Show More
@@ -1,2152 +1,2156 b'' | |||||
1 | # revlog.py - storage back-end for mercurial |
|
1 | # revlog.py - storage back-end for mercurial | |
2 | # |
|
2 | # | |
3 | # Copyright 2005-2007 Matt Mackall <mpm@selenic.com> |
|
3 | # Copyright 2005-2007 Matt Mackall <mpm@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 | """Storage back-end for Mercurial. |
|
8 | """Storage back-end for Mercurial. | |
9 |
|
9 | |||
10 | This provides efficient delta storage with O(1) retrieve and append |
|
10 | This provides efficient delta storage with O(1) retrieve and append | |
11 | and O(changes) merge between branches. |
|
11 | and O(changes) merge between branches. | |
12 | """ |
|
12 | """ | |
13 |
|
13 | |||
14 | from __future__ import absolute_import |
|
14 | from __future__ import absolute_import | |
15 |
|
15 | |||
16 | import collections |
|
16 | import collections | |
17 | import errno |
|
17 | import errno | |
18 | import hashlib |
|
18 | import hashlib | |
19 | import os |
|
19 | import os | |
20 | import struct |
|
20 | import struct | |
21 | import zlib |
|
21 | import zlib | |
22 |
|
22 | |||
23 | # import stuff from node for others to import from revlog |
|
23 | # import stuff from node for others to import from revlog | |
24 | from .node import ( |
|
24 | from .node import ( | |
25 | bin, |
|
25 | bin, | |
26 | hex, |
|
26 | hex, | |
27 | nullid, |
|
27 | nullid, | |
28 | nullrev, |
|
28 | nullrev, | |
29 | ) |
|
29 | ) | |
30 | from .i18n import _ |
|
30 | from .i18n import _ | |
31 | from . import ( |
|
31 | from . import ( | |
32 | ancestor, |
|
32 | ancestor, | |
33 | error, |
|
33 | error, | |
34 | mdiff, |
|
34 | mdiff, | |
35 | policy, |
|
35 | policy, | |
36 | pycompat, |
|
36 | pycompat, | |
37 | templatefilters, |
|
37 | templatefilters, | |
38 | util, |
|
38 | util, | |
39 | ) |
|
39 | ) | |
40 |
|
40 | |||
41 | parsers = policy.importmod(r'parsers') |
|
41 | parsers = policy.importmod(r'parsers') | |
42 |
|
42 | |||
43 | _pack = struct.pack |
|
43 | _pack = struct.pack | |
44 | _unpack = struct.unpack |
|
44 | _unpack = struct.unpack | |
45 | # Aliased for performance. |
|
45 | # Aliased for performance. | |
46 | _zlibdecompress = zlib.decompress |
|
46 | _zlibdecompress = zlib.decompress | |
47 |
|
47 | |||
48 | # revlog header flags |
|
48 | # revlog header flags | |
49 | REVLOGV0 = 0 |
|
49 | REVLOGV0 = 0 | |
50 | REVLOGV1 = 1 |
|
50 | REVLOGV1 = 1 | |
51 | FLAG_INLINE_DATA = (1 << 16) |
|
51 | FLAG_INLINE_DATA = (1 << 16) | |
52 | FLAG_GENERALDELTA = (1 << 17) |
|
52 | FLAG_GENERALDELTA = (1 << 17) | |
53 | REVLOG_DEFAULT_FLAGS = FLAG_INLINE_DATA |
|
53 | REVLOG_DEFAULT_FLAGS = FLAG_INLINE_DATA | |
54 | REVLOG_DEFAULT_FORMAT = REVLOGV1 |
|
54 | REVLOG_DEFAULT_FORMAT = REVLOGV1 | |
55 | REVLOG_DEFAULT_VERSION = REVLOG_DEFAULT_FORMAT | REVLOG_DEFAULT_FLAGS |
|
55 | REVLOG_DEFAULT_VERSION = REVLOG_DEFAULT_FORMAT | REVLOG_DEFAULT_FLAGS | |
56 | REVLOGV1_FLAGS = FLAG_INLINE_DATA | FLAG_GENERALDELTA |
|
56 | REVLOGV1_FLAGS = FLAG_INLINE_DATA | FLAG_GENERALDELTA | |
57 |
|
57 | |||
58 | # revlog index flags |
|
58 | # revlog index flags | |
59 | REVIDX_ISCENSORED = (1 << 15) # revision has censor metadata, must be verified |
|
59 | REVIDX_ISCENSORED = (1 << 15) # revision has censor metadata, must be verified | |
60 | REVIDX_ELLIPSIS = (1 << 14) # revision hash does not match data (narrowhg) |
|
60 | REVIDX_ELLIPSIS = (1 << 14) # revision hash does not match data (narrowhg) | |
61 | REVIDX_EXTSTORED = (1 << 13) # revision data is stored externally |
|
61 | REVIDX_EXTSTORED = (1 << 13) # revision data is stored externally | |
62 | REVIDX_DEFAULT_FLAGS = 0 |
|
62 | REVIDX_DEFAULT_FLAGS = 0 | |
63 | # stable order in which flags need to be processed and their processors applied |
|
63 | # stable order in which flags need to be processed and their processors applied | |
64 | REVIDX_FLAGS_ORDER = [ |
|
64 | REVIDX_FLAGS_ORDER = [ | |
65 | REVIDX_ISCENSORED, |
|
65 | REVIDX_ISCENSORED, | |
66 | REVIDX_ELLIPSIS, |
|
66 | REVIDX_ELLIPSIS, | |
67 | REVIDX_EXTSTORED, |
|
67 | REVIDX_EXTSTORED, | |
68 | ] |
|
68 | ] | |
69 | REVIDX_KNOWN_FLAGS = util.bitsfrom(REVIDX_FLAGS_ORDER) |
|
69 | REVIDX_KNOWN_FLAGS = util.bitsfrom(REVIDX_FLAGS_ORDER) | |
70 |
|
70 | |||
71 | # max size of revlog with inline data |
|
71 | # max size of revlog with inline data | |
72 | _maxinline = 131072 |
|
72 | _maxinline = 131072 | |
73 | _chunksize = 1048576 |
|
73 | _chunksize = 1048576 | |
74 |
|
74 | |||
75 | RevlogError = error.RevlogError |
|
75 | RevlogError = error.RevlogError | |
76 | LookupError = error.LookupError |
|
76 | LookupError = error.LookupError | |
77 | CensoredNodeError = error.CensoredNodeError |
|
77 | CensoredNodeError = error.CensoredNodeError | |
78 | ProgrammingError = error.ProgrammingError |
|
78 | ProgrammingError = error.ProgrammingError | |
79 |
|
79 | |||
80 | # Store flag processors (cf. 'addflagprocessor()' to register) |
|
80 | # Store flag processors (cf. 'addflagprocessor()' to register) | |
81 | _flagprocessors = { |
|
81 | _flagprocessors = { | |
82 | REVIDX_ISCENSORED: None, |
|
82 | REVIDX_ISCENSORED: None, | |
83 | } |
|
83 | } | |
84 |
|
84 | |||
85 | def addflagprocessor(flag, processor): |
|
85 | def addflagprocessor(flag, processor): | |
86 | """Register a flag processor on a revision data flag. |
|
86 | """Register a flag processor on a revision data flag. | |
87 |
|
87 | |||
88 | Invariant: |
|
88 | Invariant: | |
89 | - Flags need to be defined in REVIDX_KNOWN_FLAGS and REVIDX_FLAGS_ORDER. |
|
89 | - Flags need to be defined in REVIDX_KNOWN_FLAGS and REVIDX_FLAGS_ORDER. | |
90 | - Only one flag processor can be registered on a specific flag. |
|
90 | - Only one flag processor can be registered on a specific flag. | |
91 | - flagprocessors must be 3-tuples of functions (read, write, raw) with the |
|
91 | - flagprocessors must be 3-tuples of functions (read, write, raw) with the | |
92 | following signatures: |
|
92 | following signatures: | |
93 | - (read) f(self, rawtext) -> text, bool |
|
93 | - (read) f(self, rawtext) -> text, bool | |
94 | - (write) f(self, text) -> rawtext, bool |
|
94 | - (write) f(self, text) -> rawtext, bool | |
95 | - (raw) f(self, rawtext) -> bool |
|
95 | - (raw) f(self, rawtext) -> bool | |
96 | "text" is presented to the user. "rawtext" is stored in revlog data, not |
|
96 | "text" is presented to the user. "rawtext" is stored in revlog data, not | |
97 | directly visible to the user. |
|
97 | directly visible to the user. | |
98 | The boolean returned by these transforms is used to determine whether |
|
98 | The boolean returned by these transforms is used to determine whether | |
99 | the returned text can be used for hash integrity checking. For example, |
|
99 | the returned text can be used for hash integrity checking. For example, | |
100 | if "write" returns False, then "text" is used to generate hash. If |
|
100 | if "write" returns False, then "text" is used to generate hash. If | |
101 | "write" returns True, that basically means "rawtext" returned by "write" |
|
101 | "write" returns True, that basically means "rawtext" returned by "write" | |
102 | should be used to generate hash. Usually, "write" and "read" return |
|
102 | should be used to generate hash. Usually, "write" and "read" return | |
103 | different booleans. And "raw" returns a same boolean as "write". |
|
103 | different booleans. And "raw" returns a same boolean as "write". | |
104 |
|
104 | |||
105 | Note: The 'raw' transform is used for changegroup generation and in some |
|
105 | Note: The 'raw' transform is used for changegroup generation and in some | |
106 | debug commands. In this case the transform only indicates whether the |
|
106 | debug commands. In this case the transform only indicates whether the | |
107 | contents can be used for hash integrity checks. |
|
107 | contents can be used for hash integrity checks. | |
108 | """ |
|
108 | """ | |
109 | if not flag & REVIDX_KNOWN_FLAGS: |
|
109 | if not flag & REVIDX_KNOWN_FLAGS: | |
110 | msg = _("cannot register processor on unknown flag '%#x'.") % (flag) |
|
110 | msg = _("cannot register processor on unknown flag '%#x'.") % (flag) | |
111 | raise ProgrammingError(msg) |
|
111 | raise ProgrammingError(msg) | |
112 | if flag not in REVIDX_FLAGS_ORDER: |
|
112 | if flag not in REVIDX_FLAGS_ORDER: | |
113 | msg = _("flag '%#x' undefined in REVIDX_FLAGS_ORDER.") % (flag) |
|
113 | msg = _("flag '%#x' undefined in REVIDX_FLAGS_ORDER.") % (flag) | |
114 | raise ProgrammingError(msg) |
|
114 | raise ProgrammingError(msg) | |
115 | if flag in _flagprocessors: |
|
115 | if flag in _flagprocessors: | |
116 | msg = _("cannot register multiple processors on flag '%#x'.") % (flag) |
|
116 | msg = _("cannot register multiple processors on flag '%#x'.") % (flag) | |
117 | raise error.Abort(msg) |
|
117 | raise error.Abort(msg) | |
118 | _flagprocessors[flag] = processor |
|
118 | _flagprocessors[flag] = processor | |
119 |
|
119 | |||
120 | def getoffset(q): |
|
120 | def getoffset(q): | |
121 | return int(q >> 16) |
|
121 | return int(q >> 16) | |
122 |
|
122 | |||
123 | def gettype(q): |
|
123 | def gettype(q): | |
124 | return int(q & 0xFFFF) |
|
124 | return int(q & 0xFFFF) | |
125 |
|
125 | |||
126 | def offset_type(offset, type): |
|
126 | def offset_type(offset, type): | |
127 | if (type & ~REVIDX_KNOWN_FLAGS) != 0: |
|
127 | if (type & ~REVIDX_KNOWN_FLAGS) != 0: | |
128 | raise ValueError('unknown revlog index flags') |
|
128 | raise ValueError('unknown revlog index flags') | |
129 | return int(int(offset) << 16 | type) |
|
129 | return int(int(offset) << 16 | type) | |
130 |
|
130 | |||
131 | _nullhash = hashlib.sha1(nullid) |
|
131 | _nullhash = hashlib.sha1(nullid) | |
132 |
|
132 | |||
133 | def hash(text, p1, p2): |
|
133 | def hash(text, p1, p2): | |
134 | """generate a hash from the given text and its parent hashes |
|
134 | """generate a hash from the given text and its parent hashes | |
135 |
|
135 | |||
136 | This hash combines both the current file contents and its history |
|
136 | This hash combines both the current file contents and its history | |
137 | in a manner that makes it easy to distinguish nodes with the same |
|
137 | in a manner that makes it easy to distinguish nodes with the same | |
138 | content in the revision graph. |
|
138 | content in the revision graph. | |
139 | """ |
|
139 | """ | |
140 | # As of now, if one of the parent node is null, p2 is null |
|
140 | # As of now, if one of the parent node is null, p2 is null | |
141 | if p2 == nullid: |
|
141 | if p2 == nullid: | |
142 | # deep copy of a hash is faster than creating one |
|
142 | # deep copy of a hash is faster than creating one | |
143 | s = _nullhash.copy() |
|
143 | s = _nullhash.copy() | |
144 | s.update(p1) |
|
144 | s.update(p1) | |
145 | else: |
|
145 | else: | |
146 | # none of the parent nodes are nullid |
|
146 | # none of the parent nodes are nullid | |
147 | l = [p1, p2] |
|
147 | l = [p1, p2] | |
148 | l.sort() |
|
148 | l.sort() | |
149 | s = hashlib.sha1(l[0]) |
|
149 | s = hashlib.sha1(l[0]) | |
150 | s.update(l[1]) |
|
150 | s.update(l[1]) | |
151 | s.update(text) |
|
151 | s.update(text) | |
152 | return s.digest() |
|
152 | return s.digest() | |
153 |
|
153 | |||
154 | # index v0: |
|
154 | # index v0: | |
155 | # 4 bytes: offset |
|
155 | # 4 bytes: offset | |
156 | # 4 bytes: compressed length |
|
156 | # 4 bytes: compressed length | |
157 | # 4 bytes: base rev |
|
157 | # 4 bytes: base rev | |
158 | # 4 bytes: link rev |
|
158 | # 4 bytes: link rev | |
159 | # 20 bytes: parent 1 nodeid |
|
159 | # 20 bytes: parent 1 nodeid | |
160 | # 20 bytes: parent 2 nodeid |
|
160 | # 20 bytes: parent 2 nodeid | |
161 | # 20 bytes: nodeid |
|
161 | # 20 bytes: nodeid | |
162 | indexformatv0 = ">4l20s20s20s" |
|
162 | indexformatv0 = ">4l20s20s20s" | |
163 |
|
163 | |||
164 | class revlogoldio(object): |
|
164 | class revlogoldio(object): | |
165 | def __init__(self): |
|
165 | def __init__(self): | |
166 | self.size = struct.calcsize(indexformatv0) |
|
166 | self.size = struct.calcsize(indexformatv0) | |
167 |
|
167 | |||
168 | def parseindex(self, data, inline): |
|
168 | def parseindex(self, data, inline): | |
169 | s = self.size |
|
169 | s = self.size | |
170 | index = [] |
|
170 | index = [] | |
171 | nodemap = {nullid: nullrev} |
|
171 | nodemap = {nullid: nullrev} | |
172 | n = off = 0 |
|
172 | n = off = 0 | |
173 | l = len(data) |
|
173 | l = len(data) | |
174 | while off + s <= l: |
|
174 | while off + s <= l: | |
175 | cur = data[off:off + s] |
|
175 | cur = data[off:off + s] | |
176 | off += s |
|
176 | off += s | |
177 | e = _unpack(indexformatv0, cur) |
|
177 | e = _unpack(indexformatv0, cur) | |
178 | # transform to revlogv1 format |
|
178 | # transform to revlogv1 format | |
179 | e2 = (offset_type(e[0], 0), e[1], -1, e[2], e[3], |
|
179 | e2 = (offset_type(e[0], 0), e[1], -1, e[2], e[3], | |
180 | nodemap.get(e[4], nullrev), nodemap.get(e[5], nullrev), e[6]) |
|
180 | nodemap.get(e[4], nullrev), nodemap.get(e[5], nullrev), e[6]) | |
181 | index.append(e2) |
|
181 | index.append(e2) | |
182 | nodemap[e[6]] = n |
|
182 | nodemap[e[6]] = n | |
183 | n += 1 |
|
183 | n += 1 | |
184 |
|
184 | |||
185 | # add the magic null revision at -1 |
|
185 | # add the magic null revision at -1 | |
186 | index.append((0, 0, 0, -1, -1, -1, -1, nullid)) |
|
186 | index.append((0, 0, 0, -1, -1, -1, -1, nullid)) | |
187 |
|
187 | |||
188 | return index, nodemap, None |
|
188 | return index, nodemap, None | |
189 |
|
189 | |||
190 | def packentry(self, entry, node, version, rev): |
|
190 | def packentry(self, entry, node, version, rev): | |
191 | if gettype(entry[0]): |
|
191 | if gettype(entry[0]): | |
192 | raise RevlogError(_("index entry flags need RevlogNG")) |
|
192 | raise RevlogError(_("index entry flags need RevlogNG")) | |
193 | e2 = (getoffset(entry[0]), entry[1], entry[3], entry[4], |
|
193 | e2 = (getoffset(entry[0]), entry[1], entry[3], entry[4], | |
194 | node(entry[5]), node(entry[6]), entry[7]) |
|
194 | node(entry[5]), node(entry[6]), entry[7]) | |
195 | return _pack(indexformatv0, *e2) |
|
195 | return _pack(indexformatv0, *e2) | |
196 |
|
196 | |||
197 | # index ng: |
|
197 | # index ng: | |
198 | # 6 bytes: offset |
|
198 | # 6 bytes: offset | |
199 | # 2 bytes: flags |
|
199 | # 2 bytes: flags | |
200 | # 4 bytes: compressed length |
|
200 | # 4 bytes: compressed length | |
201 | # 4 bytes: uncompressed length |
|
201 | # 4 bytes: uncompressed length | |
202 | # 4 bytes: base rev |
|
202 | # 4 bytes: base rev | |
203 | # 4 bytes: link rev |
|
203 | # 4 bytes: link rev | |
204 | # 4 bytes: parent 1 rev |
|
204 | # 4 bytes: parent 1 rev | |
205 | # 4 bytes: parent 2 rev |
|
205 | # 4 bytes: parent 2 rev | |
206 | # 32 bytes: nodeid |
|
206 | # 32 bytes: nodeid | |
207 | indexformatng = ">Qiiiiii20s12x" |
|
207 | indexformatng = ">Qiiiiii20s12x" | |
208 | versionformat = ">I" |
|
208 | versionformat = ">I" | |
209 |
|
209 | |||
210 | # corresponds to uncompressed length of indexformatng (2 gigs, 4-byte |
|
210 | # corresponds to uncompressed length of indexformatng (2 gigs, 4-byte | |
211 | # signed integer) |
|
211 | # signed integer) | |
212 | _maxentrysize = 0x7fffffff |
|
212 | _maxentrysize = 0x7fffffff | |
213 |
|
213 | |||
214 | class revlogio(object): |
|
214 | class revlogio(object): | |
215 | def __init__(self): |
|
215 | def __init__(self): | |
216 | self.size = struct.calcsize(indexformatng) |
|
216 | self.size = struct.calcsize(indexformatng) | |
217 |
|
217 | |||
218 | def parseindex(self, data, inline): |
|
218 | def parseindex(self, data, inline): | |
219 | # call the C implementation to parse the index data |
|
219 | # call the C implementation to parse the index data | |
220 | index, cache = parsers.parse_index2(data, inline) |
|
220 | index, cache = parsers.parse_index2(data, inline) | |
221 | return index, getattr(index, 'nodemap', None), cache |
|
221 | return index, getattr(index, 'nodemap', None), cache | |
222 |
|
222 | |||
223 | def packentry(self, entry, node, version, rev): |
|
223 | def packentry(self, entry, node, version, rev): | |
224 | p = _pack(indexformatng, *entry) |
|
224 | p = _pack(indexformatng, *entry) | |
225 | if rev == 0: |
|
225 | if rev == 0: | |
226 | p = _pack(versionformat, version) + p[4:] |
|
226 | p = _pack(versionformat, version) + p[4:] | |
227 | return p |
|
227 | return p | |
228 |
|
228 | |||
229 | class revlog(object): |
|
229 | class revlog(object): | |
230 | """ |
|
230 | """ | |
231 | the underlying revision storage object |
|
231 | the underlying revision storage object | |
232 |
|
232 | |||
233 | A revlog consists of two parts, an index and the revision data. |
|
233 | A revlog consists of two parts, an index and the revision data. | |
234 |
|
234 | |||
235 | The index is a file with a fixed record size containing |
|
235 | The index is a file with a fixed record size containing | |
236 | information on each revision, including its nodeid (hash), the |
|
236 | information on each revision, including its nodeid (hash), the | |
237 | nodeids of its parents, the position and offset of its data within |
|
237 | nodeids of its parents, the position and offset of its data within | |
238 | the data file, and the revision it's based on. Finally, each entry |
|
238 | the data file, and the revision it's based on. Finally, each entry | |
239 | contains a linkrev entry that can serve as a pointer to external |
|
239 | contains a linkrev entry that can serve as a pointer to external | |
240 | data. |
|
240 | data. | |
241 |
|
241 | |||
242 | The revision data itself is a linear collection of data chunks. |
|
242 | The revision data itself is a linear collection of data chunks. | |
243 | Each chunk represents a revision and is usually represented as a |
|
243 | Each chunk represents a revision and is usually represented as a | |
244 | delta against the previous chunk. To bound lookup time, runs of |
|
244 | delta against the previous chunk. To bound lookup time, runs of | |
245 | deltas are limited to about 2 times the length of the original |
|
245 | deltas are limited to about 2 times the length of the original | |
246 | version data. This makes retrieval of a version proportional to |
|
246 | version data. This makes retrieval of a version proportional to | |
247 | its size, or O(1) relative to the number of revisions. |
|
247 | its size, or O(1) relative to the number of revisions. | |
248 |
|
248 | |||
249 | Both pieces of the revlog are written to in an append-only |
|
249 | Both pieces of the revlog are written to in an append-only | |
250 | fashion, which means we never need to rewrite a file to insert or |
|
250 | fashion, which means we never need to rewrite a file to insert or | |
251 | remove data, and can use some simple techniques to avoid the need |
|
251 | remove data, and can use some simple techniques to avoid the need | |
252 | for locking while reading. |
|
252 | for locking while reading. | |
253 |
|
253 | |||
254 | If checkambig, indexfile is opened with checkambig=True at |
|
254 | If checkambig, indexfile is opened with checkambig=True at | |
255 | writing, to avoid file stat ambiguity. |
|
255 | writing, to avoid file stat ambiguity. | |
256 | """ |
|
256 | """ | |
257 | def __init__(self, opener, indexfile, datafile=None, checkambig=False): |
|
257 | def __init__(self, opener, indexfile, datafile=None, checkambig=False): | |
258 | """ |
|
258 | """ | |
259 | create a revlog object |
|
259 | create a revlog object | |
260 |
|
260 | |||
261 | opener is a function that abstracts the file opening operation |
|
261 | opener is a function that abstracts the file opening operation | |
262 | and can be used to implement COW semantics or the like. |
|
262 | and can be used to implement COW semantics or the like. | |
263 | """ |
|
263 | """ | |
264 | self.indexfile = indexfile |
|
264 | self.indexfile = indexfile | |
265 | self.datafile = datafile or (indexfile[:-2] + ".d") |
|
265 | self.datafile = datafile or (indexfile[:-2] + ".d") | |
266 | self.opener = opener |
|
266 | self.opener = opener | |
267 | # When True, indexfile is opened with checkambig=True at writing, to |
|
267 | # When True, indexfile is opened with checkambig=True at writing, to | |
268 | # avoid file stat ambiguity. |
|
268 | # avoid file stat ambiguity. | |
269 | self._checkambig = checkambig |
|
269 | self._checkambig = checkambig | |
270 | # 3-tuple of (node, rev, text) for a raw revision. |
|
270 | # 3-tuple of (node, rev, text) for a raw revision. | |
271 | self._cache = None |
|
271 | self._cache = None | |
272 | # Maps rev to chain base rev. |
|
272 | # Maps rev to chain base rev. | |
273 | self._chainbasecache = util.lrucachedict(100) |
|
273 | self._chainbasecache = util.lrucachedict(100) | |
274 | # 2-tuple of (offset, data) of raw data from the revlog at an offset. |
|
274 | # 2-tuple of (offset, data) of raw data from the revlog at an offset. | |
275 | self._chunkcache = (0, '') |
|
275 | self._chunkcache = (0, '') | |
276 | # How much data to read and cache into the raw revlog data cache. |
|
276 | # How much data to read and cache into the raw revlog data cache. | |
277 | self._chunkcachesize = 65536 |
|
277 | self._chunkcachesize = 65536 | |
278 | self._maxchainlen = None |
|
278 | self._maxchainlen = None | |
279 | self._aggressivemergedeltas = False |
|
279 | self._aggressivemergedeltas = False | |
280 | self.index = [] |
|
280 | self.index = [] | |
281 | # Mapping of partial identifiers to full nodes. |
|
281 | # Mapping of partial identifiers to full nodes. | |
282 | self._pcache = {} |
|
282 | self._pcache = {} | |
283 | # Mapping of revision integer to full node. |
|
283 | # Mapping of revision integer to full node. | |
284 | self._nodecache = {nullid: nullrev} |
|
284 | self._nodecache = {nullid: nullrev} | |
285 | self._nodepos = None |
|
285 | self._nodepos = None | |
286 | self._compengine = 'zlib' |
|
286 | self._compengine = 'zlib' | |
287 |
|
287 | |||
288 | v = REVLOG_DEFAULT_VERSION |
|
288 | v = REVLOG_DEFAULT_VERSION | |
289 | opts = getattr(opener, 'options', None) |
|
289 | opts = getattr(opener, 'options', None) | |
290 | if opts is not None: |
|
290 | if opts is not None: | |
291 | if 'revlogv1' in opts: |
|
291 | if 'revlogv1' in opts: | |
292 | if 'generaldelta' in opts: |
|
292 | if 'generaldelta' in opts: | |
293 | v |= FLAG_GENERALDELTA |
|
293 | v |= FLAG_GENERALDELTA | |
294 | else: |
|
294 | else: | |
295 | v = 0 |
|
295 | v = 0 | |
296 | if 'chunkcachesize' in opts: |
|
296 | if 'chunkcachesize' in opts: | |
297 | self._chunkcachesize = opts['chunkcachesize'] |
|
297 | self._chunkcachesize = opts['chunkcachesize'] | |
298 | if 'maxchainlen' in opts: |
|
298 | if 'maxchainlen' in opts: | |
299 | self._maxchainlen = opts['maxchainlen'] |
|
299 | self._maxchainlen = opts['maxchainlen'] | |
300 | if 'aggressivemergedeltas' in opts: |
|
300 | if 'aggressivemergedeltas' in opts: | |
301 | self._aggressivemergedeltas = opts['aggressivemergedeltas'] |
|
301 | self._aggressivemergedeltas = opts['aggressivemergedeltas'] | |
302 | self._lazydeltabase = bool(opts.get('lazydeltabase', False)) |
|
302 | self._lazydeltabase = bool(opts.get('lazydeltabase', False)) | |
303 | if 'compengine' in opts: |
|
303 | if 'compengine' in opts: | |
304 | self._compengine = opts['compengine'] |
|
304 | self._compengine = opts['compengine'] | |
305 |
|
305 | |||
306 | if self._chunkcachesize <= 0: |
|
306 | if self._chunkcachesize <= 0: | |
307 | raise RevlogError(_('revlog chunk cache size %r is not greater ' |
|
307 | raise RevlogError(_('revlog chunk cache size %r is not greater ' | |
308 | 'than 0') % self._chunkcachesize) |
|
308 | 'than 0') % self._chunkcachesize) | |
309 | elif self._chunkcachesize & (self._chunkcachesize - 1): |
|
309 | elif self._chunkcachesize & (self._chunkcachesize - 1): | |
310 | raise RevlogError(_('revlog chunk cache size %r is not a power ' |
|
310 | raise RevlogError(_('revlog chunk cache size %r is not a power ' | |
311 | 'of 2') % self._chunkcachesize) |
|
311 | 'of 2') % self._chunkcachesize) | |
312 |
|
312 | |||
313 | indexdata = '' |
|
313 | indexdata = '' | |
314 | self._initempty = True |
|
314 | self._initempty = True | |
315 | try: |
|
315 | try: | |
316 | f = self.opener(self.indexfile) |
|
316 | f = self.opener(self.indexfile) | |
317 | indexdata = f.read() |
|
317 | indexdata = f.read() | |
318 | f.close() |
|
318 | f.close() | |
319 | if len(indexdata) > 0: |
|
319 | if len(indexdata) > 0: | |
320 | v = struct.unpack(versionformat, indexdata[:4])[0] |
|
320 | v = struct.unpack(versionformat, indexdata[:4])[0] | |
321 | self._initempty = False |
|
321 | self._initempty = False | |
322 | except IOError as inst: |
|
322 | except IOError as inst: | |
323 | if inst.errno != errno.ENOENT: |
|
323 | if inst.errno != errno.ENOENT: | |
324 | raise |
|
324 | raise | |
325 |
|
325 | |||
326 | self.version = v |
|
326 | self.version = v | |
327 | self._inline = v & FLAG_INLINE_DATA |
|
327 | self._inline = v & FLAG_INLINE_DATA | |
328 | self._generaldelta = v & FLAG_GENERALDELTA |
|
328 | self._generaldelta = v & FLAG_GENERALDELTA | |
329 | flags = v & ~0xFFFF |
|
329 | flags = v & ~0xFFFF | |
330 | fmt = v & 0xFFFF |
|
330 | fmt = v & 0xFFFF | |
331 |
if fmt == REVLOGV0 |
|
331 | if fmt == REVLOGV0: | |
332 | raise RevlogError(_("index %s unknown flags %#04x for format v0") |
|
332 | if flags: | |
333 | % (self.indexfile, flags >> 16)) |
|
333 | raise RevlogError(_('unknown flags (%#04x) in version %d ' | |
334 | elif fmt == REVLOGV1 and flags & ~REVLOGV1_FLAGS: |
|
334 | 'revlog %s') % | |
335 | raise RevlogError(_("index %s unknown flags %#04x for revlogng") |
|
335 | (flags >> 16, fmt, self.indexfile)) | |
336 | % (self.indexfile, flags >> 16)) |
|
336 | elif fmt == REVLOGV1: | |
337 |
|
|
337 | if flags & ~REVLOGV1_FLAGS: | |
338 |
raise RevlogError(_( |
|
338 | raise RevlogError(_('unknown flags (%#04x) in version %d ' | |
339 |
% |
|
339 | 'revlog %s') % | |
|
340 | (flags >> 16, fmt, self.indexfile)) | |||
|
341 | else: | |||
|
342 | raise RevlogError(_('unknown version (%d) in revlog %s') % | |||
|
343 | (fmt, self.indexfile)) | |||
340 |
|
344 | |||
341 | self.storedeltachains = True |
|
345 | self.storedeltachains = True | |
342 |
|
346 | |||
343 | self._io = revlogio() |
|
347 | self._io = revlogio() | |
344 | if self.version == REVLOGV0: |
|
348 | if self.version == REVLOGV0: | |
345 | self._io = revlogoldio() |
|
349 | self._io = revlogoldio() | |
346 | try: |
|
350 | try: | |
347 | d = self._io.parseindex(indexdata, self._inline) |
|
351 | d = self._io.parseindex(indexdata, self._inline) | |
348 | except (ValueError, IndexError): |
|
352 | except (ValueError, IndexError): | |
349 | raise RevlogError(_("index %s is corrupted") % (self.indexfile)) |
|
353 | raise RevlogError(_("index %s is corrupted") % (self.indexfile)) | |
350 | self.index, nodemap, self._chunkcache = d |
|
354 | self.index, nodemap, self._chunkcache = d | |
351 | if nodemap is not None: |
|
355 | if nodemap is not None: | |
352 | self.nodemap = self._nodecache = nodemap |
|
356 | self.nodemap = self._nodecache = nodemap | |
353 | if not self._chunkcache: |
|
357 | if not self._chunkcache: | |
354 | self._chunkclear() |
|
358 | self._chunkclear() | |
355 | # revnum -> (chain-length, sum-delta-length) |
|
359 | # revnum -> (chain-length, sum-delta-length) | |
356 | self._chaininfocache = {} |
|
360 | self._chaininfocache = {} | |
357 | # revlog header -> revlog compressor |
|
361 | # revlog header -> revlog compressor | |
358 | self._decompressors = {} |
|
362 | self._decompressors = {} | |
359 |
|
363 | |||
360 | @util.propertycache |
|
364 | @util.propertycache | |
361 | def _compressor(self): |
|
365 | def _compressor(self): | |
362 | return util.compengines[self._compengine].revlogcompressor() |
|
366 | return util.compengines[self._compengine].revlogcompressor() | |
363 |
|
367 | |||
364 | def tip(self): |
|
368 | def tip(self): | |
365 | return self.node(len(self.index) - 2) |
|
369 | return self.node(len(self.index) - 2) | |
366 | def __contains__(self, rev): |
|
370 | def __contains__(self, rev): | |
367 | return 0 <= rev < len(self) |
|
371 | return 0 <= rev < len(self) | |
368 | def __len__(self): |
|
372 | def __len__(self): | |
369 | return len(self.index) - 1 |
|
373 | return len(self.index) - 1 | |
370 | def __iter__(self): |
|
374 | def __iter__(self): | |
371 | return iter(xrange(len(self))) |
|
375 | return iter(xrange(len(self))) | |
372 | def revs(self, start=0, stop=None): |
|
376 | def revs(self, start=0, stop=None): | |
373 | """iterate over all rev in this revlog (from start to stop)""" |
|
377 | """iterate over all rev in this revlog (from start to stop)""" | |
374 | step = 1 |
|
378 | step = 1 | |
375 | if stop is not None: |
|
379 | if stop is not None: | |
376 | if start > stop: |
|
380 | if start > stop: | |
377 | step = -1 |
|
381 | step = -1 | |
378 | stop += step |
|
382 | stop += step | |
379 | else: |
|
383 | else: | |
380 | stop = len(self) |
|
384 | stop = len(self) | |
381 | return xrange(start, stop, step) |
|
385 | return xrange(start, stop, step) | |
382 |
|
386 | |||
383 | @util.propertycache |
|
387 | @util.propertycache | |
384 | def nodemap(self): |
|
388 | def nodemap(self): | |
385 | self.rev(self.node(0)) |
|
389 | self.rev(self.node(0)) | |
386 | return self._nodecache |
|
390 | return self._nodecache | |
387 |
|
391 | |||
388 | def hasnode(self, node): |
|
392 | def hasnode(self, node): | |
389 | try: |
|
393 | try: | |
390 | self.rev(node) |
|
394 | self.rev(node) | |
391 | return True |
|
395 | return True | |
392 | except KeyError: |
|
396 | except KeyError: | |
393 | return False |
|
397 | return False | |
394 |
|
398 | |||
395 | def clearcaches(self): |
|
399 | def clearcaches(self): | |
396 | self._cache = None |
|
400 | self._cache = None | |
397 | self._chainbasecache.clear() |
|
401 | self._chainbasecache.clear() | |
398 | self._chunkcache = (0, '') |
|
402 | self._chunkcache = (0, '') | |
399 | self._pcache = {} |
|
403 | self._pcache = {} | |
400 |
|
404 | |||
401 | try: |
|
405 | try: | |
402 | self._nodecache.clearcaches() |
|
406 | self._nodecache.clearcaches() | |
403 | except AttributeError: |
|
407 | except AttributeError: | |
404 | self._nodecache = {nullid: nullrev} |
|
408 | self._nodecache = {nullid: nullrev} | |
405 | self._nodepos = None |
|
409 | self._nodepos = None | |
406 |
|
410 | |||
407 | def rev(self, node): |
|
411 | def rev(self, node): | |
408 | try: |
|
412 | try: | |
409 | return self._nodecache[node] |
|
413 | return self._nodecache[node] | |
410 | except TypeError: |
|
414 | except TypeError: | |
411 | raise |
|
415 | raise | |
412 | except RevlogError: |
|
416 | except RevlogError: | |
413 | # parsers.c radix tree lookup failed |
|
417 | # parsers.c radix tree lookup failed | |
414 | raise LookupError(node, self.indexfile, _('no node')) |
|
418 | raise LookupError(node, self.indexfile, _('no node')) | |
415 | except KeyError: |
|
419 | except KeyError: | |
416 | # pure python cache lookup failed |
|
420 | # pure python cache lookup failed | |
417 | n = self._nodecache |
|
421 | n = self._nodecache | |
418 | i = self.index |
|
422 | i = self.index | |
419 | p = self._nodepos |
|
423 | p = self._nodepos | |
420 | if p is None: |
|
424 | if p is None: | |
421 | p = len(i) - 2 |
|
425 | p = len(i) - 2 | |
422 | for r in xrange(p, -1, -1): |
|
426 | for r in xrange(p, -1, -1): | |
423 | v = i[r][7] |
|
427 | v = i[r][7] | |
424 | n[v] = r |
|
428 | n[v] = r | |
425 | if v == node: |
|
429 | if v == node: | |
426 | self._nodepos = r - 1 |
|
430 | self._nodepos = r - 1 | |
427 | return r |
|
431 | return r | |
428 | raise LookupError(node, self.indexfile, _('no node')) |
|
432 | raise LookupError(node, self.indexfile, _('no node')) | |
429 |
|
433 | |||
430 | # Accessors for index entries. |
|
434 | # Accessors for index entries. | |
431 |
|
435 | |||
432 | # First tuple entry is 8 bytes. First 6 bytes are offset. Last 2 bytes |
|
436 | # First tuple entry is 8 bytes. First 6 bytes are offset. Last 2 bytes | |
433 | # are flags. |
|
437 | # are flags. | |
434 | def start(self, rev): |
|
438 | def start(self, rev): | |
435 | return int(self.index[rev][0] >> 16) |
|
439 | return int(self.index[rev][0] >> 16) | |
436 |
|
440 | |||
437 | def flags(self, rev): |
|
441 | def flags(self, rev): | |
438 | return self.index[rev][0] & 0xFFFF |
|
442 | return self.index[rev][0] & 0xFFFF | |
439 |
|
443 | |||
440 | def length(self, rev): |
|
444 | def length(self, rev): | |
441 | return self.index[rev][1] |
|
445 | return self.index[rev][1] | |
442 |
|
446 | |||
443 | def rawsize(self, rev): |
|
447 | def rawsize(self, rev): | |
444 | """return the length of the uncompressed text for a given revision""" |
|
448 | """return the length of the uncompressed text for a given revision""" | |
445 | l = self.index[rev][2] |
|
449 | l = self.index[rev][2] | |
446 | if l >= 0: |
|
450 | if l >= 0: | |
447 | return l |
|
451 | return l | |
448 |
|
452 | |||
449 | t = self.revision(rev, raw=True) |
|
453 | t = self.revision(rev, raw=True) | |
450 | return len(t) |
|
454 | return len(t) | |
451 |
|
455 | |||
452 | def size(self, rev): |
|
456 | def size(self, rev): | |
453 | """length of non-raw text (processed by a "read" flag processor)""" |
|
457 | """length of non-raw text (processed by a "read" flag processor)""" | |
454 | # fast path: if no "read" flag processor could change the content, |
|
458 | # fast path: if no "read" flag processor could change the content, | |
455 | # size is rawsize. note: ELLIPSIS is known to not change the content. |
|
459 | # size is rawsize. note: ELLIPSIS is known to not change the content. | |
456 | flags = self.flags(rev) |
|
460 | flags = self.flags(rev) | |
457 | if flags & (REVIDX_KNOWN_FLAGS ^ REVIDX_ELLIPSIS) == 0: |
|
461 | if flags & (REVIDX_KNOWN_FLAGS ^ REVIDX_ELLIPSIS) == 0: | |
458 | return self.rawsize(rev) |
|
462 | return self.rawsize(rev) | |
459 |
|
463 | |||
460 | return len(self.revision(rev, raw=False)) |
|
464 | return len(self.revision(rev, raw=False)) | |
461 |
|
465 | |||
462 | def chainbase(self, rev): |
|
466 | def chainbase(self, rev): | |
463 | base = self._chainbasecache.get(rev) |
|
467 | base = self._chainbasecache.get(rev) | |
464 | if base is not None: |
|
468 | if base is not None: | |
465 | return base |
|
469 | return base | |
466 |
|
470 | |||
467 | index = self.index |
|
471 | index = self.index | |
468 | base = index[rev][3] |
|
472 | base = index[rev][3] | |
469 | while base != rev: |
|
473 | while base != rev: | |
470 | rev = base |
|
474 | rev = base | |
471 | base = index[rev][3] |
|
475 | base = index[rev][3] | |
472 |
|
476 | |||
473 | self._chainbasecache[rev] = base |
|
477 | self._chainbasecache[rev] = base | |
474 | return base |
|
478 | return base | |
475 |
|
479 | |||
476 | def linkrev(self, rev): |
|
480 | def linkrev(self, rev): | |
477 | return self.index[rev][4] |
|
481 | return self.index[rev][4] | |
478 |
|
482 | |||
479 | def parentrevs(self, rev): |
|
483 | def parentrevs(self, rev): | |
480 | return self.index[rev][5:7] |
|
484 | return self.index[rev][5:7] | |
481 |
|
485 | |||
482 | def node(self, rev): |
|
486 | def node(self, rev): | |
483 | return self.index[rev][7] |
|
487 | return self.index[rev][7] | |
484 |
|
488 | |||
485 | # Derived from index values. |
|
489 | # Derived from index values. | |
486 |
|
490 | |||
487 | def end(self, rev): |
|
491 | def end(self, rev): | |
488 | return self.start(rev) + self.length(rev) |
|
492 | return self.start(rev) + self.length(rev) | |
489 |
|
493 | |||
490 | def parents(self, node): |
|
494 | def parents(self, node): | |
491 | i = self.index |
|
495 | i = self.index | |
492 | d = i[self.rev(node)] |
|
496 | d = i[self.rev(node)] | |
493 | return i[d[5]][7], i[d[6]][7] # map revisions to nodes inline |
|
497 | return i[d[5]][7], i[d[6]][7] # map revisions to nodes inline | |
494 |
|
498 | |||
495 | def chainlen(self, rev): |
|
499 | def chainlen(self, rev): | |
496 | return self._chaininfo(rev)[0] |
|
500 | return self._chaininfo(rev)[0] | |
497 |
|
501 | |||
498 | def _chaininfo(self, rev): |
|
502 | def _chaininfo(self, rev): | |
499 | chaininfocache = self._chaininfocache |
|
503 | chaininfocache = self._chaininfocache | |
500 | if rev in chaininfocache: |
|
504 | if rev in chaininfocache: | |
501 | return chaininfocache[rev] |
|
505 | return chaininfocache[rev] | |
502 | index = self.index |
|
506 | index = self.index | |
503 | generaldelta = self._generaldelta |
|
507 | generaldelta = self._generaldelta | |
504 | iterrev = rev |
|
508 | iterrev = rev | |
505 | e = index[iterrev] |
|
509 | e = index[iterrev] | |
506 | clen = 0 |
|
510 | clen = 0 | |
507 | compresseddeltalen = 0 |
|
511 | compresseddeltalen = 0 | |
508 | while iterrev != e[3]: |
|
512 | while iterrev != e[3]: | |
509 | clen += 1 |
|
513 | clen += 1 | |
510 | compresseddeltalen += e[1] |
|
514 | compresseddeltalen += e[1] | |
511 | if generaldelta: |
|
515 | if generaldelta: | |
512 | iterrev = e[3] |
|
516 | iterrev = e[3] | |
513 | else: |
|
517 | else: | |
514 | iterrev -= 1 |
|
518 | iterrev -= 1 | |
515 | if iterrev in chaininfocache: |
|
519 | if iterrev in chaininfocache: | |
516 | t = chaininfocache[iterrev] |
|
520 | t = chaininfocache[iterrev] | |
517 | clen += t[0] |
|
521 | clen += t[0] | |
518 | compresseddeltalen += t[1] |
|
522 | compresseddeltalen += t[1] | |
519 | break |
|
523 | break | |
520 | e = index[iterrev] |
|
524 | e = index[iterrev] | |
521 | else: |
|
525 | else: | |
522 | # Add text length of base since decompressing that also takes |
|
526 | # Add text length of base since decompressing that also takes | |
523 | # work. For cache hits the length is already included. |
|
527 | # work. For cache hits the length is already included. | |
524 | compresseddeltalen += e[1] |
|
528 | compresseddeltalen += e[1] | |
525 | r = (clen, compresseddeltalen) |
|
529 | r = (clen, compresseddeltalen) | |
526 | chaininfocache[rev] = r |
|
530 | chaininfocache[rev] = r | |
527 | return r |
|
531 | return r | |
528 |
|
532 | |||
529 | def _deltachain(self, rev, stoprev=None): |
|
533 | def _deltachain(self, rev, stoprev=None): | |
530 | """Obtain the delta chain for a revision. |
|
534 | """Obtain the delta chain for a revision. | |
531 |
|
535 | |||
532 | ``stoprev`` specifies a revision to stop at. If not specified, we |
|
536 | ``stoprev`` specifies a revision to stop at. If not specified, we | |
533 | stop at the base of the chain. |
|
537 | stop at the base of the chain. | |
534 |
|
538 | |||
535 | Returns a 2-tuple of (chain, stopped) where ``chain`` is a list of |
|
539 | Returns a 2-tuple of (chain, stopped) where ``chain`` is a list of | |
536 | revs in ascending order and ``stopped`` is a bool indicating whether |
|
540 | revs in ascending order and ``stopped`` is a bool indicating whether | |
537 | ``stoprev`` was hit. |
|
541 | ``stoprev`` was hit. | |
538 | """ |
|
542 | """ | |
539 | chain = [] |
|
543 | chain = [] | |
540 |
|
544 | |||
541 | # Alias to prevent attribute lookup in tight loop. |
|
545 | # Alias to prevent attribute lookup in tight loop. | |
542 | index = self.index |
|
546 | index = self.index | |
543 | generaldelta = self._generaldelta |
|
547 | generaldelta = self._generaldelta | |
544 |
|
548 | |||
545 | iterrev = rev |
|
549 | iterrev = rev | |
546 | e = index[iterrev] |
|
550 | e = index[iterrev] | |
547 | while iterrev != e[3] and iterrev != stoprev: |
|
551 | while iterrev != e[3] and iterrev != stoprev: | |
548 | chain.append(iterrev) |
|
552 | chain.append(iterrev) | |
549 | if generaldelta: |
|
553 | if generaldelta: | |
550 | iterrev = e[3] |
|
554 | iterrev = e[3] | |
551 | else: |
|
555 | else: | |
552 | iterrev -= 1 |
|
556 | iterrev -= 1 | |
553 | e = index[iterrev] |
|
557 | e = index[iterrev] | |
554 |
|
558 | |||
555 | if iterrev == stoprev: |
|
559 | if iterrev == stoprev: | |
556 | stopped = True |
|
560 | stopped = True | |
557 | else: |
|
561 | else: | |
558 | chain.append(iterrev) |
|
562 | chain.append(iterrev) | |
559 | stopped = False |
|
563 | stopped = False | |
560 |
|
564 | |||
561 | chain.reverse() |
|
565 | chain.reverse() | |
562 | return chain, stopped |
|
566 | return chain, stopped | |
563 |
|
567 | |||
564 | def ancestors(self, revs, stoprev=0, inclusive=False): |
|
568 | def ancestors(self, revs, stoprev=0, inclusive=False): | |
565 | """Generate the ancestors of 'revs' in reverse topological order. |
|
569 | """Generate the ancestors of 'revs' in reverse topological order. | |
566 | Does not generate revs lower than stoprev. |
|
570 | Does not generate revs lower than stoprev. | |
567 |
|
571 | |||
568 | See the documentation for ancestor.lazyancestors for more details.""" |
|
572 | See the documentation for ancestor.lazyancestors for more details.""" | |
569 |
|
573 | |||
570 | return ancestor.lazyancestors(self.parentrevs, revs, stoprev=stoprev, |
|
574 | return ancestor.lazyancestors(self.parentrevs, revs, stoprev=stoprev, | |
571 | inclusive=inclusive) |
|
575 | inclusive=inclusive) | |
572 |
|
576 | |||
573 | def descendants(self, revs): |
|
577 | def descendants(self, revs): | |
574 | """Generate the descendants of 'revs' in revision order. |
|
578 | """Generate the descendants of 'revs' in revision order. | |
575 |
|
579 | |||
576 | Yield a sequence of revision numbers starting with a child of |
|
580 | Yield a sequence of revision numbers starting with a child of | |
577 | some rev in revs, i.e., each revision is *not* considered a |
|
581 | some rev in revs, i.e., each revision is *not* considered a | |
578 | descendant of itself. Results are ordered by revision number (a |
|
582 | descendant of itself. Results are ordered by revision number (a | |
579 | topological sort).""" |
|
583 | topological sort).""" | |
580 | first = min(revs) |
|
584 | first = min(revs) | |
581 | if first == nullrev: |
|
585 | if first == nullrev: | |
582 | for i in self: |
|
586 | for i in self: | |
583 | yield i |
|
587 | yield i | |
584 | return |
|
588 | return | |
585 |
|
589 | |||
586 | seen = set(revs) |
|
590 | seen = set(revs) | |
587 | for i in self.revs(start=first + 1): |
|
591 | for i in self.revs(start=first + 1): | |
588 | for x in self.parentrevs(i): |
|
592 | for x in self.parentrevs(i): | |
589 | if x != nullrev and x in seen: |
|
593 | if x != nullrev and x in seen: | |
590 | seen.add(i) |
|
594 | seen.add(i) | |
591 | yield i |
|
595 | yield i | |
592 | break |
|
596 | break | |
593 |
|
597 | |||
594 | def findcommonmissing(self, common=None, heads=None): |
|
598 | def findcommonmissing(self, common=None, heads=None): | |
595 | """Return a tuple of the ancestors of common and the ancestors of heads |
|
599 | """Return a tuple of the ancestors of common and the ancestors of heads | |
596 | that are not ancestors of common. In revset terminology, we return the |
|
600 | that are not ancestors of common. In revset terminology, we return the | |
597 | tuple: |
|
601 | tuple: | |
598 |
|
602 | |||
599 | ::common, (::heads) - (::common) |
|
603 | ::common, (::heads) - (::common) | |
600 |
|
604 | |||
601 | The list is sorted by revision number, meaning it is |
|
605 | The list is sorted by revision number, meaning it is | |
602 | topologically sorted. |
|
606 | topologically sorted. | |
603 |
|
607 | |||
604 | 'heads' and 'common' are both lists of node IDs. If heads is |
|
608 | 'heads' and 'common' are both lists of node IDs. If heads is | |
605 | not supplied, uses all of the revlog's heads. If common is not |
|
609 | not supplied, uses all of the revlog's heads. If common is not | |
606 | supplied, uses nullid.""" |
|
610 | supplied, uses nullid.""" | |
607 | if common is None: |
|
611 | if common is None: | |
608 | common = [nullid] |
|
612 | common = [nullid] | |
609 | if heads is None: |
|
613 | if heads is None: | |
610 | heads = self.heads() |
|
614 | heads = self.heads() | |
611 |
|
615 | |||
612 | common = [self.rev(n) for n in common] |
|
616 | common = [self.rev(n) for n in common] | |
613 | heads = [self.rev(n) for n in heads] |
|
617 | heads = [self.rev(n) for n in heads] | |
614 |
|
618 | |||
615 | # we want the ancestors, but inclusive |
|
619 | # we want the ancestors, but inclusive | |
616 | class lazyset(object): |
|
620 | class lazyset(object): | |
617 | def __init__(self, lazyvalues): |
|
621 | def __init__(self, lazyvalues): | |
618 | self.addedvalues = set() |
|
622 | self.addedvalues = set() | |
619 | self.lazyvalues = lazyvalues |
|
623 | self.lazyvalues = lazyvalues | |
620 |
|
624 | |||
621 | def __contains__(self, value): |
|
625 | def __contains__(self, value): | |
622 | return value in self.addedvalues or value in self.lazyvalues |
|
626 | return value in self.addedvalues or value in self.lazyvalues | |
623 |
|
627 | |||
624 | def __iter__(self): |
|
628 | def __iter__(self): | |
625 | added = self.addedvalues |
|
629 | added = self.addedvalues | |
626 | for r in added: |
|
630 | for r in added: | |
627 | yield r |
|
631 | yield r | |
628 | for r in self.lazyvalues: |
|
632 | for r in self.lazyvalues: | |
629 | if not r in added: |
|
633 | if not r in added: | |
630 | yield r |
|
634 | yield r | |
631 |
|
635 | |||
632 | def add(self, value): |
|
636 | def add(self, value): | |
633 | self.addedvalues.add(value) |
|
637 | self.addedvalues.add(value) | |
634 |
|
638 | |||
635 | def update(self, values): |
|
639 | def update(self, values): | |
636 | self.addedvalues.update(values) |
|
640 | self.addedvalues.update(values) | |
637 |
|
641 | |||
638 | has = lazyset(self.ancestors(common)) |
|
642 | has = lazyset(self.ancestors(common)) | |
639 | has.add(nullrev) |
|
643 | has.add(nullrev) | |
640 | has.update(common) |
|
644 | has.update(common) | |
641 |
|
645 | |||
642 | # take all ancestors from heads that aren't in has |
|
646 | # take all ancestors from heads that aren't in has | |
643 | missing = set() |
|
647 | missing = set() | |
644 | visit = collections.deque(r for r in heads if r not in has) |
|
648 | visit = collections.deque(r for r in heads if r not in has) | |
645 | while visit: |
|
649 | while visit: | |
646 | r = visit.popleft() |
|
650 | r = visit.popleft() | |
647 | if r in missing: |
|
651 | if r in missing: | |
648 | continue |
|
652 | continue | |
649 | else: |
|
653 | else: | |
650 | missing.add(r) |
|
654 | missing.add(r) | |
651 | for p in self.parentrevs(r): |
|
655 | for p in self.parentrevs(r): | |
652 | if p not in has: |
|
656 | if p not in has: | |
653 | visit.append(p) |
|
657 | visit.append(p) | |
654 | missing = list(missing) |
|
658 | missing = list(missing) | |
655 | missing.sort() |
|
659 | missing.sort() | |
656 | return has, [self.node(miss) for miss in missing] |
|
660 | return has, [self.node(miss) for miss in missing] | |
657 |
|
661 | |||
658 | def incrementalmissingrevs(self, common=None): |
|
662 | def incrementalmissingrevs(self, common=None): | |
659 | """Return an object that can be used to incrementally compute the |
|
663 | """Return an object that can be used to incrementally compute the | |
660 | revision numbers of the ancestors of arbitrary sets that are not |
|
664 | revision numbers of the ancestors of arbitrary sets that are not | |
661 | ancestors of common. This is an ancestor.incrementalmissingancestors |
|
665 | ancestors of common. This is an ancestor.incrementalmissingancestors | |
662 | object. |
|
666 | object. | |
663 |
|
667 | |||
664 | 'common' is a list of revision numbers. If common is not supplied, uses |
|
668 | 'common' is a list of revision numbers. If common is not supplied, uses | |
665 | nullrev. |
|
669 | nullrev. | |
666 | """ |
|
670 | """ | |
667 | if common is None: |
|
671 | if common is None: | |
668 | common = [nullrev] |
|
672 | common = [nullrev] | |
669 |
|
673 | |||
670 | return ancestor.incrementalmissingancestors(self.parentrevs, common) |
|
674 | return ancestor.incrementalmissingancestors(self.parentrevs, common) | |
671 |
|
675 | |||
672 | def findmissingrevs(self, common=None, heads=None): |
|
676 | def findmissingrevs(self, common=None, heads=None): | |
673 | """Return the revision numbers of the ancestors of heads that |
|
677 | """Return the revision numbers of the ancestors of heads that | |
674 | are not ancestors of common. |
|
678 | are not ancestors of common. | |
675 |
|
679 | |||
676 | More specifically, return a list of revision numbers corresponding to |
|
680 | More specifically, return a list of revision numbers corresponding to | |
677 | nodes N such that every N satisfies the following constraints: |
|
681 | nodes N such that every N satisfies the following constraints: | |
678 |
|
682 | |||
679 | 1. N is an ancestor of some node in 'heads' |
|
683 | 1. N is an ancestor of some node in 'heads' | |
680 | 2. N is not an ancestor of any node in 'common' |
|
684 | 2. N is not an ancestor of any node in 'common' | |
681 |
|
685 | |||
682 | The list is sorted by revision number, meaning it is |
|
686 | The list is sorted by revision number, meaning it is | |
683 | topologically sorted. |
|
687 | topologically sorted. | |
684 |
|
688 | |||
685 | 'heads' and 'common' are both lists of revision numbers. If heads is |
|
689 | 'heads' and 'common' are both lists of revision numbers. If heads is | |
686 | not supplied, uses all of the revlog's heads. If common is not |
|
690 | not supplied, uses all of the revlog's heads. If common is not | |
687 | supplied, uses nullid.""" |
|
691 | supplied, uses nullid.""" | |
688 | if common is None: |
|
692 | if common is None: | |
689 | common = [nullrev] |
|
693 | common = [nullrev] | |
690 | if heads is None: |
|
694 | if heads is None: | |
691 | heads = self.headrevs() |
|
695 | heads = self.headrevs() | |
692 |
|
696 | |||
693 | inc = self.incrementalmissingrevs(common=common) |
|
697 | inc = self.incrementalmissingrevs(common=common) | |
694 | return inc.missingancestors(heads) |
|
698 | return inc.missingancestors(heads) | |
695 |
|
699 | |||
696 | def findmissing(self, common=None, heads=None): |
|
700 | def findmissing(self, common=None, heads=None): | |
697 | """Return the ancestors of heads that are not ancestors of common. |
|
701 | """Return the ancestors of heads that are not ancestors of common. | |
698 |
|
702 | |||
699 | More specifically, return a list of nodes N such that every N |
|
703 | More specifically, return a list of nodes N such that every N | |
700 | satisfies the following constraints: |
|
704 | satisfies the following constraints: | |
701 |
|
705 | |||
702 | 1. N is an ancestor of some node in 'heads' |
|
706 | 1. N is an ancestor of some node in 'heads' | |
703 | 2. N is not an ancestor of any node in 'common' |
|
707 | 2. N is not an ancestor of any node in 'common' | |
704 |
|
708 | |||
705 | The list is sorted by revision number, meaning it is |
|
709 | The list is sorted by revision number, meaning it is | |
706 | topologically sorted. |
|
710 | topologically sorted. | |
707 |
|
711 | |||
708 | 'heads' and 'common' are both lists of node IDs. If heads is |
|
712 | 'heads' and 'common' are both lists of node IDs. If heads is | |
709 | not supplied, uses all of the revlog's heads. If common is not |
|
713 | not supplied, uses all of the revlog's heads. If common is not | |
710 | supplied, uses nullid.""" |
|
714 | supplied, uses nullid.""" | |
711 | if common is None: |
|
715 | if common is None: | |
712 | common = [nullid] |
|
716 | common = [nullid] | |
713 | if heads is None: |
|
717 | if heads is None: | |
714 | heads = self.heads() |
|
718 | heads = self.heads() | |
715 |
|
719 | |||
716 | common = [self.rev(n) for n in common] |
|
720 | common = [self.rev(n) for n in common] | |
717 | heads = [self.rev(n) for n in heads] |
|
721 | heads = [self.rev(n) for n in heads] | |
718 |
|
722 | |||
719 | inc = self.incrementalmissingrevs(common=common) |
|
723 | inc = self.incrementalmissingrevs(common=common) | |
720 | return [self.node(r) for r in inc.missingancestors(heads)] |
|
724 | return [self.node(r) for r in inc.missingancestors(heads)] | |
721 |
|
725 | |||
722 | def nodesbetween(self, roots=None, heads=None): |
|
726 | def nodesbetween(self, roots=None, heads=None): | |
723 | """Return a topological path from 'roots' to 'heads'. |
|
727 | """Return a topological path from 'roots' to 'heads'. | |
724 |
|
728 | |||
725 | Return a tuple (nodes, outroots, outheads) where 'nodes' is a |
|
729 | Return a tuple (nodes, outroots, outheads) where 'nodes' is a | |
726 | topologically sorted list of all nodes N that satisfy both of |
|
730 | topologically sorted list of all nodes N that satisfy both of | |
727 | these constraints: |
|
731 | these constraints: | |
728 |
|
732 | |||
729 | 1. N is a descendant of some node in 'roots' |
|
733 | 1. N is a descendant of some node in 'roots' | |
730 | 2. N is an ancestor of some node in 'heads' |
|
734 | 2. N is an ancestor of some node in 'heads' | |
731 |
|
735 | |||
732 | Every node is considered to be both a descendant and an ancestor |
|
736 | Every node is considered to be both a descendant and an ancestor | |
733 | of itself, so every reachable node in 'roots' and 'heads' will be |
|
737 | of itself, so every reachable node in 'roots' and 'heads' will be | |
734 | included in 'nodes'. |
|
738 | included in 'nodes'. | |
735 |
|
739 | |||
736 | 'outroots' is the list of reachable nodes in 'roots', i.e., the |
|
740 | 'outroots' is the list of reachable nodes in 'roots', i.e., the | |
737 | subset of 'roots' that is returned in 'nodes'. Likewise, |
|
741 | subset of 'roots' that is returned in 'nodes'. Likewise, | |
738 | 'outheads' is the subset of 'heads' that is also in 'nodes'. |
|
742 | 'outheads' is the subset of 'heads' that is also in 'nodes'. | |
739 |
|
743 | |||
740 | 'roots' and 'heads' are both lists of node IDs. If 'roots' is |
|
744 | 'roots' and 'heads' are both lists of node IDs. If 'roots' is | |
741 | unspecified, uses nullid as the only root. If 'heads' is |
|
745 | unspecified, uses nullid as the only root. If 'heads' is | |
742 | unspecified, uses list of all of the revlog's heads.""" |
|
746 | unspecified, uses list of all of the revlog's heads.""" | |
743 | nonodes = ([], [], []) |
|
747 | nonodes = ([], [], []) | |
744 | if roots is not None: |
|
748 | if roots is not None: | |
745 | roots = list(roots) |
|
749 | roots = list(roots) | |
746 | if not roots: |
|
750 | if not roots: | |
747 | return nonodes |
|
751 | return nonodes | |
748 | lowestrev = min([self.rev(n) for n in roots]) |
|
752 | lowestrev = min([self.rev(n) for n in roots]) | |
749 | else: |
|
753 | else: | |
750 | roots = [nullid] # Everybody's a descendant of nullid |
|
754 | roots = [nullid] # Everybody's a descendant of nullid | |
751 | lowestrev = nullrev |
|
755 | lowestrev = nullrev | |
752 | if (lowestrev == nullrev) and (heads is None): |
|
756 | if (lowestrev == nullrev) and (heads is None): | |
753 | # We want _all_ the nodes! |
|
757 | # We want _all_ the nodes! | |
754 | return ([self.node(r) for r in self], [nullid], list(self.heads())) |
|
758 | return ([self.node(r) for r in self], [nullid], list(self.heads())) | |
755 | if heads is None: |
|
759 | if heads is None: | |
756 | # All nodes are ancestors, so the latest ancestor is the last |
|
760 | # All nodes are ancestors, so the latest ancestor is the last | |
757 | # node. |
|
761 | # node. | |
758 | highestrev = len(self) - 1 |
|
762 | highestrev = len(self) - 1 | |
759 | # Set ancestors to None to signal that every node is an ancestor. |
|
763 | # Set ancestors to None to signal that every node is an ancestor. | |
760 | ancestors = None |
|
764 | ancestors = None | |
761 | # Set heads to an empty dictionary for later discovery of heads |
|
765 | # Set heads to an empty dictionary for later discovery of heads | |
762 | heads = {} |
|
766 | heads = {} | |
763 | else: |
|
767 | else: | |
764 | heads = list(heads) |
|
768 | heads = list(heads) | |
765 | if not heads: |
|
769 | if not heads: | |
766 | return nonodes |
|
770 | return nonodes | |
767 | ancestors = set() |
|
771 | ancestors = set() | |
768 | # Turn heads into a dictionary so we can remove 'fake' heads. |
|
772 | # Turn heads into a dictionary so we can remove 'fake' heads. | |
769 | # Also, later we will be using it to filter out the heads we can't |
|
773 | # Also, later we will be using it to filter out the heads we can't | |
770 | # find from roots. |
|
774 | # find from roots. | |
771 | heads = dict.fromkeys(heads, False) |
|
775 | heads = dict.fromkeys(heads, False) | |
772 | # Start at the top and keep marking parents until we're done. |
|
776 | # Start at the top and keep marking parents until we're done. | |
773 | nodestotag = set(heads) |
|
777 | nodestotag = set(heads) | |
774 | # Remember where the top was so we can use it as a limit later. |
|
778 | # Remember where the top was so we can use it as a limit later. | |
775 | highestrev = max([self.rev(n) for n in nodestotag]) |
|
779 | highestrev = max([self.rev(n) for n in nodestotag]) | |
776 | while nodestotag: |
|
780 | while nodestotag: | |
777 | # grab a node to tag |
|
781 | # grab a node to tag | |
778 | n = nodestotag.pop() |
|
782 | n = nodestotag.pop() | |
779 | # Never tag nullid |
|
783 | # Never tag nullid | |
780 | if n == nullid: |
|
784 | if n == nullid: | |
781 | continue |
|
785 | continue | |
782 | # A node's revision number represents its place in a |
|
786 | # A node's revision number represents its place in a | |
783 | # topologically sorted list of nodes. |
|
787 | # topologically sorted list of nodes. | |
784 | r = self.rev(n) |
|
788 | r = self.rev(n) | |
785 | if r >= lowestrev: |
|
789 | if r >= lowestrev: | |
786 | if n not in ancestors: |
|
790 | if n not in ancestors: | |
787 | # If we are possibly a descendant of one of the roots |
|
791 | # If we are possibly a descendant of one of the roots | |
788 | # and we haven't already been marked as an ancestor |
|
792 | # and we haven't already been marked as an ancestor | |
789 | ancestors.add(n) # Mark as ancestor |
|
793 | ancestors.add(n) # Mark as ancestor | |
790 | # Add non-nullid parents to list of nodes to tag. |
|
794 | # Add non-nullid parents to list of nodes to tag. | |
791 | nodestotag.update([p for p in self.parents(n) if |
|
795 | nodestotag.update([p for p in self.parents(n) if | |
792 | p != nullid]) |
|
796 | p != nullid]) | |
793 | elif n in heads: # We've seen it before, is it a fake head? |
|
797 | elif n in heads: # We've seen it before, is it a fake head? | |
794 | # So it is, real heads should not be the ancestors of |
|
798 | # So it is, real heads should not be the ancestors of | |
795 | # any other heads. |
|
799 | # any other heads. | |
796 | heads.pop(n) |
|
800 | heads.pop(n) | |
797 | if not ancestors: |
|
801 | if not ancestors: | |
798 | return nonodes |
|
802 | return nonodes | |
799 | # Now that we have our set of ancestors, we want to remove any |
|
803 | # Now that we have our set of ancestors, we want to remove any | |
800 | # roots that are not ancestors. |
|
804 | # roots that are not ancestors. | |
801 |
|
805 | |||
802 | # If one of the roots was nullid, everything is included anyway. |
|
806 | # If one of the roots was nullid, everything is included anyway. | |
803 | if lowestrev > nullrev: |
|
807 | if lowestrev > nullrev: | |
804 | # But, since we weren't, let's recompute the lowest rev to not |
|
808 | # But, since we weren't, let's recompute the lowest rev to not | |
805 | # include roots that aren't ancestors. |
|
809 | # include roots that aren't ancestors. | |
806 |
|
810 | |||
807 | # Filter out roots that aren't ancestors of heads |
|
811 | # Filter out roots that aren't ancestors of heads | |
808 | roots = [root for root in roots if root in ancestors] |
|
812 | roots = [root for root in roots if root in ancestors] | |
809 | # Recompute the lowest revision |
|
813 | # Recompute the lowest revision | |
810 | if roots: |
|
814 | if roots: | |
811 | lowestrev = min([self.rev(root) for root in roots]) |
|
815 | lowestrev = min([self.rev(root) for root in roots]) | |
812 | else: |
|
816 | else: | |
813 | # No more roots? Return empty list |
|
817 | # No more roots? Return empty list | |
814 | return nonodes |
|
818 | return nonodes | |
815 | else: |
|
819 | else: | |
816 | # We are descending from nullid, and don't need to care about |
|
820 | # We are descending from nullid, and don't need to care about | |
817 | # any other roots. |
|
821 | # any other roots. | |
818 | lowestrev = nullrev |
|
822 | lowestrev = nullrev | |
819 | roots = [nullid] |
|
823 | roots = [nullid] | |
820 | # Transform our roots list into a set. |
|
824 | # Transform our roots list into a set. | |
821 | descendants = set(roots) |
|
825 | descendants = set(roots) | |
822 | # Also, keep the original roots so we can filter out roots that aren't |
|
826 | # Also, keep the original roots so we can filter out roots that aren't | |
823 | # 'real' roots (i.e. are descended from other roots). |
|
827 | # 'real' roots (i.e. are descended from other roots). | |
824 | roots = descendants.copy() |
|
828 | roots = descendants.copy() | |
825 | # Our topologically sorted list of output nodes. |
|
829 | # Our topologically sorted list of output nodes. | |
826 | orderedout = [] |
|
830 | orderedout = [] | |
827 | # Don't start at nullid since we don't want nullid in our output list, |
|
831 | # Don't start at nullid since we don't want nullid in our output list, | |
828 | # and if nullid shows up in descendants, empty parents will look like |
|
832 | # and if nullid shows up in descendants, empty parents will look like | |
829 | # they're descendants. |
|
833 | # they're descendants. | |
830 | for r in self.revs(start=max(lowestrev, 0), stop=highestrev + 1): |
|
834 | for r in self.revs(start=max(lowestrev, 0), stop=highestrev + 1): | |
831 | n = self.node(r) |
|
835 | n = self.node(r) | |
832 | isdescendant = False |
|
836 | isdescendant = False | |
833 | if lowestrev == nullrev: # Everybody is a descendant of nullid |
|
837 | if lowestrev == nullrev: # Everybody is a descendant of nullid | |
834 | isdescendant = True |
|
838 | isdescendant = True | |
835 | elif n in descendants: |
|
839 | elif n in descendants: | |
836 | # n is already a descendant |
|
840 | # n is already a descendant | |
837 | isdescendant = True |
|
841 | isdescendant = True | |
838 | # This check only needs to be done here because all the roots |
|
842 | # This check only needs to be done here because all the roots | |
839 | # will start being marked is descendants before the loop. |
|
843 | # will start being marked is descendants before the loop. | |
840 | if n in roots: |
|
844 | if n in roots: | |
841 | # If n was a root, check if it's a 'real' root. |
|
845 | # If n was a root, check if it's a 'real' root. | |
842 | p = tuple(self.parents(n)) |
|
846 | p = tuple(self.parents(n)) | |
843 | # If any of its parents are descendants, it's not a root. |
|
847 | # If any of its parents are descendants, it's not a root. | |
844 | if (p[0] in descendants) or (p[1] in descendants): |
|
848 | if (p[0] in descendants) or (p[1] in descendants): | |
845 | roots.remove(n) |
|
849 | roots.remove(n) | |
846 | else: |
|
850 | else: | |
847 | p = tuple(self.parents(n)) |
|
851 | p = tuple(self.parents(n)) | |
848 | # A node is a descendant if either of its parents are |
|
852 | # A node is a descendant if either of its parents are | |
849 | # descendants. (We seeded the dependents list with the roots |
|
853 | # descendants. (We seeded the dependents list with the roots | |
850 | # up there, remember?) |
|
854 | # up there, remember?) | |
851 | if (p[0] in descendants) or (p[1] in descendants): |
|
855 | if (p[0] in descendants) or (p[1] in descendants): | |
852 | descendants.add(n) |
|
856 | descendants.add(n) | |
853 | isdescendant = True |
|
857 | isdescendant = True | |
854 | if isdescendant and ((ancestors is None) or (n in ancestors)): |
|
858 | if isdescendant and ((ancestors is None) or (n in ancestors)): | |
855 | # Only include nodes that are both descendants and ancestors. |
|
859 | # Only include nodes that are both descendants and ancestors. | |
856 | orderedout.append(n) |
|
860 | orderedout.append(n) | |
857 | if (ancestors is not None) and (n in heads): |
|
861 | if (ancestors is not None) and (n in heads): | |
858 | # We're trying to figure out which heads are reachable |
|
862 | # We're trying to figure out which heads are reachable | |
859 | # from roots. |
|
863 | # from roots. | |
860 | # Mark this head as having been reached |
|
864 | # Mark this head as having been reached | |
861 | heads[n] = True |
|
865 | heads[n] = True | |
862 | elif ancestors is None: |
|
866 | elif ancestors is None: | |
863 | # Otherwise, we're trying to discover the heads. |
|
867 | # Otherwise, we're trying to discover the heads. | |
864 | # Assume this is a head because if it isn't, the next step |
|
868 | # Assume this is a head because if it isn't, the next step | |
865 | # will eventually remove it. |
|
869 | # will eventually remove it. | |
866 | heads[n] = True |
|
870 | heads[n] = True | |
867 | # But, obviously its parents aren't. |
|
871 | # But, obviously its parents aren't. | |
868 | for p in self.parents(n): |
|
872 | for p in self.parents(n): | |
869 | heads.pop(p, None) |
|
873 | heads.pop(p, None) | |
870 | heads = [head for head, flag in heads.iteritems() if flag] |
|
874 | heads = [head for head, flag in heads.iteritems() if flag] | |
871 | roots = list(roots) |
|
875 | roots = list(roots) | |
872 | assert orderedout |
|
876 | assert orderedout | |
873 | assert roots |
|
877 | assert roots | |
874 | assert heads |
|
878 | assert heads | |
875 | return (orderedout, roots, heads) |
|
879 | return (orderedout, roots, heads) | |
876 |
|
880 | |||
877 | def headrevs(self): |
|
881 | def headrevs(self): | |
878 | try: |
|
882 | try: | |
879 | return self.index.headrevs() |
|
883 | return self.index.headrevs() | |
880 | except AttributeError: |
|
884 | except AttributeError: | |
881 | return self._headrevs() |
|
885 | return self._headrevs() | |
882 |
|
886 | |||
883 | def computephases(self, roots): |
|
887 | def computephases(self, roots): | |
884 | return self.index.computephasesmapsets(roots) |
|
888 | return self.index.computephasesmapsets(roots) | |
885 |
|
889 | |||
886 | def _headrevs(self): |
|
890 | def _headrevs(self): | |
887 | count = len(self) |
|
891 | count = len(self) | |
888 | if not count: |
|
892 | if not count: | |
889 | return [nullrev] |
|
893 | return [nullrev] | |
890 | # we won't iter over filtered rev so nobody is a head at start |
|
894 | # we won't iter over filtered rev so nobody is a head at start | |
891 | ishead = [0] * (count + 1) |
|
895 | ishead = [0] * (count + 1) | |
892 | index = self.index |
|
896 | index = self.index | |
893 | for r in self: |
|
897 | for r in self: | |
894 | ishead[r] = 1 # I may be an head |
|
898 | ishead[r] = 1 # I may be an head | |
895 | e = index[r] |
|
899 | e = index[r] | |
896 | ishead[e[5]] = ishead[e[6]] = 0 # my parent are not |
|
900 | ishead[e[5]] = ishead[e[6]] = 0 # my parent are not | |
897 | return [r for r, val in enumerate(ishead) if val] |
|
901 | return [r for r, val in enumerate(ishead) if val] | |
898 |
|
902 | |||
899 | def heads(self, start=None, stop=None): |
|
903 | def heads(self, start=None, stop=None): | |
900 | """return the list of all nodes that have no children |
|
904 | """return the list of all nodes that have no children | |
901 |
|
905 | |||
902 | if start is specified, only heads that are descendants of |
|
906 | if start is specified, only heads that are descendants of | |
903 | start will be returned |
|
907 | start will be returned | |
904 | if stop is specified, it will consider all the revs from stop |
|
908 | if stop is specified, it will consider all the revs from stop | |
905 | as if they had no children |
|
909 | as if they had no children | |
906 | """ |
|
910 | """ | |
907 | if start is None and stop is None: |
|
911 | if start is None and stop is None: | |
908 | if not len(self): |
|
912 | if not len(self): | |
909 | return [nullid] |
|
913 | return [nullid] | |
910 | return [self.node(r) for r in self.headrevs()] |
|
914 | return [self.node(r) for r in self.headrevs()] | |
911 |
|
915 | |||
912 | if start is None: |
|
916 | if start is None: | |
913 | start = nullid |
|
917 | start = nullid | |
914 | if stop is None: |
|
918 | if stop is None: | |
915 | stop = [] |
|
919 | stop = [] | |
916 | stoprevs = set([self.rev(n) for n in stop]) |
|
920 | stoprevs = set([self.rev(n) for n in stop]) | |
917 | startrev = self.rev(start) |
|
921 | startrev = self.rev(start) | |
918 | reachable = {startrev} |
|
922 | reachable = {startrev} | |
919 | heads = {startrev} |
|
923 | heads = {startrev} | |
920 |
|
924 | |||
921 | parentrevs = self.parentrevs |
|
925 | parentrevs = self.parentrevs | |
922 | for r in self.revs(start=startrev + 1): |
|
926 | for r in self.revs(start=startrev + 1): | |
923 | for p in parentrevs(r): |
|
927 | for p in parentrevs(r): | |
924 | if p in reachable: |
|
928 | if p in reachable: | |
925 | if r not in stoprevs: |
|
929 | if r not in stoprevs: | |
926 | reachable.add(r) |
|
930 | reachable.add(r) | |
927 | heads.add(r) |
|
931 | heads.add(r) | |
928 | if p in heads and p not in stoprevs: |
|
932 | if p in heads and p not in stoprevs: | |
929 | heads.remove(p) |
|
933 | heads.remove(p) | |
930 |
|
934 | |||
931 | return [self.node(r) for r in heads] |
|
935 | return [self.node(r) for r in heads] | |
932 |
|
936 | |||
933 | def children(self, node): |
|
937 | def children(self, node): | |
934 | """find the children of a given node""" |
|
938 | """find the children of a given node""" | |
935 | c = [] |
|
939 | c = [] | |
936 | p = self.rev(node) |
|
940 | p = self.rev(node) | |
937 | for r in self.revs(start=p + 1): |
|
941 | for r in self.revs(start=p + 1): | |
938 | prevs = [pr for pr in self.parentrevs(r) if pr != nullrev] |
|
942 | prevs = [pr for pr in self.parentrevs(r) if pr != nullrev] | |
939 | if prevs: |
|
943 | if prevs: | |
940 | for pr in prevs: |
|
944 | for pr in prevs: | |
941 | if pr == p: |
|
945 | if pr == p: | |
942 | c.append(self.node(r)) |
|
946 | c.append(self.node(r)) | |
943 | elif p == nullrev: |
|
947 | elif p == nullrev: | |
944 | c.append(self.node(r)) |
|
948 | c.append(self.node(r)) | |
945 | return c |
|
949 | return c | |
946 |
|
950 | |||
947 | def descendant(self, start, end): |
|
951 | def descendant(self, start, end): | |
948 | if start == nullrev: |
|
952 | if start == nullrev: | |
949 | return True |
|
953 | return True | |
950 | for i in self.descendants([start]): |
|
954 | for i in self.descendants([start]): | |
951 | if i == end: |
|
955 | if i == end: | |
952 | return True |
|
956 | return True | |
953 | elif i > end: |
|
957 | elif i > end: | |
954 | break |
|
958 | break | |
955 | return False |
|
959 | return False | |
956 |
|
960 | |||
957 | def commonancestorsheads(self, a, b): |
|
961 | def commonancestorsheads(self, a, b): | |
958 | """calculate all the heads of the common ancestors of nodes a and b""" |
|
962 | """calculate all the heads of the common ancestors of nodes a and b""" | |
959 | a, b = self.rev(a), self.rev(b) |
|
963 | a, b = self.rev(a), self.rev(b) | |
960 | try: |
|
964 | try: | |
961 | ancs = self.index.commonancestorsheads(a, b) |
|
965 | ancs = self.index.commonancestorsheads(a, b) | |
962 | except (AttributeError, OverflowError): # C implementation failed |
|
966 | except (AttributeError, OverflowError): # C implementation failed | |
963 | ancs = ancestor.commonancestorsheads(self.parentrevs, a, b) |
|
967 | ancs = ancestor.commonancestorsheads(self.parentrevs, a, b) | |
964 | return pycompat.maplist(self.node, ancs) |
|
968 | return pycompat.maplist(self.node, ancs) | |
965 |
|
969 | |||
966 | def isancestor(self, a, b): |
|
970 | def isancestor(self, a, b): | |
967 | """return True if node a is an ancestor of node b |
|
971 | """return True if node a is an ancestor of node b | |
968 |
|
972 | |||
969 | The implementation of this is trivial but the use of |
|
973 | The implementation of this is trivial but the use of | |
970 | commonancestorsheads is not.""" |
|
974 | commonancestorsheads is not.""" | |
971 | return a in self.commonancestorsheads(a, b) |
|
975 | return a in self.commonancestorsheads(a, b) | |
972 |
|
976 | |||
973 | def ancestor(self, a, b): |
|
977 | def ancestor(self, a, b): | |
974 | """calculate the "best" common ancestor of nodes a and b""" |
|
978 | """calculate the "best" common ancestor of nodes a and b""" | |
975 |
|
979 | |||
976 | a, b = self.rev(a), self.rev(b) |
|
980 | a, b = self.rev(a), self.rev(b) | |
977 | try: |
|
981 | try: | |
978 | ancs = self.index.ancestors(a, b) |
|
982 | ancs = self.index.ancestors(a, b) | |
979 | except (AttributeError, OverflowError): |
|
983 | except (AttributeError, OverflowError): | |
980 | ancs = ancestor.ancestors(self.parentrevs, a, b) |
|
984 | ancs = ancestor.ancestors(self.parentrevs, a, b) | |
981 | if ancs: |
|
985 | if ancs: | |
982 | # choose a consistent winner when there's a tie |
|
986 | # choose a consistent winner when there's a tie | |
983 | return min(map(self.node, ancs)) |
|
987 | return min(map(self.node, ancs)) | |
984 | return nullid |
|
988 | return nullid | |
985 |
|
989 | |||
986 | def _match(self, id): |
|
990 | def _match(self, id): | |
987 | if isinstance(id, int): |
|
991 | if isinstance(id, int): | |
988 | # rev |
|
992 | # rev | |
989 | return self.node(id) |
|
993 | return self.node(id) | |
990 | if len(id) == 20: |
|
994 | if len(id) == 20: | |
991 | # possibly a binary node |
|
995 | # possibly a binary node | |
992 | # odds of a binary node being all hex in ASCII are 1 in 10**25 |
|
996 | # odds of a binary node being all hex in ASCII are 1 in 10**25 | |
993 | try: |
|
997 | try: | |
994 | node = id |
|
998 | node = id | |
995 | self.rev(node) # quick search the index |
|
999 | self.rev(node) # quick search the index | |
996 | return node |
|
1000 | return node | |
997 | except LookupError: |
|
1001 | except LookupError: | |
998 | pass # may be partial hex id |
|
1002 | pass # may be partial hex id | |
999 | try: |
|
1003 | try: | |
1000 | # str(rev) |
|
1004 | # str(rev) | |
1001 | rev = int(id) |
|
1005 | rev = int(id) | |
1002 | if str(rev) != id: |
|
1006 | if str(rev) != id: | |
1003 | raise ValueError |
|
1007 | raise ValueError | |
1004 | if rev < 0: |
|
1008 | if rev < 0: | |
1005 | rev = len(self) + rev |
|
1009 | rev = len(self) + rev | |
1006 | if rev < 0 or rev >= len(self): |
|
1010 | if rev < 0 or rev >= len(self): | |
1007 | raise ValueError |
|
1011 | raise ValueError | |
1008 | return self.node(rev) |
|
1012 | return self.node(rev) | |
1009 | except (ValueError, OverflowError): |
|
1013 | except (ValueError, OverflowError): | |
1010 | pass |
|
1014 | pass | |
1011 | if len(id) == 40: |
|
1015 | if len(id) == 40: | |
1012 | try: |
|
1016 | try: | |
1013 | # a full hex nodeid? |
|
1017 | # a full hex nodeid? | |
1014 | node = bin(id) |
|
1018 | node = bin(id) | |
1015 | self.rev(node) |
|
1019 | self.rev(node) | |
1016 | return node |
|
1020 | return node | |
1017 | except (TypeError, LookupError): |
|
1021 | except (TypeError, LookupError): | |
1018 | pass |
|
1022 | pass | |
1019 |
|
1023 | |||
1020 | def _partialmatch(self, id): |
|
1024 | def _partialmatch(self, id): | |
1021 | try: |
|
1025 | try: | |
1022 | partial = self.index.partialmatch(id) |
|
1026 | partial = self.index.partialmatch(id) | |
1023 | if partial and self.hasnode(partial): |
|
1027 | if partial and self.hasnode(partial): | |
1024 | return partial |
|
1028 | return partial | |
1025 | return None |
|
1029 | return None | |
1026 | except RevlogError: |
|
1030 | except RevlogError: | |
1027 | # parsers.c radix tree lookup gave multiple matches |
|
1031 | # parsers.c radix tree lookup gave multiple matches | |
1028 | # fast path: for unfiltered changelog, radix tree is accurate |
|
1032 | # fast path: for unfiltered changelog, radix tree is accurate | |
1029 | if not getattr(self, 'filteredrevs', None): |
|
1033 | if not getattr(self, 'filteredrevs', None): | |
1030 | raise LookupError(id, self.indexfile, |
|
1034 | raise LookupError(id, self.indexfile, | |
1031 | _('ambiguous identifier')) |
|
1035 | _('ambiguous identifier')) | |
1032 | # fall through to slow path that filters hidden revisions |
|
1036 | # fall through to slow path that filters hidden revisions | |
1033 | except (AttributeError, ValueError): |
|
1037 | except (AttributeError, ValueError): | |
1034 | # we are pure python, or key was too short to search radix tree |
|
1038 | # we are pure python, or key was too short to search radix tree | |
1035 | pass |
|
1039 | pass | |
1036 |
|
1040 | |||
1037 | if id in self._pcache: |
|
1041 | if id in self._pcache: | |
1038 | return self._pcache[id] |
|
1042 | return self._pcache[id] | |
1039 |
|
1043 | |||
1040 | if len(id) < 40: |
|
1044 | if len(id) < 40: | |
1041 | try: |
|
1045 | try: | |
1042 | # hex(node)[:...] |
|
1046 | # hex(node)[:...] | |
1043 | l = len(id) // 2 # grab an even number of digits |
|
1047 | l = len(id) // 2 # grab an even number of digits | |
1044 | prefix = bin(id[:l * 2]) |
|
1048 | prefix = bin(id[:l * 2]) | |
1045 | nl = [e[7] for e in self.index if e[7].startswith(prefix)] |
|
1049 | nl = [e[7] for e in self.index if e[7].startswith(prefix)] | |
1046 | nl = [n for n in nl if hex(n).startswith(id) and |
|
1050 | nl = [n for n in nl if hex(n).startswith(id) and | |
1047 | self.hasnode(n)] |
|
1051 | self.hasnode(n)] | |
1048 | if len(nl) > 0: |
|
1052 | if len(nl) > 0: | |
1049 | if len(nl) == 1: |
|
1053 | if len(nl) == 1: | |
1050 | self._pcache[id] = nl[0] |
|
1054 | self._pcache[id] = nl[0] | |
1051 | return nl[0] |
|
1055 | return nl[0] | |
1052 | raise LookupError(id, self.indexfile, |
|
1056 | raise LookupError(id, self.indexfile, | |
1053 | _('ambiguous identifier')) |
|
1057 | _('ambiguous identifier')) | |
1054 | return None |
|
1058 | return None | |
1055 | except TypeError: |
|
1059 | except TypeError: | |
1056 | pass |
|
1060 | pass | |
1057 |
|
1061 | |||
1058 | def lookup(self, id): |
|
1062 | def lookup(self, id): | |
1059 | """locate a node based on: |
|
1063 | """locate a node based on: | |
1060 | - revision number or str(revision number) |
|
1064 | - revision number or str(revision number) | |
1061 | - nodeid or subset of hex nodeid |
|
1065 | - nodeid or subset of hex nodeid | |
1062 | """ |
|
1066 | """ | |
1063 | n = self._match(id) |
|
1067 | n = self._match(id) | |
1064 | if n is not None: |
|
1068 | if n is not None: | |
1065 | return n |
|
1069 | return n | |
1066 | n = self._partialmatch(id) |
|
1070 | n = self._partialmatch(id) | |
1067 | if n: |
|
1071 | if n: | |
1068 | return n |
|
1072 | return n | |
1069 |
|
1073 | |||
1070 | raise LookupError(id, self.indexfile, _('no match found')) |
|
1074 | raise LookupError(id, self.indexfile, _('no match found')) | |
1071 |
|
1075 | |||
1072 | def cmp(self, node, text): |
|
1076 | def cmp(self, node, text): | |
1073 | """compare text with a given file revision |
|
1077 | """compare text with a given file revision | |
1074 |
|
1078 | |||
1075 | returns True if text is different than what is stored. |
|
1079 | returns True if text is different than what is stored. | |
1076 | """ |
|
1080 | """ | |
1077 | p1, p2 = self.parents(node) |
|
1081 | p1, p2 = self.parents(node) | |
1078 | return hash(text, p1, p2) != node |
|
1082 | return hash(text, p1, p2) != node | |
1079 |
|
1083 | |||
1080 | def _cachesegment(self, offset, data): |
|
1084 | def _cachesegment(self, offset, data): | |
1081 | """Add a segment to the revlog cache. |
|
1085 | """Add a segment to the revlog cache. | |
1082 |
|
1086 | |||
1083 | Accepts an absolute offset and the data that is at that location. |
|
1087 | Accepts an absolute offset and the data that is at that location. | |
1084 | """ |
|
1088 | """ | |
1085 | o, d = self._chunkcache |
|
1089 | o, d = self._chunkcache | |
1086 | # try to add to existing cache |
|
1090 | # try to add to existing cache | |
1087 | if o + len(d) == offset and len(d) + len(data) < _chunksize: |
|
1091 | if o + len(d) == offset and len(d) + len(data) < _chunksize: | |
1088 | self._chunkcache = o, d + data |
|
1092 | self._chunkcache = o, d + data | |
1089 | else: |
|
1093 | else: | |
1090 | self._chunkcache = offset, data |
|
1094 | self._chunkcache = offset, data | |
1091 |
|
1095 | |||
1092 | def _readsegment(self, offset, length, df=None): |
|
1096 | def _readsegment(self, offset, length, df=None): | |
1093 | """Load a segment of raw data from the revlog. |
|
1097 | """Load a segment of raw data from the revlog. | |
1094 |
|
1098 | |||
1095 | Accepts an absolute offset, length to read, and an optional existing |
|
1099 | Accepts an absolute offset, length to read, and an optional existing | |
1096 | file handle to read from. |
|
1100 | file handle to read from. | |
1097 |
|
1101 | |||
1098 | If an existing file handle is passed, it will be seeked and the |
|
1102 | If an existing file handle is passed, it will be seeked and the | |
1099 | original seek position will NOT be restored. |
|
1103 | original seek position will NOT be restored. | |
1100 |
|
1104 | |||
1101 | Returns a str or buffer of raw byte data. |
|
1105 | Returns a str or buffer of raw byte data. | |
1102 | """ |
|
1106 | """ | |
1103 | if df is not None: |
|
1107 | if df is not None: | |
1104 | closehandle = False |
|
1108 | closehandle = False | |
1105 | else: |
|
1109 | else: | |
1106 | if self._inline: |
|
1110 | if self._inline: | |
1107 | df = self.opener(self.indexfile) |
|
1111 | df = self.opener(self.indexfile) | |
1108 | else: |
|
1112 | else: | |
1109 | df = self.opener(self.datafile) |
|
1113 | df = self.opener(self.datafile) | |
1110 | closehandle = True |
|
1114 | closehandle = True | |
1111 |
|
1115 | |||
1112 | # Cache data both forward and backward around the requested |
|
1116 | # Cache data both forward and backward around the requested | |
1113 | # data, in a fixed size window. This helps speed up operations |
|
1117 | # data, in a fixed size window. This helps speed up operations | |
1114 | # involving reading the revlog backwards. |
|
1118 | # involving reading the revlog backwards. | |
1115 | cachesize = self._chunkcachesize |
|
1119 | cachesize = self._chunkcachesize | |
1116 | realoffset = offset & ~(cachesize - 1) |
|
1120 | realoffset = offset & ~(cachesize - 1) | |
1117 | reallength = (((offset + length + cachesize) & ~(cachesize - 1)) |
|
1121 | reallength = (((offset + length + cachesize) & ~(cachesize - 1)) | |
1118 | - realoffset) |
|
1122 | - realoffset) | |
1119 | df.seek(realoffset) |
|
1123 | df.seek(realoffset) | |
1120 | d = df.read(reallength) |
|
1124 | d = df.read(reallength) | |
1121 | if closehandle: |
|
1125 | if closehandle: | |
1122 | df.close() |
|
1126 | df.close() | |
1123 | self._cachesegment(realoffset, d) |
|
1127 | self._cachesegment(realoffset, d) | |
1124 | if offset != realoffset or reallength != length: |
|
1128 | if offset != realoffset or reallength != length: | |
1125 | return util.buffer(d, offset - realoffset, length) |
|
1129 | return util.buffer(d, offset - realoffset, length) | |
1126 | return d |
|
1130 | return d | |
1127 |
|
1131 | |||
1128 | def _getsegment(self, offset, length, df=None): |
|
1132 | def _getsegment(self, offset, length, df=None): | |
1129 | """Obtain a segment of raw data from the revlog. |
|
1133 | """Obtain a segment of raw data from the revlog. | |
1130 |
|
1134 | |||
1131 | Accepts an absolute offset, length of bytes to obtain, and an |
|
1135 | Accepts an absolute offset, length of bytes to obtain, and an | |
1132 | optional file handle to the already-opened revlog. If the file |
|
1136 | optional file handle to the already-opened revlog. If the file | |
1133 | handle is used, it's original seek position will not be preserved. |
|
1137 | handle is used, it's original seek position will not be preserved. | |
1134 |
|
1138 | |||
1135 | Requests for data may be returned from a cache. |
|
1139 | Requests for data may be returned from a cache. | |
1136 |
|
1140 | |||
1137 | Returns a str or a buffer instance of raw byte data. |
|
1141 | Returns a str or a buffer instance of raw byte data. | |
1138 | """ |
|
1142 | """ | |
1139 | o, d = self._chunkcache |
|
1143 | o, d = self._chunkcache | |
1140 | l = len(d) |
|
1144 | l = len(d) | |
1141 |
|
1145 | |||
1142 | # is it in the cache? |
|
1146 | # is it in the cache? | |
1143 | cachestart = offset - o |
|
1147 | cachestart = offset - o | |
1144 | cacheend = cachestart + length |
|
1148 | cacheend = cachestart + length | |
1145 | if cachestart >= 0 and cacheend <= l: |
|
1149 | if cachestart >= 0 and cacheend <= l: | |
1146 | if cachestart == 0 and cacheend == l: |
|
1150 | if cachestart == 0 and cacheend == l: | |
1147 | return d # avoid a copy |
|
1151 | return d # avoid a copy | |
1148 | return util.buffer(d, cachestart, cacheend - cachestart) |
|
1152 | return util.buffer(d, cachestart, cacheend - cachestart) | |
1149 |
|
1153 | |||
1150 | return self._readsegment(offset, length, df=df) |
|
1154 | return self._readsegment(offset, length, df=df) | |
1151 |
|
1155 | |||
1152 | def _getsegmentforrevs(self, startrev, endrev, df=None): |
|
1156 | def _getsegmentforrevs(self, startrev, endrev, df=None): | |
1153 | """Obtain a segment of raw data corresponding to a range of revisions. |
|
1157 | """Obtain a segment of raw data corresponding to a range of revisions. | |
1154 |
|
1158 | |||
1155 | Accepts the start and end revisions and an optional already-open |
|
1159 | Accepts the start and end revisions and an optional already-open | |
1156 | file handle to be used for reading. If the file handle is read, its |
|
1160 | file handle to be used for reading. If the file handle is read, its | |
1157 | seek position will not be preserved. |
|
1161 | seek position will not be preserved. | |
1158 |
|
1162 | |||
1159 | Requests for data may be satisfied by a cache. |
|
1163 | Requests for data may be satisfied by a cache. | |
1160 |
|
1164 | |||
1161 | Returns a 2-tuple of (offset, data) for the requested range of |
|
1165 | Returns a 2-tuple of (offset, data) for the requested range of | |
1162 | revisions. Offset is the integer offset from the beginning of the |
|
1166 | revisions. Offset is the integer offset from the beginning of the | |
1163 | revlog and data is a str or buffer of the raw byte data. |
|
1167 | revlog and data is a str or buffer of the raw byte data. | |
1164 |
|
1168 | |||
1165 | Callers will need to call ``self.start(rev)`` and ``self.length(rev)`` |
|
1169 | Callers will need to call ``self.start(rev)`` and ``self.length(rev)`` | |
1166 | to determine where each revision's data begins and ends. |
|
1170 | to determine where each revision's data begins and ends. | |
1167 | """ |
|
1171 | """ | |
1168 | # Inlined self.start(startrev) & self.end(endrev) for perf reasons |
|
1172 | # Inlined self.start(startrev) & self.end(endrev) for perf reasons | |
1169 | # (functions are expensive). |
|
1173 | # (functions are expensive). | |
1170 | index = self.index |
|
1174 | index = self.index | |
1171 | istart = index[startrev] |
|
1175 | istart = index[startrev] | |
1172 | start = int(istart[0] >> 16) |
|
1176 | start = int(istart[0] >> 16) | |
1173 | if startrev == endrev: |
|
1177 | if startrev == endrev: | |
1174 | end = start + istart[1] |
|
1178 | end = start + istart[1] | |
1175 | else: |
|
1179 | else: | |
1176 | iend = index[endrev] |
|
1180 | iend = index[endrev] | |
1177 | end = int(iend[0] >> 16) + iend[1] |
|
1181 | end = int(iend[0] >> 16) + iend[1] | |
1178 |
|
1182 | |||
1179 | if self._inline: |
|
1183 | if self._inline: | |
1180 | start += (startrev + 1) * self._io.size |
|
1184 | start += (startrev + 1) * self._io.size | |
1181 | end += (endrev + 1) * self._io.size |
|
1185 | end += (endrev + 1) * self._io.size | |
1182 | length = end - start |
|
1186 | length = end - start | |
1183 |
|
1187 | |||
1184 | return start, self._getsegment(start, length, df=df) |
|
1188 | return start, self._getsegment(start, length, df=df) | |
1185 |
|
1189 | |||
1186 | def _chunk(self, rev, df=None): |
|
1190 | def _chunk(self, rev, df=None): | |
1187 | """Obtain a single decompressed chunk for a revision. |
|
1191 | """Obtain a single decompressed chunk for a revision. | |
1188 |
|
1192 | |||
1189 | Accepts an integer revision and an optional already-open file handle |
|
1193 | Accepts an integer revision and an optional already-open file handle | |
1190 | to be used for reading. If used, the seek position of the file will not |
|
1194 | to be used for reading. If used, the seek position of the file will not | |
1191 | be preserved. |
|
1195 | be preserved. | |
1192 |
|
1196 | |||
1193 | Returns a str holding uncompressed data for the requested revision. |
|
1197 | Returns a str holding uncompressed data for the requested revision. | |
1194 | """ |
|
1198 | """ | |
1195 | return self.decompress(self._getsegmentforrevs(rev, rev, df=df)[1]) |
|
1199 | return self.decompress(self._getsegmentforrevs(rev, rev, df=df)[1]) | |
1196 |
|
1200 | |||
1197 | def _chunks(self, revs, df=None): |
|
1201 | def _chunks(self, revs, df=None): | |
1198 | """Obtain decompressed chunks for the specified revisions. |
|
1202 | """Obtain decompressed chunks for the specified revisions. | |
1199 |
|
1203 | |||
1200 | Accepts an iterable of numeric revisions that are assumed to be in |
|
1204 | Accepts an iterable of numeric revisions that are assumed to be in | |
1201 | ascending order. Also accepts an optional already-open file handle |
|
1205 | ascending order. Also accepts an optional already-open file handle | |
1202 | to be used for reading. If used, the seek position of the file will |
|
1206 | to be used for reading. If used, the seek position of the file will | |
1203 | not be preserved. |
|
1207 | not be preserved. | |
1204 |
|
1208 | |||
1205 | This function is similar to calling ``self._chunk()`` multiple times, |
|
1209 | This function is similar to calling ``self._chunk()`` multiple times, | |
1206 | but is faster. |
|
1210 | but is faster. | |
1207 |
|
1211 | |||
1208 | Returns a list with decompressed data for each requested revision. |
|
1212 | Returns a list with decompressed data for each requested revision. | |
1209 | """ |
|
1213 | """ | |
1210 | if not revs: |
|
1214 | if not revs: | |
1211 | return [] |
|
1215 | return [] | |
1212 | start = self.start |
|
1216 | start = self.start | |
1213 | length = self.length |
|
1217 | length = self.length | |
1214 | inline = self._inline |
|
1218 | inline = self._inline | |
1215 | iosize = self._io.size |
|
1219 | iosize = self._io.size | |
1216 | buffer = util.buffer |
|
1220 | buffer = util.buffer | |
1217 |
|
1221 | |||
1218 | l = [] |
|
1222 | l = [] | |
1219 | ladd = l.append |
|
1223 | ladd = l.append | |
1220 |
|
1224 | |||
1221 | try: |
|
1225 | try: | |
1222 | offset, data = self._getsegmentforrevs(revs[0], revs[-1], df=df) |
|
1226 | offset, data = self._getsegmentforrevs(revs[0], revs[-1], df=df) | |
1223 | except OverflowError: |
|
1227 | except OverflowError: | |
1224 | # issue4215 - we can't cache a run of chunks greater than |
|
1228 | # issue4215 - we can't cache a run of chunks greater than | |
1225 | # 2G on Windows |
|
1229 | # 2G on Windows | |
1226 | return [self._chunk(rev, df=df) for rev in revs] |
|
1230 | return [self._chunk(rev, df=df) for rev in revs] | |
1227 |
|
1231 | |||
1228 | decomp = self.decompress |
|
1232 | decomp = self.decompress | |
1229 | for rev in revs: |
|
1233 | for rev in revs: | |
1230 | chunkstart = start(rev) |
|
1234 | chunkstart = start(rev) | |
1231 | if inline: |
|
1235 | if inline: | |
1232 | chunkstart += (rev + 1) * iosize |
|
1236 | chunkstart += (rev + 1) * iosize | |
1233 | chunklength = length(rev) |
|
1237 | chunklength = length(rev) | |
1234 | ladd(decomp(buffer(data, chunkstart - offset, chunklength))) |
|
1238 | ladd(decomp(buffer(data, chunkstart - offset, chunklength))) | |
1235 |
|
1239 | |||
1236 | return l |
|
1240 | return l | |
1237 |
|
1241 | |||
1238 | def _chunkclear(self): |
|
1242 | def _chunkclear(self): | |
1239 | """Clear the raw chunk cache.""" |
|
1243 | """Clear the raw chunk cache.""" | |
1240 | self._chunkcache = (0, '') |
|
1244 | self._chunkcache = (0, '') | |
1241 |
|
1245 | |||
1242 | def deltaparent(self, rev): |
|
1246 | def deltaparent(self, rev): | |
1243 | """return deltaparent of the given revision""" |
|
1247 | """return deltaparent of the given revision""" | |
1244 | base = self.index[rev][3] |
|
1248 | base = self.index[rev][3] | |
1245 | if base == rev: |
|
1249 | if base == rev: | |
1246 | return nullrev |
|
1250 | return nullrev | |
1247 | elif self._generaldelta: |
|
1251 | elif self._generaldelta: | |
1248 | return base |
|
1252 | return base | |
1249 | else: |
|
1253 | else: | |
1250 | return rev - 1 |
|
1254 | return rev - 1 | |
1251 |
|
1255 | |||
1252 | def revdiff(self, rev1, rev2): |
|
1256 | def revdiff(self, rev1, rev2): | |
1253 | """return or calculate a delta between two revisions |
|
1257 | """return or calculate a delta between two revisions | |
1254 |
|
1258 | |||
1255 | The delta calculated is in binary form and is intended to be written to |
|
1259 | The delta calculated is in binary form and is intended to be written to | |
1256 | revlog data directly. So this function needs raw revision data. |
|
1260 | revlog data directly. So this function needs raw revision data. | |
1257 | """ |
|
1261 | """ | |
1258 | if rev1 != nullrev and self.deltaparent(rev2) == rev1: |
|
1262 | if rev1 != nullrev and self.deltaparent(rev2) == rev1: | |
1259 | return bytes(self._chunk(rev2)) |
|
1263 | return bytes(self._chunk(rev2)) | |
1260 |
|
1264 | |||
1261 | return mdiff.textdiff(self.revision(rev1, raw=True), |
|
1265 | return mdiff.textdiff(self.revision(rev1, raw=True), | |
1262 | self.revision(rev2, raw=True)) |
|
1266 | self.revision(rev2, raw=True)) | |
1263 |
|
1267 | |||
1264 | def revision(self, nodeorrev, _df=None, raw=False): |
|
1268 | def revision(self, nodeorrev, _df=None, raw=False): | |
1265 | """return an uncompressed revision of a given node or revision |
|
1269 | """return an uncompressed revision of a given node or revision | |
1266 | number. |
|
1270 | number. | |
1267 |
|
1271 | |||
1268 | _df - an existing file handle to read from. (internal-only) |
|
1272 | _df - an existing file handle to read from. (internal-only) | |
1269 | raw - an optional argument specifying if the revision data is to be |
|
1273 | raw - an optional argument specifying if the revision data is to be | |
1270 | treated as raw data when applying flag transforms. 'raw' should be set |
|
1274 | treated as raw data when applying flag transforms. 'raw' should be set | |
1271 | to True when generating changegroups or in debug commands. |
|
1275 | to True when generating changegroups or in debug commands. | |
1272 | """ |
|
1276 | """ | |
1273 | if isinstance(nodeorrev, int): |
|
1277 | if isinstance(nodeorrev, int): | |
1274 | rev = nodeorrev |
|
1278 | rev = nodeorrev | |
1275 | node = self.node(rev) |
|
1279 | node = self.node(rev) | |
1276 | else: |
|
1280 | else: | |
1277 | node = nodeorrev |
|
1281 | node = nodeorrev | |
1278 | rev = None |
|
1282 | rev = None | |
1279 |
|
1283 | |||
1280 | cachedrev = None |
|
1284 | cachedrev = None | |
1281 | flags = None |
|
1285 | flags = None | |
1282 | rawtext = None |
|
1286 | rawtext = None | |
1283 | if node == nullid: |
|
1287 | if node == nullid: | |
1284 | return "" |
|
1288 | return "" | |
1285 | if self._cache: |
|
1289 | if self._cache: | |
1286 | if self._cache[0] == node: |
|
1290 | if self._cache[0] == node: | |
1287 | # _cache only stores rawtext |
|
1291 | # _cache only stores rawtext | |
1288 | if raw: |
|
1292 | if raw: | |
1289 | return self._cache[2] |
|
1293 | return self._cache[2] | |
1290 | # duplicated, but good for perf |
|
1294 | # duplicated, but good for perf | |
1291 | if rev is None: |
|
1295 | if rev is None: | |
1292 | rev = self.rev(node) |
|
1296 | rev = self.rev(node) | |
1293 | if flags is None: |
|
1297 | if flags is None: | |
1294 | flags = self.flags(rev) |
|
1298 | flags = self.flags(rev) | |
1295 | # no extra flags set, no flag processor runs, text = rawtext |
|
1299 | # no extra flags set, no flag processor runs, text = rawtext | |
1296 | if flags == REVIDX_DEFAULT_FLAGS: |
|
1300 | if flags == REVIDX_DEFAULT_FLAGS: | |
1297 | return self._cache[2] |
|
1301 | return self._cache[2] | |
1298 | # rawtext is reusable. need to run flag processor |
|
1302 | # rawtext is reusable. need to run flag processor | |
1299 | rawtext = self._cache[2] |
|
1303 | rawtext = self._cache[2] | |
1300 |
|
1304 | |||
1301 | cachedrev = self._cache[1] |
|
1305 | cachedrev = self._cache[1] | |
1302 |
|
1306 | |||
1303 | # look up what we need to read |
|
1307 | # look up what we need to read | |
1304 | if rawtext is None: |
|
1308 | if rawtext is None: | |
1305 | if rev is None: |
|
1309 | if rev is None: | |
1306 | rev = self.rev(node) |
|
1310 | rev = self.rev(node) | |
1307 |
|
1311 | |||
1308 | chain, stopped = self._deltachain(rev, stoprev=cachedrev) |
|
1312 | chain, stopped = self._deltachain(rev, stoprev=cachedrev) | |
1309 | if stopped: |
|
1313 | if stopped: | |
1310 | rawtext = self._cache[2] |
|
1314 | rawtext = self._cache[2] | |
1311 |
|
1315 | |||
1312 | # drop cache to save memory |
|
1316 | # drop cache to save memory | |
1313 | self._cache = None |
|
1317 | self._cache = None | |
1314 |
|
1318 | |||
1315 | bins = self._chunks(chain, df=_df) |
|
1319 | bins = self._chunks(chain, df=_df) | |
1316 | if rawtext is None: |
|
1320 | if rawtext is None: | |
1317 | rawtext = bytes(bins[0]) |
|
1321 | rawtext = bytes(bins[0]) | |
1318 | bins = bins[1:] |
|
1322 | bins = bins[1:] | |
1319 |
|
1323 | |||
1320 | rawtext = mdiff.patches(rawtext, bins) |
|
1324 | rawtext = mdiff.patches(rawtext, bins) | |
1321 | self._cache = (node, rev, rawtext) |
|
1325 | self._cache = (node, rev, rawtext) | |
1322 |
|
1326 | |||
1323 | if flags is None: |
|
1327 | if flags is None: | |
1324 | if rev is None: |
|
1328 | if rev is None: | |
1325 | rev = self.rev(node) |
|
1329 | rev = self.rev(node) | |
1326 | flags = self.flags(rev) |
|
1330 | flags = self.flags(rev) | |
1327 |
|
1331 | |||
1328 | text, validatehash = self._processflags(rawtext, flags, 'read', raw=raw) |
|
1332 | text, validatehash = self._processflags(rawtext, flags, 'read', raw=raw) | |
1329 | if validatehash: |
|
1333 | if validatehash: | |
1330 | self.checkhash(text, node, rev=rev) |
|
1334 | self.checkhash(text, node, rev=rev) | |
1331 |
|
1335 | |||
1332 | return text |
|
1336 | return text | |
1333 |
|
1337 | |||
1334 | def hash(self, text, p1, p2): |
|
1338 | def hash(self, text, p1, p2): | |
1335 | """Compute a node hash. |
|
1339 | """Compute a node hash. | |
1336 |
|
1340 | |||
1337 | Available as a function so that subclasses can replace the hash |
|
1341 | Available as a function so that subclasses can replace the hash | |
1338 | as needed. |
|
1342 | as needed. | |
1339 | """ |
|
1343 | """ | |
1340 | return hash(text, p1, p2) |
|
1344 | return hash(text, p1, p2) | |
1341 |
|
1345 | |||
1342 | def _processflags(self, text, flags, operation, raw=False): |
|
1346 | def _processflags(self, text, flags, operation, raw=False): | |
1343 | """Inspect revision data flags and applies transforms defined by |
|
1347 | """Inspect revision data flags and applies transforms defined by | |
1344 | registered flag processors. |
|
1348 | registered flag processors. | |
1345 |
|
1349 | |||
1346 | ``text`` - the revision data to process |
|
1350 | ``text`` - the revision data to process | |
1347 | ``flags`` - the revision flags |
|
1351 | ``flags`` - the revision flags | |
1348 | ``operation`` - the operation being performed (read or write) |
|
1352 | ``operation`` - the operation being performed (read or write) | |
1349 | ``raw`` - an optional argument describing if the raw transform should be |
|
1353 | ``raw`` - an optional argument describing if the raw transform should be | |
1350 | applied. |
|
1354 | applied. | |
1351 |
|
1355 | |||
1352 | This method processes the flags in the order (or reverse order if |
|
1356 | This method processes the flags in the order (or reverse order if | |
1353 | ``operation`` is 'write') defined by REVIDX_FLAGS_ORDER, applying the |
|
1357 | ``operation`` is 'write') defined by REVIDX_FLAGS_ORDER, applying the | |
1354 | flag processors registered for present flags. The order of flags defined |
|
1358 | flag processors registered for present flags. The order of flags defined | |
1355 | in REVIDX_FLAGS_ORDER needs to be stable to allow non-commutativity. |
|
1359 | in REVIDX_FLAGS_ORDER needs to be stable to allow non-commutativity. | |
1356 |
|
1360 | |||
1357 | Returns a 2-tuple of ``(text, validatehash)`` where ``text`` is the |
|
1361 | Returns a 2-tuple of ``(text, validatehash)`` where ``text`` is the | |
1358 | processed text and ``validatehash`` is a bool indicating whether the |
|
1362 | processed text and ``validatehash`` is a bool indicating whether the | |
1359 | returned text should be checked for hash integrity. |
|
1363 | returned text should be checked for hash integrity. | |
1360 |
|
1364 | |||
1361 | Note: If the ``raw`` argument is set, it has precedence over the |
|
1365 | Note: If the ``raw`` argument is set, it has precedence over the | |
1362 | operation and will only update the value of ``validatehash``. |
|
1366 | operation and will only update the value of ``validatehash``. | |
1363 | """ |
|
1367 | """ | |
1364 | # fast path: no flag processors will run |
|
1368 | # fast path: no flag processors will run | |
1365 | if flags == 0: |
|
1369 | if flags == 0: | |
1366 | return text, True |
|
1370 | return text, True | |
1367 | if not operation in ('read', 'write'): |
|
1371 | if not operation in ('read', 'write'): | |
1368 | raise ProgrammingError(_("invalid '%s' operation ") % (operation)) |
|
1372 | raise ProgrammingError(_("invalid '%s' operation ") % (operation)) | |
1369 | # Check all flags are known. |
|
1373 | # Check all flags are known. | |
1370 | if flags & ~REVIDX_KNOWN_FLAGS: |
|
1374 | if flags & ~REVIDX_KNOWN_FLAGS: | |
1371 | raise RevlogError(_("incompatible revision flag '%#x'") % |
|
1375 | raise RevlogError(_("incompatible revision flag '%#x'") % | |
1372 | (flags & ~REVIDX_KNOWN_FLAGS)) |
|
1376 | (flags & ~REVIDX_KNOWN_FLAGS)) | |
1373 | validatehash = True |
|
1377 | validatehash = True | |
1374 | # Depending on the operation (read or write), the order might be |
|
1378 | # Depending on the operation (read or write), the order might be | |
1375 | # reversed due to non-commutative transforms. |
|
1379 | # reversed due to non-commutative transforms. | |
1376 | orderedflags = REVIDX_FLAGS_ORDER |
|
1380 | orderedflags = REVIDX_FLAGS_ORDER | |
1377 | if operation == 'write': |
|
1381 | if operation == 'write': | |
1378 | orderedflags = reversed(orderedflags) |
|
1382 | orderedflags = reversed(orderedflags) | |
1379 |
|
1383 | |||
1380 | for flag in orderedflags: |
|
1384 | for flag in orderedflags: | |
1381 | # If a flagprocessor has been registered for a known flag, apply the |
|
1385 | # If a flagprocessor has been registered for a known flag, apply the | |
1382 | # related operation transform and update result tuple. |
|
1386 | # related operation transform and update result tuple. | |
1383 | if flag & flags: |
|
1387 | if flag & flags: | |
1384 | vhash = True |
|
1388 | vhash = True | |
1385 |
|
1389 | |||
1386 | if flag not in _flagprocessors: |
|
1390 | if flag not in _flagprocessors: | |
1387 | message = _("missing processor for flag '%#x'") % (flag) |
|
1391 | message = _("missing processor for flag '%#x'") % (flag) | |
1388 | raise RevlogError(message) |
|
1392 | raise RevlogError(message) | |
1389 |
|
1393 | |||
1390 | processor = _flagprocessors[flag] |
|
1394 | processor = _flagprocessors[flag] | |
1391 | if processor is not None: |
|
1395 | if processor is not None: | |
1392 | readtransform, writetransform, rawtransform = processor |
|
1396 | readtransform, writetransform, rawtransform = processor | |
1393 |
|
1397 | |||
1394 | if raw: |
|
1398 | if raw: | |
1395 | vhash = rawtransform(self, text) |
|
1399 | vhash = rawtransform(self, text) | |
1396 | elif operation == 'read': |
|
1400 | elif operation == 'read': | |
1397 | text, vhash = readtransform(self, text) |
|
1401 | text, vhash = readtransform(self, text) | |
1398 | else: # write operation |
|
1402 | else: # write operation | |
1399 | text, vhash = writetransform(self, text) |
|
1403 | text, vhash = writetransform(self, text) | |
1400 | validatehash = validatehash and vhash |
|
1404 | validatehash = validatehash and vhash | |
1401 |
|
1405 | |||
1402 | return text, validatehash |
|
1406 | return text, validatehash | |
1403 |
|
1407 | |||
1404 | def checkhash(self, text, node, p1=None, p2=None, rev=None): |
|
1408 | def checkhash(self, text, node, p1=None, p2=None, rev=None): | |
1405 | """Check node hash integrity. |
|
1409 | """Check node hash integrity. | |
1406 |
|
1410 | |||
1407 | Available as a function so that subclasses can extend hash mismatch |
|
1411 | Available as a function so that subclasses can extend hash mismatch | |
1408 | behaviors as needed. |
|
1412 | behaviors as needed. | |
1409 | """ |
|
1413 | """ | |
1410 | if p1 is None and p2 is None: |
|
1414 | if p1 is None and p2 is None: | |
1411 | p1, p2 = self.parents(node) |
|
1415 | p1, p2 = self.parents(node) | |
1412 | if node != self.hash(text, p1, p2): |
|
1416 | if node != self.hash(text, p1, p2): | |
1413 | revornode = rev |
|
1417 | revornode = rev | |
1414 | if revornode is None: |
|
1418 | if revornode is None: | |
1415 | revornode = templatefilters.short(hex(node)) |
|
1419 | revornode = templatefilters.short(hex(node)) | |
1416 | raise RevlogError(_("integrity check failed on %s:%s") |
|
1420 | raise RevlogError(_("integrity check failed on %s:%s") | |
1417 | % (self.indexfile, revornode)) |
|
1421 | % (self.indexfile, revornode)) | |
1418 |
|
1422 | |||
1419 | def checkinlinesize(self, tr, fp=None): |
|
1423 | def checkinlinesize(self, tr, fp=None): | |
1420 | """Check if the revlog is too big for inline and convert if so. |
|
1424 | """Check if the revlog is too big for inline and convert if so. | |
1421 |
|
1425 | |||
1422 | This should be called after revisions are added to the revlog. If the |
|
1426 | This should be called after revisions are added to the revlog. If the | |
1423 | revlog has grown too large to be an inline revlog, it will convert it |
|
1427 | revlog has grown too large to be an inline revlog, it will convert it | |
1424 | to use multiple index and data files. |
|
1428 | to use multiple index and data files. | |
1425 | """ |
|
1429 | """ | |
1426 | if not self._inline or (self.start(-2) + self.length(-2)) < _maxinline: |
|
1430 | if not self._inline or (self.start(-2) + self.length(-2)) < _maxinline: | |
1427 | return |
|
1431 | return | |
1428 |
|
1432 | |||
1429 | trinfo = tr.find(self.indexfile) |
|
1433 | trinfo = tr.find(self.indexfile) | |
1430 | if trinfo is None: |
|
1434 | if trinfo is None: | |
1431 | raise RevlogError(_("%s not found in the transaction") |
|
1435 | raise RevlogError(_("%s not found in the transaction") | |
1432 | % self.indexfile) |
|
1436 | % self.indexfile) | |
1433 |
|
1437 | |||
1434 | trindex = trinfo[2] |
|
1438 | trindex = trinfo[2] | |
1435 | if trindex is not None: |
|
1439 | if trindex is not None: | |
1436 | dataoff = self.start(trindex) |
|
1440 | dataoff = self.start(trindex) | |
1437 | else: |
|
1441 | else: | |
1438 | # revlog was stripped at start of transaction, use all leftover data |
|
1442 | # revlog was stripped at start of transaction, use all leftover data | |
1439 | trindex = len(self) - 1 |
|
1443 | trindex = len(self) - 1 | |
1440 | dataoff = self.end(-2) |
|
1444 | dataoff = self.end(-2) | |
1441 |
|
1445 | |||
1442 | tr.add(self.datafile, dataoff) |
|
1446 | tr.add(self.datafile, dataoff) | |
1443 |
|
1447 | |||
1444 | if fp: |
|
1448 | if fp: | |
1445 | fp.flush() |
|
1449 | fp.flush() | |
1446 | fp.close() |
|
1450 | fp.close() | |
1447 |
|
1451 | |||
1448 | df = self.opener(self.datafile, 'w') |
|
1452 | df = self.opener(self.datafile, 'w') | |
1449 | try: |
|
1453 | try: | |
1450 | for r in self: |
|
1454 | for r in self: | |
1451 | df.write(self._getsegmentforrevs(r, r)[1]) |
|
1455 | df.write(self._getsegmentforrevs(r, r)[1]) | |
1452 | finally: |
|
1456 | finally: | |
1453 | df.close() |
|
1457 | df.close() | |
1454 |
|
1458 | |||
1455 | fp = self.opener(self.indexfile, 'w', atomictemp=True, |
|
1459 | fp = self.opener(self.indexfile, 'w', atomictemp=True, | |
1456 | checkambig=self._checkambig) |
|
1460 | checkambig=self._checkambig) | |
1457 | self.version &= ~FLAG_INLINE_DATA |
|
1461 | self.version &= ~FLAG_INLINE_DATA | |
1458 | self._inline = False |
|
1462 | self._inline = False | |
1459 | for i in self: |
|
1463 | for i in self: | |
1460 | e = self._io.packentry(self.index[i], self.node, self.version, i) |
|
1464 | e = self._io.packentry(self.index[i], self.node, self.version, i) | |
1461 | fp.write(e) |
|
1465 | fp.write(e) | |
1462 |
|
1466 | |||
1463 | # if we don't call close, the temp file will never replace the |
|
1467 | # if we don't call close, the temp file will never replace the | |
1464 | # real index |
|
1468 | # real index | |
1465 | fp.close() |
|
1469 | fp.close() | |
1466 |
|
1470 | |||
1467 | tr.replace(self.indexfile, trindex * self._io.size) |
|
1471 | tr.replace(self.indexfile, trindex * self._io.size) | |
1468 | self._chunkclear() |
|
1472 | self._chunkclear() | |
1469 |
|
1473 | |||
1470 | def addrevision(self, text, transaction, link, p1, p2, cachedelta=None, |
|
1474 | def addrevision(self, text, transaction, link, p1, p2, cachedelta=None, | |
1471 | node=None, flags=REVIDX_DEFAULT_FLAGS): |
|
1475 | node=None, flags=REVIDX_DEFAULT_FLAGS): | |
1472 | """add a revision to the log |
|
1476 | """add a revision to the log | |
1473 |
|
1477 | |||
1474 | text - the revision data to add |
|
1478 | text - the revision data to add | |
1475 | transaction - the transaction object used for rollback |
|
1479 | transaction - the transaction object used for rollback | |
1476 | link - the linkrev data to add |
|
1480 | link - the linkrev data to add | |
1477 | p1, p2 - the parent nodeids of the revision |
|
1481 | p1, p2 - the parent nodeids of the revision | |
1478 | cachedelta - an optional precomputed delta |
|
1482 | cachedelta - an optional precomputed delta | |
1479 | node - nodeid of revision; typically node is not specified, and it is |
|
1483 | node - nodeid of revision; typically node is not specified, and it is | |
1480 | computed by default as hash(text, p1, p2), however subclasses might |
|
1484 | computed by default as hash(text, p1, p2), however subclasses might | |
1481 | use different hashing method (and override checkhash() in such case) |
|
1485 | use different hashing method (and override checkhash() in such case) | |
1482 | flags - the known flags to set on the revision |
|
1486 | flags - the known flags to set on the revision | |
1483 | """ |
|
1487 | """ | |
1484 | if link == nullrev: |
|
1488 | if link == nullrev: | |
1485 | raise RevlogError(_("attempted to add linkrev -1 to %s") |
|
1489 | raise RevlogError(_("attempted to add linkrev -1 to %s") | |
1486 | % self.indexfile) |
|
1490 | % self.indexfile) | |
1487 |
|
1491 | |||
1488 | if flags: |
|
1492 | if flags: | |
1489 | node = node or self.hash(text, p1, p2) |
|
1493 | node = node or self.hash(text, p1, p2) | |
1490 |
|
1494 | |||
1491 | rawtext, validatehash = self._processflags(text, flags, 'write') |
|
1495 | rawtext, validatehash = self._processflags(text, flags, 'write') | |
1492 |
|
1496 | |||
1493 | # If the flag processor modifies the revision data, ignore any provided |
|
1497 | # If the flag processor modifies the revision data, ignore any provided | |
1494 | # cachedelta. |
|
1498 | # cachedelta. | |
1495 | if rawtext != text: |
|
1499 | if rawtext != text: | |
1496 | cachedelta = None |
|
1500 | cachedelta = None | |
1497 |
|
1501 | |||
1498 | if len(rawtext) > _maxentrysize: |
|
1502 | if len(rawtext) > _maxentrysize: | |
1499 | raise RevlogError( |
|
1503 | raise RevlogError( | |
1500 | _("%s: size of %d bytes exceeds maximum revlog storage of 2GiB") |
|
1504 | _("%s: size of %d bytes exceeds maximum revlog storage of 2GiB") | |
1501 | % (self.indexfile, len(rawtext))) |
|
1505 | % (self.indexfile, len(rawtext))) | |
1502 |
|
1506 | |||
1503 | node = node or self.hash(rawtext, p1, p2) |
|
1507 | node = node or self.hash(rawtext, p1, p2) | |
1504 | if node in self.nodemap: |
|
1508 | if node in self.nodemap: | |
1505 | return node |
|
1509 | return node | |
1506 |
|
1510 | |||
1507 | if validatehash: |
|
1511 | if validatehash: | |
1508 | self.checkhash(rawtext, node, p1=p1, p2=p2) |
|
1512 | self.checkhash(rawtext, node, p1=p1, p2=p2) | |
1509 |
|
1513 | |||
1510 | return self.addrawrevision(rawtext, transaction, link, p1, p2, node, |
|
1514 | return self.addrawrevision(rawtext, transaction, link, p1, p2, node, | |
1511 | flags, cachedelta=cachedelta) |
|
1515 | flags, cachedelta=cachedelta) | |
1512 |
|
1516 | |||
1513 | def addrawrevision(self, rawtext, transaction, link, p1, p2, node, flags, |
|
1517 | def addrawrevision(self, rawtext, transaction, link, p1, p2, node, flags, | |
1514 | cachedelta=None): |
|
1518 | cachedelta=None): | |
1515 | """add a raw revision with known flags, node and parents |
|
1519 | """add a raw revision with known flags, node and parents | |
1516 | useful when reusing a revision not stored in this revlog (ex: received |
|
1520 | useful when reusing a revision not stored in this revlog (ex: received | |
1517 | over wire, or read from an external bundle). |
|
1521 | over wire, or read from an external bundle). | |
1518 | """ |
|
1522 | """ | |
1519 | dfh = None |
|
1523 | dfh = None | |
1520 | if not self._inline: |
|
1524 | if not self._inline: | |
1521 | dfh = self.opener(self.datafile, "a+") |
|
1525 | dfh = self.opener(self.datafile, "a+") | |
1522 | ifh = self.opener(self.indexfile, "a+", checkambig=self._checkambig) |
|
1526 | ifh = self.opener(self.indexfile, "a+", checkambig=self._checkambig) | |
1523 | try: |
|
1527 | try: | |
1524 | return self._addrevision(node, rawtext, transaction, link, p1, p2, |
|
1528 | return self._addrevision(node, rawtext, transaction, link, p1, p2, | |
1525 | flags, cachedelta, ifh, dfh) |
|
1529 | flags, cachedelta, ifh, dfh) | |
1526 | finally: |
|
1530 | finally: | |
1527 | if dfh: |
|
1531 | if dfh: | |
1528 | dfh.close() |
|
1532 | dfh.close() | |
1529 | ifh.close() |
|
1533 | ifh.close() | |
1530 |
|
1534 | |||
1531 | def compress(self, data): |
|
1535 | def compress(self, data): | |
1532 | """Generate a possibly-compressed representation of data.""" |
|
1536 | """Generate a possibly-compressed representation of data.""" | |
1533 | if not data: |
|
1537 | if not data: | |
1534 | return '', data |
|
1538 | return '', data | |
1535 |
|
1539 | |||
1536 | compressed = self._compressor.compress(data) |
|
1540 | compressed = self._compressor.compress(data) | |
1537 |
|
1541 | |||
1538 | if compressed: |
|
1542 | if compressed: | |
1539 | # The revlog compressor added the header in the returned data. |
|
1543 | # The revlog compressor added the header in the returned data. | |
1540 | return '', compressed |
|
1544 | return '', compressed | |
1541 |
|
1545 | |||
1542 | if data[0:1] == '\0': |
|
1546 | if data[0:1] == '\0': | |
1543 | return '', data |
|
1547 | return '', data | |
1544 | return 'u', data |
|
1548 | return 'u', data | |
1545 |
|
1549 | |||
1546 | def decompress(self, data): |
|
1550 | def decompress(self, data): | |
1547 | """Decompress a revlog chunk. |
|
1551 | """Decompress a revlog chunk. | |
1548 |
|
1552 | |||
1549 | The chunk is expected to begin with a header identifying the |
|
1553 | The chunk is expected to begin with a header identifying the | |
1550 | format type so it can be routed to an appropriate decompressor. |
|
1554 | format type so it can be routed to an appropriate decompressor. | |
1551 | """ |
|
1555 | """ | |
1552 | if not data: |
|
1556 | if not data: | |
1553 | return data |
|
1557 | return data | |
1554 |
|
1558 | |||
1555 | # Revlogs are read much more frequently than they are written and many |
|
1559 | # Revlogs are read much more frequently than they are written and many | |
1556 | # chunks only take microseconds to decompress, so performance is |
|
1560 | # chunks only take microseconds to decompress, so performance is | |
1557 | # important here. |
|
1561 | # important here. | |
1558 | # |
|
1562 | # | |
1559 | # We can make a few assumptions about revlogs: |
|
1563 | # We can make a few assumptions about revlogs: | |
1560 | # |
|
1564 | # | |
1561 | # 1) the majority of chunks will be compressed (as opposed to inline |
|
1565 | # 1) the majority of chunks will be compressed (as opposed to inline | |
1562 | # raw data). |
|
1566 | # raw data). | |
1563 | # 2) decompressing *any* data will likely by at least 10x slower than |
|
1567 | # 2) decompressing *any* data will likely by at least 10x slower than | |
1564 | # returning raw inline data. |
|
1568 | # returning raw inline data. | |
1565 | # 3) we want to prioritize common and officially supported compression |
|
1569 | # 3) we want to prioritize common and officially supported compression | |
1566 | # engines |
|
1570 | # engines | |
1567 | # |
|
1571 | # | |
1568 | # It follows that we want to optimize for "decompress compressed data |
|
1572 | # It follows that we want to optimize for "decompress compressed data | |
1569 | # when encoded with common and officially supported compression engines" |
|
1573 | # when encoded with common and officially supported compression engines" | |
1570 | # case over "raw data" and "data encoded by less common or non-official |
|
1574 | # case over "raw data" and "data encoded by less common or non-official | |
1571 | # compression engines." That is why we have the inline lookup first |
|
1575 | # compression engines." That is why we have the inline lookup first | |
1572 | # followed by the compengines lookup. |
|
1576 | # followed by the compengines lookup. | |
1573 | # |
|
1577 | # | |
1574 | # According to `hg perfrevlogchunks`, this is ~0.5% faster for zlib |
|
1578 | # According to `hg perfrevlogchunks`, this is ~0.5% faster for zlib | |
1575 | # compressed chunks. And this matters for changelog and manifest reads. |
|
1579 | # compressed chunks. And this matters for changelog and manifest reads. | |
1576 | t = data[0:1] |
|
1580 | t = data[0:1] | |
1577 |
|
1581 | |||
1578 | if t == 'x': |
|
1582 | if t == 'x': | |
1579 | try: |
|
1583 | try: | |
1580 | return _zlibdecompress(data) |
|
1584 | return _zlibdecompress(data) | |
1581 | except zlib.error as e: |
|
1585 | except zlib.error as e: | |
1582 | raise RevlogError(_('revlog decompress error: %s') % str(e)) |
|
1586 | raise RevlogError(_('revlog decompress error: %s') % str(e)) | |
1583 | # '\0' is more common than 'u' so it goes first. |
|
1587 | # '\0' is more common than 'u' so it goes first. | |
1584 | elif t == '\0': |
|
1588 | elif t == '\0': | |
1585 | return data |
|
1589 | return data | |
1586 | elif t == 'u': |
|
1590 | elif t == 'u': | |
1587 | return util.buffer(data, 1) |
|
1591 | return util.buffer(data, 1) | |
1588 |
|
1592 | |||
1589 | try: |
|
1593 | try: | |
1590 | compressor = self._decompressors[t] |
|
1594 | compressor = self._decompressors[t] | |
1591 | except KeyError: |
|
1595 | except KeyError: | |
1592 | try: |
|
1596 | try: | |
1593 | engine = util.compengines.forrevlogheader(t) |
|
1597 | engine = util.compengines.forrevlogheader(t) | |
1594 | compressor = engine.revlogcompressor() |
|
1598 | compressor = engine.revlogcompressor() | |
1595 | self._decompressors[t] = compressor |
|
1599 | self._decompressors[t] = compressor | |
1596 | except KeyError: |
|
1600 | except KeyError: | |
1597 | raise RevlogError(_('unknown compression type %r') % t) |
|
1601 | raise RevlogError(_('unknown compression type %r') % t) | |
1598 |
|
1602 | |||
1599 | return compressor.decompress(data) |
|
1603 | return compressor.decompress(data) | |
1600 |
|
1604 | |||
1601 | def _isgooddelta(self, d, textlen): |
|
1605 | def _isgooddelta(self, d, textlen): | |
1602 | """Returns True if the given delta is good. Good means that it is within |
|
1606 | """Returns True if the given delta is good. Good means that it is within | |
1603 | the disk span, disk size, and chain length bounds that we know to be |
|
1607 | the disk span, disk size, and chain length bounds that we know to be | |
1604 | performant.""" |
|
1608 | performant.""" | |
1605 | if d is None: |
|
1609 | if d is None: | |
1606 | return False |
|
1610 | return False | |
1607 |
|
1611 | |||
1608 | # - 'dist' is the distance from the base revision -- bounding it limits |
|
1612 | # - 'dist' is the distance from the base revision -- bounding it limits | |
1609 | # the amount of I/O we need to do. |
|
1613 | # the amount of I/O we need to do. | |
1610 | # - 'compresseddeltalen' is the sum of the total size of deltas we need |
|
1614 | # - 'compresseddeltalen' is the sum of the total size of deltas we need | |
1611 | # to apply -- bounding it limits the amount of CPU we consume. |
|
1615 | # to apply -- bounding it limits the amount of CPU we consume. | |
1612 | dist, l, data, base, chainbase, chainlen, compresseddeltalen = d |
|
1616 | dist, l, data, base, chainbase, chainlen, compresseddeltalen = d | |
1613 | if (dist > textlen * 4 or l > textlen or |
|
1617 | if (dist > textlen * 4 or l > textlen or | |
1614 | compresseddeltalen > textlen * 2 or |
|
1618 | compresseddeltalen > textlen * 2 or | |
1615 | (self._maxchainlen and chainlen > self._maxchainlen)): |
|
1619 | (self._maxchainlen and chainlen > self._maxchainlen)): | |
1616 | return False |
|
1620 | return False | |
1617 |
|
1621 | |||
1618 | return True |
|
1622 | return True | |
1619 |
|
1623 | |||
1620 | def _addrevision(self, node, rawtext, transaction, link, p1, p2, flags, |
|
1624 | def _addrevision(self, node, rawtext, transaction, link, p1, p2, flags, | |
1621 | cachedelta, ifh, dfh, alwayscache=False): |
|
1625 | cachedelta, ifh, dfh, alwayscache=False): | |
1622 | """internal function to add revisions to the log |
|
1626 | """internal function to add revisions to the log | |
1623 |
|
1627 | |||
1624 | see addrevision for argument descriptions. |
|
1628 | see addrevision for argument descriptions. | |
1625 |
|
1629 | |||
1626 | note: "addrevision" takes non-raw text, "_addrevision" takes raw text. |
|
1630 | note: "addrevision" takes non-raw text, "_addrevision" takes raw text. | |
1627 |
|
1631 | |||
1628 | invariants: |
|
1632 | invariants: | |
1629 | - rawtext is optional (can be None); if not set, cachedelta must be set. |
|
1633 | - rawtext is optional (can be None); if not set, cachedelta must be set. | |
1630 | if both are set, they must correspond to each other. |
|
1634 | if both are set, they must correspond to each other. | |
1631 | """ |
|
1635 | """ | |
1632 | btext = [rawtext] |
|
1636 | btext = [rawtext] | |
1633 | def buildtext(): |
|
1637 | def buildtext(): | |
1634 | if btext[0] is not None: |
|
1638 | if btext[0] is not None: | |
1635 | return btext[0] |
|
1639 | return btext[0] | |
1636 | baserev = cachedelta[0] |
|
1640 | baserev = cachedelta[0] | |
1637 | delta = cachedelta[1] |
|
1641 | delta = cachedelta[1] | |
1638 | # special case deltas which replace entire base; no need to decode |
|
1642 | # special case deltas which replace entire base; no need to decode | |
1639 | # base revision. this neatly avoids censored bases, which throw when |
|
1643 | # base revision. this neatly avoids censored bases, which throw when | |
1640 | # they're decoded. |
|
1644 | # they're decoded. | |
1641 | hlen = struct.calcsize(">lll") |
|
1645 | hlen = struct.calcsize(">lll") | |
1642 | if delta[:hlen] == mdiff.replacediffheader(self.rawsize(baserev), |
|
1646 | if delta[:hlen] == mdiff.replacediffheader(self.rawsize(baserev), | |
1643 | len(delta) - hlen): |
|
1647 | len(delta) - hlen): | |
1644 | btext[0] = delta[hlen:] |
|
1648 | btext[0] = delta[hlen:] | |
1645 | else: |
|
1649 | else: | |
1646 | if self._inline: |
|
1650 | if self._inline: | |
1647 | fh = ifh |
|
1651 | fh = ifh | |
1648 | else: |
|
1652 | else: | |
1649 | fh = dfh |
|
1653 | fh = dfh | |
1650 | basetext = self.revision(baserev, _df=fh, raw=True) |
|
1654 | basetext = self.revision(baserev, _df=fh, raw=True) | |
1651 | btext[0] = mdiff.patch(basetext, delta) |
|
1655 | btext[0] = mdiff.patch(basetext, delta) | |
1652 |
|
1656 | |||
1653 | try: |
|
1657 | try: | |
1654 | res = self._processflags(btext[0], flags, 'read', raw=True) |
|
1658 | res = self._processflags(btext[0], flags, 'read', raw=True) | |
1655 | btext[0], validatehash = res |
|
1659 | btext[0], validatehash = res | |
1656 | if validatehash: |
|
1660 | if validatehash: | |
1657 | self.checkhash(btext[0], node, p1=p1, p2=p2) |
|
1661 | self.checkhash(btext[0], node, p1=p1, p2=p2) | |
1658 | if flags & REVIDX_ISCENSORED: |
|
1662 | if flags & REVIDX_ISCENSORED: | |
1659 | raise RevlogError(_('node %s is not censored') % node) |
|
1663 | raise RevlogError(_('node %s is not censored') % node) | |
1660 | except CensoredNodeError: |
|
1664 | except CensoredNodeError: | |
1661 | # must pass the censored index flag to add censored revisions |
|
1665 | # must pass the censored index flag to add censored revisions | |
1662 | if not flags & REVIDX_ISCENSORED: |
|
1666 | if not flags & REVIDX_ISCENSORED: | |
1663 | raise |
|
1667 | raise | |
1664 | return btext[0] |
|
1668 | return btext[0] | |
1665 |
|
1669 | |||
1666 | def builddelta(rev): |
|
1670 | def builddelta(rev): | |
1667 | # can we use the cached delta? |
|
1671 | # can we use the cached delta? | |
1668 | if cachedelta and cachedelta[0] == rev: |
|
1672 | if cachedelta and cachedelta[0] == rev: | |
1669 | delta = cachedelta[1] |
|
1673 | delta = cachedelta[1] | |
1670 | else: |
|
1674 | else: | |
1671 | t = buildtext() |
|
1675 | t = buildtext() | |
1672 | if self.iscensored(rev): |
|
1676 | if self.iscensored(rev): | |
1673 | # deltas based on a censored revision must replace the |
|
1677 | # deltas based on a censored revision must replace the | |
1674 | # full content in one patch, so delta works everywhere |
|
1678 | # full content in one patch, so delta works everywhere | |
1675 | header = mdiff.replacediffheader(self.rawsize(rev), len(t)) |
|
1679 | header = mdiff.replacediffheader(self.rawsize(rev), len(t)) | |
1676 | delta = header + t |
|
1680 | delta = header + t | |
1677 | else: |
|
1681 | else: | |
1678 | if self._inline: |
|
1682 | if self._inline: | |
1679 | fh = ifh |
|
1683 | fh = ifh | |
1680 | else: |
|
1684 | else: | |
1681 | fh = dfh |
|
1685 | fh = dfh | |
1682 | ptext = self.revision(rev, _df=fh, raw=True) |
|
1686 | ptext = self.revision(rev, _df=fh, raw=True) | |
1683 | delta = mdiff.textdiff(ptext, t) |
|
1687 | delta = mdiff.textdiff(ptext, t) | |
1684 | header, data = self.compress(delta) |
|
1688 | header, data = self.compress(delta) | |
1685 | deltalen = len(header) + len(data) |
|
1689 | deltalen = len(header) + len(data) | |
1686 | chainbase = self.chainbase(rev) |
|
1690 | chainbase = self.chainbase(rev) | |
1687 | dist = deltalen + offset - self.start(chainbase) |
|
1691 | dist = deltalen + offset - self.start(chainbase) | |
1688 | if self._generaldelta: |
|
1692 | if self._generaldelta: | |
1689 | base = rev |
|
1693 | base = rev | |
1690 | else: |
|
1694 | else: | |
1691 | base = chainbase |
|
1695 | base = chainbase | |
1692 | chainlen, compresseddeltalen = self._chaininfo(rev) |
|
1696 | chainlen, compresseddeltalen = self._chaininfo(rev) | |
1693 | chainlen += 1 |
|
1697 | chainlen += 1 | |
1694 | compresseddeltalen += deltalen |
|
1698 | compresseddeltalen += deltalen | |
1695 | return (dist, deltalen, (header, data), base, |
|
1699 | return (dist, deltalen, (header, data), base, | |
1696 | chainbase, chainlen, compresseddeltalen) |
|
1700 | chainbase, chainlen, compresseddeltalen) | |
1697 |
|
1701 | |||
1698 | curr = len(self) |
|
1702 | curr = len(self) | |
1699 | prev = curr - 1 |
|
1703 | prev = curr - 1 | |
1700 | offset = self.end(prev) |
|
1704 | offset = self.end(prev) | |
1701 | delta = None |
|
1705 | delta = None | |
1702 | p1r, p2r = self.rev(p1), self.rev(p2) |
|
1706 | p1r, p2r = self.rev(p1), self.rev(p2) | |
1703 |
|
1707 | |||
1704 | # full versions are inserted when the needed deltas |
|
1708 | # full versions are inserted when the needed deltas | |
1705 | # become comparable to the uncompressed text |
|
1709 | # become comparable to the uncompressed text | |
1706 | if rawtext is None: |
|
1710 | if rawtext is None: | |
1707 | textlen = mdiff.patchedsize(self.rawsize(cachedelta[0]), |
|
1711 | textlen = mdiff.patchedsize(self.rawsize(cachedelta[0]), | |
1708 | cachedelta[1]) |
|
1712 | cachedelta[1]) | |
1709 | else: |
|
1713 | else: | |
1710 | textlen = len(rawtext) |
|
1714 | textlen = len(rawtext) | |
1711 |
|
1715 | |||
1712 | # should we try to build a delta? |
|
1716 | # should we try to build a delta? | |
1713 | if prev != nullrev and self.storedeltachains: |
|
1717 | if prev != nullrev and self.storedeltachains: | |
1714 | tested = set() |
|
1718 | tested = set() | |
1715 | # This condition is true most of the time when processing |
|
1719 | # This condition is true most of the time when processing | |
1716 | # changegroup data into a generaldelta repo. The only time it |
|
1720 | # changegroup data into a generaldelta repo. The only time it | |
1717 | # isn't true is if this is the first revision in a delta chain |
|
1721 | # isn't true is if this is the first revision in a delta chain | |
1718 | # or if ``format.generaldelta=true`` disabled ``lazydeltabase``. |
|
1722 | # or if ``format.generaldelta=true`` disabled ``lazydeltabase``. | |
1719 | if cachedelta and self._generaldelta and self._lazydeltabase: |
|
1723 | if cachedelta and self._generaldelta and self._lazydeltabase: | |
1720 | # Assume what we received from the server is a good choice |
|
1724 | # Assume what we received from the server is a good choice | |
1721 | # build delta will reuse the cache |
|
1725 | # build delta will reuse the cache | |
1722 | candidatedelta = builddelta(cachedelta[0]) |
|
1726 | candidatedelta = builddelta(cachedelta[0]) | |
1723 | tested.add(cachedelta[0]) |
|
1727 | tested.add(cachedelta[0]) | |
1724 | if self._isgooddelta(candidatedelta, textlen): |
|
1728 | if self._isgooddelta(candidatedelta, textlen): | |
1725 | delta = candidatedelta |
|
1729 | delta = candidatedelta | |
1726 | if delta is None and self._generaldelta: |
|
1730 | if delta is None and self._generaldelta: | |
1727 | # exclude already lazy tested base if any |
|
1731 | # exclude already lazy tested base if any | |
1728 | parents = [p for p in (p1r, p2r) |
|
1732 | parents = [p for p in (p1r, p2r) | |
1729 | if p != nullrev and p not in tested] |
|
1733 | if p != nullrev and p not in tested] | |
1730 | if parents and not self._aggressivemergedeltas: |
|
1734 | if parents and not self._aggressivemergedeltas: | |
1731 | # Pick whichever parent is closer to us (to minimize the |
|
1735 | # Pick whichever parent is closer to us (to minimize the | |
1732 | # chance of having to build a fulltext). |
|
1736 | # chance of having to build a fulltext). | |
1733 | parents = [max(parents)] |
|
1737 | parents = [max(parents)] | |
1734 | tested.update(parents) |
|
1738 | tested.update(parents) | |
1735 | pdeltas = [] |
|
1739 | pdeltas = [] | |
1736 | for p in parents: |
|
1740 | for p in parents: | |
1737 | pd = builddelta(p) |
|
1741 | pd = builddelta(p) | |
1738 | if self._isgooddelta(pd, textlen): |
|
1742 | if self._isgooddelta(pd, textlen): | |
1739 | pdeltas.append(pd) |
|
1743 | pdeltas.append(pd) | |
1740 | if pdeltas: |
|
1744 | if pdeltas: | |
1741 | delta = min(pdeltas, key=lambda x: x[1]) |
|
1745 | delta = min(pdeltas, key=lambda x: x[1]) | |
1742 | if delta is None and prev not in tested: |
|
1746 | if delta is None and prev not in tested: | |
1743 | # other approach failed try against prev to hopefully save us a |
|
1747 | # other approach failed try against prev to hopefully save us a | |
1744 | # fulltext. |
|
1748 | # fulltext. | |
1745 | candidatedelta = builddelta(prev) |
|
1749 | candidatedelta = builddelta(prev) | |
1746 | if self._isgooddelta(candidatedelta, textlen): |
|
1750 | if self._isgooddelta(candidatedelta, textlen): | |
1747 | delta = candidatedelta |
|
1751 | delta = candidatedelta | |
1748 | if delta is not None: |
|
1752 | if delta is not None: | |
1749 | dist, l, data, base, chainbase, chainlen, compresseddeltalen = delta |
|
1753 | dist, l, data, base, chainbase, chainlen, compresseddeltalen = delta | |
1750 | else: |
|
1754 | else: | |
1751 | rawtext = buildtext() |
|
1755 | rawtext = buildtext() | |
1752 | data = self.compress(rawtext) |
|
1756 | data = self.compress(rawtext) | |
1753 | l = len(data[1]) + len(data[0]) |
|
1757 | l = len(data[1]) + len(data[0]) | |
1754 | base = chainbase = curr |
|
1758 | base = chainbase = curr | |
1755 |
|
1759 | |||
1756 | e = (offset_type(offset, flags), l, textlen, |
|
1760 | e = (offset_type(offset, flags), l, textlen, | |
1757 | base, link, p1r, p2r, node) |
|
1761 | base, link, p1r, p2r, node) | |
1758 | self.index.insert(-1, e) |
|
1762 | self.index.insert(-1, e) | |
1759 | self.nodemap[node] = curr |
|
1763 | self.nodemap[node] = curr | |
1760 |
|
1764 | |||
1761 | entry = self._io.packentry(e, self.node, self.version, curr) |
|
1765 | entry = self._io.packentry(e, self.node, self.version, curr) | |
1762 | self._writeentry(transaction, ifh, dfh, entry, data, link, offset) |
|
1766 | self._writeentry(transaction, ifh, dfh, entry, data, link, offset) | |
1763 |
|
1767 | |||
1764 | if alwayscache and rawtext is None: |
|
1768 | if alwayscache and rawtext is None: | |
1765 | rawtext = buildtext() |
|
1769 | rawtext = buildtext() | |
1766 |
|
1770 | |||
1767 | if type(rawtext) == str: # only accept immutable objects |
|
1771 | if type(rawtext) == str: # only accept immutable objects | |
1768 | self._cache = (node, curr, rawtext) |
|
1772 | self._cache = (node, curr, rawtext) | |
1769 | self._chainbasecache[curr] = chainbase |
|
1773 | self._chainbasecache[curr] = chainbase | |
1770 | return node |
|
1774 | return node | |
1771 |
|
1775 | |||
1772 | def _writeentry(self, transaction, ifh, dfh, entry, data, link, offset): |
|
1776 | def _writeentry(self, transaction, ifh, dfh, entry, data, link, offset): | |
1773 | # Files opened in a+ mode have inconsistent behavior on various |
|
1777 | # Files opened in a+ mode have inconsistent behavior on various | |
1774 | # platforms. Windows requires that a file positioning call be made |
|
1778 | # platforms. Windows requires that a file positioning call be made | |
1775 | # when the file handle transitions between reads and writes. See |
|
1779 | # when the file handle transitions between reads and writes. See | |
1776 | # 3686fa2b8eee and the mixedfilemodewrapper in windows.py. On other |
|
1780 | # 3686fa2b8eee and the mixedfilemodewrapper in windows.py. On other | |
1777 | # platforms, Python or the platform itself can be buggy. Some versions |
|
1781 | # platforms, Python or the platform itself can be buggy. Some versions | |
1778 | # of Solaris have been observed to not append at the end of the file |
|
1782 | # of Solaris have been observed to not append at the end of the file | |
1779 | # if the file was seeked to before the end. See issue4943 for more. |
|
1783 | # if the file was seeked to before the end. See issue4943 for more. | |
1780 | # |
|
1784 | # | |
1781 | # We work around this issue by inserting a seek() before writing. |
|
1785 | # We work around this issue by inserting a seek() before writing. | |
1782 | # Note: This is likely not necessary on Python 3. |
|
1786 | # Note: This is likely not necessary on Python 3. | |
1783 | ifh.seek(0, os.SEEK_END) |
|
1787 | ifh.seek(0, os.SEEK_END) | |
1784 | if dfh: |
|
1788 | if dfh: | |
1785 | dfh.seek(0, os.SEEK_END) |
|
1789 | dfh.seek(0, os.SEEK_END) | |
1786 |
|
1790 | |||
1787 | curr = len(self) - 1 |
|
1791 | curr = len(self) - 1 | |
1788 | if not self._inline: |
|
1792 | if not self._inline: | |
1789 | transaction.add(self.datafile, offset) |
|
1793 | transaction.add(self.datafile, offset) | |
1790 | transaction.add(self.indexfile, curr * len(entry)) |
|
1794 | transaction.add(self.indexfile, curr * len(entry)) | |
1791 | if data[0]: |
|
1795 | if data[0]: | |
1792 | dfh.write(data[0]) |
|
1796 | dfh.write(data[0]) | |
1793 | dfh.write(data[1]) |
|
1797 | dfh.write(data[1]) | |
1794 | ifh.write(entry) |
|
1798 | ifh.write(entry) | |
1795 | else: |
|
1799 | else: | |
1796 | offset += curr * self._io.size |
|
1800 | offset += curr * self._io.size | |
1797 | transaction.add(self.indexfile, offset, curr) |
|
1801 | transaction.add(self.indexfile, offset, curr) | |
1798 | ifh.write(entry) |
|
1802 | ifh.write(entry) | |
1799 | ifh.write(data[0]) |
|
1803 | ifh.write(data[0]) | |
1800 | ifh.write(data[1]) |
|
1804 | ifh.write(data[1]) | |
1801 | self.checkinlinesize(transaction, ifh) |
|
1805 | self.checkinlinesize(transaction, ifh) | |
1802 |
|
1806 | |||
1803 | def addgroup(self, cg, linkmapper, transaction, addrevisioncb=None): |
|
1807 | def addgroup(self, cg, linkmapper, transaction, addrevisioncb=None): | |
1804 | """ |
|
1808 | """ | |
1805 | add a delta group |
|
1809 | add a delta group | |
1806 |
|
1810 | |||
1807 | given a set of deltas, add them to the revision log. the |
|
1811 | given a set of deltas, add them to the revision log. the | |
1808 | first delta is against its parent, which should be in our |
|
1812 | first delta is against its parent, which should be in our | |
1809 | log, the rest are against the previous delta. |
|
1813 | log, the rest are against the previous delta. | |
1810 |
|
1814 | |||
1811 | If ``addrevisioncb`` is defined, it will be called with arguments of |
|
1815 | If ``addrevisioncb`` is defined, it will be called with arguments of | |
1812 | this revlog and the node that was added. |
|
1816 | this revlog and the node that was added. | |
1813 | """ |
|
1817 | """ | |
1814 |
|
1818 | |||
1815 | # track the base of the current delta log |
|
1819 | # track the base of the current delta log | |
1816 | content = [] |
|
1820 | content = [] | |
1817 | node = None |
|
1821 | node = None | |
1818 |
|
1822 | |||
1819 | r = len(self) |
|
1823 | r = len(self) | |
1820 | end = 0 |
|
1824 | end = 0 | |
1821 | if r: |
|
1825 | if r: | |
1822 | end = self.end(r - 1) |
|
1826 | end = self.end(r - 1) | |
1823 | ifh = self.opener(self.indexfile, "a+", checkambig=self._checkambig) |
|
1827 | ifh = self.opener(self.indexfile, "a+", checkambig=self._checkambig) | |
1824 | isize = r * self._io.size |
|
1828 | isize = r * self._io.size | |
1825 | if self._inline: |
|
1829 | if self._inline: | |
1826 | transaction.add(self.indexfile, end + isize, r) |
|
1830 | transaction.add(self.indexfile, end + isize, r) | |
1827 | dfh = None |
|
1831 | dfh = None | |
1828 | else: |
|
1832 | else: | |
1829 | transaction.add(self.indexfile, isize, r) |
|
1833 | transaction.add(self.indexfile, isize, r) | |
1830 | transaction.add(self.datafile, end) |
|
1834 | transaction.add(self.datafile, end) | |
1831 | dfh = self.opener(self.datafile, "a+") |
|
1835 | dfh = self.opener(self.datafile, "a+") | |
1832 | def flush(): |
|
1836 | def flush(): | |
1833 | if dfh: |
|
1837 | if dfh: | |
1834 | dfh.flush() |
|
1838 | dfh.flush() | |
1835 | ifh.flush() |
|
1839 | ifh.flush() | |
1836 | try: |
|
1840 | try: | |
1837 | # loop through our set of deltas |
|
1841 | # loop through our set of deltas | |
1838 | chain = None |
|
1842 | chain = None | |
1839 | for chunkdata in iter(lambda: cg.deltachunk(chain), {}): |
|
1843 | for chunkdata in iter(lambda: cg.deltachunk(chain), {}): | |
1840 | node = chunkdata['node'] |
|
1844 | node = chunkdata['node'] | |
1841 | p1 = chunkdata['p1'] |
|
1845 | p1 = chunkdata['p1'] | |
1842 | p2 = chunkdata['p2'] |
|
1846 | p2 = chunkdata['p2'] | |
1843 | cs = chunkdata['cs'] |
|
1847 | cs = chunkdata['cs'] | |
1844 | deltabase = chunkdata['deltabase'] |
|
1848 | deltabase = chunkdata['deltabase'] | |
1845 | delta = chunkdata['delta'] |
|
1849 | delta = chunkdata['delta'] | |
1846 | flags = chunkdata['flags'] or REVIDX_DEFAULT_FLAGS |
|
1850 | flags = chunkdata['flags'] or REVIDX_DEFAULT_FLAGS | |
1847 |
|
1851 | |||
1848 | content.append(node) |
|
1852 | content.append(node) | |
1849 |
|
1853 | |||
1850 | link = linkmapper(cs) |
|
1854 | link = linkmapper(cs) | |
1851 | if node in self.nodemap: |
|
1855 | if node in self.nodemap: | |
1852 | # this can happen if two branches make the same change |
|
1856 | # this can happen if two branches make the same change | |
1853 | chain = node |
|
1857 | chain = node | |
1854 | continue |
|
1858 | continue | |
1855 |
|
1859 | |||
1856 | for p in (p1, p2): |
|
1860 | for p in (p1, p2): | |
1857 | if p not in self.nodemap: |
|
1861 | if p not in self.nodemap: | |
1858 | raise LookupError(p, self.indexfile, |
|
1862 | raise LookupError(p, self.indexfile, | |
1859 | _('unknown parent')) |
|
1863 | _('unknown parent')) | |
1860 |
|
1864 | |||
1861 | if deltabase not in self.nodemap: |
|
1865 | if deltabase not in self.nodemap: | |
1862 | raise LookupError(deltabase, self.indexfile, |
|
1866 | raise LookupError(deltabase, self.indexfile, | |
1863 | _('unknown delta base')) |
|
1867 | _('unknown delta base')) | |
1864 |
|
1868 | |||
1865 | baserev = self.rev(deltabase) |
|
1869 | baserev = self.rev(deltabase) | |
1866 |
|
1870 | |||
1867 | if baserev != nullrev and self.iscensored(baserev): |
|
1871 | if baserev != nullrev and self.iscensored(baserev): | |
1868 | # if base is censored, delta must be full replacement in a |
|
1872 | # if base is censored, delta must be full replacement in a | |
1869 | # single patch operation |
|
1873 | # single patch operation | |
1870 | hlen = struct.calcsize(">lll") |
|
1874 | hlen = struct.calcsize(">lll") | |
1871 | oldlen = self.rawsize(baserev) |
|
1875 | oldlen = self.rawsize(baserev) | |
1872 | newlen = len(delta) - hlen |
|
1876 | newlen = len(delta) - hlen | |
1873 | if delta[:hlen] != mdiff.replacediffheader(oldlen, newlen): |
|
1877 | if delta[:hlen] != mdiff.replacediffheader(oldlen, newlen): | |
1874 | raise error.CensoredBaseError(self.indexfile, |
|
1878 | raise error.CensoredBaseError(self.indexfile, | |
1875 | self.node(baserev)) |
|
1879 | self.node(baserev)) | |
1876 |
|
1880 | |||
1877 | if not flags and self._peek_iscensored(baserev, delta, flush): |
|
1881 | if not flags and self._peek_iscensored(baserev, delta, flush): | |
1878 | flags |= REVIDX_ISCENSORED |
|
1882 | flags |= REVIDX_ISCENSORED | |
1879 |
|
1883 | |||
1880 | # We assume consumers of addrevisioncb will want to retrieve |
|
1884 | # We assume consumers of addrevisioncb will want to retrieve | |
1881 | # the added revision, which will require a call to |
|
1885 | # the added revision, which will require a call to | |
1882 | # revision(). revision() will fast path if there is a cache |
|
1886 | # revision(). revision() will fast path if there is a cache | |
1883 | # hit. So, we tell _addrevision() to always cache in this case. |
|
1887 | # hit. So, we tell _addrevision() to always cache in this case. | |
1884 | # We're only using addgroup() in the context of changegroup |
|
1888 | # We're only using addgroup() in the context of changegroup | |
1885 | # generation so the revision data can always be handled as raw |
|
1889 | # generation so the revision data can always be handled as raw | |
1886 | # by the flagprocessor. |
|
1890 | # by the flagprocessor. | |
1887 | chain = self._addrevision(node, None, transaction, link, |
|
1891 | chain = self._addrevision(node, None, transaction, link, | |
1888 | p1, p2, flags, (baserev, delta), |
|
1892 | p1, p2, flags, (baserev, delta), | |
1889 | ifh, dfh, |
|
1893 | ifh, dfh, | |
1890 | alwayscache=bool(addrevisioncb)) |
|
1894 | alwayscache=bool(addrevisioncb)) | |
1891 |
|
1895 | |||
1892 | if addrevisioncb: |
|
1896 | if addrevisioncb: | |
1893 | addrevisioncb(self, chain) |
|
1897 | addrevisioncb(self, chain) | |
1894 |
|
1898 | |||
1895 | if not dfh and not self._inline: |
|
1899 | if not dfh and not self._inline: | |
1896 | # addrevision switched from inline to conventional |
|
1900 | # addrevision switched from inline to conventional | |
1897 | # reopen the index |
|
1901 | # reopen the index | |
1898 | ifh.close() |
|
1902 | ifh.close() | |
1899 | dfh = self.opener(self.datafile, "a+") |
|
1903 | dfh = self.opener(self.datafile, "a+") | |
1900 | ifh = self.opener(self.indexfile, "a+", |
|
1904 | ifh = self.opener(self.indexfile, "a+", | |
1901 | checkambig=self._checkambig) |
|
1905 | checkambig=self._checkambig) | |
1902 | finally: |
|
1906 | finally: | |
1903 | if dfh: |
|
1907 | if dfh: | |
1904 | dfh.close() |
|
1908 | dfh.close() | |
1905 | ifh.close() |
|
1909 | ifh.close() | |
1906 |
|
1910 | |||
1907 | return content |
|
1911 | return content | |
1908 |
|
1912 | |||
1909 | def iscensored(self, rev): |
|
1913 | def iscensored(self, rev): | |
1910 | """Check if a file revision is censored.""" |
|
1914 | """Check if a file revision is censored.""" | |
1911 | return False |
|
1915 | return False | |
1912 |
|
1916 | |||
1913 | def _peek_iscensored(self, baserev, delta, flush): |
|
1917 | def _peek_iscensored(self, baserev, delta, flush): | |
1914 | """Quickly check if a delta produces a censored revision.""" |
|
1918 | """Quickly check if a delta produces a censored revision.""" | |
1915 | return False |
|
1919 | return False | |
1916 |
|
1920 | |||
1917 | def getstrippoint(self, minlink): |
|
1921 | def getstrippoint(self, minlink): | |
1918 | """find the minimum rev that must be stripped to strip the linkrev |
|
1922 | """find the minimum rev that must be stripped to strip the linkrev | |
1919 |
|
1923 | |||
1920 | Returns a tuple containing the minimum rev and a set of all revs that |
|
1924 | Returns a tuple containing the minimum rev and a set of all revs that | |
1921 | have linkrevs that will be broken by this strip. |
|
1925 | have linkrevs that will be broken by this strip. | |
1922 | """ |
|
1926 | """ | |
1923 | brokenrevs = set() |
|
1927 | brokenrevs = set() | |
1924 | strippoint = len(self) |
|
1928 | strippoint = len(self) | |
1925 |
|
1929 | |||
1926 | heads = {} |
|
1930 | heads = {} | |
1927 | futurelargelinkrevs = set() |
|
1931 | futurelargelinkrevs = set() | |
1928 | for head in self.headrevs(): |
|
1932 | for head in self.headrevs(): | |
1929 | headlinkrev = self.linkrev(head) |
|
1933 | headlinkrev = self.linkrev(head) | |
1930 | heads[head] = headlinkrev |
|
1934 | heads[head] = headlinkrev | |
1931 | if headlinkrev >= minlink: |
|
1935 | if headlinkrev >= minlink: | |
1932 | futurelargelinkrevs.add(headlinkrev) |
|
1936 | futurelargelinkrevs.add(headlinkrev) | |
1933 |
|
1937 | |||
1934 | # This algorithm involves walking down the rev graph, starting at the |
|
1938 | # This algorithm involves walking down the rev graph, starting at the | |
1935 | # heads. Since the revs are topologically sorted according to linkrev, |
|
1939 | # heads. Since the revs are topologically sorted according to linkrev, | |
1936 | # once all head linkrevs are below the minlink, we know there are |
|
1940 | # once all head linkrevs are below the minlink, we know there are | |
1937 | # no more revs that could have a linkrev greater than minlink. |
|
1941 | # no more revs that could have a linkrev greater than minlink. | |
1938 | # So we can stop walking. |
|
1942 | # So we can stop walking. | |
1939 | while futurelargelinkrevs: |
|
1943 | while futurelargelinkrevs: | |
1940 | strippoint -= 1 |
|
1944 | strippoint -= 1 | |
1941 | linkrev = heads.pop(strippoint) |
|
1945 | linkrev = heads.pop(strippoint) | |
1942 |
|
1946 | |||
1943 | if linkrev < minlink: |
|
1947 | if linkrev < minlink: | |
1944 | brokenrevs.add(strippoint) |
|
1948 | brokenrevs.add(strippoint) | |
1945 | else: |
|
1949 | else: | |
1946 | futurelargelinkrevs.remove(linkrev) |
|
1950 | futurelargelinkrevs.remove(linkrev) | |
1947 |
|
1951 | |||
1948 | for p in self.parentrevs(strippoint): |
|
1952 | for p in self.parentrevs(strippoint): | |
1949 | if p != nullrev: |
|
1953 | if p != nullrev: | |
1950 | plinkrev = self.linkrev(p) |
|
1954 | plinkrev = self.linkrev(p) | |
1951 | heads[p] = plinkrev |
|
1955 | heads[p] = plinkrev | |
1952 | if plinkrev >= minlink: |
|
1956 | if plinkrev >= minlink: | |
1953 | futurelargelinkrevs.add(plinkrev) |
|
1957 | futurelargelinkrevs.add(plinkrev) | |
1954 |
|
1958 | |||
1955 | return strippoint, brokenrevs |
|
1959 | return strippoint, brokenrevs | |
1956 |
|
1960 | |||
1957 | def strip(self, minlink, transaction): |
|
1961 | def strip(self, minlink, transaction): | |
1958 | """truncate the revlog on the first revision with a linkrev >= minlink |
|
1962 | """truncate the revlog on the first revision with a linkrev >= minlink | |
1959 |
|
1963 | |||
1960 | This function is called when we're stripping revision minlink and |
|
1964 | This function is called when we're stripping revision minlink and | |
1961 | its descendants from the repository. |
|
1965 | its descendants from the repository. | |
1962 |
|
1966 | |||
1963 | We have to remove all revisions with linkrev >= minlink, because |
|
1967 | We have to remove all revisions with linkrev >= minlink, because | |
1964 | the equivalent changelog revisions will be renumbered after the |
|
1968 | the equivalent changelog revisions will be renumbered after the | |
1965 | strip. |
|
1969 | strip. | |
1966 |
|
1970 | |||
1967 | So we truncate the revlog on the first of these revisions, and |
|
1971 | So we truncate the revlog on the first of these revisions, and | |
1968 | trust that the caller has saved the revisions that shouldn't be |
|
1972 | trust that the caller has saved the revisions that shouldn't be | |
1969 | removed and that it'll re-add them after this truncation. |
|
1973 | removed and that it'll re-add them after this truncation. | |
1970 | """ |
|
1974 | """ | |
1971 | if len(self) == 0: |
|
1975 | if len(self) == 0: | |
1972 | return |
|
1976 | return | |
1973 |
|
1977 | |||
1974 | rev, _ = self.getstrippoint(minlink) |
|
1978 | rev, _ = self.getstrippoint(minlink) | |
1975 | if rev == len(self): |
|
1979 | if rev == len(self): | |
1976 | return |
|
1980 | return | |
1977 |
|
1981 | |||
1978 | # first truncate the files on disk |
|
1982 | # first truncate the files on disk | |
1979 | end = self.start(rev) |
|
1983 | end = self.start(rev) | |
1980 | if not self._inline: |
|
1984 | if not self._inline: | |
1981 | transaction.add(self.datafile, end) |
|
1985 | transaction.add(self.datafile, end) | |
1982 | end = rev * self._io.size |
|
1986 | end = rev * self._io.size | |
1983 | else: |
|
1987 | else: | |
1984 | end += rev * self._io.size |
|
1988 | end += rev * self._io.size | |
1985 |
|
1989 | |||
1986 | transaction.add(self.indexfile, end) |
|
1990 | transaction.add(self.indexfile, end) | |
1987 |
|
1991 | |||
1988 | # then reset internal state in memory to forget those revisions |
|
1992 | # then reset internal state in memory to forget those revisions | |
1989 | self._cache = None |
|
1993 | self._cache = None | |
1990 | self._chaininfocache = {} |
|
1994 | self._chaininfocache = {} | |
1991 | self._chunkclear() |
|
1995 | self._chunkclear() | |
1992 | for x in xrange(rev, len(self)): |
|
1996 | for x in xrange(rev, len(self)): | |
1993 | del self.nodemap[self.node(x)] |
|
1997 | del self.nodemap[self.node(x)] | |
1994 |
|
1998 | |||
1995 | del self.index[rev:-1] |
|
1999 | del self.index[rev:-1] | |
1996 |
|
2000 | |||
1997 | def checksize(self): |
|
2001 | def checksize(self): | |
1998 | expected = 0 |
|
2002 | expected = 0 | |
1999 | if len(self): |
|
2003 | if len(self): | |
2000 | expected = max(0, self.end(len(self) - 1)) |
|
2004 | expected = max(0, self.end(len(self) - 1)) | |
2001 |
|
2005 | |||
2002 | try: |
|
2006 | try: | |
2003 | f = self.opener(self.datafile) |
|
2007 | f = self.opener(self.datafile) | |
2004 | f.seek(0, 2) |
|
2008 | f.seek(0, 2) | |
2005 | actual = f.tell() |
|
2009 | actual = f.tell() | |
2006 | f.close() |
|
2010 | f.close() | |
2007 | dd = actual - expected |
|
2011 | dd = actual - expected | |
2008 | except IOError as inst: |
|
2012 | except IOError as inst: | |
2009 | if inst.errno != errno.ENOENT: |
|
2013 | if inst.errno != errno.ENOENT: | |
2010 | raise |
|
2014 | raise | |
2011 | dd = 0 |
|
2015 | dd = 0 | |
2012 |
|
2016 | |||
2013 | try: |
|
2017 | try: | |
2014 | f = self.opener(self.indexfile) |
|
2018 | f = self.opener(self.indexfile) | |
2015 | f.seek(0, 2) |
|
2019 | f.seek(0, 2) | |
2016 | actual = f.tell() |
|
2020 | actual = f.tell() | |
2017 | f.close() |
|
2021 | f.close() | |
2018 | s = self._io.size |
|
2022 | s = self._io.size | |
2019 | i = max(0, actual // s) |
|
2023 | i = max(0, actual // s) | |
2020 | di = actual - (i * s) |
|
2024 | di = actual - (i * s) | |
2021 | if self._inline: |
|
2025 | if self._inline: | |
2022 | databytes = 0 |
|
2026 | databytes = 0 | |
2023 | for r in self: |
|
2027 | for r in self: | |
2024 | databytes += max(0, self.length(r)) |
|
2028 | databytes += max(0, self.length(r)) | |
2025 | dd = 0 |
|
2029 | dd = 0 | |
2026 | di = actual - len(self) * s - databytes |
|
2030 | di = actual - len(self) * s - databytes | |
2027 | except IOError as inst: |
|
2031 | except IOError as inst: | |
2028 | if inst.errno != errno.ENOENT: |
|
2032 | if inst.errno != errno.ENOENT: | |
2029 | raise |
|
2033 | raise | |
2030 | di = 0 |
|
2034 | di = 0 | |
2031 |
|
2035 | |||
2032 | return (dd, di) |
|
2036 | return (dd, di) | |
2033 |
|
2037 | |||
2034 | def files(self): |
|
2038 | def files(self): | |
2035 | res = [self.indexfile] |
|
2039 | res = [self.indexfile] | |
2036 | if not self._inline: |
|
2040 | if not self._inline: | |
2037 | res.append(self.datafile) |
|
2041 | res.append(self.datafile) | |
2038 | return res |
|
2042 | return res | |
2039 |
|
2043 | |||
2040 | DELTAREUSEALWAYS = 'always' |
|
2044 | DELTAREUSEALWAYS = 'always' | |
2041 | DELTAREUSESAMEREVS = 'samerevs' |
|
2045 | DELTAREUSESAMEREVS = 'samerevs' | |
2042 | DELTAREUSENEVER = 'never' |
|
2046 | DELTAREUSENEVER = 'never' | |
2043 |
|
2047 | |||
2044 | DELTAREUSEALL = {'always', 'samerevs', 'never'} |
|
2048 | DELTAREUSEALL = {'always', 'samerevs', 'never'} | |
2045 |
|
2049 | |||
2046 | def clone(self, tr, destrevlog, addrevisioncb=None, |
|
2050 | def clone(self, tr, destrevlog, addrevisioncb=None, | |
2047 | deltareuse=DELTAREUSESAMEREVS, aggressivemergedeltas=None): |
|
2051 | deltareuse=DELTAREUSESAMEREVS, aggressivemergedeltas=None): | |
2048 | """Copy this revlog to another, possibly with format changes. |
|
2052 | """Copy this revlog to another, possibly with format changes. | |
2049 |
|
2053 | |||
2050 | The destination revlog will contain the same revisions and nodes. |
|
2054 | The destination revlog will contain the same revisions and nodes. | |
2051 | However, it may not be bit-for-bit identical due to e.g. delta encoding |
|
2055 | However, it may not be bit-for-bit identical due to e.g. delta encoding | |
2052 | differences. |
|
2056 | differences. | |
2053 |
|
2057 | |||
2054 | The ``deltareuse`` argument control how deltas from the existing revlog |
|
2058 | The ``deltareuse`` argument control how deltas from the existing revlog | |
2055 | are preserved in the destination revlog. The argument can have the |
|
2059 | are preserved in the destination revlog. The argument can have the | |
2056 | following values: |
|
2060 | following values: | |
2057 |
|
2061 | |||
2058 | DELTAREUSEALWAYS |
|
2062 | DELTAREUSEALWAYS | |
2059 | Deltas will always be reused (if possible), even if the destination |
|
2063 | Deltas will always be reused (if possible), even if the destination | |
2060 | revlog would not select the same revisions for the delta. This is the |
|
2064 | revlog would not select the same revisions for the delta. This is the | |
2061 | fastest mode of operation. |
|
2065 | fastest mode of operation. | |
2062 | DELTAREUSESAMEREVS |
|
2066 | DELTAREUSESAMEREVS | |
2063 | Deltas will be reused if the destination revlog would pick the same |
|
2067 | Deltas will be reused if the destination revlog would pick the same | |
2064 | revisions for the delta. This mode strikes a balance between speed |
|
2068 | revisions for the delta. This mode strikes a balance between speed | |
2065 | and optimization. |
|
2069 | and optimization. | |
2066 | DELTAREUSENEVER |
|
2070 | DELTAREUSENEVER | |
2067 | Deltas will never be reused. This is the slowest mode of execution. |
|
2071 | Deltas will never be reused. This is the slowest mode of execution. | |
2068 | This mode can be used to recompute deltas (e.g. if the diff/delta |
|
2072 | This mode can be used to recompute deltas (e.g. if the diff/delta | |
2069 | algorithm changes). |
|
2073 | algorithm changes). | |
2070 |
|
2074 | |||
2071 | Delta computation can be slow, so the choice of delta reuse policy can |
|
2075 | Delta computation can be slow, so the choice of delta reuse policy can | |
2072 | significantly affect run time. |
|
2076 | significantly affect run time. | |
2073 |
|
2077 | |||
2074 | The default policy (``DELTAREUSESAMEREVS``) strikes a balance between |
|
2078 | The default policy (``DELTAREUSESAMEREVS``) strikes a balance between | |
2075 | two extremes. Deltas will be reused if they are appropriate. But if the |
|
2079 | two extremes. Deltas will be reused if they are appropriate. But if the | |
2076 | delta could choose a better revision, it will do so. This means if you |
|
2080 | delta could choose a better revision, it will do so. This means if you | |
2077 | are converting a non-generaldelta revlog to a generaldelta revlog, |
|
2081 | are converting a non-generaldelta revlog to a generaldelta revlog, | |
2078 | deltas will be recomputed if the delta's parent isn't a parent of the |
|
2082 | deltas will be recomputed if the delta's parent isn't a parent of the | |
2079 | revision. |
|
2083 | revision. | |
2080 |
|
2084 | |||
2081 | In addition to the delta policy, the ``aggressivemergedeltas`` argument |
|
2085 | In addition to the delta policy, the ``aggressivemergedeltas`` argument | |
2082 | controls whether to compute deltas against both parents for merges. |
|
2086 | controls whether to compute deltas against both parents for merges. | |
2083 | By default, the current default is used. |
|
2087 | By default, the current default is used. | |
2084 | """ |
|
2088 | """ | |
2085 | if deltareuse not in self.DELTAREUSEALL: |
|
2089 | if deltareuse not in self.DELTAREUSEALL: | |
2086 | raise ValueError(_('value for deltareuse invalid: %s') % deltareuse) |
|
2090 | raise ValueError(_('value for deltareuse invalid: %s') % deltareuse) | |
2087 |
|
2091 | |||
2088 | if len(destrevlog): |
|
2092 | if len(destrevlog): | |
2089 | raise ValueError(_('destination revlog is not empty')) |
|
2093 | raise ValueError(_('destination revlog is not empty')) | |
2090 |
|
2094 | |||
2091 | if getattr(self, 'filteredrevs', None): |
|
2095 | if getattr(self, 'filteredrevs', None): | |
2092 | raise ValueError(_('source revlog has filtered revisions')) |
|
2096 | raise ValueError(_('source revlog has filtered revisions')) | |
2093 | if getattr(destrevlog, 'filteredrevs', None): |
|
2097 | if getattr(destrevlog, 'filteredrevs', None): | |
2094 | raise ValueError(_('destination revlog has filtered revisions')) |
|
2098 | raise ValueError(_('destination revlog has filtered revisions')) | |
2095 |
|
2099 | |||
2096 | # lazydeltabase controls whether to reuse a cached delta, if possible. |
|
2100 | # lazydeltabase controls whether to reuse a cached delta, if possible. | |
2097 | oldlazydeltabase = destrevlog._lazydeltabase |
|
2101 | oldlazydeltabase = destrevlog._lazydeltabase | |
2098 | oldamd = destrevlog._aggressivemergedeltas |
|
2102 | oldamd = destrevlog._aggressivemergedeltas | |
2099 |
|
2103 | |||
2100 | try: |
|
2104 | try: | |
2101 | if deltareuse == self.DELTAREUSEALWAYS: |
|
2105 | if deltareuse == self.DELTAREUSEALWAYS: | |
2102 | destrevlog._lazydeltabase = True |
|
2106 | destrevlog._lazydeltabase = True | |
2103 | elif deltareuse == self.DELTAREUSESAMEREVS: |
|
2107 | elif deltareuse == self.DELTAREUSESAMEREVS: | |
2104 | destrevlog._lazydeltabase = False |
|
2108 | destrevlog._lazydeltabase = False | |
2105 |
|
2109 | |||
2106 | destrevlog._aggressivemergedeltas = aggressivemergedeltas or oldamd |
|
2110 | destrevlog._aggressivemergedeltas = aggressivemergedeltas or oldamd | |
2107 |
|
2111 | |||
2108 | populatecachedelta = deltareuse in (self.DELTAREUSEALWAYS, |
|
2112 | populatecachedelta = deltareuse in (self.DELTAREUSEALWAYS, | |
2109 | self.DELTAREUSESAMEREVS) |
|
2113 | self.DELTAREUSESAMEREVS) | |
2110 |
|
2114 | |||
2111 | index = self.index |
|
2115 | index = self.index | |
2112 | for rev in self: |
|
2116 | for rev in self: | |
2113 | entry = index[rev] |
|
2117 | entry = index[rev] | |
2114 |
|
2118 | |||
2115 | # Some classes override linkrev to take filtered revs into |
|
2119 | # Some classes override linkrev to take filtered revs into | |
2116 | # account. Use raw entry from index. |
|
2120 | # account. Use raw entry from index. | |
2117 | flags = entry[0] & 0xffff |
|
2121 | flags = entry[0] & 0xffff | |
2118 | linkrev = entry[4] |
|
2122 | linkrev = entry[4] | |
2119 | p1 = index[entry[5]][7] |
|
2123 | p1 = index[entry[5]][7] | |
2120 | p2 = index[entry[6]][7] |
|
2124 | p2 = index[entry[6]][7] | |
2121 | node = entry[7] |
|
2125 | node = entry[7] | |
2122 |
|
2126 | |||
2123 | # (Possibly) reuse the delta from the revlog if allowed and |
|
2127 | # (Possibly) reuse the delta from the revlog if allowed and | |
2124 | # the revlog chunk is a delta. |
|
2128 | # the revlog chunk is a delta. | |
2125 | cachedelta = None |
|
2129 | cachedelta = None | |
2126 | rawtext = None |
|
2130 | rawtext = None | |
2127 | if populatecachedelta: |
|
2131 | if populatecachedelta: | |
2128 | dp = self.deltaparent(rev) |
|
2132 | dp = self.deltaparent(rev) | |
2129 | if dp != nullrev: |
|
2133 | if dp != nullrev: | |
2130 | cachedelta = (dp, str(self._chunk(rev))) |
|
2134 | cachedelta = (dp, str(self._chunk(rev))) | |
2131 |
|
2135 | |||
2132 | if not cachedelta: |
|
2136 | if not cachedelta: | |
2133 | rawtext = self.revision(rev, raw=True) |
|
2137 | rawtext = self.revision(rev, raw=True) | |
2134 |
|
2138 | |||
2135 | ifh = destrevlog.opener(destrevlog.indexfile, 'a+', |
|
2139 | ifh = destrevlog.opener(destrevlog.indexfile, 'a+', | |
2136 | checkambig=False) |
|
2140 | checkambig=False) | |
2137 | dfh = None |
|
2141 | dfh = None | |
2138 | if not destrevlog._inline: |
|
2142 | if not destrevlog._inline: | |
2139 | dfh = destrevlog.opener(destrevlog.datafile, 'a+') |
|
2143 | dfh = destrevlog.opener(destrevlog.datafile, 'a+') | |
2140 | try: |
|
2144 | try: | |
2141 | destrevlog._addrevision(node, rawtext, tr, linkrev, p1, p2, |
|
2145 | destrevlog._addrevision(node, rawtext, tr, linkrev, p1, p2, | |
2142 | flags, cachedelta, ifh, dfh) |
|
2146 | flags, cachedelta, ifh, dfh) | |
2143 | finally: |
|
2147 | finally: | |
2144 | if dfh: |
|
2148 | if dfh: | |
2145 | dfh.close() |
|
2149 | dfh.close() | |
2146 | ifh.close() |
|
2150 | ifh.close() | |
2147 |
|
2151 | |||
2148 | if addrevisioncb: |
|
2152 | if addrevisioncb: | |
2149 | addrevisioncb(self, rev, node) |
|
2153 | addrevisioncb(self, rev, node) | |
2150 | finally: |
|
2154 | finally: | |
2151 | destrevlog._lazydeltabase = oldlazydeltabase |
|
2155 | destrevlog._lazydeltabase = oldlazydeltabase | |
2152 | destrevlog._aggressivemergedeltas = oldamd |
|
2156 | destrevlog._aggressivemergedeltas = oldamd |
@@ -1,72 +1,72 b'' | |||||
1 | $ hg init t |
|
1 | $ hg init t | |
2 | $ cd t |
|
2 | $ cd t | |
3 | $ echo a > a |
|
3 | $ echo a > a | |
4 | $ hg add a |
|
4 | $ hg add a | |
5 | $ hg commit -m test |
|
5 | $ hg commit -m test | |
6 | $ rm .hg/requires |
|
6 | $ rm .hg/requires | |
7 | $ hg tip |
|
7 | $ hg tip | |
8 | abort: index 00changelog.i unknown format 2! |
|
8 | abort: unknown version (2) in revlog 00changelog.i! | |
9 | [255] |
|
9 | [255] | |
10 |
$ |
|
10 | $ echo indoor-pool > .hg/requires | |
11 | $ hg tip |
|
11 | $ hg tip | |
12 | abort: repository requires features unknown to this Mercurial: indoor-pool! |
|
12 | abort: repository requires features unknown to this Mercurial: indoor-pool! | |
13 | (see https://mercurial-scm.org/wiki/MissingRequirement for more information) |
|
13 | (see https://mercurial-scm.org/wiki/MissingRequirement for more information) | |
14 | [255] |
|
14 | [255] | |
15 | $ echo outdoor-pool >> .hg/requires |
|
15 | $ echo outdoor-pool >> .hg/requires | |
16 | $ hg tip |
|
16 | $ hg tip | |
17 | abort: repository requires features unknown to this Mercurial: indoor-pool outdoor-pool! |
|
17 | abort: repository requires features unknown to this Mercurial: indoor-pool outdoor-pool! | |
18 | (see https://mercurial-scm.org/wiki/MissingRequirement for more information) |
|
18 | (see https://mercurial-scm.org/wiki/MissingRequirement for more information) | |
19 | [255] |
|
19 | [255] | |
20 | $ cd .. |
|
20 | $ cd .. | |
21 |
|
21 | |||
22 | Test checking between features supported locally and ones required in |
|
22 | Test checking between features supported locally and ones required in | |
23 | another repository of push/pull/clone on localhost: |
|
23 | another repository of push/pull/clone on localhost: | |
24 |
|
24 | |||
25 | $ mkdir supported-locally |
|
25 | $ mkdir supported-locally | |
26 | $ cd supported-locally |
|
26 | $ cd supported-locally | |
27 |
|
27 | |||
28 | $ hg init supported |
|
28 | $ hg init supported | |
29 | $ echo a > supported/a |
|
29 | $ echo a > supported/a | |
30 | $ hg -R supported commit -Am '#0 at supported' |
|
30 | $ hg -R supported commit -Am '#0 at supported' | |
31 | adding a |
|
31 | adding a | |
32 |
|
32 | |||
33 | $ echo 'featuresetup-test' >> supported/.hg/requires |
|
33 | $ echo 'featuresetup-test' >> supported/.hg/requires | |
34 | $ cat > $TESTTMP/supported-locally/supportlocally.py <<EOF |
|
34 | $ cat > $TESTTMP/supported-locally/supportlocally.py <<EOF | |
35 | > from mercurial import localrepo, extensions |
|
35 | > from mercurial import localrepo, extensions | |
36 | > def featuresetup(ui, supported): |
|
36 | > def featuresetup(ui, supported): | |
37 | > for name, module in extensions.extensions(ui): |
|
37 | > for name, module in extensions.extensions(ui): | |
38 | > if __name__ == module.__name__: |
|
38 | > if __name__ == module.__name__: | |
39 | > # support specific feature locally |
|
39 | > # support specific feature locally | |
40 | > supported |= {'featuresetup-test'} |
|
40 | > supported |= {'featuresetup-test'} | |
41 | > return |
|
41 | > return | |
42 | > def uisetup(ui): |
|
42 | > def uisetup(ui): | |
43 | > localrepo.localrepository.featuresetupfuncs.add(featuresetup) |
|
43 | > localrepo.localrepository.featuresetupfuncs.add(featuresetup) | |
44 | > EOF |
|
44 | > EOF | |
45 | $ cat > supported/.hg/hgrc <<EOF |
|
45 | $ cat > supported/.hg/hgrc <<EOF | |
46 | > [extensions] |
|
46 | > [extensions] | |
47 | > # enable extension locally |
|
47 | > # enable extension locally | |
48 | > supportlocally = $TESTTMP/supported-locally/supportlocally.py |
|
48 | > supportlocally = $TESTTMP/supported-locally/supportlocally.py | |
49 | > EOF |
|
49 | > EOF | |
50 | $ hg -R supported status |
|
50 | $ hg -R supported status | |
51 |
|
51 | |||
52 | $ hg init push-dst |
|
52 | $ hg init push-dst | |
53 | $ hg -R supported push push-dst |
|
53 | $ hg -R supported push push-dst | |
54 | pushing to push-dst |
|
54 | pushing to push-dst | |
55 | abort: required features are not supported in the destination: featuresetup-test |
|
55 | abort: required features are not supported in the destination: featuresetup-test | |
56 | [255] |
|
56 | [255] | |
57 |
|
57 | |||
58 | $ hg init pull-src |
|
58 | $ hg init pull-src | |
59 | $ hg -R pull-src pull supported |
|
59 | $ hg -R pull-src pull supported | |
60 | pulling from supported |
|
60 | pulling from supported | |
61 | abort: required features are not supported in the destination: featuresetup-test |
|
61 | abort: required features are not supported in the destination: featuresetup-test | |
62 | [255] |
|
62 | [255] | |
63 |
|
63 | |||
64 | $ hg clone supported clone-dst |
|
64 | $ hg clone supported clone-dst | |
65 | abort: repository requires features unknown to this Mercurial: featuresetup-test! |
|
65 | abort: repository requires features unknown to this Mercurial: featuresetup-test! | |
66 | (see https://mercurial-scm.org/wiki/MissingRequirement for more information) |
|
66 | (see https://mercurial-scm.org/wiki/MissingRequirement for more information) | |
67 | [255] |
|
67 | [255] | |
68 | $ hg clone --pull supported clone-dst |
|
68 | $ hg clone --pull supported clone-dst | |
69 | abort: required features are not supported in the destination: featuresetup-test |
|
69 | abort: required features are not supported in the destination: featuresetup-test | |
70 | [255] |
|
70 | [255] | |
71 |
|
71 | |||
72 | $ cd .. |
|
72 | $ cd .. |
@@ -1,47 +1,47 b'' | |||||
1 | $ hg init empty-repo |
|
1 | $ hg init empty-repo | |
2 | $ cd empty-repo |
|
2 | $ cd empty-repo | |
3 |
|
3 | |||
4 | Flags on revlog version 0 are rejected |
|
4 | Flags on revlog version 0 are rejected | |
5 |
|
5 | |||
6 | >>> with open('.hg/store/00changelog.i', 'wb') as fh: |
|
6 | >>> with open('.hg/store/00changelog.i', 'wb') as fh: | |
7 | ... fh.write('\x00\x01\x00\x00') |
|
7 | ... fh.write('\x00\x01\x00\x00') | |
8 |
|
8 | |||
9 | $ hg log |
|
9 | $ hg log | |
10 | abort: index 00changelog.i unknown flags 0x01 for format v0! |
|
10 | abort: unknown flags (0x01) in version 0 revlog 00changelog.i! | |
11 | [255] |
|
11 | [255] | |
12 |
|
12 | |||
13 | Unknown flags on revlog version 1 are rejected |
|
13 | Unknown flags on revlog version 1 are rejected | |
14 |
|
14 | |||
15 | >>> with open('.hg/store/00changelog.i', 'wb') as fh: |
|
15 | >>> with open('.hg/store/00changelog.i', 'wb') as fh: | |
16 | ... fh.write('\x00\x04\x00\x01') |
|
16 | ... fh.write('\x00\x04\x00\x01') | |
17 |
|
17 | |||
18 | $ hg log |
|
18 | $ hg log | |
19 |
abort: |
|
19 | abort: unknown flags (0x04) in version 1 revlog 00changelog.i! | |
20 | [255] |
|
20 | [255] | |
21 |
|
21 | |||
22 | Unknown version is rejected |
|
22 | Unknown version is rejected | |
23 |
|
23 | |||
24 | >>> with open('.hg/store/00changelog.i', 'wb') as fh: |
|
24 | >>> with open('.hg/store/00changelog.i', 'wb') as fh: | |
25 | ... fh.write('\x00\x00\x00\x02') |
|
25 | ... fh.write('\x00\x00\x00\x02') | |
26 |
|
26 | |||
27 | $ hg log |
|
27 | $ hg log | |
28 | abort: index 00changelog.i unknown format 2! |
|
28 | abort: unknown version (2) in revlog 00changelog.i! | |
29 | [255] |
|
29 | [255] | |
30 |
|
30 | |||
31 |
$ |
|
31 | $ cd .. | |
32 |
|
32 | |||
33 | Test for CVE-2016-3630 |
|
33 | Test for CVE-2016-3630 | |
34 |
|
34 | |||
35 | $ hg init |
|
35 | $ hg init | |
36 |
|
36 | |||
37 | >>> open("a.i", "w").write( |
|
37 | >>> open("a.i", "w").write( | |
38 | ... """eJxjYGZgZIAAYQYGxhgom+k/FMx8YKx9ZUaKSOyqo4cnuKb8mbqHV5cBCVTMWb1Cwqkhe4Gsg9AD |
|
38 | ... """eJxjYGZgZIAAYQYGxhgom+k/FMx8YKx9ZUaKSOyqo4cnuKb8mbqHV5cBCVTMWb1Cwqkhe4Gsg9AD | |
39 | ... Joa3dYtcYYYBAQ8Qr4OqZAYRICPTSr5WKd/42rV36d+8/VmrNpv7NP1jQAXrQE4BqQUARngwVA==""" |
|
39 | ... Joa3dYtcYYYBAQ8Qr4OqZAYRICPTSr5WKd/42rV36d+8/VmrNpv7NP1jQAXrQE4BqQUARngwVA==""" | |
40 | ... .decode("base64").decode("zlib")) |
|
40 | ... .decode("base64").decode("zlib")) | |
41 |
|
41 | |||
42 | $ hg debugindex a.i |
|
42 | $ hg debugindex a.i | |
43 | rev offset length delta linkrev nodeid p1 p2 |
|
43 | rev offset length delta linkrev nodeid p1 p2 | |
44 | 0 0 19 -1 2 99e0332bd498 000000000000 000000000000 |
|
44 | 0 0 19 -1 2 99e0332bd498 000000000000 000000000000 | |
45 | 1 19 12 0 3 6674f57a23d8 99e0332bd498 000000000000 |
|
45 | 1 19 12 0 3 6674f57a23d8 99e0332bd498 000000000000 | |
46 | $ hg debugdata a.i 1 2>&1 | egrep 'Error:.*decoded' |
|
46 | $ hg debugdata a.i 1 2>&1 | egrep 'Error:.*decoded' | |
47 | (mercurial\.\w+\.mpatch\.)?mpatchError: patch cannot be decoded (re) |
|
47 | (mercurial\.\w+\.mpatch\.)?mpatchError: patch cannot be decoded (re) |
General Comments 0
You need to be logged in to leave comments.
Login now