*nodemap_data) | |
942 |
|
942 | |||
943 | def rev(self, node): |
|
943 | def rev(self, node): | |
944 | """return the revision number associated with a <nodeid>""" |
|
944 | """return the revision number associated with a <nodeid>""" | |
945 | try: |
|
945 | try: | |
946 | return self.index.rev(node) |
|
946 | return self.index.rev(node) | |
947 | except TypeError: |
|
947 | except TypeError: | |
948 | raise |
|
948 | raise | |
949 | except error.RevlogError: |
|
949 | except error.RevlogError: | |
950 | # parsers.c radix tree lookup failed |
|
950 | # parsers.c radix tree lookup failed | |
951 | if ( |
|
951 | if ( | |
952 | node == self.nodeconstants.wdirid |
|
952 | node == self.nodeconstants.wdirid | |
953 | or node in self.nodeconstants.wdirfilenodeids |
|
953 | or node in self.nodeconstants.wdirfilenodeids | |
954 | ): |
|
954 | ): | |
955 | raise error.WdirUnsupported |
|
955 | raise error.WdirUnsupported | |
956 | raise error.LookupError(node, self.display_id, _(b'no node')) |
|
956 | raise error.LookupError(node, self.display_id, _(b'no node')) | |
957 |
|
957 | |||
958 | # Accessors for index entries. |
|
958 | # Accessors for index entries. | |
959 |
|
959 | |||
960 | # First tuple entry is 8 bytes. First 6 bytes are offset. Last 2 bytes |
|
960 | # First tuple entry is 8 bytes. First 6 bytes are offset. Last 2 bytes | |
961 | # are flags. |
|
961 | # are flags. | |
962 | def start(self, rev): |
|
962 | def start(self, rev): | |
963 | return int(self.index[rev][0] >> 16) |
|
963 | return int(self.index[rev][0] >> 16) | |
964 |
|
964 | |||
965 | def sidedata_cut_off(self, rev): |
|
965 | def sidedata_cut_off(self, rev): | |
966 | sd_cut_off = self.index[rev][8] |
|
966 | sd_cut_off = self.index[rev][8] | |
967 | if sd_cut_off != 0: |
|
967 | if sd_cut_off != 0: | |
968 | return sd_cut_off |
|
968 | return sd_cut_off | |
969 | # This is some annoying dance, because entries without sidedata |
|
969 | # This is some annoying dance, because entries without sidedata | |
970 | # currently use 0 as their ofsset. (instead of previous-offset + |
|
970 | # currently use 0 as their ofsset. (instead of previous-offset + | |
971 | # previous-size) |
|
971 | # previous-size) | |
972 | # |
|
972 | # | |
973 | # We should reconsider this sidedata β 0 sidata_offset policy. |
|
973 | # We should reconsider this sidedata β 0 sidata_offset policy. | |
974 | # In the meantime, we need this. |
|
974 | # In the meantime, we need this. | |
975 | while 0 <= rev: |
|
975 | while 0 <= rev: | |
976 | e = self.index[rev] |
|
976 | e = self.index[rev] | |
977 | if e[9] != 0: |
|
977 | if e[9] != 0: | |
978 | return e[8] + e[9] |
|
978 | return e[8] + e[9] | |
979 | rev -= 1 |
|
979 | rev -= 1 | |
980 | return 0 |
|
980 | return 0 | |
981 |
|
981 | |||
982 | def flags(self, rev): |
|
982 | def flags(self, rev): | |
983 | return self.index[rev][0] & 0xFFFF |
|
983 | return self.index[rev][0] & 0xFFFF | |
984 |
|
984 | |||
985 | def length(self, rev): |
|
985 | def length(self, rev): | |
986 | return self.index[rev][1] |
|
986 | return self.index[rev][1] | |
987 |
|
987 | |||
988 | def sidedata_length(self, rev): |
|
988 | def sidedata_length(self, rev): | |
989 | if not self.hassidedata: |
|
989 | if not self.hassidedata: | |
990 | return 0 |
|
990 | return 0 | |
991 | return self.index[rev][9] |
|
991 | return self.index[rev][9] | |
992 |
|
992 | |||
993 | def rawsize(self, rev): |
|
993 | def rawsize(self, rev): | |
994 | """return the length of the uncompressed text for a given revision""" |
|
994 | """return the length of the uncompressed text for a given revision""" | |
995 | l = self.index[rev][2] |
|
995 | l = self.index[rev][2] | |
996 | if l >= 0: |
|
996 | if l >= 0: | |
997 | return l |
|
997 | return l | |
998 |
|
998 | |||
999 | t = self.rawdata(rev) |
|
999 | t = self.rawdata(rev) | |
1000 | return len(t) |
|
1000 | return len(t) | |
1001 |
|
1001 | |||
1002 | def size(self, rev): |
|
1002 | def size(self, rev): | |
1003 | """length of non-raw text (processed by a "read" flag processor)""" |
|
1003 | """length of non-raw text (processed by a "read" flag processor)""" | |
1004 | # fast path: if no "read" flag processor could change the content, |
|
1004 | # fast path: if no "read" flag processor could change the content, | |
1005 | # size is rawsize. note: ELLIPSIS is known to not change the content. |
|
1005 | # size is rawsize. note: ELLIPSIS is known to not change the content. | |
1006 | flags = self.flags(rev) |
|
1006 | flags = self.flags(rev) | |
1007 | if flags & (flagutil.REVIDX_KNOWN_FLAGS ^ REVIDX_ELLIPSIS) == 0: |
|
1007 | if flags & (flagutil.REVIDX_KNOWN_FLAGS ^ REVIDX_ELLIPSIS) == 0: | |
1008 | return self.rawsize(rev) |
|
1008 | return self.rawsize(rev) | |
1009 |
|
1009 | |||
1010 | return len(self.revision(rev)) |
|
1010 | return len(self.revision(rev)) | |
1011 |
|
1011 | |||
1012 | def fast_rank(self, rev): |
|
1012 | def fast_rank(self, rev): | |
1013 | """Return the rank of a revision if already known, or None otherwise. |
|
1013 | """Return the rank of a revision if already known, or None otherwise. | |
1014 |
|
1014 | |||
1015 | The rank of a revision is the size of the sub-graph it defines as a |
|
1015 | The rank of a revision is the size of the sub-graph it defines as a | |
1016 | head. Equivalently, the rank of a revision `r` is the size of the set |
|
1016 | head. Equivalently, the rank of a revision `r` is the size of the set | |
1017 | `ancestors(r)`, `r` included. |
|
1017 | `ancestors(r)`, `r` included. | |
1018 |
|
1018 | |||
1019 | This method returns the rank retrieved from the revlog in constant |
|
1019 | This method returns the rank retrieved from the revlog in constant | |
1020 | time. It makes no attempt at computing unknown values for versions of |
|
1020 | time. It makes no attempt at computing unknown values for versions of | |
1021 | the revlog which do not persist the rank. |
|
1021 | the revlog which do not persist the rank. | |
1022 | """ |
|
1022 | """ | |
1023 | rank = self.index[rev][ENTRY_RANK] |
|
1023 | rank = self.index[rev][ENTRY_RANK] | |
1024 | if self._format_version != CHANGELOGV2 or rank == RANK_UNKNOWN: |
|
1024 | if self._format_version != CHANGELOGV2 or rank == RANK_UNKNOWN: | |
1025 | return None |
|
1025 | return None | |
1026 | if rev == nullrev: |
|
1026 | if rev == nullrev: | |
1027 | return 0 # convention |
|
1027 | return 0 # convention | |
1028 | return rank |
|
1028 | return rank | |
1029 |
|
1029 | |||
1030 | def chainbase(self, rev): |
|
1030 | def chainbase(self, rev): | |
1031 | base = self._chainbasecache.get(rev) |
|
1031 | base = self._chainbasecache.get(rev) | |
1032 | if base is not None: |
|
1032 | if base is not None: | |
1033 | return base |
|
1033 | return base | |
1034 |
|
1034 | |||
1035 | index = self.index |
|
1035 | index = self.index | |
1036 | iterrev = rev |
|
1036 | iterrev = rev | |
1037 | base = index[iterrev][3] |
|
1037 | base = index[iterrev][3] | |
1038 | while base != iterrev: |
|
1038 | while base != iterrev: | |
1039 | iterrev = base |
|
1039 | iterrev = base | |
1040 | base = index[iterrev][3] |
|
1040 | base = index[iterrev][3] | |
1041 |
|
1041 | |||
1042 | self._chainbasecache[rev] = base |
|
1042 | self._chainbasecache[rev] = base | |
1043 | return base |
|
1043 | return base | |
1044 |
|
1044 | |||
1045 | def linkrev(self, rev): |
|
1045 | def linkrev(self, rev): | |
1046 | return self.index[rev][4] |
|
1046 | return self.index[rev][4] | |
1047 |
|
1047 | |||
1048 | def parentrevs(self, rev): |
|
1048 | def parentrevs(self, rev): | |
1049 | try: |
|
1049 | try: | |
1050 | entry = self.index[rev] |
|
1050 | entry = self.index[rev] | |
1051 | except IndexError: |
|
1051 | except IndexError: | |
1052 | if rev == wdirrev: |
|
1052 | if rev == wdirrev: | |
1053 | raise error.WdirUnsupported |
|
1053 | raise error.WdirUnsupported | |
1054 | raise |
|
1054 | raise | |
1055 |
|
1055 | |||
1056 | if self.canonical_parent_order and entry[5] == nullrev: |
|
1056 | if self.canonical_parent_order and entry[5] == nullrev: | |
1057 | return entry[6], entry[5] |
|
1057 | return entry[6], entry[5] | |
1058 | else: |
|
1058 | else: | |
1059 | return entry[5], entry[6] |
|
1059 | return entry[5], entry[6] | |
1060 |
|
1060 | |||
1061 | # fast parentrevs(rev) where rev isn't filtered |
|
1061 | # fast parentrevs(rev) where rev isn't filtered | |
1062 | _uncheckedparentrevs = parentrevs |
|
1062 | _uncheckedparentrevs = parentrevs | |
1063 |
|
1063 | |||
1064 | def node(self, rev): |
|
1064 | def node(self, rev): | |
1065 | try: |
|
1065 | try: | |
1066 | return self.index[rev][7] |
|
1066 | return self.index[rev][7] | |
1067 | except IndexError: |
|
1067 | except IndexError: | |
1068 | if rev == wdirrev: |
|
1068 | if rev == wdirrev: | |
1069 | raise error.WdirUnsupported |
|
1069 | raise error.WdirUnsupported | |
1070 | raise |
|
1070 | raise | |
1071 |
|
1071 | |||
1072 | # Derived from index values. |
|
1072 | # Derived from index values. | |
1073 |
|
1073 | |||
1074 | def end(self, rev): |
|
1074 | def end(self, rev): | |
1075 | return self.start(rev) + self.length(rev) |
|
1075 | return self.start(rev) + self.length(rev) | |
1076 |
|
1076 | |||
1077 | def parents(self, node): |
|
1077 | def parents(self, node): | |
1078 | i = self.index |
|
1078 | i = self.index | |
1079 | d = i[self.rev(node)] |
|
1079 | d = i[self.rev(node)] | |
1080 | # inline node() to avoid function call overhead |
|
1080 | # inline node() to avoid function call overhead | |
1081 | if self.canonical_parent_order and d[5] == self.nullid: |
|
1081 | if self.canonical_parent_order and d[5] == self.nullid: | |
1082 | return i[d[6]][7], i[d[5]][7] |
|
1082 | return i[d[6]][7], i[d[5]][7] | |
1083 | else: |
|
1083 | else: | |
1084 | return i[d[5]][7], i[d[6]][7] |
|
1084 | return i[d[5]][7], i[d[6]][7] | |
1085 |
|
1085 | |||
1086 | def chainlen(self, rev): |
|
1086 | def chainlen(self, rev): | |
1087 | return self._chaininfo(rev)[0] |
|
1087 | return self._chaininfo(rev)[0] | |
1088 |
|
1088 | |||
1089 | def _chaininfo(self, rev): |
|
1089 | def _chaininfo(self, rev): | |
1090 | chaininfocache = self._chaininfocache |
|
1090 | chaininfocache = self._chaininfocache | |
1091 | if rev in chaininfocache: |
|
1091 | if rev in chaininfocache: | |
1092 | return chaininfocache[rev] |
|
1092 | return chaininfocache[rev] | |
1093 | index = self.index |
|
1093 | index = self.index | |
1094 | generaldelta = self._generaldelta |
|
1094 | generaldelta = self._generaldelta | |
1095 | iterrev = rev |
|
1095 | iterrev = rev | |
1096 | e = index[iterrev] |
|
1096 | e = index[iterrev] | |
1097 | clen = 0 |
|
1097 | clen = 0 | |
1098 | compresseddeltalen = 0 |
|
1098 | compresseddeltalen = 0 | |
1099 | while iterrev != e[3]: |
|
1099 | while iterrev != e[3]: | |
1100 | clen += 1 |
|
1100 | clen += 1 | |
1101 | compresseddeltalen += e[1] |
|
1101 | compresseddeltalen += e[1] | |
1102 | if generaldelta: |
|
1102 | if generaldelta: | |
1103 | iterrev = e[3] |
|
1103 | iterrev = e[3] | |
1104 | else: |
|
1104 | else: | |
1105 | iterrev -= 1 |
|
1105 | iterrev -= 1 | |
1106 | if iterrev in chaininfocache: |
|
1106 | if iterrev in chaininfocache: | |
1107 | t = chaininfocache[iterrev] |
|
1107 | t = chaininfocache[iterrev] | |
1108 | clen += t[0] |
|
1108 | clen += t[0] | |
1109 | compresseddeltalen += t[1] |
|
1109 | compresseddeltalen += t[1] | |
1110 | break |
|
1110 | break | |
1111 | e = index[iterrev] |
|
1111 | e = index[iterrev] | |
1112 | else: |
|
1112 | else: | |
1113 | # Add text length of base since decompressing that also takes |
|
1113 | # Add text length of base since decompressing that also takes | |
1114 | # work. For cache hits the length is already included. |
|
1114 | # work. For cache hits the length is already included. | |
1115 | compresseddeltalen += e[1] |
|
1115 | compresseddeltalen += e[1] | |
1116 | r = (clen, compresseddeltalen) |
|
1116 | r = (clen, compresseddeltalen) | |
1117 | chaininfocache[rev] = r |
|
1117 | chaininfocache[rev] = r | |
1118 | return r |
|
1118 | return r | |
1119 |
|
1119 | |||
1120 | def _deltachain(self, rev, stoprev=None): |
|
1120 | def _deltachain(self, rev, stoprev=None): | |
1121 | """Obtain the delta chain for a revision. |
|
1121 | """Obtain the delta chain for a revision. | |
1122 |
|
1122 | |||
1123 | ``stoprev`` specifies a revision to stop at. If not specified, we |
|
1123 | ``stoprev`` specifies a revision to stop at. If not specified, we | |
1124 | stop at the base of the chain. |
|
1124 | stop at the base of the chain. | |
1125 |
|
1125 | |||
1126 | Returns a 2-tuple of (chain, stopped) where ``chain`` is a list of |
|
1126 | Returns a 2-tuple of (chain, stopped) where ``chain`` is a list of | |
1127 | revs in ascending order and ``stopped`` is a bool indicating whether |
|
1127 | revs in ascending order and ``stopped`` is a bool indicating whether | |
1128 | ``stoprev`` was hit. |
|
1128 | ``stoprev`` was hit. | |
1129 | """ |
|
1129 | """ | |
1130 | # Try C implementation. |
|
1130 | # Try C implementation. | |
1131 | try: |
|
1131 | try: | |
1132 | return self.index.deltachain(rev, stoprev, self._generaldelta) |
|
1132 | return self.index.deltachain(rev, stoprev, self._generaldelta) | |
1133 | except AttributeError: |
|
1133 | except AttributeError: | |
1134 | pass |
|
1134 | pass | |
1135 |
|
1135 | |||
1136 | chain = [] |
|
1136 | chain = [] | |
1137 |
|
1137 | |||
1138 | # Alias to prevent attribute lookup in tight loop. |
|
1138 | # Alias to prevent attribute lookup in tight loop. | |
1139 | index = self.index |
|
1139 | index = self.index | |
1140 | generaldelta = self._generaldelta |
|
1140 | generaldelta = self._generaldelta | |
1141 |
|
1141 | |||
1142 | iterrev = rev |
|
1142 | iterrev = rev | |
1143 | e = index[iterrev] |
|
1143 | e = index[iterrev] | |
1144 | while iterrev != e[3] and iterrev != stoprev: |
|
1144 | while iterrev != e[3] and iterrev != stoprev: | |
1145 | chain.append(iterrev) |
|
1145 | chain.append(iterrev) | |
1146 | if generaldelta: |
|
1146 | if generaldelta: | |
1147 | iterrev = e[3] |
|
1147 | iterrev = e[3] | |
1148 | else: |
|
1148 | else: | |
1149 | iterrev -= 1 |
|
1149 | iterrev -= 1 | |
1150 | e = index[iterrev] |
|
1150 | e = index[iterrev] | |
1151 |
|
1151 | |||
1152 | if iterrev == stoprev: |
|
1152 | if iterrev == stoprev: | |
1153 | stopped = True |
|
1153 | stopped = True | |
1154 | else: |
|
1154 | else: | |
1155 | chain.append(iterrev) |
|
1155 | chain.append(iterrev) | |
1156 | stopped = False |
|
1156 | stopped = False | |
1157 |
|
1157 | |||
1158 | chain.reverse() |
|
1158 | chain.reverse() | |
1159 | return chain, stopped |
|
1159 | return chain, stopped | |
1160 |
|
1160 | |||
1161 | def ancestors(self, revs, stoprev=0, inclusive=False): |
|
1161 | def ancestors(self, revs, stoprev=0, inclusive=False): | |
1162 | """Generate the ancestors of 'revs' in reverse revision order. |
|
1162 | """Generate the ancestors of 'revs' in reverse revision order. | |
1163 | Does not generate revs lower than stoprev. |
|
1163 | Does not generate revs lower than stoprev. | |
1164 |
|
1164 | |||
1165 | See the documentation for ancestor.lazyancestors for more details.""" |
|
1165 | See the documentation for ancestor.lazyancestors for more details.""" | |
1166 |
|
1166 | |||
1167 | # first, make sure start revisions aren't filtered |
|
1167 | # first, make sure start revisions aren't filtered | |
1168 | revs = list(revs) |
|
1168 | revs = list(revs) | |
1169 | checkrev = self.node |
|
1169 | checkrev = self.node | |
1170 | for r in revs: |
|
1170 | for r in revs: | |
1171 | checkrev(r) |
|
1171 | checkrev(r) | |
1172 | # and we're sure ancestors aren't filtered as well |
|
1172 | # and we're sure ancestors aren't filtered as well | |
1173 |
|
1173 | |||
1174 | if rustancestor is not None and self.index.rust_ext_compat: |
|
1174 | if rustancestor is not None and self.index.rust_ext_compat: | |
1175 | lazyancestors = rustancestor.LazyAncestors |
|
1175 | lazyancestors = rustancestor.LazyAncestors | |
1176 | arg = self.index |
|
1176 | arg = self.index | |
1177 | else: |
|
1177 | else: | |
1178 | lazyancestors = ancestor.lazyancestors |
|
1178 | lazyancestors = ancestor.lazyancestors | |
1179 | arg = self._uncheckedparentrevs |
|
1179 | arg = self._uncheckedparentrevs | |
1180 | return lazyancestors(arg, revs, stoprev=stoprev, inclusive=inclusive) |
|
1180 | return lazyancestors(arg, revs, stoprev=stoprev, inclusive=inclusive) | |
1181 |
|
1181 | |||
1182 | def descendants(self, revs): |
|
1182 | def descendants(self, revs): | |
1183 | return dagop.descendantrevs(revs, self.revs, self.parentrevs) |
|
1183 | return dagop.descendantrevs(revs, self.revs, self.parentrevs) | |
1184 |
|
1184 | |||
1185 | def findcommonmissing(self, common=None, heads=None): |
|
1185 | def findcommonmissing(self, common=None, heads=None): | |
1186 | """Return a tuple of the ancestors of common and the ancestors of heads |
|
1186 | """Return a tuple of the ancestors of common and the ancestors of heads | |
1187 | that are not ancestors of common. In revset terminology, we return the |
|
1187 | that are not ancestors of common. In revset terminology, we return the | |
1188 | tuple: |
|
1188 | tuple: | |
1189 |
|
1189 | |||
1190 | ::common, (::heads) - (::common) |
|
1190 | ::common, (::heads) - (::common) | |
1191 |
|
1191 | |||
1192 | The list is sorted by revision number, meaning it is |
|
1192 | The list is sorted by revision number, meaning it is | |
1193 | topologically sorted. |
|
1193 | topologically sorted. | |
1194 |
|
1194 | |||
1195 | 'heads' and 'common' are both lists of node IDs. If heads is |
|
1195 | 'heads' and 'common' are both lists of node IDs. If heads is | |
1196 | not supplied, uses all of the revlog's heads. If common is not |
|
1196 | not supplied, uses all of the revlog's heads. If common is not | |
1197 | supplied, uses nullid.""" |
|
1197 | supplied, uses nullid.""" | |
1198 | if common is None: |
|
1198 | if common is None: | |
1199 | common = [self.nullid] |
|
1199 | common = [self.nullid] | |
1200 | if heads is None: |
|
1200 | if heads is None: | |
1201 | heads = self.heads() |
|
1201 | heads = self.heads() | |
1202 |
|
1202 | |||
1203 | common = [self.rev(n) for n in common] |
|
1203 | common = [self.rev(n) for n in common] | |
1204 | heads = [self.rev(n) for n in heads] |
|
1204 | heads = [self.rev(n) for n in heads] | |
1205 |
|
1205 | |||
1206 | # we want the ancestors, but inclusive |
|
1206 | # we want the ancestors, but inclusive | |
1207 | class lazyset: |
|
1207 | class lazyset: | |
1208 | def __init__(self, lazyvalues): |
|
1208 | def __init__(self, lazyvalues): | |
1209 | self.addedvalues = set() |
|
1209 | self.addedvalues = set() | |
1210 | self.lazyvalues = lazyvalues |
|
1210 | self.lazyvalues = lazyvalues | |
1211 |
|
1211 | |||
1212 | def __contains__(self, value): |
|
1212 | def __contains__(self, value): | |
1213 | return value in self.addedvalues or value in self.lazyvalues |
|
1213 | return value in self.addedvalues or value in self.lazyvalues | |
1214 |
|
1214 | |||
1215 | def __iter__(self): |
|
1215 | def __iter__(self): | |
1216 | added = self.addedvalues |
|
1216 | added = self.addedvalues | |
1217 | for r in added: |
|
1217 | for r in added: | |
1218 | yield r |
|
1218 | yield r | |
1219 | for r in self.lazyvalues: |
|
1219 | for r in self.lazyvalues: | |
1220 | if not r in added: |
|
1220 | if not r in added: | |
1221 | yield r |
|
1221 | yield r | |
1222 |
|
1222 | |||
1223 | def add(self, value): |
|
1223 | def add(self, value): | |
1224 | self.addedvalues.add(value) |
|
1224 | self.addedvalues.add(value) | |
1225 |
|
1225 | |||
1226 | def update(self, values): |
|
1226 | def update(self, values): | |
1227 | self.addedvalues.update(values) |
|
1227 | self.addedvalues.update(values) | |
1228 |
|
1228 | |||
1229 | has = lazyset(self.ancestors(common)) |
|
1229 | has = lazyset(self.ancestors(common)) | |
1230 | has.add(nullrev) |
|
1230 | has.add(nullrev) | |
1231 | has.update(common) |
|
1231 | has.update(common) | |
1232 |
|
1232 | |||
1233 | # take all ancestors from heads that aren't in has |
|
1233 | # take all ancestors from heads that aren't in has | |
1234 | missing = set() |
|
1234 | missing = set() | |
1235 | visit = collections.deque(r for r in heads if r not in has) |
|
1235 | visit = collections.deque(r for r in heads if r not in has) | |
1236 | while visit: |
|
1236 | while visit: | |
1237 | r = visit.popleft() |
|
1237 | r = visit.popleft() | |
1238 | if r in missing: |
|
1238 | if r in missing: | |
1239 | continue |
|
1239 | continue | |
1240 | else: |
|
1240 | else: | |
1241 | missing.add(r) |
|
1241 | missing.add(r) | |
1242 | for p in self.parentrevs(r): |
|
1242 | for p in self.parentrevs(r): | |
1243 | if p not in has: |
|
1243 | if p not in has: | |
1244 | visit.append(p) |
|
1244 | visit.append(p) | |
1245 | missing = list(missing) |
|
1245 | missing = list(missing) | |
1246 | missing.sort() |
|
1246 | missing.sort() | |
1247 | return has, [self.node(miss) for miss in missing] |
|
1247 | return has, [self.node(miss) for miss in missing] | |
1248 |
|
1248 | |||
1249 | def incrementalmissingrevs(self, common=None): |
|
1249 | def incrementalmissingrevs(self, common=None): | |
1250 | """Return an object that can be used to incrementally compute the |
|
1250 | """Return an object that can be used to incrementally compute the | |
1251 | revision numbers of the ancestors of arbitrary sets that are not |
|
1251 | revision numbers of the ancestors of arbitrary sets that are not | |
1252 | ancestors of common. This is an ancestor.incrementalmissingancestors |
|
1252 | ancestors of common. This is an ancestor.incrementalmissingancestors | |
1253 | object. |
|
1253 | object. | |
1254 |
|
1254 | |||
1255 | 'common' is a list of revision numbers. If common is not supplied, uses |
|
1255 | 'common' is a list of revision numbers. If common is not supplied, uses | |
1256 | nullrev. |
|
1256 | nullrev. | |
1257 | """ |
|
1257 | """ | |
1258 | if common is None: |
|
1258 | if common is None: | |
1259 | common = [nullrev] |
|
1259 | common = [nullrev] | |
1260 |
|
1260 | |||
1261 | if rustancestor is not None and self.index.rust_ext_compat: |
|
1261 | if rustancestor is not None and self.index.rust_ext_compat: | |
1262 | return rustancestor.MissingAncestors(self.index, common) |
|
1262 | return rustancestor.MissingAncestors(self.index, common) | |
1263 | return ancestor.incrementalmissingancestors(self.parentrevs, common) |
|
1263 | return ancestor.incrementalmissingancestors(self.parentrevs, common) | |
1264 |
|
1264 | |||
1265 | def findmissingrevs(self, common=None, heads=None): |
|
1265 | def findmissingrevs(self, common=None, heads=None): | |
1266 | """Return the revision numbers of the ancestors of heads that |
|
1266 | """Return the revision numbers of the ancestors of heads that | |
1267 | are not ancestors of common. |
|
1267 | are not ancestors of common. | |
1268 |
|
1268 | |||
1269 | More specifically, return a list of revision numbers corresponding to |
|
1269 | More specifically, return a list of revision numbers corresponding to | |
1270 | nodes N such that every N satisfies the following constraints: |
|
1270 | nodes N such that every N satisfies the following constraints: | |
1271 |
|
1271 | |||
1272 | 1. N is an ancestor of some node in 'heads' |
|
1272 | 1. N is an ancestor of some node in 'heads' | |
1273 | 2. N is not an ancestor of any node in 'common' |
|
1273 | 2. N is not an ancestor of any node in 'common' | |
1274 |
|
1274 | |||
1275 | The list is sorted by revision number, meaning it is |
|
1275 | The list is sorted by revision number, meaning it is | |
1276 | topologically sorted. |
|
1276 | topologically sorted. | |
1277 |
|
1277 | |||
1278 | 'heads' and 'common' are both lists of revision numbers. If heads is |
|
1278 | 'heads' and 'common' are both lists of revision numbers. If heads is | |
1279 | not supplied, uses all of the revlog's heads. If common is not |
|
1279 | not supplied, uses all of the revlog's heads. If common is not | |
1280 | supplied, uses nullid.""" |
|
1280 | supplied, uses nullid.""" | |
1281 | if common is None: |
|
1281 | if common is None: | |
1282 | common = [nullrev] |
|
1282 | common = [nullrev] | |
1283 | if heads is None: |
|
1283 | if heads is None: | |
1284 | heads = self.headrevs() |
|
1284 | heads = self.headrevs() | |
1285 |
|
1285 | |||
1286 | inc = self.incrementalmissingrevs(common=common) |
|
1286 | inc = self.incrementalmissingrevs(common=common) | |
1287 | return inc.missingancestors(heads) |
|
1287 | return inc.missingancestors(heads) | |
1288 |
|
1288 | |||
1289 | def findmissing(self, common=None, heads=None): |
|
1289 | def findmissing(self, common=None, heads=None): | |
1290 | """Return the ancestors of heads that are not ancestors of common. |
|
1290 | """Return the ancestors of heads that are not ancestors of common. | |
1291 |
|
1291 | |||
1292 | More specifically, return a list of nodes N such that every N |
|
1292 | More specifically, return a list of nodes N such that every N | |
1293 | satisfies the following constraints: |
|
1293 | satisfies the following constraints: | |
1294 |
|
1294 | |||
1295 | 1. N is an ancestor of some node in 'heads' |
|
1295 | 1. N is an ancestor of some node in 'heads' | |
1296 | 2. N is not an ancestor of any node in 'common' |
|
1296 | 2. N is not an ancestor of any node in 'common' | |
1297 |
|
1297 | |||
1298 | The list is sorted by revision number, meaning it is |
|
1298 | The list is sorted by revision number, meaning it is | |
1299 | topologically sorted. |
|
1299 | topologically sorted. | |
1300 |
|
1300 | |||
1301 | 'heads' and 'common' are both lists of node IDs. If heads is |
|
1301 | 'heads' and 'common' are both lists of node IDs. If heads is | |
1302 | not supplied, uses all of the revlog's heads. If common is not |
|
1302 | not supplied, uses all of the revlog's heads. If common is not | |
1303 | supplied, uses nullid.""" |
|
1303 | supplied, uses nullid.""" | |
1304 | if common is None: |
|
1304 | if common is None: | |
1305 | common = [self.nullid] |
|
1305 | common = [self.nullid] | |
1306 | if heads is None: |
|
1306 | if heads is None: | |
1307 | heads = self.heads() |
|
1307 | heads = self.heads() | |
1308 |
|
1308 | |||
1309 | common = [self.rev(n) for n in common] |
|
1309 | common = [self.rev(n) for n in common] | |
1310 | heads = [self.rev(n) for n in heads] |
|
1310 | heads = [self.rev(n) for n in heads] | |
1311 |
|
1311 | |||
1312 | inc = self.incrementalmissingrevs(common=common) |
|
1312 | inc = self.incrementalmissingrevs(common=common) | |
1313 | return [self.node(r) for r in inc.missingancestors(heads)] |
|
1313 | return [self.node(r) for r in inc.missingancestors(heads)] | |
1314 |
|
1314 | |||
1315 | def nodesbetween(self, roots=None, heads=None): |
|
1315 | def nodesbetween(self, roots=None, heads=None): | |
1316 | """Return a topological path from 'roots' to 'heads'. |
|
1316 | """Return a topological path from 'roots' to 'heads'. | |
1317 |
|
1317 | |||
1318 | Return a tuple (nodes, outroots, outheads) where 'nodes' is a |
|
1318 | Return a tuple (nodes, outroots, outheads) where 'nodes' is a | |
1319 | topologically sorted list of all nodes N that satisfy both of |
|
1319 | topologically sorted list of all nodes N that satisfy both of | |
1320 | these constraints: |
|
1320 | these constraints: | |
1321 |
|
1321 | |||
1322 | 1. N is a descendant of some node in 'roots' |
|
1322 | 1. N is a descendant of some node in 'roots' | |
1323 | 2. N is an ancestor of some node in 'heads' |
|
1323 | 2. N is an ancestor of some node in 'heads' | |
1324 |
|
1324 | |||
1325 | Every node is considered to be both a descendant and an ancestor |
|
1325 | Every node is considered to be both a descendant and an ancestor | |
1326 | of itself, so every reachable node in 'roots' and 'heads' will be |
|
1326 | of itself, so every reachable node in 'roots' and 'heads' will be | |
1327 | included in 'nodes'. |
|
1327 | included in 'nodes'. | |
1328 |
|
1328 | |||
1329 | 'outroots' is the list of reachable nodes in 'roots', i.e., the |
|
1329 | 'outroots' is the list of reachable nodes in 'roots', i.e., the | |
1330 | subset of 'roots' that is returned in 'nodes'. Likewise, |
|
1330 | subset of 'roots' that is returned in 'nodes'. Likewise, | |
1331 | 'outheads' is the subset of 'heads' that is also in 'nodes'. |
|
1331 | 'outheads' is the subset of 'heads' that is also in 'nodes'. | |
1332 |
|
1332 | |||
1333 | 'roots' and 'heads' are both lists of node IDs. If 'roots' is |
|
1333 | 'roots' and 'heads' are both lists of node IDs. If 'roots' is | |
1334 | unspecified, uses nullid as the only root. If 'heads' is |
|
1334 | unspecified, uses nullid as the only root. If 'heads' is | |
1335 | unspecified, uses list of all of the revlog's heads.""" |
|
1335 | unspecified, uses list of all of the revlog's heads.""" | |
1336 | nonodes = ([], [], []) |
|
1336 | nonodes = ([], [], []) | |
1337 | if roots is not None: |
|
1337 | if roots is not None: | |
1338 | roots = list(roots) |
|
1338 | roots = list(roots) | |
1339 | if not roots: |
|
1339 | if not roots: | |
1340 | return nonodes |
|
1340 | return nonodes | |
1341 | lowestrev = min([self.rev(n) for n in roots]) |
|
1341 | lowestrev = min([self.rev(n) for n in roots]) | |
1342 | else: |
|
1342 | else: | |
1343 | roots = [self.nullid] # Everybody's a descendant of nullid |
|
1343 | roots = [self.nullid] # Everybody's a descendant of nullid | |
1344 | lowestrev = nullrev |
|
1344 | lowestrev = nullrev | |
1345 | if (lowestrev == nullrev) and (heads is None): |
|
1345 | if (lowestrev == nullrev) and (heads is None): | |
1346 | # We want _all_ the nodes! |
|
1346 | # We want _all_ the nodes! | |
1347 | return ( |
|
1347 | return ( | |
1348 | [self.node(r) for r in self], |
|
1348 | [self.node(r) for r in self], | |
1349 | [self.nullid], |
|
1349 | [self.nullid], | |
1350 | list(self.heads()), |
|
1350 | list(self.heads()), | |
1351 | ) |
|
1351 | ) | |
1352 | if heads is None: |
|
1352 | if heads is None: | |
1353 | # All nodes are ancestors, so the latest ancestor is the last |
|
1353 | # All nodes are ancestors, so the latest ancestor is the last | |
1354 | # node. |
|
1354 | # node. | |
1355 | highestrev = len(self) - 1 |
|
1355 | highestrev = len(self) - 1 | |
1356 | # Set ancestors to None to signal that every node is an ancestor. |
|
1356 | # Set ancestors to None to signal that every node is an ancestor. | |
1357 | ancestors = None |
|
1357 | ancestors = None | |
1358 | # Set heads to an empty dictionary for later discovery of heads |
|
1358 | # Set heads to an empty dictionary for later discovery of heads | |
1359 | heads = {} |
|
1359 | heads = {} | |
1360 | else: |
|
1360 | else: | |
1361 | heads = list(heads) |
|
1361 | heads = list(heads) | |
1362 | if not heads: |
|
1362 | if not heads: | |
1363 | return nonodes |
|
1363 | return nonodes | |
1364 | ancestors = set() |
|
1364 | ancestors = set() | |
1365 | # Turn heads into a dictionary so we can remove 'fake' heads. |
|
1365 | # Turn heads into a dictionary so we can remove 'fake' heads. | |
1366 | # Also, later we will be using it to filter out the heads we can't |
|
1366 | # Also, later we will be using it to filter out the heads we can't | |
1367 | # find from roots. |
|
1367 | # find from roots. | |
1368 | heads = dict.fromkeys(heads, False) |
|
1368 | heads = dict.fromkeys(heads, False) | |
1369 | # Start at the top and keep marking parents until we're done. |
|
1369 | # Start at the top and keep marking parents until we're done. | |
1370 | nodestotag = set(heads) |
|
1370 | nodestotag = set(heads) | |
1371 | # Remember where the top was so we can use it as a limit later. |
|
1371 | # Remember where the top was so we can use it as a limit later. | |
1372 | highestrev = max([self.rev(n) for n in nodestotag]) |
|
1372 | highestrev = max([self.rev(n) for n in nodestotag]) | |
1373 | while nodestotag: |
|
1373 | while nodestotag: | |
1374 | # grab a node to tag |
|
1374 | # grab a node to tag | |
1375 | n = nodestotag.pop() |
|
1375 | n = nodestotag.pop() | |
1376 | # Never tag nullid |
|
1376 | # Never tag nullid | |
1377 | if n == self.nullid: |
|
1377 | if n == self.nullid: | |
1378 | continue |
|
1378 | continue | |
1379 | # A node's revision number represents its place in a |
|
1379 | # A node's revision number represents its place in a | |
1380 | # topologically sorted list of nodes. |
|
1380 | # topologically sorted list of nodes. | |
1381 | r = self.rev(n) |
|
1381 | r = self.rev(n) | |
1382 | if r >= lowestrev: |
|
1382 | if r >= lowestrev: | |
1383 | if n not in ancestors: |
|
1383 | if n not in ancestors: | |
1384 | # If we are possibly a descendant of one of the roots |
|
1384 | # If we are possibly a descendant of one of the roots | |
1385 | # and we haven't already been marked as an ancestor |
|
1385 | # and we haven't already been marked as an ancestor | |
1386 | ancestors.add(n) # Mark as ancestor |
|
1386 | ancestors.add(n) # Mark as ancestor | |
1387 | # Add non-nullid parents to list of nodes to tag. |
|
1387 | # Add non-nullid parents to list of nodes to tag. | |
1388 | nodestotag.update( |
|
1388 | nodestotag.update( | |
1389 | [p for p in self.parents(n) if p != self.nullid] |
|
1389 | [p for p in self.parents(n) if p != self.nullid] | |
1390 | ) |
|
1390 | ) | |
1391 | elif n in heads: # We've seen it before, is it a fake head? |
|
1391 | elif n in heads: # We've seen it before, is it a fake head? | |
1392 | # So it is, real heads should not be the ancestors of |
|
1392 | # So it is, real heads should not be the ancestors of | |
1393 | # any other heads. |
|
1393 | # any other heads. | |
1394 | heads.pop(n) |
|
1394 | heads.pop(n) | |
1395 | if not ancestors: |
|
1395 | if not ancestors: | |
1396 | return nonodes |
|
1396 | return nonodes | |
1397 | # Now that we have our set of ancestors, we want to remove any |
|
1397 | # Now that we have our set of ancestors, we want to remove any | |
1398 | # roots that are not ancestors. |
|
1398 | # roots that are not ancestors. | |
1399 |
|
1399 | |||
1400 | # If one of the roots was nullid, everything is included anyway. |
|
1400 | # If one of the roots was nullid, everything is included anyway. | |
1401 | if lowestrev > nullrev: |
|
1401 | if lowestrev > nullrev: | |
1402 | # But, since we weren't, let's recompute the lowest rev to not |
|
1402 | # But, since we weren't, let's recompute the lowest rev to not | |
1403 | # include roots that aren't ancestors. |
|
1403 | # include roots that aren't ancestors. | |
1404 |
|
1404 | |||
1405 | # Filter out roots that aren't ancestors of heads |
|
1405 | # Filter out roots that aren't ancestors of heads | |
1406 | roots = [root for root in roots if root in ancestors] |
|
1406 | roots = [root for root in roots if root in ancestors] | |
1407 | # Recompute the lowest revision |
|
1407 | # Recompute the lowest revision | |
1408 | if roots: |
|
1408 | if roots: | |
1409 | lowestrev = min([self.rev(root) for root in roots]) |
|
1409 | lowestrev = min([self.rev(root) for root in roots]) | |
1410 | else: |
|
1410 | else: | |
1411 | # No more roots? Return empty list |
|
1411 | # No more roots? Return empty list | |
1412 | return nonodes |
|
1412 | return nonodes | |
1413 | else: |
|
1413 | else: | |
1414 | # We are descending from nullid, and don't need to care about |
|
1414 | # We are descending from nullid, and don't need to care about | |
1415 | # any other roots. |
|
1415 | # any other roots. | |
1416 | lowestrev = nullrev |
|
1416 | lowestrev = nullrev | |
1417 | roots = [self.nullid] |
|
1417 | roots = [self.nullid] | |
1418 | # Transform our roots list into a set. |
|
1418 | # Transform our roots list into a set. | |
1419 | descendants = set(roots) |
|
1419 | descendants = set(roots) | |
1420 | # Also, keep the original roots so we can filter out roots that aren't |
|
1420 | # Also, keep the original roots so we can filter out roots that aren't | |
1421 | # 'real' roots (i.e. are descended from other roots). |
|
1421 | # 'real' roots (i.e. are descended from other roots). | |
1422 | roots = descendants.copy() |
|
1422 | roots = descendants.copy() | |
1423 | # Our topologically sorted list of output nodes. |
|
1423 | # Our topologically sorted list of output nodes. | |
1424 | orderedout = [] |
|
1424 | orderedout = [] | |
1425 | # Don't start at nullid since we don't want nullid in our output list, |
|
1425 | # Don't start at nullid since we don't want nullid in our output list, | |
1426 | # and if nullid shows up in descendants, empty parents will look like |
|
1426 | # and if nullid shows up in descendants, empty parents will look like | |
1427 | # they're descendants. |
|
1427 | # they're descendants. | |
1428 | for r in self.revs(start=max(lowestrev, 0), stop=highestrev + 1): |
|
1428 | for r in self.revs(start=max(lowestrev, 0), stop=highestrev + 1): | |
1429 | n = self.node(r) |
|
1429 | n = self.node(r) | |
1430 | isdescendant = False |
|
1430 | isdescendant = False | |
1431 | if lowestrev == nullrev: # Everybody is a descendant of nullid |
|
1431 | if lowestrev == nullrev: # Everybody is a descendant of nullid | |
1432 | isdescendant = True |
|
1432 | isdescendant = True | |
1433 | elif n in descendants: |
|
1433 | elif n in descendants: | |
1434 | # n is already a descendant |
|
1434 | # n is already a descendant | |
1435 | isdescendant = True |
|
1435 | isdescendant = True | |
1436 | # This check only needs to be done here because all the roots |
|
1436 | # This check only needs to be done here because all the roots | |
1437 | # will start being marked is descendants before the loop. |
|
1437 | # will start being marked is descendants before the loop. | |
1438 | if n in roots: |
|
1438 | if n in roots: | |
1439 | # If n was a root, check if it's a 'real' root. |
|
1439 | # If n was a root, check if it's a 'real' root. | |
1440 | p = tuple(self.parents(n)) |
|
1440 | p = tuple(self.parents(n)) | |
1441 | # If any of its parents are descendants, it's not a root. |
|
1441 | # If any of its parents are descendants, it's not a root. | |
1442 | if (p[0] in descendants) or (p[1] in descendants): |
|
1442 | if (p[0] in descendants) or (p[1] in descendants): | |
1443 | roots.remove(n) |
|
1443 | roots.remove(n) | |
1444 | else: |
|
1444 | else: | |
1445 | p = tuple(self.parents(n)) |
|
1445 | p = tuple(self.parents(n)) | |
1446 | # A node is a descendant if either of its parents are |
|
1446 | # A node is a descendant if either of its parents are | |
1447 | # descendants. (We seeded the dependents list with the roots |
|
1447 | # descendants. (We seeded the dependents list with the roots | |
1448 | # up there, remember?) |
|
1448 | # up there, remember?) | |
1449 | if (p[0] in descendants) or (p[1] in descendants): |
|
1449 | if (p[0] in descendants) or (p[1] in descendants): | |
1450 | descendants.add(n) |
|
1450 | descendants.add(n) | |
1451 | isdescendant = True |
|
1451 | isdescendant = True | |
1452 | if isdescendant and ((ancestors is None) or (n in ancestors)): |
|
1452 | if isdescendant and ((ancestors is None) or (n in ancestors)): | |
1453 | # Only include nodes that are both descendants and ancestors. |
|
1453 | # Only include nodes that are both descendants and ancestors. | |
1454 | orderedout.append(n) |
|
1454 | orderedout.append(n) | |
1455 | if (ancestors is not None) and (n in heads): |
|
1455 | if (ancestors is not None) and (n in heads): | |
1456 | # We're trying to figure out which heads are reachable |
|
1456 | # We're trying to figure out which heads are reachable | |
1457 | # from roots. |
|
1457 | # from roots. | |
1458 | # Mark this head as having been reached |
|
1458 | # Mark this head as having been reached | |
1459 | heads[n] = True |
|
1459 | heads[n] = True | |
1460 | elif ancestors is None: |
|
1460 | elif ancestors is None: | |
1461 | # Otherwise, we're trying to discover the heads. |
|
1461 | # Otherwise, we're trying to discover the heads. | |
1462 | # Assume this is a head because if it isn't, the next step |
|
1462 | # Assume this is a head because if it isn't, the next step | |
1463 | # will eventually remove it. |
|
1463 | # will eventually remove it. | |
1464 | heads[n] = True |
|
1464 | heads[n] = True | |
1465 | # But, obviously its parents aren't. |
|
1465 | # But, obviously its parents aren't. | |
1466 | for p in self.parents(n): |
|
1466 | for p in self.parents(n): | |
1467 | heads.pop(p, None) |
|
1467 | heads.pop(p, None) | |
1468 | heads = [head for head, flag in heads.items() if flag] |
|
1468 | heads = [head for head, flag in heads.items() if flag] | |
1469 | roots = list(roots) |
|
1469 | roots = list(roots) | |
1470 | assert orderedout |
|
1470 | assert orderedout | |
1471 | assert roots |
|
1471 | assert roots | |
1472 | assert heads |
|
1472 | assert heads | |
1473 | return (orderedout, roots, heads) |
|
1473 | return (orderedout, roots, heads) | |
1474 |
|
1474 | |||
1475 | def headrevs(self, revs=None): |
|
1475 | def headrevs(self, revs=None): | |
1476 | if revs is None: |
|
1476 | if revs is None: | |
1477 | try: |
|
1477 | try: | |
1478 | return self.index.headrevs() |
|
1478 | return self.index.headrevs() | |
1479 | except AttributeError: |
|
1479 | except AttributeError: | |
1480 | return self._headrevs() |
|
1480 | return self._headrevs() | |
1481 | if rustdagop is not None and self.index.rust_ext_compat: |
|
1481 | if rustdagop is not None and self.index.rust_ext_compat: | |
1482 | return rustdagop.headrevs(self.index, revs) |
|
1482 | return rustdagop.headrevs(self.index, revs) | |
1483 | return dagop.headrevs(revs, self._uncheckedparentrevs) |
|
1483 | return dagop.headrevs(revs, self._uncheckedparentrevs) | |
1484 |
|
1484 | |||
1485 | def computephases(self, roots): |
|
1485 | def computephases(self, roots): | |
1486 | return self.index.computephasesmapsets(roots) |
|
1486 | return self.index.computephasesmapsets(roots) | |
1487 |
|
1487 | |||
1488 | def _headrevs(self): |
|
1488 | def _headrevs(self): | |
1489 | count = len(self) |
|
1489 | count = len(self) | |
1490 | if not count: |
|
1490 | if not count: | |
1491 | return [nullrev] |
|
1491 | return [nullrev] | |
1492 | # we won't iter over filtered rev so nobody is a head at start |
|
1492 | # we won't iter over filtered rev so nobody is a head at start | |
1493 | ishead = [0] * (count + 1) |
|
1493 | ishead = [0] * (count + 1) | |
1494 | index = self.index |
|
1494 | index = self.index | |
1495 | for r in self: |
|
1495 | for r in self: | |
1496 | ishead[r] = 1 # I may be an head |
|
1496 | ishead[r] = 1 # I may be an head | |
1497 | e = index[r] |
|
1497 | e = index[r] | |
1498 | ishead[e[5]] = ishead[e[6]] = 0 # my parent are not |
|
1498 | ishead[e[5]] = ishead[e[6]] = 0 # my parent are not | |
1499 | return [r for r, val in enumerate(ishead) if val] |
|
1499 | return [r for r, val in enumerate(ishead) if val] | |
1500 |
|
1500 | |||
1501 | def heads(self, start=None, stop=None): |
|
1501 | def heads(self, start=None, stop=None): | |
1502 | """return the list of all nodes that have no children |
|
1502 | """return the list of all nodes that have no children | |
1503 |
|
1503 | |||
1504 | if start is specified, only heads that are descendants of |
|
1504 | if start is specified, only heads that are descendants of | |
1505 | start will be returned |
|
1505 | start will be returned | |
1506 | if stop is specified, it will consider all the revs from stop |
|
1506 | if stop is specified, it will consider all the revs from stop | |
1507 | as if they had no children |
|
1507 | as if they had no children | |
1508 | """ |
|
1508 | """ | |
1509 | if start is None and stop is None: |
|
1509 | if start is None and stop is None: | |
1510 | if not len(self): |
|
1510 | if not len(self): | |
1511 | return [self.nullid] |
|
1511 | return [self.nullid] | |
1512 | return [self.node(r) for r in self.headrevs()] |
|
1512 | return [self.node(r) for r in self.headrevs()] | |
1513 |
|
1513 | |||
1514 | if start is None: |
|
1514 | if start is None: | |
1515 | start = nullrev |
|
1515 | start = nullrev | |
1516 | else: |
|
1516 | else: | |
1517 | start = self.rev(start) |
|
1517 | start = self.rev(start) | |
1518 |
|
1518 | |||
1519 | stoprevs = {self.rev(n) for n in stop or []} |
|
1519 | stoprevs = {self.rev(n) for n in stop or []} | |
1520 |
|
1520 | |||
1521 | revs = dagop.headrevssubset( |
|
1521 | revs = dagop.headrevssubset( | |
1522 | self.revs, self.parentrevs, startrev=start, stoprevs=stoprevs |
|
1522 | self.revs, self.parentrevs, startrev=start, stoprevs=stoprevs | |
1523 | ) |
|
1523 | ) | |
1524 |
|
1524 | |||
1525 | return [self.node(rev) for rev in revs] |
|
1525 | return [self.node(rev) for rev in revs] | |
1526 |
|
1526 | |||
1527 | def children(self, node): |
|
1527 | def children(self, node): | |
1528 | """find the children of a given node""" |
|
1528 | """find the children of a given node""" | |
1529 | c = [] |
|
1529 | c = [] | |
1530 | p = self.rev(node) |
|
1530 | p = self.rev(node) | |
1531 | for r in self.revs(start=p + 1): |
|
1531 | for r in self.revs(start=p + 1): | |
1532 | prevs = [pr for pr in self.parentrevs(r) if pr != nullrev] |
|
1532 | prevs = [pr for pr in self.parentrevs(r) if pr != nullrev] | |
1533 | if prevs: |
|
1533 | if prevs: | |
1534 | for pr in prevs: |
|
1534 | for pr in prevs: | |
1535 | if pr == p: |
|
1535 | if pr == p: | |
1536 | c.append(self.node(r)) |
|
1536 | c.append(self.node(r)) | |
1537 | elif p == nullrev: |
|
1537 | elif p == nullrev: | |
1538 | c.append(self.node(r)) |
|
1538 | c.append(self.node(r)) | |
1539 | return c |
|
1539 | return c | |
1540 |
|
1540 | |||
1541 | def commonancestorsheads(self, a, b): |
|
1541 | def commonancestorsheads(self, a, b): | |
1542 | """calculate all the heads of the common ancestors of nodes a and b""" |
|
1542 | """calculate all the heads of the common ancestors of nodes a and b""" | |
1543 | a, b = self.rev(a), self.rev(b) |
|
1543 | a, b = self.rev(a), self.rev(b) | |
1544 | ancs = self._commonancestorsheads(a, b) |
|
1544 | ancs = self._commonancestorsheads(a, b) | |
1545 | return pycompat.maplist(self.node, ancs) |
|
1545 | return pycompat.maplist(self.node, ancs) | |
1546 |
|
1546 | |||
1547 | def _commonancestorsheads(self, *revs): |
|
1547 | def _commonancestorsheads(self, *revs): | |
1548 | """calculate all the heads of the common ancestors of revs""" |
|
1548 | """calculate all the heads of the common ancestors of revs""" | |
1549 | try: |
|
1549 | try: | |
1550 | ancs = self.index.commonancestorsheads(*revs) |
|
1550 | ancs = self.index.commonancestorsheads(*revs) | |
1551 | except (AttributeError, OverflowError): # C implementation failed |
|
1551 | except (AttributeError, OverflowError): # C implementation failed | |
1552 | ancs = ancestor.commonancestorsheads(self.parentrevs, *revs) |
|
1552 | ancs = ancestor.commonancestorsheads(self.parentrevs, *revs) | |
1553 | return ancs |
|
1553 | return ancs | |
1554 |
|
1554 | |||
1555 | def isancestor(self, a, b): |
|
1555 | def isancestor(self, a, b): | |
1556 | """return True if node a is an ancestor of node b |
|
1556 | """return True if node a is an ancestor of node b | |
1557 |
|
1557 | |||
1558 | A revision is considered an ancestor of itself.""" |
|
1558 | A revision is considered an ancestor of itself.""" | |
1559 | a, b = self.rev(a), self.rev(b) |
|
1559 | a, b = self.rev(a), self.rev(b) | |
1560 | return self.isancestorrev(a, b) |
|
1560 | return self.isancestorrev(a, b) | |
1561 |
|
1561 | |||
1562 | def isancestorrev(self, a, b): |
|
1562 | def isancestorrev(self, a, b): | |
1563 | """return True if revision a is an ancestor of revision b |
|
1563 | """return True if revision a is an ancestor of revision b | |
1564 |
|
1564 | |||
1565 | A revision is considered an ancestor of itself. |
|
1565 | A revision is considered an ancestor of itself. | |
1566 |
|
1566 | |||
1567 | The implementation of this is trivial but the use of |
|
1567 | The implementation of this is trivial but the use of | |
1568 | reachableroots is not.""" |
|
1568 | reachableroots is not.""" | |
1569 | if a == nullrev: |
|
1569 | if a == nullrev: | |
1570 | return True |
|
1570 | return True | |
1571 | elif a == b: |
|
1571 | elif a == b: | |
1572 | return True |
|
1572 | return True | |
1573 | elif a > b: |
|
1573 | elif a > b: | |
1574 | return False |
|
1574 | return False | |
1575 | return bool(self.reachableroots(a, [b], [a], includepath=False)) |
|
1575 | return bool(self.reachableroots(a, [b], [a], includepath=False)) | |
1576 |
|
1576 | |||
1577 | def reachableroots(self, minroot, heads, roots, includepath=False): |
|
1577 | def reachableroots(self, minroot, heads, roots, includepath=False): | |
1578 | """return (heads(::(<roots> and <roots>::<heads>))) |
|
1578 | """return (heads(::(<roots> and <roots>::<heads>))) | |
1579 |
|
1579 | |||
1580 | If includepath is True, return (<roots>::<heads>).""" |
|
1580 | If includepath is True, return (<roots>::<heads>).""" | |
1581 | try: |
|
1581 | try: | |
1582 | return self.index.reachableroots2( |
|
1582 | return self.index.reachableroots2( | |
1583 | minroot, heads, roots, includepath |
|
1583 | minroot, heads, roots, includepath | |
1584 | ) |
|
1584 | ) | |
1585 | except AttributeError: |
|
1585 | except AttributeError: | |
1586 | return dagop._reachablerootspure( |
|
1586 | return dagop._reachablerootspure( | |
1587 | self.parentrevs, minroot, roots, heads, includepath |
|
1587 | self.parentrevs, minroot, roots, heads, includepath | |
1588 | ) |
|
1588 | ) | |
1589 |
|
1589 | |||
1590 | def ancestor(self, a, b): |
|
1590 | def ancestor(self, a, b): | |
1591 | """calculate the "best" common ancestor of nodes a and b""" |
|
1591 | """calculate the "best" common ancestor of nodes a and b""" | |
1592 |
|
1592 | |||
1593 | a, b = self.rev(a), self.rev(b) |
|
1593 | a, b = self.rev(a), self.rev(b) | |
1594 | try: |
|
1594 | try: | |
1595 | ancs = self.index.ancestors(a, b) |
|
1595 | ancs = self.index.ancestors(a, b) | |
1596 | except (AttributeError, OverflowError): |
|
1596 | except (AttributeError, OverflowError): | |
1597 | ancs = ancestor.ancestors(self.parentrevs, a, b) |
|
1597 | ancs = ancestor.ancestors(self.parentrevs, a, b) | |
1598 | if ancs: |
|
1598 | if ancs: | |
1599 | # choose a consistent winner when there's a tie |
|
1599 | # choose a consistent winner when there's a tie | |
1600 | return min(map(self.node, ancs)) |
|
1600 | return min(map(self.node, ancs)) | |
1601 | return self.nullid |
|
1601 | return self.nullid | |
1602 |
|
1602 | |||
1603 | def _match(self, id): |
|
1603 | def _match(self, id): | |
1604 | if isinstance(id, int): |
|
1604 | if isinstance(id, int): | |
1605 | # rev |
|
1605 | # rev | |
1606 | return self.node(id) |
|
1606 | return self.node(id) | |
1607 | if len(id) == self.nodeconstants.nodelen: |
|
1607 | if len(id) == self.nodeconstants.nodelen: | |
1608 | # possibly a binary node |
|
1608 | # possibly a binary node | |
1609 | # odds of a binary node being all hex in ASCII are 1 in 10**25 |
|
1609 | # odds of a binary node being all hex in ASCII are 1 in 10**25 | |
1610 | try: |
|
1610 | try: | |
1611 | node = id |
|
1611 | node = id | |
1612 | self.rev(node) # quick search the index |
|
1612 | self.rev(node) # quick search the index | |
1613 | return node |
|
1613 | return node | |
1614 | except error.LookupError: |
|
1614 | except error.LookupError: | |
1615 | pass # may be partial hex id |
|
1615 | pass # may be partial hex id | |
1616 | try: |
|
1616 | try: | |
1617 | # str(rev) |
|
1617 | # str(rev) | |
1618 | rev = int(id) |
|
1618 | rev = int(id) | |
1619 | if b"%d" % rev != id: |
|
1619 | if b"%d" % rev != id: | |
1620 | raise ValueError |
|
1620 | raise ValueError | |
1621 | if rev < 0: |
|
1621 | if rev < 0: | |
1622 | rev = len(self) + rev |
|
1622 | rev = len(self) + rev | |
1623 | if rev < 0 or rev >= len(self): |
|
1623 | if rev < 0 or rev >= len(self): | |
1624 | raise ValueError |
|
1624 | raise ValueError | |
1625 | return self.node(rev) |
|
1625 | return self.node(rev) | |
1626 | except (ValueError, OverflowError): |
|
1626 | except (ValueError, OverflowError): | |
1627 | pass |
|
1627 | pass | |
1628 | if len(id) == 2 * self.nodeconstants.nodelen: |
|
1628 | if len(id) == 2 * self.nodeconstants.nodelen: | |
1629 | try: |
|
1629 | try: | |
1630 | # a full hex nodeid? |
|
1630 | # a full hex nodeid? | |
1631 | node = bin(id) |
|
1631 | node = bin(id) | |
1632 | self.rev(node) |
|
1632 | self.rev(node) | |
1633 | return node |
|
1633 | return node | |
1634 | except (binascii.Error, error.LookupError): |
|
1634 | except (binascii.Error, error.LookupError): | |
1635 | pass |
|
1635 | pass | |
1636 |
|
1636 | |||
1637 | def _partialmatch(self, id): |
|
1637 | def _partialmatch(self, id): | |
1638 | # we don't care wdirfilenodeids as they should be always full hash |
|
1638 | # we don't care wdirfilenodeids as they should be always full hash | |
1639 | maybewdir = self.nodeconstants.wdirhex.startswith(id) |
|
1639 | maybewdir = self.nodeconstants.wdirhex.startswith(id) | |
1640 | ambiguous = False |
|
1640 | ambiguous = False | |
1641 | try: |
|
1641 | try: | |
1642 | partial = self.index.partialmatch(id) |
|
1642 | partial = self.index.partialmatch(id) | |
1643 | if partial and self.hasnode(partial): |
|
1643 | if partial and self.hasnode(partial): | |
1644 | if maybewdir: |
|
1644 | if maybewdir: | |
1645 | # single 'ff...' match in radix tree, ambiguous with wdir |
|
1645 | # single 'ff...' match in radix tree, ambiguous with wdir | |
1646 | ambiguous = True |
|
1646 | ambiguous = True | |
1647 | else: |
|
1647 | else: | |
1648 | return partial |
|
1648 | return partial | |
1649 | elif maybewdir: |
|
1649 | elif maybewdir: | |
1650 | # no 'ff...' match in radix tree, wdir identified |
|
1650 | # no 'ff...' match in radix tree, wdir identified | |
1651 | raise error.WdirUnsupported |
|
1651 | raise error.WdirUnsupported | |
1652 | else: |
|
1652 | else: | |
1653 | return None |
|
1653 | return None | |
1654 | except error.RevlogError: |
|
1654 | except error.RevlogError: | |
1655 | # parsers.c radix tree lookup gave multiple matches |
|
1655 | # parsers.c radix tree lookup gave multiple matches | |
1656 | # fast path: for unfiltered changelog, radix tree is accurate |
|
1656 | # fast path: for unfiltered changelog, radix tree is accurate | |
1657 | if not getattr(self, 'filteredrevs', None): |
|
1657 | if not getattr(self, 'filteredrevs', None): | |
1658 | ambiguous = True |
|
1658 | ambiguous = True | |
1659 | # fall through to slow path that filters hidden revisions |
|
1659 | # fall through to slow path that filters hidden revisions | |
1660 | except (AttributeError, ValueError): |
|
1660 | except (AttributeError, ValueError): | |
1661 | # we are pure python, or key is not hex |
|
1661 | # we are pure python, or key is not hex | |
1662 | pass |
|
1662 | pass | |
1663 | if ambiguous: |
|
1663 | if ambiguous: | |
1664 | raise error.AmbiguousPrefixLookupError( |
|
1664 | raise error.AmbiguousPrefixLookupError( | |
1665 | id, self.display_id, _(b'ambiguous identifier') |
|
1665 | id, self.display_id, _(b'ambiguous identifier') | |
1666 | ) |
|
1666 | ) | |
1667 |
|
1667 | |||
1668 | if id in self._pcache: |
|
1668 | if id in self._pcache: | |
1669 | return self._pcache[id] |
|
1669 | return self._pcache[id] | |
1670 |
|
1670 | |||
1671 | if len(id) <= 40: |
|
1671 | if len(id) <= 40: | |
1672 | # hex(node)[:...] |
|
1672 | # hex(node)[:...] | |
1673 | l = len(id) // 2 * 2 # grab an even number of digits |
|
1673 | l = len(id) // 2 * 2 # grab an even number of digits | |
1674 | try: |
|
1674 | try: | |
1675 | # we're dropping the last digit, so let's check that it's hex, |
|
1675 | # we're dropping the last digit, so let's check that it's hex, | |
1676 | # to avoid the expensive computation below if it's not |
|
1676 | # to avoid the expensive computation below if it's not | |
1677 | if len(id) % 2 > 0: |
|
1677 | if len(id) % 2 > 0: | |
1678 | if not (id[-1] in hexdigits): |
|
1678 | if not (id[-1] in hexdigits): | |
1679 | return None |
|
1679 | return None | |
1680 | prefix = bin(id[:l]) |
|
1680 | prefix = bin(id[:l]) | |
1681 | except binascii.Error: |
|
1681 | except binascii.Error: | |
1682 | pass |
|
1682 | pass | |
1683 | else: |
|
1683 | else: | |
1684 | nl = [e[7] for e in self.index if e[7].startswith(prefix)] |
|
1684 | nl = [e[7] for e in self.index if e[7].startswith(prefix)] | |
1685 | nl = [ |
|
1685 | nl = [ | |
1686 | n for n in nl if hex(n).startswith(id) and self.hasnode(n) |
|
1686 | n for n in nl if hex(n).startswith(id) and self.hasnode(n) | |
1687 | ] |
|
1687 | ] | |
1688 | if self.nodeconstants.nullhex.startswith(id): |
|
1688 | if self.nodeconstants.nullhex.startswith(id): | |
1689 | nl.append(self.nullid) |
|
1689 | nl.append(self.nullid) | |
1690 | if len(nl) > 0: |
|
1690 | if len(nl) > 0: | |
1691 | if len(nl) == 1 and not maybewdir: |
|
1691 | if len(nl) == 1 and not maybewdir: | |
1692 | self._pcache[id] = nl[0] |
|
1692 | self._pcache[id] = nl[0] | |
1693 | return nl[0] |
|
1693 | return nl[0] | |
1694 | raise error.AmbiguousPrefixLookupError( |
|
1694 | raise error.AmbiguousPrefixLookupError( | |
1695 | id, self.display_id, _(b'ambiguous identifier') |
|
1695 | id, self.display_id, _(b'ambiguous identifier') | |
1696 | ) |
|
1696 | ) | |
1697 | if maybewdir: |
|
1697 | if maybewdir: | |
1698 | raise error.WdirUnsupported |
|
1698 | raise error.WdirUnsupported | |
1699 | return None |
|
1699 | return None | |
1700 |
|
1700 | |||
1701 | def lookup(self, id): |
|
1701 | def lookup(self, id): | |
1702 | """locate a node based on: |
|
1702 | """locate a node based on: | |
1703 | - revision number or str(revision number) |
|
1703 | - revision number or str(revision number) | |
1704 | - nodeid or subset of hex nodeid |
|
1704 | - nodeid or subset of hex nodeid | |
1705 | """ |
|
1705 | """ | |
1706 | n = self._match(id) |
|
1706 | n = self._match(id) | |
1707 | if n is not None: |
|
1707 | if n is not None: | |
1708 | return n |
|
1708 | return n | |
1709 | n = self._partialmatch(id) |
|
1709 | n = self._partialmatch(id) | |
1710 | if n: |
|
1710 | if n: | |
1711 | return n |
|
1711 | return n | |
1712 |
|
1712 | |||
1713 | raise error.LookupError(id, self.display_id, _(b'no match found')) |
|
1713 | raise error.LookupError(id, self.display_id, _(b'no match found')) | |
1714 |
|
1714 | |||
1715 | def shortest(self, node, minlength=1): |
|
1715 | def shortest(self, node, minlength=1): | |
1716 | """Find the shortest unambiguous prefix that matches node.""" |
|
1716 | """Find the shortest unambiguous prefix that matches node.""" | |
1717 |
|
1717 | |||
1718 | def isvalid(prefix): |
|
1718 | def isvalid(prefix): | |
1719 | try: |
|
1719 | try: | |
1720 | matchednode = self._partialmatch(prefix) |
|
1720 | matchednode = self._partialmatch(prefix) | |
1721 | except error.AmbiguousPrefixLookupError: |
|
1721 | except error.AmbiguousPrefixLookupError: | |
1722 | return False |
|
1722 | return False | |
1723 | except error.WdirUnsupported: |
|
1723 | except error.WdirUnsupported: | |
1724 | # single 'ff...' match |
|
1724 | # single 'ff...' match | |
1725 | return True |
|
1725 | return True | |
1726 | if matchednode is None: |
|
1726 | if matchednode is None: | |
1727 | raise error.LookupError(node, self.display_id, _(b'no node')) |
|
1727 | raise error.LookupError(node, self.display_id, _(b'no node')) | |
1728 | return True |
|
1728 | return True | |
1729 |
|
1729 | |||
1730 | def maybewdir(prefix): |
|
1730 | def maybewdir(prefix): | |
1731 | return all(c == b'f' for c in pycompat.iterbytestr(prefix)) |
|
1731 | return all(c == b'f' for c in pycompat.iterbytestr(prefix)) | |
1732 |
|
1732 | |||
1733 | hexnode = hex(node) |
|
1733 | hexnode = hex(node) | |
1734 |
|
1734 | |||
1735 | def disambiguate(hexnode, minlength): |
|
1735 | def disambiguate(hexnode, minlength): | |
1736 | """Disambiguate against wdirid.""" |
|
1736 | """Disambiguate against wdirid.""" | |
1737 | for length in range(minlength, len(hexnode) + 1): |
|
1737 | for length in range(minlength, len(hexnode) + 1): | |
1738 | prefix = hexnode[:length] |
|
1738 | prefix = hexnode[:length] | |
1739 | if not maybewdir(prefix): |
|
1739 | if not maybewdir(prefix): | |
1740 | return prefix |
|
1740 | return prefix | |
1741 |
|
1741 | |||
1742 | if not getattr(self, 'filteredrevs', None): |
|
1742 | if not getattr(self, 'filteredrevs', None): | |
1743 | try: |
|
1743 | try: | |
1744 | length = max(self.index.shortest(node), minlength) |
|
1744 | length = max(self.index.shortest(node), minlength) | |
1745 | return disambiguate(hexnode, length) |
|
1745 | return disambiguate(hexnode, length) | |
1746 | except error.RevlogError: |
|
1746 | except error.RevlogError: | |
1747 | if node != self.nodeconstants.wdirid: |
|
1747 | if node != self.nodeconstants.wdirid: | |
1748 | raise error.LookupError( |
|
1748 | raise error.LookupError( | |
1749 | node, self.display_id, _(b'no node') |
|
1749 | node, self.display_id, _(b'no node') | |
1750 | ) |
|
1750 | ) | |
1751 | except AttributeError: |
|
1751 | except AttributeError: | |
1752 | # Fall through to pure code |
|
1752 | # Fall through to pure code | |
1753 | pass |
|
1753 | pass | |
1754 |
|
1754 | |||
1755 | if node == self.nodeconstants.wdirid: |
|
1755 | if node == self.nodeconstants.wdirid: | |
1756 | for length in range(minlength, len(hexnode) + 1): |
|
1756 | for length in range(minlength, len(hexnode) + 1): | |
1757 | prefix = hexnode[:length] |
|
1757 | prefix = hexnode[:length] | |
1758 | if isvalid(prefix): |
|
1758 | if isvalid(prefix): | |
1759 | return prefix |
|
1759 | return prefix | |
1760 |
|
1760 | |||
1761 | for length in range(minlength, len(hexnode) + 1): |
|
1761 | for length in range(minlength, len(hexnode) + 1): | |
1762 | prefix = hexnode[:length] |
|
1762 | prefix = hexnode[:length] | |
1763 | if isvalid(prefix): |
|
1763 | if isvalid(prefix): | |
1764 | return disambiguate(hexnode, length) |
|
1764 | return disambiguate(hexnode, length) | |
1765 |
|
1765 | |||
1766 | def cmp(self, node, text): |
|
1766 | def cmp(self, node, text): | |
1767 | """compare text with a given file revision |
|
1767 | """compare text with a given file revision | |
1768 |
|
1768 | |||
1769 | returns True if text is different than what is stored. |
|
1769 | returns True if text is different than what is stored. | |
1770 | """ |
|
1770 | """ | |
1771 | p1, p2 = self.parents(node) |
|
1771 | p1, p2 = self.parents(node) | |
1772 | return storageutil.hashrevisionsha1(text, p1, p2) != node |
|
1772 | return storageutil.hashrevisionsha1(text, p1, p2) != node | |
1773 |
|
1773 | |||
1774 | def _getsegmentforrevs(self, startrev, endrev, df=None): |
|
1774 | def _getsegmentforrevs(self, startrev, endrev, df=None): | |
1775 | """Obtain a segment of raw data corresponding to a range of revisions. |
|
1775 | """Obtain a segment of raw data corresponding to a range of revisions. | |
1776 |
|
1776 | |||
1777 | Accepts the start and end revisions and an optional already-open |
|
1777 | Accepts the start and end revisions and an optional already-open | |
1778 | file handle to be used for reading. If the file handle is read, its |
|
1778 | file handle to be used for reading. If the file handle is read, its | |
1779 | seek position will not be preserved. |
|
1779 | seek position will not be preserved. | |
1780 |
|
1780 | |||
1781 | Requests for data may be satisfied by a cache. |
|
1781 | Requests for data may be satisfied by a cache. | |
1782 |
|
1782 | |||
1783 | Returns a 2-tuple of (offset, data) for the requested range of |
|
1783 | Returns a 2-tuple of (offset, data) for the requested range of | |
1784 | revisions. Offset is the integer offset from the beginning of the |
|
1784 | revisions. Offset is the integer offset from the beginning of the | |
1785 | revlog and data is a str or buffer of the raw byte data. |
|
1785 | revlog and data is a str or buffer of the raw byte data. | |
1786 |
|
1786 | |||
1787 | Callers will need to call ``self.start(rev)`` and ``self.length(rev)`` |
|
1787 | Callers will need to call ``self.start(rev)`` and ``self.length(rev)`` | |
1788 | to determine where each revision's data begins and ends. |
|
1788 | to determine where each revision's data begins and ends. | |
1789 | """ |
|
1789 | """ | |
1790 | # Inlined self.start(startrev) & self.end(endrev) for perf reasons |
|
1790 | # Inlined self.start(startrev) & self.end(endrev) for perf reasons | |
1791 | # (functions are expensive). |
|
1791 | # (functions are expensive). | |
1792 | index = self.index |
|
1792 | index = self.index | |
1793 | istart = index[startrev] |
|
1793 | istart = index[startrev] | |
1794 | start = int(istart[0] >> 16) |
|
1794 | start = int(istart[0] >> 16) | |
1795 | if startrev == endrev: |
|
1795 | if startrev == endrev: | |
1796 | end = start + istart[1] |
|
1796 | end = start + istart[1] | |
1797 | else: |
|
1797 | else: | |
1798 | iend = index[endrev] |
|
1798 | iend = index[endrev] | |
1799 | end = int(iend[0] >> 16) + iend[1] |
|
1799 | end = int(iend[0] >> 16) + iend[1] | |
1800 |
|
1800 | |||
1801 | if self._inline: |
|
1801 | if self._inline: | |
1802 | start += (startrev + 1) * self.index.entry_size |
|
1802 | start += (startrev + 1) * self.index.entry_size | |
1803 | end += (endrev + 1) * self.index.entry_size |
|
1803 | end += (endrev + 1) * self.index.entry_size | |
1804 | length = end - start |
|
1804 | length = end - start | |
1805 |
|
1805 | |||
1806 | return start, self._segmentfile.read_chunk(start, length, df) |
|
1806 | return start, self._segmentfile.read_chunk(start, length, df) | |
1807 |
|
1807 | |||
1808 | def _chunk(self, rev, df=None): |
|
1808 | def _chunk(self, rev, df=None): | |
1809 | """Obtain a single decompressed chunk for a revision. |
|
1809 | """Obtain a single decompressed chunk for a revision. | |
1810 |
|
1810 | |||
1811 | Accepts an integer revision and an optional already-open file handle |
|
1811 | Accepts an integer revision and an optional already-open file handle | |
1812 | to be used for reading. If used, the seek position of the file will not |
|
1812 | to be used for reading. If used, the seek position of the file will not | |
1813 | be preserved. |
|
1813 | be preserved. | |
1814 |
|
1814 | |||
1815 | Returns a str holding uncompressed data for the requested revision. |
|
1815 | Returns a str holding uncompressed data for the requested revision. | |
1816 | """ |
|
1816 | """ | |
1817 | compression_mode = self.index[rev][10] |
|
1817 | compression_mode = self.index[rev][10] | |
1818 | data = self._getsegmentforrevs(rev, rev, df=df)[1] |
|
1818 | data = self._getsegmentforrevs(rev, rev, df=df)[1] | |
1819 | if compression_mode == COMP_MODE_PLAIN: |
|
1819 | if compression_mode == COMP_MODE_PLAIN: | |
1820 | return data |
|
1820 | return data | |
1821 | elif compression_mode == COMP_MODE_DEFAULT: |
|
1821 | elif compression_mode == COMP_MODE_DEFAULT: | |
1822 | return self._decompressor(data) |
|
1822 | return self._decompressor(data) | |
1823 | elif compression_mode == COMP_MODE_INLINE: |
|
1823 | elif compression_mode == COMP_MODE_INLINE: | |
1824 | return self.decompress(data) |
|
1824 | return self.decompress(data) | |
1825 | else: |
|
1825 | else: | |
1826 | msg = b'unknown compression mode %d' |
|
1826 | msg = b'unknown compression mode %d' | |
1827 | msg %= compression_mode |
|
1827 | msg %= compression_mode | |
1828 | raise error.RevlogError(msg) |
|
1828 | raise error.RevlogError(msg) | |
1829 |
|
1829 | |||
1830 | def _chunks(self, revs, df=None, targetsize=None): |
|
1830 | def _chunks(self, revs, df=None, targetsize=None): | |
1831 | """Obtain decompressed chunks for the specified revisions. |
|
1831 | """Obtain decompressed chunks for the specified revisions. | |
1832 |
|
1832 | |||
1833 | Accepts an iterable of numeric revisions that are assumed to be in |
|
1833 | Accepts an iterable of numeric revisions that are assumed to be in | |
1834 | ascending order. Also accepts an optional already-open file handle |
|
1834 | ascending order. Also accepts an optional already-open file handle | |
1835 | to be used for reading. If used, the seek position of the file will |
|
1835 | to be used for reading. If used, the seek position of the file will | |
1836 | not be preserved. |
|
1836 | not be preserved. | |
1837 |
|
1837 | |||
1838 | This function is similar to calling ``self._chunk()`` multiple times, |
|
1838 | This function is similar to calling ``self._chunk()`` multiple times, | |
1839 | but is faster. |
|
1839 | but is faster. | |
1840 |
|
1840 | |||
1841 | Returns a list with decompressed data for each requested revision. |
|
1841 | Returns a list with decompressed data for each requested revision. | |
1842 | """ |
|
1842 | """ | |
1843 | if not revs: |
|
1843 | if not revs: | |
1844 | return [] |
|
1844 | return [] | |
1845 | start = self.start |
|
1845 | start = self.start | |
1846 | length = self.length |
|
1846 | length = self.length | |
1847 | inline = self._inline |
|
1847 | inline = self._inline | |
1848 | iosize = self.index.entry_size |
|
1848 | iosize = self.index.entry_size | |
1849 | buffer = util.buffer |
|
1849 | buffer = util.buffer | |
1850 |
|
1850 | |||
1851 | l = [] |
|
1851 | l = [] | |
1852 | ladd = l.append |
|
1852 | ladd = l.append | |
1853 |
|
1853 | |||
1854 | if not self._withsparseread: |
|
1854 | if not self._withsparseread: | |
1855 | slicedchunks = (revs,) |
|
1855 | slicedchunks = (revs,) | |
1856 | else: |
|
1856 | else: | |
1857 | slicedchunks = deltautil.slicechunk( |
|
1857 | slicedchunks = deltautil.slicechunk( | |
1858 | self, revs, targetsize=targetsize |
|
1858 | self, revs, targetsize=targetsize | |
1859 | ) |
|
1859 | ) | |
1860 |
|
1860 | |||
1861 | for revschunk in slicedchunks: |
|
1861 | for revschunk in slicedchunks: | |
1862 | firstrev = revschunk[0] |
|
1862 | firstrev = revschunk[0] | |
1863 | # Skip trailing revisions with empty diff |
|
1863 | # Skip trailing revisions with empty diff | |
1864 | for lastrev in revschunk[::-1]: |
|
1864 | for lastrev in revschunk[::-1]: | |
1865 | if length(lastrev) != 0: |
|
1865 | if length(lastrev) != 0: | |
1866 | break |
|
1866 | break | |
1867 |
|
1867 | |||
1868 | try: |
|
1868 | try: | |
1869 | offset, data = self._getsegmentforrevs(firstrev, lastrev, df=df) |
|
1869 | offset, data = self._getsegmentforrevs(firstrev, lastrev, df=df) | |
1870 | except OverflowError: |
|
1870 | except OverflowError: | |
1871 | # issue4215 - we can't cache a run of chunks greater than |
|
1871 | # issue4215 - we can't cache a run of chunks greater than | |
1872 | # 2G on Windows |
|
1872 | # 2G on Windows | |
1873 | return [self._chunk(rev, df=df) for rev in revschunk] |
|
1873 | return [self._chunk(rev, df=df) for rev in revschunk] | |
1874 |
|
1874 | |||
1875 | decomp = self.decompress |
|
1875 | decomp = self.decompress | |
1876 | # self._decompressor might be None, but will not be used in that case |
|
1876 | # self._decompressor might be None, but will not be used in that case | |
1877 | def_decomp = self._decompressor |
|
1877 | def_decomp = self._decompressor | |
1878 | for rev in revschunk: |
|
1878 | for rev in revschunk: | |
1879 | chunkstart = start(rev) |
|
1879 | chunkstart = start(rev) | |
1880 | if inline: |
|
1880 | if inline: | |
1881 | chunkstart += (rev + 1) * iosize |
|
1881 | chunkstart += (rev + 1) * iosize | |
1882 | chunklength = length(rev) |
|
1882 | chunklength = length(rev) | |
1883 | comp_mode = self.index[rev][10] |
|
1883 | comp_mode = self.index[rev][10] | |
1884 | c = buffer(data, chunkstart - offset, chunklength) |
|
1884 | c = buffer(data, chunkstart - offset, chunklength) | |
1885 | if comp_mode == COMP_MODE_PLAIN: |
|
1885 | if comp_mode == COMP_MODE_PLAIN: | |
1886 | ladd(c) |
|
1886 | ladd(c) | |
1887 | elif comp_mode == COMP_MODE_INLINE: |
|
1887 | elif comp_mode == COMP_MODE_INLINE: | |
1888 | ladd(decomp(c)) |
|
1888 | ladd(decomp(c)) | |
1889 | elif comp_mode == COMP_MODE_DEFAULT: |
|
1889 | elif comp_mode == COMP_MODE_DEFAULT: | |
1890 | ladd(def_decomp(c)) |
|
1890 | ladd(def_decomp(c)) | |
1891 | else: |
|
1891 | else: | |
1892 | msg = b'unknown compression mode %d' |
|
1892 | msg = b'unknown compression mode %d' | |
1893 | msg %= comp_mode |
|
1893 | msg %= comp_mode | |
1894 | raise error.RevlogError(msg) |
|
1894 | raise error.RevlogError(msg) | |
1895 |
|
1895 | |||
1896 | return l |
|
1896 | return l | |
1897 |
|
1897 | |||
1898 | def deltaparent(self, rev): |
|
1898 | def deltaparent(self, rev): | |
1899 | """return deltaparent of the given revision""" |
|
1899 | """return deltaparent of the given revision""" | |
1900 | base = self.index[rev][3] |
|
1900 | base = self.index[rev][3] | |
1901 | if base == rev: |
|
1901 | if base == rev: | |
1902 | return nullrev |
|
1902 | return nullrev | |
1903 | elif self._generaldelta: |
|
1903 | elif self._generaldelta: | |
1904 | return base |
|
1904 | return base | |
1905 | else: |
|
1905 | else: | |
1906 | return rev - 1 |
|
1906 | return rev - 1 | |
1907 |
|
1907 | |||
1908 | def issnapshot(self, rev): |
|
1908 | def issnapshot(self, rev): | |
1909 | """tells whether rev is a snapshot""" |
|
1909 | """tells whether rev is a snapshot""" | |
1910 | if not self._sparserevlog: |
|
1910 | if not self._sparserevlog: | |
1911 | return self.deltaparent(rev) == nullrev |
|
1911 | return self.deltaparent(rev) == nullrev | |
1912 | elif hasattr(self.index, 'issnapshot'): |
|
1912 | elif hasattr(self.index, 'issnapshot'): | |
1913 | # directly assign the method to cache the testing and access |
|
1913 | # directly assign the method to cache the testing and access | |
1914 | self.issnapshot = self.index.issnapshot |
|
1914 | self.issnapshot = self.index.issnapshot | |
1915 | return self.issnapshot(rev) |
|
1915 | return self.issnapshot(rev) | |
1916 | if rev == nullrev: |
|
1916 | if rev == nullrev: | |
1917 | return True |
|
1917 | return True | |
1918 | entry = self.index[rev] |
|
1918 | entry = self.index[rev] | |
1919 | base = entry[3] |
|
1919 | base = entry[3] | |
1920 | if base == rev: |
|
1920 | if base == rev: | |
1921 | return True |
|
1921 | return True | |
1922 | if base == nullrev: |
|
1922 | if base == nullrev: | |
1923 | return True |
|
1923 | return True | |
1924 | p1 = entry[5] |
|
1924 | p1 = entry[5] | |
1925 | while self.length(p1) == 0: |
|
1925 | while self.length(p1) == 0: | |
1926 | b = self.deltaparent(p1) |
|
1926 | b = self.deltaparent(p1) | |
1927 | if b == p1: |
|
1927 | if b == p1: | |
1928 | break |
|
1928 | break | |
1929 | p1 = b |
|
1929 | p1 = b | |
1930 | p2 = entry[6] |
|
1930 | p2 = entry[6] | |
1931 | while self.length(p2) == 0: |
|
1931 | while self.length(p2) == 0: | |
1932 | b = self.deltaparent(p2) |
|
1932 | b = self.deltaparent(p2) | |
1933 | if b == p2: |
|
1933 | if b == p2: | |
1934 | break |
|
1934 | break | |
1935 | p2 = b |
|
1935 | p2 = b | |
1936 | if base == p1 or base == p2: |
|
1936 | if base == p1 or base == p2: | |
1937 | return False |
|
1937 | return False | |
1938 | return self.issnapshot(base) |
|
1938 | return self.issnapshot(base) | |
1939 |
|
1939 | |||
1940 | def snapshotdepth(self, rev): |
|
1940 | def snapshotdepth(self, rev): | |
1941 | """number of snapshot in the chain before this one""" |
|
1941 | """number of snapshot in the chain before this one""" | |
1942 | if not self.issnapshot(rev): |
|
1942 | if not self.issnapshot(rev): | |
1943 | raise error.ProgrammingError(b'revision %d not a snapshot') |
|
1943 | raise error.ProgrammingError(b'revision %d not a snapshot') | |
1944 | return len(self._deltachain(rev)[0]) - 1 |
|
1944 | return len(self._deltachain(rev)[0]) - 1 | |
1945 |
|
1945 | |||
1946 | def revdiff(self, rev1, rev2): |
|
1946 | def revdiff(self, rev1, rev2): | |
1947 | """return or calculate a delta between two revisions |
|
1947 | """return or calculate a delta between two revisions | |
1948 |
|
1948 | |||
1949 | The delta calculated is in binary form and is intended to be written to |
|
1949 | The delta calculated is in binary form and is intended to be written to | |
1950 | revlog data directly. So this function needs raw revision data. |
|
1950 | revlog data directly. So this function needs raw revision data. | |
1951 | """ |
|
1951 | """ | |
1952 | if rev1 != nullrev and self.deltaparent(rev2) == rev1: |
|
1952 | if rev1 != nullrev and self.deltaparent(rev2) == rev1: | |
1953 | return bytes(self._chunk(rev2)) |
|
1953 | return bytes(self._chunk(rev2)) | |
1954 |
|
1954 | |||
1955 | return mdiff.textdiff(self.rawdata(rev1), self.rawdata(rev2)) |
|
1955 | return mdiff.textdiff(self.rawdata(rev1), self.rawdata(rev2)) | |
1956 |
|
1956 | |||
1957 | def revision(self, nodeorrev): |
|
1957 | def revision(self, nodeorrev): | |
1958 | """return an uncompressed revision of a given node or revision |
|
1958 | """return an uncompressed revision of a given node or revision | |
1959 | number. |
|
1959 | number. | |
1960 | """ |
|
1960 | """ | |
1961 | return self._revisiondata(nodeorrev) |
|
1961 | return self._revisiondata(nodeorrev) | |
1962 |
|
1962 | |||
1963 | def sidedata(self, nodeorrev, _df=None): |
|
1963 | def sidedata(self, nodeorrev, _df=None): | |
1964 | """a map of extra data related to the changeset but not part of the hash |
|
1964 | """a map of extra data related to the changeset but not part of the hash | |
1965 |
|
1965 | |||
1966 | This function currently return a dictionary. However, more advanced |
|
1966 | This function currently return a dictionary. However, more advanced | |
1967 | mapping object will likely be used in the future for a more |
|
1967 | mapping object will likely be used in the future for a more | |
1968 | efficient/lazy code. |
|
1968 | efficient/lazy code. | |
1969 | """ |
|
1969 | """ | |
1970 | # deal with <nodeorrev> argument type |
|
1970 | # deal with <nodeorrev> argument type | |
1971 | if isinstance(nodeorrev, int): |
|
1971 | if isinstance(nodeorrev, int): | |
1972 | rev = nodeorrev |
|
1972 | rev = nodeorrev | |
1973 | else: |
|
1973 | else: | |
1974 | rev = self.rev(nodeorrev) |
|
1974 | rev = self.rev(nodeorrev) | |
1975 | return self._sidedata(rev) |
|
1975 | return self._sidedata(rev) | |
1976 |
|
1976 | |||
1977 | def _revisiondata(self, nodeorrev, _df=None, raw=False): |
|
1977 | def _revisiondata(self, nodeorrev, _df=None, raw=False): | |
1978 | # deal with <nodeorrev> argument type |
|
1978 | # deal with <nodeorrev> argument type | |
1979 | if isinstance(nodeorrev, int): |
|
1979 | if isinstance(nodeorrev, int): | |
1980 | rev = nodeorrev |
|
1980 | rev = nodeorrev | |
1981 | node = self.node(rev) |
|
1981 | node = self.node(rev) | |
1982 | else: |
|
1982 | else: | |
1983 | node = nodeorrev |
|
1983 | node = nodeorrev | |
1984 | rev = None |
|
1984 | rev = None | |
1985 |
|
1985 | |||
1986 | # fast path the special `nullid` rev |
|
1986 | # fast path the special `nullid` rev | |
1987 | if node == self.nullid: |
|
1987 | if node == self.nullid: | |
1988 | return b"" |
|
1988 | return b"" | |
1989 |
|
1989 | |||
1990 | # ``rawtext`` is the text as stored inside the revlog. Might be the |
|
1990 | # ``rawtext`` is the text as stored inside the revlog. Might be the | |
1991 | # revision or might need to be processed to retrieve the revision. |
|
1991 | # revision or might need to be processed to retrieve the revision. | |
1992 | rev, rawtext, validated = self._rawtext(node, rev, _df=_df) |
|
1992 | rev, rawtext, validated = self._rawtext(node, rev, _df=_df) | |
1993 |
|
1993 | |||
1994 | if raw and validated: |
|
1994 | if raw and validated: | |
1995 | # if we don't want to process the raw text and that raw |
|
1995 | # if we don't want to process the raw text and that raw | |
1996 | # text is cached, we can exit early. |
|
1996 | # text is cached, we can exit early. | |
1997 | return rawtext |
|
1997 | return rawtext | |
1998 | if rev is None: |
|
1998 | if rev is None: | |
1999 | rev = self.rev(node) |
|
1999 | rev = self.rev(node) | |
2000 | # the revlog's flag for this revision |
|
2000 | # the revlog's flag for this revision | |
2001 | # (usually alter its state or content) |
|
2001 | # (usually alter its state or content) | |
2002 | flags = self.flags(rev) |
|
2002 | flags = self.flags(rev) | |
2003 |
|
2003 | |||
2004 | if validated and flags == REVIDX_DEFAULT_FLAGS: |
|
2004 | if validated and flags == REVIDX_DEFAULT_FLAGS: | |
2005 | # no extra flags set, no flag processor runs, text = rawtext |
|
2005 | # no extra flags set, no flag processor runs, text = rawtext | |
2006 | return rawtext |
|
2006 | return rawtext | |
2007 |
|
2007 | |||
2008 | if raw: |
|
2008 | if raw: | |
2009 | validatehash = flagutil.processflagsraw(self, rawtext, flags) |
|
2009 | validatehash = flagutil.processflagsraw(self, rawtext, flags) | |
2010 | text = rawtext |
|
2010 | text = rawtext | |
2011 | else: |
|
2011 | else: | |
2012 | r = flagutil.processflagsread(self, rawtext, flags) |
|
2012 | r = flagutil.processflagsread(self, rawtext, flags) | |
2013 | text, validatehash = r |
|
2013 | text, validatehash = r | |
2014 | if validatehash: |
|
2014 | if validatehash: | |
2015 | self.checkhash(text, node, rev=rev) |
|
2015 | self.checkhash(text, node, rev=rev) | |
2016 | if not validated: |
|
2016 | if not validated: | |
2017 | self._revisioncache = (node, rev, rawtext) |
|
2017 | self._revisioncache = (node, rev, rawtext) | |
2018 |
|
2018 | |||
2019 | return text |
|
2019 | return text | |
2020 |
|
2020 | |||
2021 | def _rawtext(self, node, rev, _df=None): |
|
2021 | def _rawtext(self, node, rev, _df=None): | |
2022 | """return the possibly unvalidated rawtext for a revision |
|
2022 | """return the possibly unvalidated rawtext for a revision | |
2023 |
|
2023 | |||
2024 | returns (rev, rawtext, validated) |
|
2024 | returns (rev, rawtext, validated) | |
2025 | """ |
|
2025 | """ | |
2026 |
|
2026 | |||
2027 | # revision in the cache (could be useful to apply delta) |
|
2027 | # revision in the cache (could be useful to apply delta) | |
2028 | cachedrev = None |
|
2028 | cachedrev = None | |
2029 | # An intermediate text to apply deltas to |
|
2029 | # An intermediate text to apply deltas to | |
2030 | basetext = None |
|
2030 | basetext = None | |
2031 |
|
2031 | |||
2032 | # Check if we have the entry in cache |
|
2032 | # Check if we have the entry in cache | |
2033 | # The cache entry looks like (node, rev, rawtext) |
|
2033 | # The cache entry looks like (node, rev, rawtext) | |
2034 | if self._revisioncache: |
|
2034 | if self._revisioncache: | |
2035 | if self._revisioncache[0] == node: |
|
2035 | if self._revisioncache[0] == node: | |
2036 | return (rev, self._revisioncache[2], True) |
|
2036 | return (rev, self._revisioncache[2], True) | |
2037 | cachedrev = self._revisioncache[1] |
|
2037 | cachedrev = self._revisioncache[1] | |
2038 |
|
2038 | |||
2039 | if rev is None: |
|
2039 | if rev is None: | |
2040 | rev = self.rev(node) |
|
2040 | rev = self.rev(node) | |
2041 |
|
2041 | |||
2042 | chain, stopped = self._deltachain(rev, stoprev=cachedrev) |
|
2042 | chain, stopped = self._deltachain(rev, stoprev=cachedrev) | |
2043 | if stopped: |
|
2043 | if stopped: | |
2044 | basetext = self._revisioncache[2] |
|
2044 | basetext = self._revisioncache[2] | |
2045 |
|
2045 | |||
2046 | # drop cache to save memory, the caller is expected to |
|
2046 | # drop cache to save memory, the caller is expected to | |
2047 | # update self._revisioncache after validating the text |
|
2047 | # update self._revisioncache after validating the text | |
2048 | self._revisioncache = None |
|
2048 | self._revisioncache = None | |
2049 |
|
2049 | |||
2050 | targetsize = None |
|
2050 | targetsize = None | |
2051 | rawsize = self.index[rev][2] |
|
2051 | rawsize = self.index[rev][2] | |
2052 | if 0 <= rawsize: |
|
2052 | if 0 <= rawsize: | |
2053 | targetsize = 4 * rawsize |
|
2053 | targetsize = 4 * rawsize | |
2054 |
|
2054 | |||
2055 | bins = self._chunks(chain, df=_df, targetsize=targetsize) |
|
2055 | bins = self._chunks(chain, df=_df, targetsize=targetsize) | |
2056 | if basetext is None: |
|
2056 | if basetext is None: | |
2057 | basetext = bytes(bins[0]) |
|
2057 | basetext = bytes(bins[0]) | |
2058 | bins = bins[1:] |
|
2058 | bins = bins[1:] | |
2059 |
|
2059 | |||
2060 | rawtext = mdiff.patches(basetext, bins) |
|
2060 | rawtext = mdiff.patches(basetext, bins) | |
2061 | del basetext # let us have a chance to free memory early |
|
2061 | del basetext # let us have a chance to free memory early | |
2062 | return (rev, rawtext, False) |
|
2062 | return (rev, rawtext, False) | |
2063 |
|
2063 | |||
2064 | def _sidedata(self, rev): |
|
2064 | def _sidedata(self, rev): | |
2065 | """Return the sidedata for a given revision number.""" |
|
2065 | """Return the sidedata for a given revision number.""" | |
2066 | index_entry = self.index[rev] |
|
2066 | index_entry = self.index[rev] | |
2067 | sidedata_offset = index_entry[8] |
|
2067 | sidedata_offset = index_entry[8] | |
2068 | sidedata_size = index_entry[9] |
|
2068 | sidedata_size = index_entry[9] | |
2069 |
|
2069 | |||
2070 | if self._inline: |
|
2070 | if self._inline: | |
2071 | sidedata_offset += self.index.entry_size * (1 + rev) |
|
2071 | sidedata_offset += self.index.entry_size * (1 + rev) | |
2072 | if sidedata_size == 0: |
|
2072 | if sidedata_size == 0: | |
2073 | return {} |
|
2073 | return {} | |
2074 |
|
2074 | |||
2075 | if self._docket.sidedata_end < sidedata_offset + sidedata_size: |
|
2075 | if self._docket.sidedata_end < sidedata_offset + sidedata_size: | |
2076 | filename = self._sidedatafile |
|
2076 | filename = self._sidedatafile | |
2077 | end = self._docket.sidedata_end |
|
2077 | end = self._docket.sidedata_end | |
2078 | offset = sidedata_offset |
|
2078 | offset = sidedata_offset | |
2079 | length = sidedata_size |
|
2079 | length = sidedata_size | |
2080 | m = FILE_TOO_SHORT_MSG % (filename, length, offset, end) |
|
2080 | m = FILE_TOO_SHORT_MSG % (filename, length, offset, end) | |
2081 | raise error.RevlogError(m) |
|
2081 | raise error.RevlogError(m) | |
2082 |
|
2082 | |||
2083 | comp_segment = self._segmentfile_sidedata.read_chunk( |
|
2083 | comp_segment = self._segmentfile_sidedata.read_chunk( | |
2084 | sidedata_offset, sidedata_size |
|
2084 | sidedata_offset, sidedata_size | |
2085 | ) |
|
2085 | ) | |
2086 |
|
2086 | |||
2087 | comp = self.index[rev][11] |
|
2087 | comp = self.index[rev][11] | |
2088 | if comp == COMP_MODE_PLAIN: |
|
2088 | if comp == COMP_MODE_PLAIN: | |
2089 | segment = comp_segment |
|
2089 | segment = comp_segment | |
2090 | elif comp == COMP_MODE_DEFAULT: |
|
2090 | elif comp == COMP_MODE_DEFAULT: | |
2091 | segment = self._decompressor(comp_segment) |
|
2091 | segment = self._decompressor(comp_segment) | |
2092 | elif comp == COMP_MODE_INLINE: |
|
2092 | elif comp == COMP_MODE_INLINE: | |
2093 | segment = self.decompress(comp_segment) |
|
2093 | segment = self.decompress(comp_segment) | |
2094 | else: |
|
2094 | else: | |
2095 | msg = b'unknown compression mode %d' |
|
2095 | msg = b'unknown compression mode %d' | |
2096 | msg %= comp |
|
2096 | msg %= comp | |
2097 | raise error.RevlogError(msg) |
|
2097 | raise error.RevlogError(msg) | |
2098 |
|
2098 | |||
2099 | sidedata = sidedatautil.deserialize_sidedata(segment) |
|
2099 | sidedata = sidedatautil.deserialize_sidedata(segment) | |
2100 | return sidedata |
|
2100 | return sidedata | |
2101 |
|
2101 | |||
2102 |
def rawdata(self, nodeorrev |
|
2102 | def rawdata(self, nodeorrev): | |
2103 | """return an uncompressed raw data of a given node or revision number. |
|
2103 | """return an uncompressed raw data of a given node or revision number.""" | |
2104 |
|
2104 | return self._revisiondata(nodeorrev, raw=True) | ||
2105 | _df - an existing file handle to read from. (internal-only) |
|
|||
2106 | """ |
|
|||
2107 | return self._revisiondata(nodeorrev, _df, raw=True) |
|
|||
2108 |
|
2105 | |||
2109 | def hash(self, text, p1, p2): |
|
2106 | def hash(self, text, p1, p2): | |
2110 | """Compute a node hash. |
|
2107 | """Compute a node hash. | |
2111 |
|
2108 | |||
2112 | Available as a function so that subclasses can replace the hash |
|
2109 | Available as a function so that subclasses can replace the hash | |
2113 | as needed. |
|
2110 | as needed. | |
2114 | """ |
|
2111 | """ | |
2115 | return storageutil.hashrevisionsha1(text, p1, p2) |
|
2112 | return storageutil.hashrevisionsha1(text, p1, p2) | |
2116 |
|
2113 | |||
2117 | def checkhash(self, text, node, p1=None, p2=None, rev=None): |
|
2114 | def checkhash(self, text, node, p1=None, p2=None, rev=None): | |
2118 | """Check node hash integrity. |
|
2115 | """Check node hash integrity. | |
2119 |
|
2116 | |||
2120 | Available as a function so that subclasses can extend hash mismatch |
|
2117 | Available as a function so that subclasses can extend hash mismatch | |
2121 | behaviors as needed. |
|
2118 | behaviors as needed. | |
2122 | """ |
|
2119 | """ | |
2123 | try: |
|
2120 | try: | |
2124 | if p1 is None and p2 is None: |
|
2121 | if p1 is None and p2 is None: | |
2125 | p1, p2 = self.parents(node) |
|
2122 | p1, p2 = self.parents(node) | |
2126 | if node != self.hash(text, p1, p2): |
|
2123 | if node != self.hash(text, p1, p2): | |
2127 | # Clear the revision cache on hash failure. The revision cache |
|
2124 | # Clear the revision cache on hash failure. The revision cache | |
2128 | # only stores the raw revision and clearing the cache does have |
|
2125 | # only stores the raw revision and clearing the cache does have | |
2129 | # the side-effect that we won't have a cache hit when the raw |
|
2126 | # the side-effect that we won't have a cache hit when the raw | |
2130 | # revision data is accessed. But this case should be rare and |
|
2127 | # revision data is accessed. But this case should be rare and | |
2131 | # it is extra work to teach the cache about the hash |
|
2128 | # it is extra work to teach the cache about the hash | |
2132 | # verification state. |
|
2129 | # verification state. | |
2133 | if self._revisioncache and self._revisioncache[0] == node: |
|
2130 | if self._revisioncache and self._revisioncache[0] == node: | |
2134 | self._revisioncache = None |
|
2131 | self._revisioncache = None | |
2135 |
|
2132 | |||
2136 | revornode = rev |
|
2133 | revornode = rev | |
2137 | if revornode is None: |
|
2134 | if revornode is None: | |
2138 | revornode = templatefilters.short(hex(node)) |
|
2135 | revornode = templatefilters.short(hex(node)) | |
2139 | raise error.RevlogError( |
|
2136 | raise error.RevlogError( | |
2140 | _(b"integrity check failed on %s:%s") |
|
2137 | _(b"integrity check failed on %s:%s") | |
2141 | % (self.display_id, pycompat.bytestr(revornode)) |
|
2138 | % (self.display_id, pycompat.bytestr(revornode)) | |
2142 | ) |
|
2139 | ) | |
2143 | except error.RevlogError: |
|
2140 | except error.RevlogError: | |
2144 | if self._censorable and storageutil.iscensoredtext(text): |
|
2141 | if self._censorable and storageutil.iscensoredtext(text): | |
2145 | raise error.CensoredNodeError(self.display_id, node, text) |
|
2142 | raise error.CensoredNodeError(self.display_id, node, text) | |
2146 | raise |
|
2143 | raise | |
2147 |
|
2144 | |||
2148 | @property |
|
2145 | @property | |
2149 | def _split_index_file(self): |
|
2146 | def _split_index_file(self): | |
2150 | """the path where to expect the index of an ongoing splitting operation |
|
2147 | """the path where to expect the index of an ongoing splitting operation | |
2151 |
|
2148 | |||
2152 | The file will only exist if a splitting operation is in progress, but |
|
2149 | The file will only exist if a splitting operation is in progress, but | |
2153 | it is always expected at the same location.""" |
|
2150 | it is always expected at the same location.""" | |
2154 | parts = self.radix.split(b'/') |
|
2151 | parts = self.radix.split(b'/') | |
2155 | if len(parts) > 1: |
|
2152 | if len(parts) > 1: | |
2156 | # adds a '-s' prefix to the ``data/` or `meta/` base |
|
2153 | # adds a '-s' prefix to the ``data/` or `meta/` base | |
2157 | head = parts[0] + b'-s' |
|
2154 | head = parts[0] + b'-s' | |
2158 | mids = parts[1:-1] |
|
2155 | mids = parts[1:-1] | |
2159 | tail = parts[-1] + b'.i' |
|
2156 | tail = parts[-1] + b'.i' | |
2160 | pieces = [head] + mids + [tail] |
|
2157 | pieces = [head] + mids + [tail] | |
2161 | return b'/'.join(pieces) |
|
2158 | return b'/'.join(pieces) | |
2162 | else: |
|
2159 | else: | |
2163 | # the revlog is stored at the root of the store (changelog or |
|
2160 | # the revlog is stored at the root of the store (changelog or | |
2164 | # manifest), no risk of collision. |
|
2161 | # manifest), no risk of collision. | |
2165 | return self.radix + b'.i.s' |
|
2162 | return self.radix + b'.i.s' | |
2166 |
|
2163 | |||
2167 | def _enforceinlinesize(self, tr, side_write=True): |
|
2164 | def _enforceinlinesize(self, tr, side_write=True): | |
2168 | """Check if the revlog is too big for inline and convert if so. |
|
2165 | """Check if the revlog is too big for inline and convert if so. | |
2169 |
|
2166 | |||
2170 | This should be called after revisions are added to the revlog. If the |
|
2167 | This should be called after revisions are added to the revlog. If the | |
2171 | revlog has grown too large to be an inline revlog, it will convert it |
|
2168 | revlog has grown too large to be an inline revlog, it will convert it | |
2172 | to use multiple index and data files. |
|
2169 | to use multiple index and data files. | |
2173 | """ |
|
2170 | """ | |
2174 | tiprev = len(self) - 1 |
|
2171 | tiprev = len(self) - 1 | |
2175 | total_size = self.start(tiprev) + self.length(tiprev) |
|
2172 | total_size = self.start(tiprev) + self.length(tiprev) | |
2176 | if not self._inline or total_size < _maxinline: |
|
2173 | if not self._inline or total_size < _maxinline: | |
2177 | return |
|
2174 | return | |
2178 |
|
2175 | |||
2179 | troffset = tr.findoffset(self._indexfile) |
|
2176 | troffset = tr.findoffset(self._indexfile) | |
2180 | if troffset is None: |
|
2177 | if troffset is None: | |
2181 | raise error.RevlogError( |
|
2178 | raise error.RevlogError( | |
2182 | _(b"%s not found in the transaction") % self._indexfile |
|
2179 | _(b"%s not found in the transaction") % self._indexfile | |
2183 | ) |
|
2180 | ) | |
2184 | if troffset: |
|
2181 | if troffset: | |
2185 | tr.addbackup(self._indexfile, for_offset=True) |
|
2182 | tr.addbackup(self._indexfile, for_offset=True) | |
2186 | tr.add(self._datafile, 0) |
|
2183 | tr.add(self._datafile, 0) | |
2187 |
|
2184 | |||
2188 | existing_handles = False |
|
2185 | existing_handles = False | |
2189 | if self._writinghandles is not None: |
|
2186 | if self._writinghandles is not None: | |
2190 | existing_handles = True |
|
2187 | existing_handles = True | |
2191 | fp = self._writinghandles[0] |
|
2188 | fp = self._writinghandles[0] | |
2192 | fp.flush() |
|
2189 | fp.flush() | |
2193 | fp.close() |
|
2190 | fp.close() | |
2194 | # We can't use the cached file handle after close(). So prevent |
|
2191 | # We can't use the cached file handle after close(). So prevent | |
2195 | # its usage. |
|
2192 | # its usage. | |
2196 | self._writinghandles = None |
|
2193 | self._writinghandles = None | |
2197 | self._segmentfile.writing_handle = None |
|
2194 | self._segmentfile.writing_handle = None | |
2198 | # No need to deal with sidedata writing handle as it is only |
|
2195 | # No need to deal with sidedata writing handle as it is only | |
2199 | # relevant with revlog-v2 which is never inline, not reaching |
|
2196 | # relevant with revlog-v2 which is never inline, not reaching | |
2200 | # this code |
|
2197 | # this code | |
2201 | if side_write: |
|
2198 | if side_write: | |
2202 | old_index_file_path = self._indexfile |
|
2199 | old_index_file_path = self._indexfile | |
2203 | new_index_file_path = self._split_index_file |
|
2200 | new_index_file_path = self._split_index_file | |
2204 | opener = self.opener |
|
2201 | opener = self.opener | |
2205 | weak_self = weakref.ref(self) |
|
2202 | weak_self = weakref.ref(self) | |
2206 |
|
2203 | |||
2207 | # the "split" index replace the real index when the transaction is finalized |
|
2204 | # the "split" index replace the real index when the transaction is finalized | |
2208 | def finalize_callback(tr): |
|
2205 | def finalize_callback(tr): | |
2209 | opener.rename( |
|
2206 | opener.rename( | |
2210 | new_index_file_path, |
|
2207 | new_index_file_path, | |
2211 | old_index_file_path, |
|
2208 | old_index_file_path, | |
2212 | checkambig=True, |
|
2209 | checkambig=True, | |
2213 | ) |
|
2210 | ) | |
2214 | maybe_self = weak_self() |
|
2211 | maybe_self = weak_self() | |
2215 | if maybe_self is not None: |
|
2212 | if maybe_self is not None: | |
2216 | maybe_self._indexfile = old_index_file_path |
|
2213 | maybe_self._indexfile = old_index_file_path | |
2217 |
|
2214 | |||
2218 | def abort_callback(tr): |
|
2215 | def abort_callback(tr): | |
2219 | maybe_self = weak_self() |
|
2216 | maybe_self = weak_self() | |
2220 | if maybe_self is not None: |
|
2217 | if maybe_self is not None: | |
2221 | maybe_self._indexfile = old_index_file_path |
|
2218 | maybe_self._indexfile = old_index_file_path | |
2222 |
|
2219 | |||
2223 | tr.registertmp(new_index_file_path) |
|
2220 | tr.registertmp(new_index_file_path) | |
2224 | if self.target[1] is not None: |
|
2221 | if self.target[1] is not None: | |
2225 | callback_id = b'000-revlog-split-%d-%s' % self.target |
|
2222 | callback_id = b'000-revlog-split-%d-%s' % self.target | |
2226 | else: |
|
2223 | else: | |
2227 | callback_id = b'000-revlog-split-%d' % self.target[0] |
|
2224 | callback_id = b'000-revlog-split-%d' % self.target[0] | |
2228 | tr.addfinalize(callback_id, finalize_callback) |
|
2225 | tr.addfinalize(callback_id, finalize_callback) | |
2229 | tr.addabort(callback_id, abort_callback) |
|
2226 | tr.addabort(callback_id, abort_callback) | |
2230 |
|
2227 | |||
2231 | new_dfh = self._datafp(b'w+') |
|
2228 | new_dfh = self._datafp(b'w+') | |
2232 | new_dfh.truncate(0) # drop any potentially existing data |
|
2229 | new_dfh.truncate(0) # drop any potentially existing data | |
2233 | try: |
|
2230 | try: | |
2234 | with self._indexfp() as read_ifh: |
|
2231 | with self._indexfp() as read_ifh: | |
2235 | for r in self: |
|
2232 | for r in self: | |
2236 | new_dfh.write(self._getsegmentforrevs(r, r, df=read_ifh)[1]) |
|
2233 | new_dfh.write(self._getsegmentforrevs(r, r, df=read_ifh)[1]) | |
2237 | new_dfh.flush() |
|
2234 | new_dfh.flush() | |
2238 |
|
2235 | |||
2239 | if side_write: |
|
2236 | if side_write: | |
2240 | self._indexfile = new_index_file_path |
|
2237 | self._indexfile = new_index_file_path | |
2241 | with self.__index_new_fp() as fp: |
|
2238 | with self.__index_new_fp() as fp: | |
2242 | self._format_flags &= ~FLAG_INLINE_DATA |
|
2239 | self._format_flags &= ~FLAG_INLINE_DATA | |
2243 | self._inline = False |
|
2240 | self._inline = False | |
2244 | for i in self: |
|
2241 | for i in self: | |
2245 | e = self.index.entry_binary(i) |
|
2242 | e = self.index.entry_binary(i) | |
2246 | if i == 0 and self._docket is None: |
|
2243 | if i == 0 and self._docket is None: | |
2247 | header = self._format_flags | self._format_version |
|
2244 | header = self._format_flags | self._format_version | |
2248 | header = self.index.pack_header(header) |
|
2245 | header = self.index.pack_header(header) | |
2249 | e = header + e |
|
2246 | e = header + e | |
2250 | fp.write(e) |
|
2247 | fp.write(e) | |
2251 | if self._docket is not None: |
|
2248 | if self._docket is not None: | |
2252 | self._docket.index_end = fp.tell() |
|
2249 | self._docket.index_end = fp.tell() | |
2253 |
|
2250 | |||
2254 | # If we don't use side-write, the temp file replace the real |
|
2251 | # If we don't use side-write, the temp file replace the real | |
2255 | # index when we exit the context manager |
|
2252 | # index when we exit the context manager | |
2256 |
|
2253 | |||
2257 | nodemaputil.setup_persistent_nodemap(tr, self) |
|
2254 | nodemaputil.setup_persistent_nodemap(tr, self) | |
2258 | self._segmentfile = randomaccessfile.randomaccessfile( |
|
2255 | self._segmentfile = randomaccessfile.randomaccessfile( | |
2259 | self.opener, |
|
2256 | self.opener, | |
2260 | self._datafile, |
|
2257 | self._datafile, | |
2261 | self._chunkcachesize, |
|
2258 | self._chunkcachesize, | |
2262 | ) |
|
2259 | ) | |
2263 |
|
2260 | |||
2264 | if existing_handles: |
|
2261 | if existing_handles: | |
2265 | # switched from inline to conventional reopen the index |
|
2262 | # switched from inline to conventional reopen the index | |
2266 | ifh = self.__index_write_fp() |
|
2263 | ifh = self.__index_write_fp() | |
2267 | self._writinghandles = (ifh, new_dfh, None) |
|
2264 | self._writinghandles = (ifh, new_dfh, None) | |
2268 | self._segmentfile.writing_handle = new_dfh |
|
2265 | self._segmentfile.writing_handle = new_dfh | |
2269 | new_dfh = None |
|
2266 | new_dfh = None | |
2270 | # No need to deal with sidedata writing handle as it is only |
|
2267 | # No need to deal with sidedata writing handle as it is only | |
2271 | # relevant with revlog-v2 which is never inline, not reaching |
|
2268 | # relevant with revlog-v2 which is never inline, not reaching | |
2272 | # this code |
|
2269 | # this code | |
2273 | finally: |
|
2270 | finally: | |
2274 | if new_dfh is not None: |
|
2271 | if new_dfh is not None: | |
2275 | new_dfh.close() |
|
2272 | new_dfh.close() | |
2276 |
|
2273 | |||
2277 | def _nodeduplicatecallback(self, transaction, node): |
|
2274 | def _nodeduplicatecallback(self, transaction, node): | |
2278 | """called when trying to add a node already stored.""" |
|
2275 | """called when trying to add a node already stored.""" | |
2279 |
|
2276 | |||
2280 | @contextlib.contextmanager |
|
2277 | @contextlib.contextmanager | |
2281 | def reading(self): |
|
2278 | def reading(self): | |
2282 | """Context manager that keeps data and sidedata files open for reading""" |
|
2279 | """Context manager that keeps data and sidedata files open for reading""" | |
2283 | if len(self.index) == 0: |
|
2280 | if len(self.index) == 0: | |
2284 | yield # nothing to be read |
|
2281 | yield # nothing to be read | |
2285 | else: |
|
2282 | else: | |
2286 | with self._segmentfile.reading(): |
|
2283 | with self._segmentfile.reading(): | |
2287 | with self._segmentfile_sidedata.reading(): |
|
2284 | with self._segmentfile_sidedata.reading(): | |
2288 | yield |
|
2285 | yield | |
2289 |
|
2286 | |||
2290 | @contextlib.contextmanager |
|
2287 | @contextlib.contextmanager | |
2291 | def _writing(self, transaction): |
|
2288 | def _writing(self, transaction): | |
2292 | if self._trypending: |
|
2289 | if self._trypending: | |
2293 | msg = b'try to write in a `trypending` revlog: %s' |
|
2290 | msg = b'try to write in a `trypending` revlog: %s' | |
2294 | msg %= self.display_id |
|
2291 | msg %= self.display_id | |
2295 | raise error.ProgrammingError(msg) |
|
2292 | raise error.ProgrammingError(msg) | |
2296 | if self._writinghandles is not None: |
|
2293 | if self._writinghandles is not None: | |
2297 | yield |
|
2294 | yield | |
2298 | else: |
|
2295 | else: | |
2299 | ifh = dfh = sdfh = None |
|
2296 | ifh = dfh = sdfh = None | |
2300 | try: |
|
2297 | try: | |
2301 | r = len(self) |
|
2298 | r = len(self) | |
2302 | # opening the data file. |
|
2299 | # opening the data file. | |
2303 | dsize = 0 |
|
2300 | dsize = 0 | |
2304 | if r: |
|
2301 | if r: | |
2305 | dsize = self.end(r - 1) |
|
2302 | dsize = self.end(r - 1) | |
2306 | dfh = None |
|
2303 | dfh = None | |
2307 | if not self._inline: |
|
2304 | if not self._inline: | |
2308 | try: |
|
2305 | try: | |
2309 | dfh = self._datafp(b"r+") |
|
2306 | dfh = self._datafp(b"r+") | |
2310 | if self._docket is None: |
|
2307 | if self._docket is None: | |
2311 | dfh.seek(0, os.SEEK_END) |
|
2308 | dfh.seek(0, os.SEEK_END) | |
2312 | else: |
|
2309 | else: | |
2313 | dfh.seek(self._docket.data_end, os.SEEK_SET) |
|
2310 | dfh.seek(self._docket.data_end, os.SEEK_SET) | |
2314 | except FileNotFoundError: |
|
2311 | except FileNotFoundError: | |
2315 | dfh = self._datafp(b"w+") |
|
2312 | dfh = self._datafp(b"w+") | |
2316 | transaction.add(self._datafile, dsize) |
|
2313 | transaction.add(self._datafile, dsize) | |
2317 | if self._sidedatafile is not None: |
|
2314 | if self._sidedatafile is not None: | |
2318 | # revlog-v2 does not inline, help Pytype |
|
2315 | # revlog-v2 does not inline, help Pytype | |
2319 | assert dfh is not None |
|
2316 | assert dfh is not None | |
2320 | try: |
|
2317 | try: | |
2321 | sdfh = self.opener(self._sidedatafile, mode=b"r+") |
|
2318 | sdfh = self.opener(self._sidedatafile, mode=b"r+") | |
2322 | dfh.seek(self._docket.sidedata_end, os.SEEK_SET) |
|
2319 | dfh.seek(self._docket.sidedata_end, os.SEEK_SET) | |
2323 | except FileNotFoundError: |
|
2320 | except FileNotFoundError: | |
2324 | sdfh = self.opener(self._sidedatafile, mode=b"w+") |
|
2321 | sdfh = self.opener(self._sidedatafile, mode=b"w+") | |
2325 | transaction.add( |
|
2322 | transaction.add( | |
2326 | self._sidedatafile, self._docket.sidedata_end |
|
2323 | self._sidedatafile, self._docket.sidedata_end | |
2327 | ) |
|
2324 | ) | |
2328 |
|
2325 | |||
2329 | # opening the index file. |
|
2326 | # opening the index file. | |
2330 | isize = r * self.index.entry_size |
|
2327 | isize = r * self.index.entry_size | |
2331 | ifh = self.__index_write_fp() |
|
2328 | ifh = self.__index_write_fp() | |
2332 | if self._inline: |
|
2329 | if self._inline: | |
2333 | transaction.add(self._indexfile, dsize + isize) |
|
2330 | transaction.add(self._indexfile, dsize + isize) | |
2334 | else: |
|
2331 | else: | |
2335 | transaction.add(self._indexfile, isize) |
|
2332 | transaction.add(self._indexfile, isize) | |
2336 | # exposing all file handle for writing. |
|
2333 | # exposing all file handle for writing. | |
2337 | self._writinghandles = (ifh, dfh, sdfh) |
|
2334 | self._writinghandles = (ifh, dfh, sdfh) | |
2338 | self._segmentfile.writing_handle = ifh if self._inline else dfh |
|
2335 | self._segmentfile.writing_handle = ifh if self._inline else dfh | |
2339 | self._segmentfile_sidedata.writing_handle = sdfh |
|
2336 | self._segmentfile_sidedata.writing_handle = sdfh | |
2340 | yield |
|
2337 | yield | |
2341 | if self._docket is not None: |
|
2338 | if self._docket is not None: | |
2342 | self._write_docket(transaction) |
|
2339 | self._write_docket(transaction) | |
2343 | finally: |
|
2340 | finally: | |
2344 | self._writinghandles = None |
|
2341 | self._writinghandles = None | |
2345 | self._segmentfile.writing_handle = None |
|
2342 | self._segmentfile.writing_handle = None | |
2346 | self._segmentfile_sidedata.writing_handle = None |
|
2343 | self._segmentfile_sidedata.writing_handle = None | |
2347 | if dfh is not None: |
|
2344 | if dfh is not None: | |
2348 | dfh.close() |
|
2345 | dfh.close() | |
2349 | if sdfh is not None: |
|
2346 | if sdfh is not None: | |
2350 | sdfh.close() |
|
2347 | sdfh.close() | |
2351 | # closing the index file last to avoid exposing referent to |
|
2348 | # closing the index file last to avoid exposing referent to | |
2352 | # potential unflushed data content. |
|
2349 | # potential unflushed data content. | |
2353 | if ifh is not None: |
|
2350 | if ifh is not None: | |
2354 | ifh.close() |
|
2351 | ifh.close() | |
2355 |
|
2352 | |||
2356 | def _write_docket(self, transaction): |
|
2353 | def _write_docket(self, transaction): | |
2357 | """write the current docket on disk |
|
2354 | """write the current docket on disk | |
2358 |
|
2355 | |||
2359 | Exist as a method to help changelog to implement transaction logic |
|
2356 | Exist as a method to help changelog to implement transaction logic | |
2360 |
|
2357 | |||
2361 | We could also imagine using the same transaction logic for all revlog |
|
2358 | We could also imagine using the same transaction logic for all revlog | |
2362 | since docket are cheap.""" |
|
2359 | since docket are cheap.""" | |
2363 | self._docket.write(transaction) |
|
2360 | self._docket.write(transaction) | |
2364 |
|
2361 | |||
2365 | def addrevision( |
|
2362 | def addrevision( | |
2366 | self, |
|
2363 | self, | |
2367 | text, |
|
2364 | text, | |
2368 | transaction, |
|
2365 | transaction, | |
2369 | link, |
|
2366 | link, | |
2370 | p1, |
|
2367 | p1, | |
2371 | p2, |
|
2368 | p2, | |
2372 | cachedelta=None, |
|
2369 | cachedelta=None, | |
2373 | node=None, |
|
2370 | node=None, | |
2374 | flags=REVIDX_DEFAULT_FLAGS, |
|
2371 | flags=REVIDX_DEFAULT_FLAGS, | |
2375 | deltacomputer=None, |
|
2372 | deltacomputer=None, | |
2376 | sidedata=None, |
|
2373 | sidedata=None, | |
2377 | ): |
|
2374 | ): | |
2378 | """add a revision to the log |
|
2375 | """add a revision to the log | |
2379 |
|
2376 | |||
2380 | text - the revision data to add |
|
2377 | text - the revision data to add | |
2381 | transaction - the transaction object used for rollback |
|
2378 | transaction - the transaction object used for rollback | |
2382 | link - the linkrev data to add |
|
2379 | link - the linkrev data to add | |
2383 | p1, p2 - the parent nodeids of the revision |
|
2380 | p1, p2 - the parent nodeids of the revision | |
2384 | cachedelta - an optional precomputed delta |
|
2381 | cachedelta - an optional precomputed delta | |
2385 | node - nodeid of revision; typically node is not specified, and it is |
|
2382 | node - nodeid of revision; typically node is not specified, and it is | |
2386 | computed by default as hash(text, p1, p2), however subclasses might |
|
2383 | computed by default as hash(text, p1, p2), however subclasses might | |
2387 | use different hashing method (and override checkhash() in such case) |
|
2384 | use different hashing method (and override checkhash() in such case) | |
2388 | flags - the known flags to set on the revision |
|
2385 | flags - the known flags to set on the revision | |
2389 | deltacomputer - an optional deltacomputer instance shared between |
|
2386 | deltacomputer - an optional deltacomputer instance shared between | |
2390 | multiple calls |
|
2387 | multiple calls | |
2391 | """ |
|
2388 | """ | |
2392 | if link == nullrev: |
|
2389 | if link == nullrev: | |
2393 | raise error.RevlogError( |
|
2390 | raise error.RevlogError( | |
2394 | _(b"attempted to add linkrev -1 to %s") % self.display_id |
|
2391 | _(b"attempted to add linkrev -1 to %s") % self.display_id | |
2395 | ) |
|
2392 | ) | |
2396 |
|
2393 | |||
2397 | if sidedata is None: |
|
2394 | if sidedata is None: | |
2398 | sidedata = {} |
|
2395 | sidedata = {} | |
2399 | elif sidedata and not self.hassidedata: |
|
2396 | elif sidedata and not self.hassidedata: | |
2400 | raise error.ProgrammingError( |
|
2397 | raise error.ProgrammingError( | |
2401 | _(b"trying to add sidedata to a revlog who don't support them") |
|
2398 | _(b"trying to add sidedata to a revlog who don't support them") | |
2402 | ) |
|
2399 | ) | |
2403 |
|
2400 | |||
2404 | if flags: |
|
2401 | if flags: | |
2405 | node = node or self.hash(text, p1, p2) |
|
2402 | node = node or self.hash(text, p1, p2) | |
2406 |
|
2403 | |||
2407 | rawtext, validatehash = flagutil.processflagswrite(self, text, flags) |
|
2404 | rawtext, validatehash = flagutil.processflagswrite(self, text, flags) | |
2408 |
|
2405 | |||
2409 | # If the flag processor modifies the revision data, ignore any provided |
|
2406 | # If the flag processor modifies the revision data, ignore any provided | |
2410 | # cachedelta. |
|
2407 | # cachedelta. | |
2411 | if rawtext != text: |
|
2408 | if rawtext != text: | |
2412 | cachedelta = None |
|
2409 | cachedelta = None | |
2413 |
|
2410 | |||
2414 | if len(rawtext) > _maxentrysize: |
|
2411 | if len(rawtext) > _maxentrysize: | |
2415 | raise error.RevlogError( |
|
2412 | raise error.RevlogError( | |
2416 | _( |
|
2413 | _( | |
2417 | b"%s: size of %d bytes exceeds maximum revlog storage of 2GiB" |
|
2414 | b"%s: size of %d bytes exceeds maximum revlog storage of 2GiB" | |
2418 | ) |
|
2415 | ) | |
2419 | % (self.display_id, len(rawtext)) |
|
2416 | % (self.display_id, len(rawtext)) | |
2420 | ) |
|
2417 | ) | |
2421 |
|
2418 | |||
2422 | node = node or self.hash(rawtext, p1, p2) |
|
2419 | node = node or self.hash(rawtext, p1, p2) | |
2423 | rev = self.index.get_rev(node) |
|
2420 | rev = self.index.get_rev(node) | |
2424 | if rev is not None: |
|
2421 | if rev is not None: | |
2425 | return rev |
|
2422 | return rev | |
2426 |
|
2423 | |||
2427 | if validatehash: |
|
2424 | if validatehash: | |
2428 | self.checkhash(rawtext, node, p1=p1, p2=p2) |
|
2425 | self.checkhash(rawtext, node, p1=p1, p2=p2) | |
2429 |
|
2426 | |||
2430 | return self.addrawrevision( |
|
2427 | return self.addrawrevision( | |
2431 | rawtext, |
|
2428 | rawtext, | |
2432 | transaction, |
|
2429 | transaction, | |
2433 | link, |
|
2430 | link, | |
2434 | p1, |
|
2431 | p1, | |
2435 | p2, |
|
2432 | p2, | |
2436 | node, |
|
2433 | node, | |
2437 | flags, |
|
2434 | flags, | |
2438 | cachedelta=cachedelta, |
|
2435 | cachedelta=cachedelta, | |
2439 | deltacomputer=deltacomputer, |
|
2436 | deltacomputer=deltacomputer, | |
2440 | sidedata=sidedata, |
|
2437 | sidedata=sidedata, | |
2441 | ) |
|
2438 | ) | |
2442 |
|
2439 | |||
2443 | def addrawrevision( |
|
2440 | def addrawrevision( | |
2444 | self, |
|
2441 | self, | |
2445 | rawtext, |
|
2442 | rawtext, | |
2446 | transaction, |
|
2443 | transaction, | |
2447 | link, |
|
2444 | link, | |
2448 | p1, |
|
2445 | p1, | |
2449 | p2, |
|
2446 | p2, | |
2450 | node, |
|
2447 | node, | |
2451 | flags, |
|
2448 | flags, | |
2452 | cachedelta=None, |
|
2449 | cachedelta=None, | |
2453 | deltacomputer=None, |
|
2450 | deltacomputer=None, | |
2454 | sidedata=None, |
|
2451 | sidedata=None, | |
2455 | ): |
|
2452 | ): | |
2456 | """add a raw revision with known flags, node and parents |
|
2453 | """add a raw revision with known flags, node and parents | |
2457 | useful when reusing a revision not stored in this revlog (ex: received |
|
2454 | useful when reusing a revision not stored in this revlog (ex: received | |
2458 | over wire, or read from an external bundle). |
|
2455 | over wire, or read from an external bundle). | |
2459 | """ |
|
2456 | """ | |
2460 | with self._writing(transaction): |
|
2457 | with self._writing(transaction): | |
2461 | return self._addrevision( |
|
2458 | return self._addrevision( | |
2462 | node, |
|
2459 | node, | |
2463 | rawtext, |
|
2460 | rawtext, | |
2464 | transaction, |
|
2461 | transaction, | |
2465 | link, |
|
2462 | link, | |
2466 | p1, |
|
2463 | p1, | |
2467 | p2, |
|
2464 | p2, | |
2468 | flags, |
|
2465 | flags, | |
2469 | cachedelta, |
|
2466 | cachedelta, | |
2470 | deltacomputer=deltacomputer, |
|
2467 | deltacomputer=deltacomputer, | |
2471 | sidedata=sidedata, |
|
2468 | sidedata=sidedata, | |
2472 | ) |
|
2469 | ) | |
2473 |
|
2470 | |||
2474 | def compress(self, data): |
|
2471 | def compress(self, data): | |
2475 | """Generate a possibly-compressed representation of data.""" |
|
2472 | """Generate a possibly-compressed representation of data.""" | |
2476 | if not data: |
|
2473 | if not data: | |
2477 | return b'', data |
|
2474 | return b'', data | |
2478 |
|
2475 | |||
2479 | compressed = self._compressor.compress(data) |
|
2476 | compressed = self._compressor.compress(data) | |
2480 |
|
2477 | |||
2481 | if compressed: |
|
2478 | if compressed: | |
2482 | # The revlog compressor added the header in the returned data. |
|
2479 | # The revlog compressor added the header in the returned data. | |
2483 | return b'', compressed |
|
2480 | return b'', compressed | |
2484 |
|
2481 | |||
2485 | if data[0:1] == b'\0': |
|
2482 | if data[0:1] == b'\0': | |
2486 | return b'', data |
|
2483 | return b'', data | |
2487 | return b'u', data |
|
2484 | return b'u', data | |
2488 |
|
2485 | |||
2489 | def decompress(self, data): |
|
2486 | def decompress(self, data): | |
2490 | """Decompress a revlog chunk. |
|
2487 | """Decompress a revlog chunk. | |
2491 |
|
2488 | |||
2492 | The chunk is expected to begin with a header identifying the |
|
2489 | The chunk is expected to begin with a header identifying the | |
2493 | format type so it can be routed to an appropriate decompressor. |
|
2490 | format type so it can be routed to an appropriate decompressor. | |
2494 | """ |
|
2491 | """ | |
2495 | if not data: |
|
2492 | if not data: | |
2496 | return data |
|
2493 | return data | |
2497 |
|
2494 | |||
2498 | # Revlogs are read much more frequently than they are written and many |
|
2495 | # Revlogs are read much more frequently than they are written and many | |
2499 | # chunks only take microseconds to decompress, so performance is |
|
2496 | # chunks only take microseconds to decompress, so performance is | |
2500 | # important here. |
|
2497 | # important here. | |
2501 | # |
|
2498 | # | |
2502 | # We can make a few assumptions about revlogs: |
|
2499 | # We can make a few assumptions about revlogs: | |
2503 | # |
|
2500 | # | |
2504 | # 1) the majority of chunks will be compressed (as opposed to inline |
|
2501 | # 1) the majority of chunks will be compressed (as opposed to inline | |
2505 | # raw data). |
|
2502 | # raw data). | |
2506 | # 2) decompressing *any* data will likely by at least 10x slower than |
|
2503 | # 2) decompressing *any* data will likely by at least 10x slower than | |
2507 | # returning raw inline data. |
|
2504 | # returning raw inline data. | |
2508 | # 3) we want to prioritize common and officially supported compression |
|
2505 | # 3) we want to prioritize common and officially supported compression | |
2509 | # engines |
|
2506 | # engines | |
2510 | # |
|
2507 | # | |
2511 | # It follows that we want to optimize for "decompress compressed data |
|
2508 | # It follows that we want to optimize for "decompress compressed data | |
2512 | # when encoded with common and officially supported compression engines" |
|
2509 | # when encoded with common and officially supported compression engines" | |
2513 | # case over "raw data" and "data encoded by less common or non-official |
|
2510 | # case over "raw data" and "data encoded by less common or non-official | |
2514 | # compression engines." That is why we have the inline lookup first |
|
2511 | # compression engines." That is why we have the inline lookup first | |
2515 | # followed by the compengines lookup. |
|
2512 | # followed by the compengines lookup. | |
2516 | # |
|
2513 | # | |
2517 | # According to `hg perfrevlogchunks`, this is ~0.5% faster for zlib |
|
2514 | # According to `hg perfrevlogchunks`, this is ~0.5% faster for zlib | |
2518 | # compressed chunks. And this matters for changelog and manifest reads. |
|
2515 | # compressed chunks. And this matters for changelog and manifest reads. | |
2519 | t = data[0:1] |
|
2516 | t = data[0:1] | |
2520 |
|
2517 | |||
2521 | if t == b'x': |
|
2518 | if t == b'x': | |
2522 | try: |
|
2519 | try: | |
2523 | return _zlibdecompress(data) |
|
2520 | return _zlibdecompress(data) | |
2524 | except zlib.error as e: |
|
2521 | except zlib.error as e: | |
2525 | raise error.RevlogError( |
|
2522 | raise error.RevlogError( | |
2526 | _(b'revlog decompress error: %s') |
|
2523 | _(b'revlog decompress error: %s') | |
2527 | % stringutil.forcebytestr(e) |
|
2524 | % stringutil.forcebytestr(e) | |
2528 | ) |
|
2525 | ) | |
2529 | # '\0' is more common than 'u' so it goes first. |
|
2526 | # '\0' is more common than 'u' so it goes first. | |
2530 | elif t == b'\0': |
|
2527 | elif t == b'\0': | |
2531 | return data |
|
2528 | return data | |
2532 | elif t == b'u': |
|
2529 | elif t == b'u': | |
2533 | return util.buffer(data, 1) |
|
2530 | return util.buffer(data, 1) | |
2534 |
|
2531 | |||
2535 | compressor = self._get_decompressor(t) |
|
2532 | compressor = self._get_decompressor(t) | |
2536 |
|
2533 | |||
2537 | return compressor.decompress(data) |
|
2534 | return compressor.decompress(data) | |
2538 |
|
2535 | |||
2539 | def _addrevision( |
|
2536 | def _addrevision( | |
2540 | self, |
|
2537 | self, | |
2541 | node, |
|
2538 | node, | |
2542 | rawtext, |
|
2539 | rawtext, | |
2543 | transaction, |
|
2540 | transaction, | |
2544 | link, |
|
2541 | link, | |
2545 | p1, |
|
2542 | p1, | |
2546 | p2, |
|
2543 | p2, | |
2547 | flags, |
|
2544 | flags, | |
2548 | cachedelta, |
|
2545 | cachedelta, | |
2549 | alwayscache=False, |
|
2546 | alwayscache=False, | |
2550 | deltacomputer=None, |
|
2547 | deltacomputer=None, | |
2551 | sidedata=None, |
|
2548 | sidedata=None, | |
2552 | ): |
|
2549 | ): | |
2553 | """internal function to add revisions to the log |
|
2550 | """internal function to add revisions to the log | |
2554 |
|
2551 | |||
2555 | see addrevision for argument descriptions. |
|
2552 | see addrevision for argument descriptions. | |
2556 |
|
2553 | |||
2557 | note: "addrevision" takes non-raw text, "_addrevision" takes raw text. |
|
2554 | note: "addrevision" takes non-raw text, "_addrevision" takes raw text. | |
2558 |
|
2555 | |||
2559 | if "deltacomputer" is not provided or None, a defaultdeltacomputer will |
|
2556 | if "deltacomputer" is not provided or None, a defaultdeltacomputer will | |
2560 | be used. |
|
2557 | be used. | |
2561 |
|
2558 | |||
2562 | invariants: |
|
2559 | invariants: | |
2563 | - rawtext is optional (can be None); if not set, cachedelta must be set. |
|
2560 | - rawtext is optional (can be None); if not set, cachedelta must be set. | |
2564 | if both are set, they must correspond to each other. |
|
2561 | if both are set, they must correspond to each other. | |
2565 | """ |
|
2562 | """ | |
2566 | if node == self.nullid: |
|
2563 | if node == self.nullid: | |
2567 | raise error.RevlogError( |
|
2564 | raise error.RevlogError( | |
2568 | _(b"%s: attempt to add null revision") % self.display_id |
|
2565 | _(b"%s: attempt to add null revision") % self.display_id | |
2569 | ) |
|
2566 | ) | |
2570 | if ( |
|
2567 | if ( | |
2571 | node == self.nodeconstants.wdirid |
|
2568 | node == self.nodeconstants.wdirid | |
2572 | or node in self.nodeconstants.wdirfilenodeids |
|
2569 | or node in self.nodeconstants.wdirfilenodeids | |
2573 | ): |
|
2570 | ): | |
2574 | raise error.RevlogError( |
|
2571 | raise error.RevlogError( | |
2575 | _(b"%s: attempt to add wdir revision") % self.display_id |
|
2572 | _(b"%s: attempt to add wdir revision") % self.display_id | |
2576 | ) |
|
2573 | ) | |
2577 | if self._writinghandles is None: |
|
2574 | if self._writinghandles is None: | |
2578 | msg = b'adding revision outside `revlog._writing` context' |
|
2575 | msg = b'adding revision outside `revlog._writing` context' | |
2579 | raise error.ProgrammingError(msg) |
|
2576 | raise error.ProgrammingError(msg) | |
2580 |
|
2577 | |||
2581 | btext = [rawtext] |
|
2578 | btext = [rawtext] | |
2582 |
|
2579 | |||
2583 | curr = len(self) |
|
2580 | curr = len(self) | |
2584 | prev = curr - 1 |
|
2581 | prev = curr - 1 | |
2585 |
|
2582 | |||
2586 | offset = self._get_data_offset(prev) |
|
2583 | offset = self._get_data_offset(prev) | |
2587 |
|
2584 | |||
2588 | if self._concurrencychecker: |
|
2585 | if self._concurrencychecker: | |
2589 | ifh, dfh, sdfh = self._writinghandles |
|
2586 | ifh, dfh, sdfh = self._writinghandles | |
2590 | # XXX no checking for the sidedata file |
|
2587 | # XXX no checking for the sidedata file | |
2591 | if self._inline: |
|
2588 | if self._inline: | |
2592 | # offset is "as if" it were in the .d file, so we need to add on |
|
2589 | # offset is "as if" it were in the .d file, so we need to add on | |
2593 | # the size of the entry metadata. |
|
2590 | # the size of the entry metadata. | |
2594 | self._concurrencychecker( |
|
2591 | self._concurrencychecker( | |
2595 | ifh, self._indexfile, offset + curr * self.index.entry_size |
|
2592 | ifh, self._indexfile, offset + curr * self.index.entry_size | |
2596 | ) |
|
2593 | ) | |
2597 | else: |
|
2594 | else: | |
2598 | # Entries in the .i are a consistent size. |
|
2595 | # Entries in the .i are a consistent size. | |
2599 | self._concurrencychecker( |
|
2596 | self._concurrencychecker( | |
2600 | ifh, self._indexfile, curr * self.index.entry_size |
|
2597 | ifh, self._indexfile, curr * self.index.entry_size | |
2601 | ) |
|
2598 | ) | |
2602 | self._concurrencychecker(dfh, self._datafile, offset) |
|
2599 | self._concurrencychecker(dfh, self._datafile, offset) | |
2603 |
|
2600 | |||
2604 | p1r, p2r = self.rev(p1), self.rev(p2) |
|
2601 | p1r, p2r = self.rev(p1), self.rev(p2) | |
2605 |
|
2602 | |||
2606 | # full versions are inserted when the needed deltas |
|
2603 | # full versions are inserted when the needed deltas | |
2607 | # become comparable to the uncompressed text |
|
2604 | # become comparable to the uncompressed text | |
2608 | if rawtext is None: |
|
2605 | if rawtext is None: | |
2609 | # need rawtext size, before changed by flag processors, which is |
|
2606 | # need rawtext size, before changed by flag processors, which is | |
2610 | # the non-raw size. use revlog explicitly to avoid filelog's extra |
|
2607 | # the non-raw size. use revlog explicitly to avoid filelog's extra | |
2611 | # logic that might remove metadata size. |
|
2608 | # logic that might remove metadata size. | |
2612 | textlen = mdiff.patchedsize( |
|
2609 | textlen = mdiff.patchedsize( | |
2613 | revlog.size(self, cachedelta[0]), cachedelta[1] |
|
2610 | revlog.size(self, cachedelta[0]), cachedelta[1] | |
2614 | ) |
|
2611 | ) | |
2615 | else: |
|
2612 | else: | |
2616 | textlen = len(rawtext) |
|
2613 | textlen = len(rawtext) | |
2617 |
|
2614 | |||
2618 | if deltacomputer is None: |
|
2615 | if deltacomputer is None: | |
2619 | write_debug = None |
|
2616 | write_debug = None | |
2620 | if self._debug_delta: |
|
2617 | if self._debug_delta: | |
2621 | write_debug = transaction._report |
|
2618 | write_debug = transaction._report | |
2622 | deltacomputer = deltautil.deltacomputer( |
|
2619 | deltacomputer = deltautil.deltacomputer( | |
2623 | self, write_debug=write_debug |
|
2620 | self, write_debug=write_debug | |
2624 | ) |
|
2621 | ) | |
2625 |
|
2622 | |||
2626 | if cachedelta is not None and len(cachedelta) == 2: |
|
2623 | if cachedelta is not None and len(cachedelta) == 2: | |
2627 | # If the cached delta has no information about how it should be |
|
2624 | # If the cached delta has no information about how it should be | |
2628 | # reused, add the default reuse instruction according to the |
|
2625 | # reused, add the default reuse instruction according to the | |
2629 | # revlog's configuration. |
|
2626 | # revlog's configuration. | |
2630 | if self._generaldelta and self._lazydeltabase: |
|
2627 | if self._generaldelta and self._lazydeltabase: | |
2631 | delta_base_reuse = DELTA_BASE_REUSE_TRY |
|
2628 | delta_base_reuse = DELTA_BASE_REUSE_TRY | |
2632 | else: |
|
2629 | else: | |
2633 | delta_base_reuse = DELTA_BASE_REUSE_NO |
|
2630 | delta_base_reuse = DELTA_BASE_REUSE_NO | |
2634 | cachedelta = (cachedelta[0], cachedelta[1], delta_base_reuse) |
|
2631 | cachedelta = (cachedelta[0], cachedelta[1], delta_base_reuse) | |
2635 |
|
2632 | |||
2636 | revinfo = revlogutils.revisioninfo( |
|
2633 | revinfo = revlogutils.revisioninfo( | |
2637 | node, |
|
2634 | node, | |
2638 | p1, |
|
2635 | p1, | |
2639 | p2, |
|
2636 | p2, | |
2640 | btext, |
|
2637 | btext, | |
2641 | textlen, |
|
2638 | textlen, | |
2642 | cachedelta, |
|
2639 | cachedelta, | |
2643 | flags, |
|
2640 | flags, | |
2644 | ) |
|
2641 | ) | |
2645 |
|
2642 | |||
2646 | deltainfo = deltacomputer.finddeltainfo(revinfo) |
|
2643 | deltainfo = deltacomputer.finddeltainfo(revinfo) | |
2647 |
|
2644 | |||
2648 | compression_mode = COMP_MODE_INLINE |
|
2645 | compression_mode = COMP_MODE_INLINE | |
2649 | if self._docket is not None: |
|
2646 | if self._docket is not None: | |
2650 | default_comp = self._docket.default_compression_header |
|
2647 | default_comp = self._docket.default_compression_header | |
2651 | r = deltautil.delta_compression(default_comp, deltainfo) |
|
2648 | r = deltautil.delta_compression(default_comp, deltainfo) | |
2652 | compression_mode, deltainfo = r |
|
2649 | compression_mode, deltainfo = r | |
2653 |
|
2650 | |||
2654 | sidedata_compression_mode = COMP_MODE_INLINE |
|
2651 | sidedata_compression_mode = COMP_MODE_INLINE | |
2655 | if sidedata and self.hassidedata: |
|
2652 | if sidedata and self.hassidedata: | |
2656 | sidedata_compression_mode = COMP_MODE_PLAIN |
|
2653 | sidedata_compression_mode = COMP_MODE_PLAIN | |
2657 | serialized_sidedata = sidedatautil.serialize_sidedata(sidedata) |
|
2654 | serialized_sidedata = sidedatautil.serialize_sidedata(sidedata) | |
2658 | sidedata_offset = self._docket.sidedata_end |
|
2655 | sidedata_offset = self._docket.sidedata_end | |
2659 | h, comp_sidedata = self.compress(serialized_sidedata) |
|
2656 | h, comp_sidedata = self.compress(serialized_sidedata) | |
2660 | if ( |
|
2657 | if ( | |
2661 | h != b'u' |
|
2658 | h != b'u' | |
2662 | and comp_sidedata[0:1] != b'\0' |
|
2659 | and comp_sidedata[0:1] != b'\0' | |
2663 | and len(comp_sidedata) < len(serialized_sidedata) |
|
2660 | and len(comp_sidedata) < len(serialized_sidedata) | |
2664 | ): |
|
2661 | ): | |
2665 | assert not h |
|
2662 | assert not h | |
2666 | if ( |
|
2663 | if ( | |
2667 | comp_sidedata[0:1] |
|
2664 | comp_sidedata[0:1] | |
2668 | == self._docket.default_compression_header |
|
2665 | == self._docket.default_compression_header | |
2669 | ): |
|
2666 | ): | |
2670 | sidedata_compression_mode = COMP_MODE_DEFAULT |
|
2667 | sidedata_compression_mode = COMP_MODE_DEFAULT | |
2671 | serialized_sidedata = comp_sidedata |
|
2668 | serialized_sidedata = comp_sidedata | |
2672 | else: |
|
2669 | else: | |
2673 | sidedata_compression_mode = COMP_MODE_INLINE |
|
2670 | sidedata_compression_mode = COMP_MODE_INLINE | |
2674 | serialized_sidedata = comp_sidedata |
|
2671 | serialized_sidedata = comp_sidedata | |
2675 | else: |
|
2672 | else: | |
2676 | serialized_sidedata = b"" |
|
2673 | serialized_sidedata = b"" | |
2677 | # Don't store the offset if the sidedata is empty, that way |
|
2674 | # Don't store the offset if the sidedata is empty, that way | |
2678 | # we can easily detect empty sidedata and they will be no different |
|
2675 | # we can easily detect empty sidedata and they will be no different | |
2679 | # than ones we manually add. |
|
2676 | # than ones we manually add. | |
2680 | sidedata_offset = 0 |
|
2677 | sidedata_offset = 0 | |
2681 |
|
2678 | |||
2682 | rank = RANK_UNKNOWN |
|
2679 | rank = RANK_UNKNOWN | |
2683 | if self._compute_rank: |
|
2680 | if self._compute_rank: | |
2684 | if (p1r, p2r) == (nullrev, nullrev): |
|
2681 | if (p1r, p2r) == (nullrev, nullrev): | |
2685 | rank = 1 |
|
2682 | rank = 1 | |
2686 | elif p1r != nullrev and p2r == nullrev: |
|
2683 | elif p1r != nullrev and p2r == nullrev: | |
2687 | rank = 1 + self.fast_rank(p1r) |
|
2684 | rank = 1 + self.fast_rank(p1r) | |
2688 | elif p1r == nullrev and p2r != nullrev: |
|
2685 | elif p1r == nullrev and p2r != nullrev: | |
2689 | rank = 1 + self.fast_rank(p2r) |
|
2686 | rank = 1 + self.fast_rank(p2r) | |
2690 | else: # merge node |
|
2687 | else: # merge node | |
2691 | if rustdagop is not None and self.index.rust_ext_compat: |
|
2688 | if rustdagop is not None and self.index.rust_ext_compat: | |
2692 | rank = rustdagop.rank(self.index, p1r, p2r) |
|
2689 | rank = rustdagop.rank(self.index, p1r, p2r) | |
2693 | else: |
|
2690 | else: | |
2694 | pmin, pmax = sorted((p1r, p2r)) |
|
2691 | pmin, pmax = sorted((p1r, p2r)) | |
2695 | rank = 1 + self.fast_rank(pmax) |
|
2692 | rank = 1 + self.fast_rank(pmax) | |
2696 | rank += sum(1 for _ in self.findmissingrevs([pmax], [pmin])) |
|
2693 | rank += sum(1 for _ in self.findmissingrevs([pmax], [pmin])) | |
2697 |
|
2694 | |||
2698 | e = revlogutils.entry( |
|
2695 | e = revlogutils.entry( | |
2699 | flags=flags, |
|
2696 | flags=flags, | |
2700 | data_offset=offset, |
|
2697 | data_offset=offset, | |
2701 | data_compressed_length=deltainfo.deltalen, |
|
2698 | data_compressed_length=deltainfo.deltalen, | |
2702 | data_uncompressed_length=textlen, |
|
2699 | data_uncompressed_length=textlen, | |
2703 | data_compression_mode=compression_mode, |
|
2700 | data_compression_mode=compression_mode, | |
2704 | data_delta_base=deltainfo.base, |
|
2701 | data_delta_base=deltainfo.base, | |
2705 | link_rev=link, |
|
2702 | link_rev=link, | |
2706 | parent_rev_1=p1r, |
|
2703 | parent_rev_1=p1r, | |
2707 | parent_rev_2=p2r, |
|
2704 | parent_rev_2=p2r, | |
2708 | node_id=node, |
|
2705 | node_id=node, | |
2709 | sidedata_offset=sidedata_offset, |
|
2706 | sidedata_offset=sidedata_offset, | |
2710 | sidedata_compressed_length=len(serialized_sidedata), |
|
2707 | sidedata_compressed_length=len(serialized_sidedata), | |
2711 | sidedata_compression_mode=sidedata_compression_mode, |
|
2708 | sidedata_compression_mode=sidedata_compression_mode, | |
2712 | rank=rank, |
|
2709 | rank=rank, | |
2713 | ) |
|
2710 | ) | |
2714 |
|
2711 | |||
2715 | self.index.append(e) |
|
2712 | self.index.append(e) | |
2716 | entry = self.index.entry_binary(curr) |
|
2713 | entry = self.index.entry_binary(curr) | |
2717 | if curr == 0 and self._docket is None: |
|
2714 | if curr == 0 and self._docket is None: | |
2718 | header = self._format_flags | self._format_version |
|
2715 | header = self._format_flags | self._format_version | |
2719 | header = self.index.pack_header(header) |
|
2716 | header = self.index.pack_header(header) | |
2720 | entry = header + entry |
|
2717 | entry = header + entry | |
2721 | self._writeentry( |
|
2718 | self._writeentry( | |
2722 | transaction, |
|
2719 | transaction, | |
2723 | entry, |
|
2720 | entry, | |
2724 | deltainfo.data, |
|
2721 | deltainfo.data, | |
2725 | link, |
|
2722 | link, | |
2726 | offset, |
|
2723 | offset, | |
2727 | serialized_sidedata, |
|
2724 | serialized_sidedata, | |
2728 | sidedata_offset, |
|
2725 | sidedata_offset, | |
2729 | ) |
|
2726 | ) | |
2730 |
|
2727 | |||
2731 | rawtext = btext[0] |
|
2728 | rawtext = btext[0] | |
2732 |
|
2729 | |||
2733 | if alwayscache and rawtext is None: |
|
2730 | if alwayscache and rawtext is None: | |
2734 | rawtext = deltacomputer.buildtext(revinfo) |
|
2731 | rawtext = deltacomputer.buildtext(revinfo) | |
2735 |
|
2732 | |||
2736 | if type(rawtext) == bytes: # only accept immutable objects |
|
2733 | if type(rawtext) == bytes: # only accept immutable objects | |
2737 | self._revisioncache = (node, curr, rawtext) |
|
2734 | self._revisioncache = (node, curr, rawtext) | |
2738 | self._chainbasecache[curr] = deltainfo.chainbase |
|
2735 | self._chainbasecache[curr] = deltainfo.chainbase | |
2739 | return curr |
|
2736 | return curr | |
2740 |
|
2737 | |||
2741 | def _get_data_offset(self, prev): |
|
2738 | def _get_data_offset(self, prev): | |
2742 | """Returns the current offset in the (in-transaction) data file. |
|
2739 | """Returns the current offset in the (in-transaction) data file. | |
2743 | Versions < 2 of the revlog can get this 0(1), revlog v2 needs a docket |
|
2740 | Versions < 2 of the revlog can get this 0(1), revlog v2 needs a docket | |
2744 | file to store that information: since sidedata can be rewritten to the |
|
2741 | file to store that information: since sidedata can be rewritten to the | |
2745 | end of the data file within a transaction, you can have cases where, for |
|
2742 | end of the data file within a transaction, you can have cases where, for | |
2746 | example, rev `n` does not have sidedata while rev `n - 1` does, leading |
|
2743 | example, rev `n` does not have sidedata while rev `n - 1` does, leading | |
2747 | to `n - 1`'s sidedata being written after `n`'s data. |
|
2744 | to `n - 1`'s sidedata being written after `n`'s data. | |
2748 |
|
2745 | |||
2749 | TODO cache this in a docket file before getting out of experimental.""" |
|
2746 | TODO cache this in a docket file before getting out of experimental.""" | |
2750 | if self._docket is None: |
|
2747 | if self._docket is None: | |
2751 | return self.end(prev) |
|
2748 | return self.end(prev) | |
2752 | else: |
|
2749 | else: | |
2753 | return self._docket.data_end |
|
2750 | return self._docket.data_end | |
2754 |
|
2751 | |||
2755 | def _writeentry( |
|
2752 | def _writeentry( | |
2756 | self, transaction, entry, data, link, offset, sidedata, sidedata_offset |
|
2753 | self, transaction, entry, data, link, offset, sidedata, sidedata_offset | |
2757 | ): |
|
2754 | ): | |
2758 | # Files opened in a+ mode have inconsistent behavior on various |
|
2755 | # Files opened in a+ mode have inconsistent behavior on various | |
2759 | # platforms. Windows requires that a file positioning call be made |
|
2756 | # platforms. Windows requires that a file positioning call be made | |
2760 | # when the file handle transitions between reads and writes. See |
|
2757 | # when the file handle transitions between reads and writes. See | |
2761 | # 3686fa2b8eee and the mixedfilemodewrapper in windows.py. On other |
|
2758 | # 3686fa2b8eee and the mixedfilemodewrapper in windows.py. On other | |
2762 | # platforms, Python or the platform itself can be buggy. Some versions |
|
2759 | # platforms, Python or the platform itself can be buggy. Some versions | |
2763 | # of Solaris have been observed to not append at the end of the file |
|
2760 | # of Solaris have been observed to not append at the end of the file | |
2764 | # if the file was seeked to before the end. See issue4943 for more. |
|
2761 | # if the file was seeked to before the end. See issue4943 for more. | |
2765 | # |
|
2762 | # | |
2766 | # We work around this issue by inserting a seek() before writing. |
|
2763 | # We work around this issue by inserting a seek() before writing. | |
2767 | # Note: This is likely not necessary on Python 3. However, because |
|
2764 | # Note: This is likely not necessary on Python 3. However, because | |
2768 | # the file handle is reused for reads and may be seeked there, we need |
|
2765 | # the file handle is reused for reads and may be seeked there, we need | |
2769 | # to be careful before changing this. |
|
2766 | # to be careful before changing this. | |
2770 | if self._writinghandles is None: |
|
2767 | if self._writinghandles is None: | |
2771 | msg = b'adding revision outside `revlog._writing` context' |
|
2768 | msg = b'adding revision outside `revlog._writing` context' | |
2772 | raise error.ProgrammingError(msg) |
|
2769 | raise error.ProgrammingError(msg) | |
2773 | ifh, dfh, sdfh = self._writinghandles |
|
2770 | ifh, dfh, sdfh = self._writinghandles | |
2774 | if self._docket is None: |
|
2771 | if self._docket is None: | |
2775 | ifh.seek(0, os.SEEK_END) |
|
2772 | ifh.seek(0, os.SEEK_END) | |
2776 | else: |
|
2773 | else: | |
2777 | ifh.seek(self._docket.index_end, os.SEEK_SET) |
|
2774 | ifh.seek(self._docket.index_end, os.SEEK_SET) | |
2778 | if dfh: |
|
2775 | if dfh: | |
2779 | if self._docket is None: |
|
2776 | if self._docket is None: | |
2780 | dfh.seek(0, os.SEEK_END) |
|
2777 | dfh.seek(0, os.SEEK_END) | |
2781 | else: |
|
2778 | else: | |
2782 | dfh.seek(self._docket.data_end, os.SEEK_SET) |
|
2779 | dfh.seek(self._docket.data_end, os.SEEK_SET) | |
2783 | if sdfh: |
|
2780 | if sdfh: | |
2784 | sdfh.seek(self._docket.sidedata_end, os.SEEK_SET) |
|
2781 | sdfh.seek(self._docket.sidedata_end, os.SEEK_SET) | |
2785 |
|
2782 | |||
2786 | curr = len(self) - 1 |
|
2783 | curr = len(self) - 1 | |
2787 | if not self._inline: |
|
2784 | if not self._inline: | |
2788 | transaction.add(self._datafile, offset) |
|
2785 | transaction.add(self._datafile, offset) | |
2789 | if self._sidedatafile: |
|
2786 | if self._sidedatafile: | |
2790 | transaction.add(self._sidedatafile, sidedata_offset) |
|
2787 | transaction.add(self._sidedatafile, sidedata_offset) | |
2791 | transaction.add(self._indexfile, curr * len(entry)) |
|
2788 | transaction.add(self._indexfile, curr * len(entry)) | |
2792 | if data[0]: |
|
2789 | if data[0]: | |
2793 | dfh.write(data[0]) |
|
2790 | dfh.write(data[0]) | |
2794 | dfh.write(data[1]) |
|
2791 | dfh.write(data[1]) | |
2795 | if sidedata: |
|
2792 | if sidedata: | |
2796 | sdfh.write(sidedata) |
|
2793 | sdfh.write(sidedata) | |
2797 | ifh.write(entry) |
|
2794 | ifh.write(entry) | |
2798 | else: |
|
2795 | else: | |
2799 | offset += curr * self.index.entry_size |
|
2796 | offset += curr * self.index.entry_size | |
2800 | transaction.add(self._indexfile, offset) |
|
2797 | transaction.add(self._indexfile, offset) | |
2801 | ifh.write(entry) |
|
2798 | ifh.write(entry) | |
2802 | ifh.write(data[0]) |
|
2799 | ifh.write(data[0]) | |
2803 | ifh.write(data[1]) |
|
2800 | ifh.write(data[1]) | |
2804 | assert not sidedata |
|
2801 | assert not sidedata | |
2805 | self._enforceinlinesize(transaction) |
|
2802 | self._enforceinlinesize(transaction) | |
2806 | if self._docket is not None: |
|
2803 | if self._docket is not None: | |
2807 | # revlog-v2 always has 3 writing handles, help Pytype |
|
2804 | # revlog-v2 always has 3 writing handles, help Pytype | |
2808 | wh1 = self._writinghandles[0] |
|
2805 | wh1 = self._writinghandles[0] | |
2809 | wh2 = self._writinghandles[1] |
|
2806 | wh2 = self._writinghandles[1] | |
2810 | wh3 = self._writinghandles[2] |
|
2807 | wh3 = self._writinghandles[2] | |
2811 | assert wh1 is not None |
|
2808 | assert wh1 is not None | |
2812 | assert wh2 is not None |
|
2809 | assert wh2 is not None | |
2813 | assert wh3 is not None |
|
2810 | assert wh3 is not None | |
2814 | self._docket.index_end = wh1.tell() |
|
2811 | self._docket.index_end = wh1.tell() | |
2815 | self._docket.data_end = wh2.tell() |
|
2812 | self._docket.data_end = wh2.tell() | |
2816 | self._docket.sidedata_end = wh3.tell() |
|
2813 | self._docket.sidedata_end = wh3.tell() | |
2817 |
|
2814 | |||
2818 | nodemaputil.setup_persistent_nodemap(transaction, self) |
|
2815 | nodemaputil.setup_persistent_nodemap(transaction, self) | |
2819 |
|
2816 | |||
2820 | def addgroup( |
|
2817 | def addgroup( | |
2821 | self, |
|
2818 | self, | |
2822 | deltas, |
|
2819 | deltas, | |
2823 | linkmapper, |
|
2820 | linkmapper, | |
2824 | transaction, |
|
2821 | transaction, | |
2825 | alwayscache=False, |
|
2822 | alwayscache=False, | |
2826 | addrevisioncb=None, |
|
2823 | addrevisioncb=None, | |
2827 | duplicaterevisioncb=None, |
|
2824 | duplicaterevisioncb=None, | |
2828 | debug_info=None, |
|
2825 | debug_info=None, | |
2829 | delta_base_reuse_policy=None, |
|
2826 | delta_base_reuse_policy=None, | |
2830 | ): |
|
2827 | ): | |
2831 | """ |
|
2828 | """ | |
2832 | add a delta group |
|
2829 | add a delta group | |
2833 |
|
2830 | |||
2834 | given a set of deltas, add them to the revision log. the |
|
2831 | given a set of deltas, add them to the revision log. the | |
2835 | first delta is against its parent, which should be in our |
|
2832 | first delta is against its parent, which should be in our | |
2836 | log, the rest are against the previous delta. |
|
2833 | log, the rest are against the previous delta. | |
2837 |
|
2834 | |||
2838 | If ``addrevisioncb`` is defined, it will be called with arguments of |
|
2835 | If ``addrevisioncb`` is defined, it will be called with arguments of | |
2839 | this revlog and the node that was added. |
|
2836 | this revlog and the node that was added. | |
2840 | """ |
|
2837 | """ | |
2841 |
|
2838 | |||
2842 | if self._adding_group: |
|
2839 | if self._adding_group: | |
2843 | raise error.ProgrammingError(b'cannot nest addgroup() calls') |
|
2840 | raise error.ProgrammingError(b'cannot nest addgroup() calls') | |
2844 |
|
2841 | |||
2845 | # read the default delta-base reuse policy from revlog config if the |
|
2842 | # read the default delta-base reuse policy from revlog config if the | |
2846 | # group did not specify one. |
|
2843 | # group did not specify one. | |
2847 | if delta_base_reuse_policy is None: |
|
2844 | if delta_base_reuse_policy is None: | |
2848 | if self._generaldelta and self._lazydeltabase: |
|
2845 | if self._generaldelta and self._lazydeltabase: | |
2849 | delta_base_reuse_policy = DELTA_BASE_REUSE_TRY |
|
2846 | delta_base_reuse_policy = DELTA_BASE_REUSE_TRY | |
2850 | else: |
|
2847 | else: | |
2851 | delta_base_reuse_policy = DELTA_BASE_REUSE_NO |
|
2848 | delta_base_reuse_policy = DELTA_BASE_REUSE_NO | |
2852 |
|
2849 | |||
2853 | self._adding_group = True |
|
2850 | self._adding_group = True | |
2854 | empty = True |
|
2851 | empty = True | |
2855 | try: |
|
2852 | try: | |
2856 | with self._writing(transaction): |
|
2853 | with self._writing(transaction): | |
2857 | write_debug = None |
|
2854 | write_debug = None | |
2858 | if self._debug_delta: |
|
2855 | if self._debug_delta: | |
2859 | write_debug = transaction._report |
|
2856 | write_debug = transaction._report | |
2860 | deltacomputer = deltautil.deltacomputer( |
|
2857 | deltacomputer = deltautil.deltacomputer( | |
2861 | self, |
|
2858 | self, | |
2862 | write_debug=write_debug, |
|
2859 | write_debug=write_debug, | |
2863 | debug_info=debug_info, |
|
2860 | debug_info=debug_info, | |
2864 | ) |
|
2861 | ) | |
2865 | # loop through our set of deltas |
|
2862 | # loop through our set of deltas | |
2866 | for data in deltas: |
|
2863 | for data in deltas: | |
2867 | ( |
|
2864 | ( | |
2868 | node, |
|
2865 | node, | |
2869 | p1, |
|
2866 | p1, | |
2870 | p2, |
|
2867 | p2, | |
2871 | linknode, |
|
2868 | linknode, | |
2872 | deltabase, |
|
2869 | deltabase, | |
2873 | delta, |
|
2870 | delta, | |
2874 | flags, |
|
2871 | flags, | |
2875 | sidedata, |
|
2872 | sidedata, | |
2876 | ) = data |
|
2873 | ) = data | |
2877 | link = linkmapper(linknode) |
|
2874 | link = linkmapper(linknode) | |
2878 | flags = flags or REVIDX_DEFAULT_FLAGS |
|
2875 | flags = flags or REVIDX_DEFAULT_FLAGS | |
2879 |
|
2876 | |||
2880 | rev = self.index.get_rev(node) |
|
2877 | rev = self.index.get_rev(node) | |
2881 | if rev is not None: |
|
2878 | if rev is not None: | |
2882 | # this can happen if two branches make the same change |
|
2879 | # this can happen if two branches make the same change | |
2883 | self._nodeduplicatecallback(transaction, rev) |
|
2880 | self._nodeduplicatecallback(transaction, rev) | |
2884 | if duplicaterevisioncb: |
|
2881 | if duplicaterevisioncb: | |
2885 | duplicaterevisioncb(self, rev) |
|
2882 | duplicaterevisioncb(self, rev) | |
2886 | empty = False |
|
2883 | empty = False | |
2887 | continue |
|
2884 | continue | |
2888 |
|
2885 | |||
2889 | for p in (p1, p2): |
|
2886 | for p in (p1, p2): | |
2890 | if not self.index.has_node(p): |
|
2887 | if not self.index.has_node(p): | |
2891 | raise error.LookupError( |
|
2888 | raise error.LookupError( | |
2892 | p, self.radix, _(b'unknown parent') |
|
2889 | p, self.radix, _(b'unknown parent') | |
2893 | ) |
|
2890 | ) | |
2894 |
|
2891 | |||
2895 | if not self.index.has_node(deltabase): |
|
2892 | if not self.index.has_node(deltabase): | |
2896 | raise error.LookupError( |
|
2893 | raise error.LookupError( | |
2897 | deltabase, self.display_id, _(b'unknown delta base') |
|
2894 | deltabase, self.display_id, _(b'unknown delta base') | |
2898 | ) |
|
2895 | ) | |
2899 |
|
2896 | |||
2900 | baserev = self.rev(deltabase) |
|
2897 | baserev = self.rev(deltabase) | |
2901 |
|
2898 | |||
2902 | if baserev != nullrev and self.iscensored(baserev): |
|
2899 | if baserev != nullrev and self.iscensored(baserev): | |
2903 | # if base is censored, delta must be full replacement in a |
|
2900 | # if base is censored, delta must be full replacement in a | |
2904 | # single patch operation |
|
2901 | # single patch operation | |
2905 | hlen = struct.calcsize(b">lll") |
|
2902 | hlen = struct.calcsize(b">lll") | |
2906 | oldlen = self.rawsize(baserev) |
|
2903 | oldlen = self.rawsize(baserev) | |
2907 | newlen = len(delta) - hlen |
|
2904 | newlen = len(delta) - hlen | |
2908 | if delta[:hlen] != mdiff.replacediffheader( |
|
2905 | if delta[:hlen] != mdiff.replacediffheader( | |
2909 | oldlen, newlen |
|
2906 | oldlen, newlen | |
2910 | ): |
|
2907 | ): | |
2911 | raise error.CensoredBaseError( |
|
2908 | raise error.CensoredBaseError( | |
2912 | self.display_id, self.node(baserev) |
|
2909 | self.display_id, self.node(baserev) | |
2913 | ) |
|
2910 | ) | |
2914 |
|
2911 | |||
2915 | if not flags and self._peek_iscensored(baserev, delta): |
|
2912 | if not flags and self._peek_iscensored(baserev, delta): | |
2916 | flags |= REVIDX_ISCENSORED |
|
2913 | flags |= REVIDX_ISCENSORED | |
2917 |
|
2914 | |||
2918 | # We assume consumers of addrevisioncb will want to retrieve |
|
2915 | # We assume consumers of addrevisioncb will want to retrieve | |
2919 | # the added revision, which will require a call to |
|
2916 | # the added revision, which will require a call to | |
2920 | # revision(). revision() will fast path if there is a cache |
|
2917 | # revision(). revision() will fast path if there is a cache | |
2921 | # hit. So, we tell _addrevision() to always cache in this case. |
|
2918 | # hit. So, we tell _addrevision() to always cache in this case. | |
2922 | # We're only using addgroup() in the context of changegroup |
|
2919 | # We're only using addgroup() in the context of changegroup | |
2923 | # generation so the revision data can always be handled as raw |
|
2920 | # generation so the revision data can always be handled as raw | |
2924 | # by the flagprocessor. |
|
2921 | # by the flagprocessor. | |
2925 | rev = self._addrevision( |
|
2922 | rev = self._addrevision( | |
2926 | node, |
|
2923 | node, | |
2927 | None, |
|
2924 | None, | |
2928 | transaction, |
|
2925 | transaction, | |
2929 | link, |
|
2926 | link, | |
2930 | p1, |
|
2927 | p1, | |
2931 | p2, |
|
2928 | p2, | |
2932 | flags, |
|
2929 | flags, | |
2933 | (baserev, delta, delta_base_reuse_policy), |
|
2930 | (baserev, delta, delta_base_reuse_policy), | |
2934 | alwayscache=alwayscache, |
|
2931 | alwayscache=alwayscache, | |
2935 | deltacomputer=deltacomputer, |
|
2932 | deltacomputer=deltacomputer, | |
2936 | sidedata=sidedata, |
|
2933 | sidedata=sidedata, | |
2937 | ) |
|
2934 | ) | |
2938 |
|
2935 | |||
2939 | if addrevisioncb: |
|
2936 | if addrevisioncb: | |
2940 | addrevisioncb(self, rev) |
|
2937 | addrevisioncb(self, rev) | |
2941 | empty = False |
|
2938 | empty = False | |
2942 | finally: |
|
2939 | finally: | |
2943 | self._adding_group = False |
|
2940 | self._adding_group = False | |
2944 | return not empty |
|
2941 | return not empty | |
2945 |
|
2942 | |||
2946 | def iscensored(self, rev): |
|
2943 | def iscensored(self, rev): | |
2947 | """Check if a file revision is censored.""" |
|
2944 | """Check if a file revision is censored.""" | |
2948 | if not self._censorable: |
|
2945 | if not self._censorable: | |
2949 | return False |
|
2946 | return False | |
2950 |
|
2947 | |||
2951 | return self.flags(rev) & REVIDX_ISCENSORED |
|
2948 | return self.flags(rev) & REVIDX_ISCENSORED | |
2952 |
|
2949 | |||
2953 | def _peek_iscensored(self, baserev, delta): |
|
2950 | def _peek_iscensored(self, baserev, delta): | |
2954 | """Quickly check if a delta produces a censored revision.""" |
|
2951 | """Quickly check if a delta produces a censored revision.""" | |
2955 | if not self._censorable: |
|
2952 | if not self._censorable: | |
2956 | return False |
|
2953 | return False | |
2957 |
|
2954 | |||
2958 | return storageutil.deltaiscensored(delta, baserev, self.rawsize) |
|
2955 | return storageutil.deltaiscensored(delta, baserev, self.rawsize) | |
2959 |
|
2956 | |||
2960 | def getstrippoint(self, minlink): |
|
2957 | def getstrippoint(self, minlink): | |
2961 | """find the minimum rev that must be stripped to strip the linkrev |
|
2958 | """find the minimum rev that must be stripped to strip the linkrev | |
2962 |
|
2959 | |||
2963 | Returns a tuple containing the minimum rev and a set of all revs that |
|
2960 | Returns a tuple containing the minimum rev and a set of all revs that | |
2964 | have linkrevs that will be broken by this strip. |
|
2961 | have linkrevs that will be broken by this strip. | |
2965 | """ |
|
2962 | """ | |
2966 | return storageutil.resolvestripinfo( |
|
2963 | return storageutil.resolvestripinfo( | |
2967 | minlink, |
|
2964 | minlink, | |
2968 | len(self) - 1, |
|
2965 | len(self) - 1, | |
2969 | self.headrevs(), |
|
2966 | self.headrevs(), | |
2970 | self.linkrev, |
|
2967 | self.linkrev, | |
2971 | self.parentrevs, |
|
2968 | self.parentrevs, | |
2972 | ) |
|
2969 | ) | |
2973 |
|
2970 | |||
2974 | def strip(self, minlink, transaction): |
|
2971 | def strip(self, minlink, transaction): | |
2975 | """truncate the revlog on the first revision with a linkrev >= minlink |
|
2972 | """truncate the revlog on the first revision with a linkrev >= minlink | |
2976 |
|
2973 | |||
2977 | This function is called when we're stripping revision minlink and |
|
2974 | This function is called when we're stripping revision minlink and | |
2978 | its descendants from the repository. |
|
2975 | its descendants from the repository. | |
2979 |
|
2976 | |||
2980 | We have to remove all revisions with linkrev >= minlink, because |
|
2977 | We have to remove all revisions with linkrev >= minlink, because | |
2981 | the equivalent changelog revisions will be renumbered after the |
|
2978 | the equivalent changelog revisions will be renumbered after the | |
2982 | strip. |
|
2979 | strip. | |
2983 |
|
2980 | |||
2984 | So we truncate the revlog on the first of these revisions, and |
|
2981 | So we truncate the revlog on the first of these revisions, and | |
2985 | trust that the caller has saved the revisions that shouldn't be |
|
2982 | trust that the caller has saved the revisions that shouldn't be | |
2986 | removed and that it'll re-add them after this truncation. |
|
2983 | removed and that it'll re-add them after this truncation. | |
2987 | """ |
|
2984 | """ | |
2988 | if len(self) == 0: |
|
2985 | if len(self) == 0: | |
2989 | return |
|
2986 | return | |
2990 |
|
2987 | |||
2991 | rev, _ = self.getstrippoint(minlink) |
|
2988 | rev, _ = self.getstrippoint(minlink) | |
2992 | if rev == len(self): |
|
2989 | if rev == len(self): | |
2993 | return |
|
2990 | return | |
2994 |
|
2991 | |||
2995 | # first truncate the files on disk |
|
2992 | # first truncate the files on disk | |
2996 | data_end = self.start(rev) |
|
2993 | data_end = self.start(rev) | |
2997 | if not self._inline: |
|
2994 | if not self._inline: | |
2998 | transaction.add(self._datafile, data_end) |
|
2995 | transaction.add(self._datafile, data_end) | |
2999 | end = rev * self.index.entry_size |
|
2996 | end = rev * self.index.entry_size | |
3000 | else: |
|
2997 | else: | |
3001 | end = data_end + (rev * self.index.entry_size) |
|
2998 | end = data_end + (rev * self.index.entry_size) | |
3002 |
|
2999 | |||
3003 | if self._sidedatafile: |
|
3000 | if self._sidedatafile: | |
3004 | sidedata_end = self.sidedata_cut_off(rev) |
|
3001 | sidedata_end = self.sidedata_cut_off(rev) | |
3005 | transaction.add(self._sidedatafile, sidedata_end) |
|
3002 | transaction.add(self._sidedatafile, sidedata_end) | |
3006 |
|
3003 | |||
3007 | transaction.add(self._indexfile, end) |
|
3004 | transaction.add(self._indexfile, end) | |
3008 | if self._docket is not None: |
|
3005 | if self._docket is not None: | |
3009 | # XXX we could, leverage the docket while stripping. However it is |
|
3006 | # XXX we could, leverage the docket while stripping. However it is | |
3010 | # not powerfull enough at the time of this comment |
|
3007 | # not powerfull enough at the time of this comment | |
3011 | self._docket.index_end = end |
|
3008 | self._docket.index_end = end | |
3012 | self._docket.data_end = data_end |
|
3009 | self._docket.data_end = data_end | |
3013 | self._docket.sidedata_end = sidedata_end |
|
3010 | self._docket.sidedata_end = sidedata_end | |
3014 | self._docket.write(transaction, stripping=True) |
|
3011 | self._docket.write(transaction, stripping=True) | |
3015 |
|
3012 | |||
3016 | # then reset internal state in memory to forget those revisions |
|
3013 | # then reset internal state in memory to forget those revisions | |
3017 | self._revisioncache = None |
|
3014 | self._revisioncache = None | |
3018 | self._chaininfocache = util.lrucachedict(500) |
|
3015 | self._chaininfocache = util.lrucachedict(500) | |
3019 | self._segmentfile.clear_cache() |
|
3016 | self._segmentfile.clear_cache() | |
3020 | self._segmentfile_sidedata.clear_cache() |
|
3017 | self._segmentfile_sidedata.clear_cache() | |
3021 |
|
3018 | |||
3022 | del self.index[rev:-1] |
|
3019 | del self.index[rev:-1] | |
3023 |
|
3020 | |||
3024 | def checksize(self): |
|
3021 | def checksize(self): | |
3025 | """Check size of index and data files |
|
3022 | """Check size of index and data files | |
3026 |
|
3023 | |||
3027 | return a (dd, di) tuple. |
|
3024 | return a (dd, di) tuple. | |
3028 | - dd: extra bytes for the "data" file |
|
3025 | - dd: extra bytes for the "data" file | |
3029 | - di: extra bytes for the "index" file |
|
3026 | - di: extra bytes for the "index" file | |
3030 |
|
3027 | |||
3031 | A healthy revlog will return (0, 0). |
|
3028 | A healthy revlog will return (0, 0). | |
3032 | """ |
|
3029 | """ | |
3033 | expected = 0 |
|
3030 | expected = 0 | |
3034 | if len(self): |
|
3031 | if len(self): | |
3035 | expected = max(0, self.end(len(self) - 1)) |
|
3032 | expected = max(0, self.end(len(self) - 1)) | |
3036 |
|
3033 | |||
3037 | try: |
|
3034 | try: | |
3038 | with self._datafp() as f: |
|
3035 | with self._datafp() as f: | |
3039 | f.seek(0, io.SEEK_END) |
|
3036 | f.seek(0, io.SEEK_END) | |
3040 | actual = f.tell() |
|
3037 | actual = f.tell() | |
3041 | dd = actual - expected |
|
3038 | dd = actual - expected | |
3042 | except FileNotFoundError: |
|
3039 | except FileNotFoundError: | |
3043 | dd = 0 |
|
3040 | dd = 0 | |
3044 |
|
3041 | |||
3045 | try: |
|
3042 | try: | |
3046 | f = self.opener(self._indexfile) |
|
3043 | f = self.opener(self._indexfile) | |
3047 | f.seek(0, io.SEEK_END) |
|
3044 | f.seek(0, io.SEEK_END) | |
3048 | actual = f.tell() |
|
3045 | actual = f.tell() | |
3049 | f.close() |
|
3046 | f.close() | |
3050 | s = self.index.entry_size |
|
3047 | s = self.index.entry_size | |
3051 | i = max(0, actual // s) |
|
3048 | i = max(0, actual // s) | |
3052 | di = actual - (i * s) |
|
3049 | di = actual - (i * s) | |
3053 | if self._inline: |
|
3050 | if self._inline: | |
3054 | databytes = 0 |
|
3051 | databytes = 0 | |
3055 | for r in self: |
|
3052 | for r in self: | |
3056 | databytes += max(0, self.length(r)) |
|
3053 | databytes += max(0, self.length(r)) | |
3057 | dd = 0 |
|
3054 | dd = 0 | |
3058 | di = actual - len(self) * s - databytes |
|
3055 | di = actual - len(self) * s - databytes | |
3059 | except FileNotFoundError: |
|
3056 | except FileNotFoundError: | |
3060 | di = 0 |
|
3057 | di = 0 | |
3061 |
|
3058 | |||
3062 | return (dd, di) |
|
3059 | return (dd, di) | |
3063 |
|
3060 | |||
3064 | def files(self): |
|
3061 | def files(self): | |
3065 | res = [self._indexfile] |
|
3062 | res = [self._indexfile] | |
3066 | if self._docket_file is None: |
|
3063 | if self._docket_file is None: | |
3067 | if not self._inline: |
|
3064 | if not self._inline: | |
3068 | res.append(self._datafile) |
|
3065 | res.append(self._datafile) | |
3069 | else: |
|
3066 | else: | |
3070 | res.append(self._docket_file) |
|
3067 | res.append(self._docket_file) | |
3071 | res.extend(self._docket.old_index_filepaths(include_empty=False)) |
|
3068 | res.extend(self._docket.old_index_filepaths(include_empty=False)) | |
3072 | if self._docket.data_end: |
|
3069 | if self._docket.data_end: | |
3073 | res.append(self._datafile) |
|
3070 | res.append(self._datafile) | |
3074 | res.extend(self._docket.old_data_filepaths(include_empty=False)) |
|
3071 | res.extend(self._docket.old_data_filepaths(include_empty=False)) | |
3075 | if self._docket.sidedata_end: |
|
3072 | if self._docket.sidedata_end: | |
3076 | res.append(self._sidedatafile) |
|
3073 | res.append(self._sidedatafile) | |
3077 | res.extend(self._docket.old_sidedata_filepaths(include_empty=False)) |
|
3074 | res.extend(self._docket.old_sidedata_filepaths(include_empty=False)) | |
3078 | return res |
|
3075 | return res | |
3079 |
|
3076 | |||
3080 | def emitrevisions( |
|
3077 | def emitrevisions( | |
3081 | self, |
|
3078 | self, | |
3082 | nodes, |
|
3079 | nodes, | |
3083 | nodesorder=None, |
|
3080 | nodesorder=None, | |
3084 | revisiondata=False, |
|
3081 | revisiondata=False, | |
3085 | assumehaveparentrevisions=False, |
|
3082 | assumehaveparentrevisions=False, | |
3086 | deltamode=repository.CG_DELTAMODE_STD, |
|
3083 | deltamode=repository.CG_DELTAMODE_STD, | |
3087 | sidedata_helpers=None, |
|
3084 | sidedata_helpers=None, | |
3088 | debug_info=None, |
|
3085 | debug_info=None, | |
3089 | ): |
|
3086 | ): | |
3090 | if nodesorder not in (b'nodes', b'storage', b'linear', None): |
|
3087 | if nodesorder not in (b'nodes', b'storage', b'linear', None): | |
3091 | raise error.ProgrammingError( |
|
3088 | raise error.ProgrammingError( | |
3092 | b'unhandled value for nodesorder: %s' % nodesorder |
|
3089 | b'unhandled value for nodesorder: %s' % nodesorder | |
3093 | ) |
|
3090 | ) | |
3094 |
|
3091 | |||
3095 | if nodesorder is None and not self._generaldelta: |
|
3092 | if nodesorder is None and not self._generaldelta: | |
3096 | nodesorder = b'storage' |
|
3093 | nodesorder = b'storage' | |
3097 |
|
3094 | |||
3098 | if ( |
|
3095 | if ( | |
3099 | not self._storedeltachains |
|
3096 | not self._storedeltachains | |
3100 | and deltamode != repository.CG_DELTAMODE_PREV |
|
3097 | and deltamode != repository.CG_DELTAMODE_PREV | |
3101 | ): |
|
3098 | ): | |
3102 | deltamode = repository.CG_DELTAMODE_FULL |
|
3099 | deltamode = repository.CG_DELTAMODE_FULL | |
3103 |
|
3100 | |||
3104 | return storageutil.emitrevisions( |
|
3101 | return storageutil.emitrevisions( | |
3105 | self, |
|
3102 | self, | |
3106 | nodes, |
|
3103 | nodes, | |
3107 | nodesorder, |
|
3104 | nodesorder, | |
3108 | revlogrevisiondelta, |
|
3105 | revlogrevisiondelta, | |
3109 | deltaparentfn=self.deltaparent, |
|
3106 | deltaparentfn=self.deltaparent, | |
3110 | candeltafn=self._candelta, |
|
3107 | candeltafn=self._candelta, | |
3111 | rawsizefn=self.rawsize, |
|
3108 | rawsizefn=self.rawsize, | |
3112 | revdifffn=self.revdiff, |
|
3109 | revdifffn=self.revdiff, | |
3113 | flagsfn=self.flags, |
|
3110 | flagsfn=self.flags, | |
3114 | deltamode=deltamode, |
|
3111 | deltamode=deltamode, | |
3115 | revisiondata=revisiondata, |
|
3112 | revisiondata=revisiondata, | |
3116 | assumehaveparentrevisions=assumehaveparentrevisions, |
|
3113 | assumehaveparentrevisions=assumehaveparentrevisions, | |
3117 | sidedata_helpers=sidedata_helpers, |
|
3114 | sidedata_helpers=sidedata_helpers, | |
3118 | debug_info=debug_info, |
|
3115 | debug_info=debug_info, | |
3119 | ) |
|
3116 | ) | |
3120 |
|
3117 | |||
3121 | DELTAREUSEALWAYS = b'always' |
|
3118 | DELTAREUSEALWAYS = b'always' | |
3122 | DELTAREUSESAMEREVS = b'samerevs' |
|
3119 | DELTAREUSESAMEREVS = b'samerevs' | |
3123 | DELTAREUSENEVER = b'never' |
|
3120 | DELTAREUSENEVER = b'never' | |
3124 |
|
3121 | |||
3125 | DELTAREUSEFULLADD = b'fulladd' |
|
3122 | DELTAREUSEFULLADD = b'fulladd' | |
3126 |
|
3123 | |||
3127 | DELTAREUSEALL = {b'always', b'samerevs', b'never', b'fulladd'} |
|
3124 | DELTAREUSEALL = {b'always', b'samerevs', b'never', b'fulladd'} | |
3128 |
|
3125 | |||
3129 | def clone( |
|
3126 | def clone( | |
3130 | self, |
|
3127 | self, | |
3131 | tr, |
|
3128 | tr, | |
3132 | destrevlog, |
|
3129 | destrevlog, | |
3133 | addrevisioncb=None, |
|
3130 | addrevisioncb=None, | |
3134 | deltareuse=DELTAREUSESAMEREVS, |
|
3131 | deltareuse=DELTAREUSESAMEREVS, | |
3135 | forcedeltabothparents=None, |
|
3132 | forcedeltabothparents=None, | |
3136 | sidedata_helpers=None, |
|
3133 | sidedata_helpers=None, | |
3137 | ): |
|
3134 | ): | |
3138 | """Copy this revlog to another, possibly with format changes. |
|
3135 | """Copy this revlog to another, possibly with format changes. | |
3139 |
|
3136 | |||
3140 | The destination revlog will contain the same revisions and nodes. |
|
3137 | The destination revlog will contain the same revisions and nodes. | |
3141 | However, it may not be bit-for-bit identical due to e.g. delta encoding |
|
3138 | However, it may not be bit-for-bit identical due to e.g. delta encoding | |
3142 | differences. |
|
3139 | differences. | |
3143 |
|
3140 | |||
3144 | The ``deltareuse`` argument control how deltas from the existing revlog |
|
3141 | The ``deltareuse`` argument control how deltas from the existing revlog | |
3145 | are preserved in the destination revlog. The argument can have the |
|
3142 | are preserved in the destination revlog. The argument can have the | |
3146 | following values: |
|
3143 | following values: | |
3147 |
|
3144 | |||
3148 | DELTAREUSEALWAYS |
|
3145 | DELTAREUSEALWAYS | |
3149 | Deltas will always be reused (if possible), even if the destination |
|
3146 | Deltas will always be reused (if possible), even if the destination | |
3150 | revlog would not select the same revisions for the delta. This is the |
|
3147 | revlog would not select the same revisions for the delta. This is the | |
3151 | fastest mode of operation. |
|
3148 | fastest mode of operation. | |
3152 | DELTAREUSESAMEREVS |
|
3149 | DELTAREUSESAMEREVS | |
3153 | Deltas will be reused if the destination revlog would pick the same |
|
3150 | Deltas will be reused if the destination revlog would pick the same | |
3154 | revisions for the delta. This mode strikes a balance between speed |
|
3151 | revisions for the delta. This mode strikes a balance between speed | |
3155 | and optimization. |
|
3152 | and optimization. | |
3156 | DELTAREUSENEVER |
|
3153 | DELTAREUSENEVER | |
3157 | Deltas will never be reused. This is the slowest mode of execution. |
|
3154 | Deltas will never be reused. This is the slowest mode of execution. | |
3158 | This mode can be used to recompute deltas (e.g. if the diff/delta |
|
3155 | This mode can be used to recompute deltas (e.g. if the diff/delta | |
3159 | algorithm changes). |
|
3156 | algorithm changes). | |
3160 | DELTAREUSEFULLADD |
|
3157 | DELTAREUSEFULLADD | |
3161 | Revision will be re-added as if their were new content. This is |
|
3158 | Revision will be re-added as if their were new content. This is | |
3162 | slower than DELTAREUSEALWAYS but allow more mechanism to kicks in. |
|
3159 | slower than DELTAREUSEALWAYS but allow more mechanism to kicks in. | |
3163 | eg: large file detection and handling. |
|
3160 | eg: large file detection and handling. | |
3164 |
|
3161 | |||
3165 | Delta computation can be slow, so the choice of delta reuse policy can |
|
3162 | Delta computation can be slow, so the choice of delta reuse policy can | |
3166 | significantly affect run time. |
|
3163 | significantly affect run time. | |
3167 |
|
3164 | |||
3168 | The default policy (``DELTAREUSESAMEREVS``) strikes a balance between |
|
3165 | The default policy (``DELTAREUSESAMEREVS``) strikes a balance between | |
3169 | two extremes. Deltas will be reused if they are appropriate. But if the |
|
3166 | two extremes. Deltas will be reused if they are appropriate. But if the | |
3170 | delta could choose a better revision, it will do so. This means if you |
|
3167 | delta could choose a better revision, it will do so. This means if you | |
3171 | are converting a non-generaldelta revlog to a generaldelta revlog, |
|
3168 | are converting a non-generaldelta revlog to a generaldelta revlog, | |
3172 | deltas will be recomputed if the delta's parent isn't a parent of the |
|
3169 | deltas will be recomputed if the delta's parent isn't a parent of the | |
3173 | revision. |
|
3170 | revision. | |
3174 |
|
3171 | |||
3175 | In addition to the delta policy, the ``forcedeltabothparents`` |
|
3172 | In addition to the delta policy, the ``forcedeltabothparents`` | |
3176 | argument controls whether to force compute deltas against both parents |
|
3173 | argument controls whether to force compute deltas against both parents | |
3177 | for merges. By default, the current default is used. |
|
3174 | for merges. By default, the current default is used. | |
3178 |
|
3175 | |||
3179 | See `revlogutil.sidedata.get_sidedata_helpers` for the doc on |
|
3176 | See `revlogutil.sidedata.get_sidedata_helpers` for the doc on | |
3180 | `sidedata_helpers`. |
|
3177 | `sidedata_helpers`. | |
3181 | """ |
|
3178 | """ | |
3182 | if deltareuse not in self.DELTAREUSEALL: |
|
3179 | if deltareuse not in self.DELTAREUSEALL: | |
3183 | raise ValueError( |
|
3180 | raise ValueError( | |
3184 | _(b'value for deltareuse invalid: %s') % deltareuse |
|
3181 | _(b'value for deltareuse invalid: %s') % deltareuse | |
3185 | ) |
|
3182 | ) | |
3186 |
|
3183 | |||
3187 | if len(destrevlog): |
|
3184 | if len(destrevlog): | |
3188 | raise ValueError(_(b'destination revlog is not empty')) |
|
3185 | raise ValueError(_(b'destination revlog is not empty')) | |
3189 |
|
3186 | |||
3190 | if getattr(self, 'filteredrevs', None): |
|
3187 | if getattr(self, 'filteredrevs', None): | |
3191 | raise ValueError(_(b'source revlog has filtered revisions')) |
|
3188 | raise ValueError(_(b'source revlog has filtered revisions')) | |
3192 | if getattr(destrevlog, 'filteredrevs', None): |
|
3189 | if getattr(destrevlog, 'filteredrevs', None): | |
3193 | raise ValueError(_(b'destination revlog has filtered revisions')) |
|
3190 | raise ValueError(_(b'destination revlog has filtered revisions')) | |
3194 |
|
3191 | |||
3195 | # lazydelta and lazydeltabase controls whether to reuse a cached delta, |
|
3192 | # lazydelta and lazydeltabase controls whether to reuse a cached delta, | |
3196 | # if possible. |
|
3193 | # if possible. | |
3197 | oldlazydelta = destrevlog._lazydelta |
|
3194 | oldlazydelta = destrevlog._lazydelta | |
3198 | oldlazydeltabase = destrevlog._lazydeltabase |
|
3195 | oldlazydeltabase = destrevlog._lazydeltabase | |
3199 | oldamd = destrevlog._deltabothparents |
|
3196 | oldamd = destrevlog._deltabothparents | |
3200 |
|
3197 | |||
3201 | try: |
|
3198 | try: | |
3202 | if deltareuse == self.DELTAREUSEALWAYS: |
|
3199 | if deltareuse == self.DELTAREUSEALWAYS: | |
3203 | destrevlog._lazydeltabase = True |
|
3200 | destrevlog._lazydeltabase = True | |
3204 | destrevlog._lazydelta = True |
|
3201 | destrevlog._lazydelta = True | |
3205 | elif deltareuse == self.DELTAREUSESAMEREVS: |
|
3202 | elif deltareuse == self.DELTAREUSESAMEREVS: | |
3206 | destrevlog._lazydeltabase = False |
|
3203 | destrevlog._lazydeltabase = False | |
3207 | destrevlog._lazydelta = True |
|
3204 | destrevlog._lazydelta = True | |
3208 | elif deltareuse == self.DELTAREUSENEVER: |
|
3205 | elif deltareuse == self.DELTAREUSENEVER: | |
3209 | destrevlog._lazydeltabase = False |
|
3206 | destrevlog._lazydeltabase = False | |
3210 | destrevlog._lazydelta = False |
|
3207 | destrevlog._lazydelta = False | |
3211 |
|
3208 | |||
3212 | destrevlog._deltabothparents = forcedeltabothparents or oldamd |
|
3209 | destrevlog._deltabothparents = forcedeltabothparents or oldamd | |
3213 |
|
3210 | |||
3214 | with self.reading(): |
|
3211 | with self.reading(): | |
3215 | self._clone( |
|
3212 | self._clone( | |
3216 | tr, |
|
3213 | tr, | |
3217 | destrevlog, |
|
3214 | destrevlog, | |
3218 | addrevisioncb, |
|
3215 | addrevisioncb, | |
3219 | deltareuse, |
|
3216 | deltareuse, | |
3220 | forcedeltabothparents, |
|
3217 | forcedeltabothparents, | |
3221 | sidedata_helpers, |
|
3218 | sidedata_helpers, | |
3222 | ) |
|
3219 | ) | |
3223 |
|
3220 | |||
3224 | finally: |
|
3221 | finally: | |
3225 | destrevlog._lazydelta = oldlazydelta |
|
3222 | destrevlog._lazydelta = oldlazydelta | |
3226 | destrevlog._lazydeltabase = oldlazydeltabase |
|
3223 | destrevlog._lazydeltabase = oldlazydeltabase | |
3227 | destrevlog._deltabothparents = oldamd |
|
3224 | destrevlog._deltabothparents = oldamd | |
3228 |
|
3225 | |||
3229 | def _clone( |
|
3226 | def _clone( | |
3230 | self, |
|
3227 | self, | |
3231 | tr, |
|
3228 | tr, | |
3232 | destrevlog, |
|
3229 | destrevlog, | |
3233 | addrevisioncb, |
|
3230 | addrevisioncb, | |
3234 | deltareuse, |
|
3231 | deltareuse, | |
3235 | forcedeltabothparents, |
|
3232 | forcedeltabothparents, | |
3236 | sidedata_helpers, |
|
3233 | sidedata_helpers, | |
3237 | ): |
|
3234 | ): | |
3238 | """perform the core duty of `revlog.clone` after parameter processing""" |
|
3235 | """perform the core duty of `revlog.clone` after parameter processing""" | |
3239 | write_debug = None |
|
3236 | write_debug = None | |
3240 | if self._debug_delta: |
|
3237 | if self._debug_delta: | |
3241 | write_debug = tr._report |
|
3238 | write_debug = tr._report | |
3242 | deltacomputer = deltautil.deltacomputer( |
|
3239 | deltacomputer = deltautil.deltacomputer( | |
3243 | destrevlog, |
|
3240 | destrevlog, | |
3244 | write_debug=write_debug, |
|
3241 | write_debug=write_debug, | |
3245 | ) |
|
3242 | ) | |
3246 | index = self.index |
|
3243 | index = self.index | |
3247 | for rev in self: |
|
3244 | for rev in self: | |
3248 | entry = index[rev] |
|
3245 | entry = index[rev] | |
3249 |
|
3246 | |||
3250 | # Some classes override linkrev to take filtered revs into |
|
3247 | # Some classes override linkrev to take filtered revs into | |
3251 | # account. Use raw entry from index. |
|
3248 | # account. Use raw entry from index. | |
3252 | flags = entry[0] & 0xFFFF |
|
3249 | flags = entry[0] & 0xFFFF | |
3253 | linkrev = entry[4] |
|
3250 | linkrev = entry[4] | |
3254 | p1 = index[entry[5]][7] |
|
3251 | p1 = index[entry[5]][7] | |
3255 | p2 = index[entry[6]][7] |
|
3252 | p2 = index[entry[6]][7] | |
3256 | node = entry[7] |
|
3253 | node = entry[7] | |
3257 |
|
3254 | |||
3258 | # (Possibly) reuse the delta from the revlog if allowed and |
|
3255 | # (Possibly) reuse the delta from the revlog if allowed and | |
3259 | # the revlog chunk is a delta. |
|
3256 | # the revlog chunk is a delta. | |
3260 | cachedelta = None |
|
3257 | cachedelta = None | |
3261 | rawtext = None |
|
3258 | rawtext = None | |
3262 | if deltareuse == self.DELTAREUSEFULLADD: |
|
3259 | if deltareuse == self.DELTAREUSEFULLADD: | |
3263 | text = self._revisiondata(rev) |
|
3260 | text = self._revisiondata(rev) | |
3264 | sidedata = self.sidedata(rev) |
|
3261 | sidedata = self.sidedata(rev) | |
3265 |
|
3262 | |||
3266 | if sidedata_helpers is not None: |
|
3263 | if sidedata_helpers is not None: | |
3267 | (sidedata, new_flags) = sidedatautil.run_sidedata_helpers( |
|
3264 | (sidedata, new_flags) = sidedatautil.run_sidedata_helpers( | |
3268 | self, sidedata_helpers, sidedata, rev |
|
3265 | self, sidedata_helpers, sidedata, rev | |
3269 | ) |
|
3266 | ) | |
3270 | flags = flags | new_flags[0] & ~new_flags[1] |
|
3267 | flags = flags | new_flags[0] & ~new_flags[1] | |
3271 |
|
3268 | |||
3272 | destrevlog.addrevision( |
|
3269 | destrevlog.addrevision( | |
3273 | text, |
|
3270 | text, | |
3274 | tr, |
|
3271 | tr, | |
3275 | linkrev, |
|
3272 | linkrev, | |
3276 | p1, |
|
3273 | p1, | |
3277 | p2, |
|
3274 | p2, | |
3278 | cachedelta=cachedelta, |
|
3275 | cachedelta=cachedelta, | |
3279 | node=node, |
|
3276 | node=node, | |
3280 | flags=flags, |
|
3277 | flags=flags, | |
3281 | deltacomputer=deltacomputer, |
|
3278 | deltacomputer=deltacomputer, | |
3282 | sidedata=sidedata, |
|
3279 | sidedata=sidedata, | |
3283 | ) |
|
3280 | ) | |
3284 | else: |
|
3281 | else: | |
3285 | if destrevlog._lazydelta: |
|
3282 | if destrevlog._lazydelta: | |
3286 | dp = self.deltaparent(rev) |
|
3283 | dp = self.deltaparent(rev) | |
3287 | if dp != nullrev: |
|
3284 | if dp != nullrev: | |
3288 | cachedelta = (dp, bytes(self._chunk(rev))) |
|
3285 | cachedelta = (dp, bytes(self._chunk(rev))) | |
3289 |
|
3286 | |||
3290 | sidedata = None |
|
3287 | sidedata = None | |
3291 | if not cachedelta: |
|
3288 | if not cachedelta: | |
3292 | rawtext = self._revisiondata(rev) |
|
3289 | rawtext = self._revisiondata(rev) | |
3293 | sidedata = self.sidedata(rev) |
|
3290 | sidedata = self.sidedata(rev) | |
3294 | if sidedata is None: |
|
3291 | if sidedata is None: | |
3295 | sidedata = self.sidedata(rev) |
|
3292 | sidedata = self.sidedata(rev) | |
3296 |
|
3293 | |||
3297 | if sidedata_helpers is not None: |
|
3294 | if sidedata_helpers is not None: | |
3298 | (sidedata, new_flags) = sidedatautil.run_sidedata_helpers( |
|
3295 | (sidedata, new_flags) = sidedatautil.run_sidedata_helpers( | |
3299 | self, sidedata_helpers, sidedata, rev |
|
3296 | self, sidedata_helpers, sidedata, rev | |
3300 | ) |
|
3297 | ) | |
3301 | flags = flags | new_flags[0] & ~new_flags[1] |
|
3298 | flags = flags | new_flags[0] & ~new_flags[1] | |
3302 |
|
3299 | |||
3303 | with destrevlog._writing(tr): |
|
3300 | with destrevlog._writing(tr): | |
3304 | destrevlog._addrevision( |
|
3301 | destrevlog._addrevision( | |
3305 | node, |
|
3302 | node, | |
3306 | rawtext, |
|
3303 | rawtext, | |
3307 | tr, |
|
3304 | tr, | |
3308 | linkrev, |
|
3305 | linkrev, | |
3309 | p1, |
|
3306 | p1, | |
3310 | p2, |
|
3307 | p2, | |
3311 | flags, |
|
3308 | flags, | |
3312 | cachedelta, |
|
3309 | cachedelta, | |
3313 | deltacomputer=deltacomputer, |
|
3310 | deltacomputer=deltacomputer, | |
3314 | sidedata=sidedata, |
|
3311 | sidedata=sidedata, | |
3315 | ) |
|
3312 | ) | |
3316 |
|
3313 | |||
3317 | if addrevisioncb: |
|
3314 | if addrevisioncb: | |
3318 | addrevisioncb(self, rev, node) |
|
3315 | addrevisioncb(self, rev, node) | |
3319 |
|
3316 | |||
3320 | def censorrevision(self, tr, censornode, tombstone=b''): |
|
3317 | def censorrevision(self, tr, censornode, tombstone=b''): | |
3321 | if self._format_version == REVLOGV0: |
|
3318 | if self._format_version == REVLOGV0: | |
3322 | raise error.RevlogError( |
|
3319 | raise error.RevlogError( | |
3323 | _(b'cannot censor with version %d revlogs') |
|
3320 | _(b'cannot censor with version %d revlogs') | |
3324 | % self._format_version |
|
3321 | % self._format_version | |
3325 | ) |
|
3322 | ) | |
3326 | elif self._format_version == REVLOGV1: |
|
3323 | elif self._format_version == REVLOGV1: | |
3327 | rewrite.v1_censor(self, tr, censornode, tombstone) |
|
3324 | rewrite.v1_censor(self, tr, censornode, tombstone) | |
3328 | else: |
|
3325 | else: | |
3329 | rewrite.v2_censor(self, tr, censornode, tombstone) |
|
3326 | rewrite.v2_censor(self, tr, censornode, tombstone) | |
3330 |
|
3327 | |||
3331 | def verifyintegrity(self, state): |
|
3328 | def verifyintegrity(self, state): | |
3332 | """Verifies the integrity of the revlog. |
|
3329 | """Verifies the integrity of the revlog. | |
3333 |
|
3330 | |||
3334 | Yields ``revlogproblem`` instances describing problems that are |
|
3331 | Yields ``revlogproblem`` instances describing problems that are | |
3335 | found. |
|
3332 | found. | |
3336 | """ |
|
3333 | """ | |
3337 | dd, di = self.checksize() |
|
3334 | dd, di = self.checksize() | |
3338 | if dd: |
|
3335 | if dd: | |
3339 | yield revlogproblem(error=_(b'data length off by %d bytes') % dd) |
|
3336 | yield revlogproblem(error=_(b'data length off by %d bytes') % dd) | |
3340 | if di: |
|
3337 | if di: | |
3341 | yield revlogproblem(error=_(b'index contains %d extra bytes') % di) |
|
3338 | yield revlogproblem(error=_(b'index contains %d extra bytes') % di) | |
3342 |
|
3339 | |||
3343 | version = self._format_version |
|
3340 | version = self._format_version | |
3344 |
|
3341 | |||
3345 | # The verifier tells us what version revlog we should be. |
|
3342 | # The verifier tells us what version revlog we should be. | |
3346 | if version != state[b'expectedversion']: |
|
3343 | if version != state[b'expectedversion']: | |
3347 | yield revlogproblem( |
|
3344 | yield revlogproblem( | |
3348 | warning=_(b"warning: '%s' uses revlog format %d; expected %d") |
|
3345 | warning=_(b"warning: '%s' uses revlog format %d; expected %d") | |
3349 | % (self.display_id, version, state[b'expectedversion']) |
|
3346 | % (self.display_id, version, state[b'expectedversion']) | |
3350 | ) |
|
3347 | ) | |
3351 |
|
3348 | |||
3352 | state[b'skipread'] = set() |
|
3349 | state[b'skipread'] = set() | |
3353 | state[b'safe_renamed'] = set() |
|
3350 | state[b'safe_renamed'] = set() | |
3354 |
|
3351 | |||
3355 | for rev in self: |
|
3352 | for rev in self: | |
3356 | node = self.node(rev) |
|
3353 | node = self.node(rev) | |
3357 |
|
3354 | |||
3358 | # Verify contents. 4 cases to care about: |
|
3355 | # Verify contents. 4 cases to care about: | |
3359 | # |
|
3356 | # | |
3360 | # common: the most common case |
|
3357 | # common: the most common case | |
3361 | # rename: with a rename |
|
3358 | # rename: with a rename | |
3362 | # meta: file content starts with b'\1\n', the metadata |
|
3359 | # meta: file content starts with b'\1\n', the metadata | |
3363 | # header defined in filelog.py, but without a rename |
|
3360 | # header defined in filelog.py, but without a rename | |
3364 | # ext: content stored externally |
|
3361 | # ext: content stored externally | |
3365 | # |
|
3362 | # | |
3366 | # More formally, their differences are shown below: |
|
3363 | # More formally, their differences are shown below: | |
3367 | # |
|
3364 | # | |
3368 | # | common | rename | meta | ext |
|
3365 | # | common | rename | meta | ext | |
3369 | # ------------------------------------------------------- |
|
3366 | # ------------------------------------------------------- | |
3370 | # flags() | 0 | 0 | 0 | not 0 |
|
3367 | # flags() | 0 | 0 | 0 | not 0 | |
3371 | # renamed() | False | True | False | ? |
|
3368 | # renamed() | False | True | False | ? | |
3372 | # rawtext[0:2]=='\1\n'| False | True | True | ? |
|
3369 | # rawtext[0:2]=='\1\n'| False | True | True | ? | |
3373 | # |
|
3370 | # | |
3374 | # "rawtext" means the raw text stored in revlog data, which |
|
3371 | # "rawtext" means the raw text stored in revlog data, which | |
3375 | # could be retrieved by "rawdata(rev)". "text" |
|
3372 | # could be retrieved by "rawdata(rev)". "text" | |
3376 | # mentioned below is "revision(rev)". |
|
3373 | # mentioned below is "revision(rev)". | |
3377 | # |
|
3374 | # | |
3378 | # There are 3 different lengths stored physically: |
|
3375 | # There are 3 different lengths stored physically: | |
3379 | # 1. L1: rawsize, stored in revlog index |
|
3376 | # 1. L1: rawsize, stored in revlog index | |
3380 | # 2. L2: len(rawtext), stored in revlog data |
|
3377 | # 2. L2: len(rawtext), stored in revlog data | |
3381 | # 3. L3: len(text), stored in revlog data if flags==0, or |
|
3378 | # 3. L3: len(text), stored in revlog data if flags==0, or | |
3382 | # possibly somewhere else if flags!=0 |
|
3379 | # possibly somewhere else if flags!=0 | |
3383 | # |
|
3380 | # | |
3384 | # L1 should be equal to L2. L3 could be different from them. |
|
3381 | # L1 should be equal to L2. L3 could be different from them. | |
3385 | # "text" may or may not affect commit hash depending on flag |
|
3382 | # "text" may or may not affect commit hash depending on flag | |
3386 | # processors (see flagutil.addflagprocessor). |
|
3383 | # processors (see flagutil.addflagprocessor). | |
3387 | # |
|
3384 | # | |
3388 | # | common | rename | meta | ext |
|
3385 | # | common | rename | meta | ext | |
3389 | # ------------------------------------------------- |
|
3386 | # ------------------------------------------------- | |
3390 | # rawsize() | L1 | L1 | L1 | L1 |
|
3387 | # rawsize() | L1 | L1 | L1 | L1 | |
3391 | # size() | L1 | L2-LM | L1(*) | L1 (?) |
|
3388 | # size() | L1 | L2-LM | L1(*) | L1 (?) | |
3392 | # len(rawtext) | L2 | L2 | L2 | L2 |
|
3389 | # len(rawtext) | L2 | L2 | L2 | L2 | |
3393 | # len(text) | L2 | L2 | L2 | L3 |
|
3390 | # len(text) | L2 | L2 | L2 | L3 | |
3394 | # len(read()) | L2 | L2-LM | L2-LM | L3 (?) |
|
3391 | # len(read()) | L2 | L2-LM | L2-LM | L3 (?) | |
3395 | # |
|
3392 | # | |
3396 | # LM: length of metadata, depending on rawtext |
|
3393 | # LM: length of metadata, depending on rawtext | |
3397 | # (*): not ideal, see comment in filelog.size |
|
3394 | # (*): not ideal, see comment in filelog.size | |
3398 | # (?): could be "- len(meta)" if the resolved content has |
|
3395 | # (?): could be "- len(meta)" if the resolved content has | |
3399 | # rename metadata |
|
3396 | # rename metadata | |
3400 | # |
|
3397 | # | |
3401 | # Checks needed to be done: |
|
3398 | # Checks needed to be done: | |
3402 | # 1. length check: L1 == L2, in all cases. |
|
3399 | # 1. length check: L1 == L2, in all cases. | |
3403 | # 2. hash check: depending on flag processor, we may need to |
|
3400 | # 2. hash check: depending on flag processor, we may need to | |
3404 | # use either "text" (external), or "rawtext" (in revlog). |
|
3401 | # use either "text" (external), or "rawtext" (in revlog). | |
3405 |
|
3402 | |||
3406 | try: |
|
3403 | try: | |
3407 | skipflags = state.get(b'skipflags', 0) |
|
3404 | skipflags = state.get(b'skipflags', 0) | |
3408 | if skipflags: |
|
3405 | if skipflags: | |
3409 | skipflags &= self.flags(rev) |
|
3406 | skipflags &= self.flags(rev) | |
3410 |
|
3407 | |||
3411 | _verify_revision(self, skipflags, state, node) |
|
3408 | _verify_revision(self, skipflags, state, node) | |
3412 |
|
3409 | |||
3413 | l1 = self.rawsize(rev) |
|
3410 | l1 = self.rawsize(rev) | |
3414 | l2 = len(self.rawdata(node)) |
|
3411 | l2 = len(self.rawdata(node)) | |
3415 |
|
3412 | |||
3416 | if l1 != l2: |
|
3413 | if l1 != l2: | |
3417 | yield revlogproblem( |
|
3414 | yield revlogproblem( | |
3418 | error=_(b'unpacked size is %d, %d expected') % (l2, l1), |
|
3415 | error=_(b'unpacked size is %d, %d expected') % (l2, l1), | |
3419 | node=node, |
|
3416 | node=node, | |
3420 | ) |
|
3417 | ) | |
3421 |
|
3418 | |||
3422 | except error.CensoredNodeError: |
|
3419 | except error.CensoredNodeError: | |
3423 | if state[b'erroroncensored']: |
|
3420 | if state[b'erroroncensored']: | |
3424 | yield revlogproblem( |
|
3421 | yield revlogproblem( | |
3425 | error=_(b'censored file data'), node=node |
|
3422 | error=_(b'censored file data'), node=node | |
3426 | ) |
|
3423 | ) | |
3427 | state[b'skipread'].add(node) |
|
3424 | state[b'skipread'].add(node) | |
3428 | except Exception as e: |
|
3425 | except Exception as e: | |
3429 | yield revlogproblem( |
|
3426 | yield revlogproblem( | |
3430 | error=_(b'unpacking %s: %s') |
|
3427 | error=_(b'unpacking %s: %s') | |
3431 | % (short(node), stringutil.forcebytestr(e)), |
|
3428 | % (short(node), stringutil.forcebytestr(e)), | |
3432 | node=node, |
|
3429 | node=node, | |
3433 | ) |
|
3430 | ) | |
3434 | state[b'skipread'].add(node) |
|
3431 | state[b'skipread'].add(node) | |
3435 |
|
3432 | |||
3436 | def storageinfo( |
|
3433 | def storageinfo( | |
3437 | self, |
|
3434 | self, | |
3438 | exclusivefiles=False, |
|
3435 | exclusivefiles=False, | |
3439 | sharedfiles=False, |
|
3436 | sharedfiles=False, | |
3440 | revisionscount=False, |
|
3437 | revisionscount=False, | |
3441 | trackedsize=False, |
|
3438 | trackedsize=False, | |
3442 | storedsize=False, |
|
3439 | storedsize=False, | |
3443 | ): |
|
3440 | ): | |
3444 | d = {} |
|
3441 | d = {} | |
3445 |
|
3442 | |||
3446 | if exclusivefiles: |
|
3443 | if exclusivefiles: | |
3447 | d[b'exclusivefiles'] = [(self.opener, self._indexfile)] |
|
3444 | d[b'exclusivefiles'] = [(self.opener, self._indexfile)] | |
3448 | if not self._inline: |
|
3445 | if not self._inline: | |
3449 | d[b'exclusivefiles'].append((self.opener, self._datafile)) |
|
3446 | d[b'exclusivefiles'].append((self.opener, self._datafile)) | |
3450 |
|
3447 | |||
3451 | if sharedfiles: |
|
3448 | if sharedfiles: | |
3452 | d[b'sharedfiles'] = [] |
|
3449 | d[b'sharedfiles'] = [] | |
3453 |
|
3450 | |||
3454 | if revisionscount: |
|
3451 | if revisionscount: | |
3455 | d[b'revisionscount'] = len(self) |
|
3452 | d[b'revisionscount'] = len(self) | |
3456 |
|
3453 | |||
3457 | if trackedsize: |
|
3454 | if trackedsize: | |
3458 | d[b'trackedsize'] = sum(map(self.rawsize, iter(self))) |
|
3455 | d[b'trackedsize'] = sum(map(self.rawsize, iter(self))) | |
3459 |
|
3456 | |||
3460 | if storedsize: |
|
3457 | if storedsize: | |
3461 | d[b'storedsize'] = sum( |
|
3458 | d[b'storedsize'] = sum( | |
3462 | self.opener.stat(path).st_size for path in self.files() |
|
3459 | self.opener.stat(path).st_size for path in self.files() | |
3463 | ) |
|
3460 | ) | |
3464 |
|
3461 | |||
3465 | return d |
|
3462 | return d | |
3466 |
|
3463 | |||
3467 | def rewrite_sidedata(self, transaction, helpers, startrev, endrev): |
|
3464 | def rewrite_sidedata(self, transaction, helpers, startrev, endrev): | |
3468 | if not self.hassidedata: |
|
3465 | if not self.hassidedata: | |
3469 | return |
|
3466 | return | |
3470 | # revlog formats with sidedata support does not support inline |
|
3467 | # revlog formats with sidedata support does not support inline | |
3471 | assert not self._inline |
|
3468 | assert not self._inline | |
3472 | if not helpers[1] and not helpers[2]: |
|
3469 | if not helpers[1] and not helpers[2]: | |
3473 | # Nothing to generate or remove |
|
3470 | # Nothing to generate or remove | |
3474 | return |
|
3471 | return | |
3475 |
|
3472 | |||
3476 | new_entries = [] |
|
3473 | new_entries = [] | |
3477 | # append the new sidedata |
|
3474 | # append the new sidedata | |
3478 | with self._writing(transaction): |
|
3475 | with self._writing(transaction): | |
3479 | ifh, dfh, sdfh = self._writinghandles |
|
3476 | ifh, dfh, sdfh = self._writinghandles | |
3480 | dfh.seek(self._docket.sidedata_end, os.SEEK_SET) |
|
3477 | dfh.seek(self._docket.sidedata_end, os.SEEK_SET) | |
3481 |
|
3478 | |||
3482 | current_offset = sdfh.tell() |
|
3479 | current_offset = sdfh.tell() | |
3483 | for rev in range(startrev, endrev + 1): |
|
3480 | for rev in range(startrev, endrev + 1): | |
3484 | entry = self.index[rev] |
|
3481 | entry = self.index[rev] | |
3485 | new_sidedata, flags = sidedatautil.run_sidedata_helpers( |
|
3482 | new_sidedata, flags = sidedatautil.run_sidedata_helpers( | |
3486 | store=self, |
|
3483 | store=self, | |
3487 | sidedata_helpers=helpers, |
|
3484 | sidedata_helpers=helpers, | |
3488 | sidedata={}, |
|
3485 | sidedata={}, | |
3489 | rev=rev, |
|
3486 | rev=rev, | |
3490 | ) |
|
3487 | ) | |
3491 |
|
3488 | |||
3492 | serialized_sidedata = sidedatautil.serialize_sidedata( |
|
3489 | serialized_sidedata = sidedatautil.serialize_sidedata( | |
3493 | new_sidedata |
|
3490 | new_sidedata | |
3494 | ) |
|
3491 | ) | |
3495 |
|
3492 | |||
3496 | sidedata_compression_mode = COMP_MODE_INLINE |
|
3493 | sidedata_compression_mode = COMP_MODE_INLINE | |
3497 | if serialized_sidedata and self.hassidedata: |
|
3494 | if serialized_sidedata and self.hassidedata: | |
3498 | sidedata_compression_mode = COMP_MODE_PLAIN |
|
3495 | sidedata_compression_mode = COMP_MODE_PLAIN | |
3499 | h, comp_sidedata = self.compress(serialized_sidedata) |
|
3496 | h, comp_sidedata = self.compress(serialized_sidedata) | |
3500 | if ( |
|
3497 | if ( | |
3501 | h != b'u' |
|
3498 | h != b'u' | |
3502 | and comp_sidedata[0] != b'\0' |
|
3499 | and comp_sidedata[0] != b'\0' | |
3503 | and len(comp_sidedata) < len(serialized_sidedata) |
|
3500 | and len(comp_sidedata) < len(serialized_sidedata) | |
3504 | ): |
|
3501 | ): | |
3505 | assert not h |
|
3502 | assert not h | |
3506 | if ( |
|
3503 | if ( | |
3507 | comp_sidedata[0] |
|
3504 | comp_sidedata[0] | |
3508 | == self._docket.default_compression_header |
|
3505 | == self._docket.default_compression_header | |
3509 | ): |
|
3506 | ): | |
3510 | sidedata_compression_mode = COMP_MODE_DEFAULT |
|
3507 | sidedata_compression_mode = COMP_MODE_DEFAULT | |
3511 | serialized_sidedata = comp_sidedata |
|
3508 | serialized_sidedata = comp_sidedata | |
3512 | else: |
|
3509 | else: | |
3513 | sidedata_compression_mode = COMP_MODE_INLINE |
|
3510 | sidedata_compression_mode = COMP_MODE_INLINE | |
3514 | serialized_sidedata = comp_sidedata |
|
3511 | serialized_sidedata = comp_sidedata | |
3515 | if entry[8] != 0 or entry[9] != 0: |
|
3512 | if entry[8] != 0 or entry[9] != 0: | |
3516 | # rewriting entries that already have sidedata is not |
|
3513 | # rewriting entries that already have sidedata is not | |
3517 | # supported yet, because it introduces garbage data in the |
|
3514 | # supported yet, because it introduces garbage data in the | |
3518 | # revlog. |
|
3515 | # revlog. | |
3519 | msg = b"rewriting existing sidedata is not supported yet" |
|
3516 | msg = b"rewriting existing sidedata is not supported yet" | |
3520 | raise error.Abort(msg) |
|
3517 | raise error.Abort(msg) | |
3521 |
|
3518 | |||
3522 | # Apply (potential) flags to add and to remove after running |
|
3519 | # Apply (potential) flags to add and to remove after running | |
3523 | # the sidedata helpers |
|
3520 | # the sidedata helpers | |
3524 | new_offset_flags = entry[0] | flags[0] & ~flags[1] |
|
3521 | new_offset_flags = entry[0] | flags[0] & ~flags[1] | |
3525 | entry_update = ( |
|
3522 | entry_update = ( | |
3526 | current_offset, |
|
3523 | current_offset, | |
3527 | len(serialized_sidedata), |
|
3524 | len(serialized_sidedata), | |
3528 | new_offset_flags, |
|
3525 | new_offset_flags, | |
3529 | sidedata_compression_mode, |
|
3526 | sidedata_compression_mode, | |
3530 | ) |
|
3527 | ) | |
3531 |
|
3528 | |||
3532 | # the sidedata computation might have move the file cursors around |
|
3529 | # the sidedata computation might have move the file cursors around | |
3533 | sdfh.seek(current_offset, os.SEEK_SET) |
|
3530 | sdfh.seek(current_offset, os.SEEK_SET) | |
3534 | sdfh.write(serialized_sidedata) |
|
3531 | sdfh.write(serialized_sidedata) | |
3535 | new_entries.append(entry_update) |
|
3532 | new_entries.append(entry_update) | |
3536 | current_offset += len(serialized_sidedata) |
|
3533 | current_offset += len(serialized_sidedata) | |
3537 | self._docket.sidedata_end = sdfh.tell() |
|
3534 | self._docket.sidedata_end = sdfh.tell() | |
3538 |
|
3535 | |||
3539 | # rewrite the new index entries |
|
3536 | # rewrite the new index entries | |
3540 | ifh.seek(startrev * self.index.entry_size) |
|
3537 | ifh.seek(startrev * self.index.entry_size) | |
3541 | for i, e in enumerate(new_entries): |
|
3538 | for i, e in enumerate(new_entries): | |
3542 | rev = startrev + i |
|
3539 | rev = startrev + i | |
3543 | self.index.replace_sidedata_info(rev, *e) |
|
3540 | self.index.replace_sidedata_info(rev, *e) | |
3544 | packed = self.index.entry_binary(rev) |
|
3541 | packed = self.index.entry_binary(rev) | |
3545 | if rev == 0 and self._docket is None: |
|
3542 | if rev == 0 and self._docket is None: | |
3546 | header = self._format_flags | self._format_version |
|
3543 | header = self._format_flags | self._format_version | |
3547 | header = self.index.pack_header(header) |
|
3544 | header = self.index.pack_header(header) | |
3548 | packed = header + packed |
|
3545 | packed = header + packed | |
3549 | ifh.write(packed) |
|
3546 | ifh.write(packed) |
General Comments 0
You need to be logged in to leave comments.
Login now