Show More
@@ -1,565 +1,566 b'' | |||||
1 | import re |
|
1 | import re | |
2 | from itertools import chain |
|
2 | from itertools import chain | |
3 | from dulwich import objects |
|
3 | from dulwich import objects | |
4 | from dulwich.config import ConfigFile |
|
4 | from dulwich.config import ConfigFile | |
5 | from subprocess import Popen, PIPE |
|
5 | from subprocess import Popen, PIPE | |
6 | from io import BytesIO |
|
6 | from io import BytesIO | |
7 |
|
7 | |||
8 | from kallithea.lib.vcs.conf import settings |
|
8 | from kallithea.lib.vcs.conf import settings | |
9 | from kallithea.lib.vcs.backends.base import BaseChangeset, EmptyChangeset |
|
9 | from kallithea.lib.vcs.backends.base import BaseChangeset, EmptyChangeset | |
10 | from kallithea.lib.vcs.exceptions import ( |
|
10 | from kallithea.lib.vcs.exceptions import ( | |
11 | RepositoryError, ChangesetError, NodeDoesNotExistError, VCSError, |
|
11 | RepositoryError, ChangesetError, NodeDoesNotExistError, VCSError, | |
12 | ChangesetDoesNotExistError, ImproperArchiveTypeError |
|
12 | ChangesetDoesNotExistError, ImproperArchiveTypeError | |
13 | ) |
|
13 | ) | |
14 | from kallithea.lib.vcs.nodes import ( |
|
14 | from kallithea.lib.vcs.nodes import ( | |
15 | FileNode, DirNode, NodeKind, RootNode, SubModuleNode, |
|
15 | FileNode, DirNode, NodeKind, RootNode, SubModuleNode, | |
16 | ChangedFileNodesGenerator, AddedFileNodesGenerator, RemovedFileNodesGenerator |
|
16 | ChangedFileNodesGenerator, AddedFileNodesGenerator, RemovedFileNodesGenerator | |
17 | ) |
|
17 | ) | |
18 | from kallithea.lib.vcs.utils import ( |
|
18 | from kallithea.lib.vcs.utils import ( | |
19 | safe_unicode, safe_str, safe_int, date_fromtimestamp |
|
19 | safe_unicode, safe_str, safe_int, date_fromtimestamp | |
20 | ) |
|
20 | ) | |
21 | from kallithea.lib.vcs.utils.lazy import LazyProperty |
|
21 | from kallithea.lib.vcs.utils.lazy import LazyProperty | |
22 |
|
22 | |||
23 |
|
23 | |||
24 | class GitChangeset(BaseChangeset): |
|
24 | class GitChangeset(BaseChangeset): | |
25 | """ |
|
25 | """ | |
26 | Represents state of the repository at single revision. |
|
26 | Represents state of the repository at single revision. | |
27 | """ |
|
27 | """ | |
28 |
|
28 | |||
29 | def __init__(self, repository, revision): |
|
29 | def __init__(self, repository, revision): | |
30 | self._stat_modes = {} |
|
30 | self._stat_modes = {} | |
31 | self.repository = repository |
|
31 | self.repository = repository | |
32 | revision = safe_str(revision) |
|
32 | revision = safe_str(revision) | |
33 | try: |
|
33 | try: | |
34 | commit = self.repository._repo[revision] |
|
34 | commit = self.repository._repo[revision] | |
35 | if isinstance(commit, objects.Tag): |
|
35 | if isinstance(commit, objects.Tag): | |
36 | revision = safe_str(commit.object[1]) |
|
36 | revision = safe_str(commit.object[1]) | |
37 | commit = self.repository._repo.get_object(commit.object[1]) |
|
37 | commit = self.repository._repo.get_object(commit.object[1]) | |
38 | except KeyError: |
|
38 | except KeyError: | |
39 | raise RepositoryError("Cannot get object with id %s" % revision) |
|
39 | raise RepositoryError("Cannot get object with id %s" % revision) | |
40 | self.raw_id = revision |
|
40 | self.raw_id = revision | |
41 | self.id = self.raw_id |
|
41 | self.id = self.raw_id | |
42 | self.short_id = self.raw_id[:12] |
|
42 | self.short_id = self.raw_id[:12] | |
43 | self._commit = commit |
|
43 | self._commit = commit | |
44 | self._tree_id = commit.tree |
|
44 | self._tree_id = commit.tree | |
45 | self._committer_property = 'committer' |
|
45 | self._committer_property = 'committer' | |
46 | self._author_property = 'author' |
|
46 | self._author_property = 'author' | |
47 | self._date_property = 'commit_time' |
|
47 | self._date_property = 'commit_time' | |
48 | self._date_tz_property = 'commit_timezone' |
|
48 | self._date_tz_property = 'commit_timezone' | |
49 | self.revision = repository.revisions.index(revision) |
|
49 | self.revision = repository.revisions.index(revision) | |
50 |
|
50 | |||
51 | self.nodes = {} |
|
51 | self.nodes = {} | |
52 | self._paths = {} |
|
52 | self._paths = {} | |
53 |
|
53 | |||
54 | @LazyProperty |
|
54 | @LazyProperty | |
55 | def bookmarks(self): |
|
55 | def bookmarks(self): | |
56 | return () |
|
56 | return () | |
57 |
|
57 | |||
58 | @LazyProperty |
|
58 | @LazyProperty | |
59 | def message(self): |
|
59 | def message(self): | |
60 | return safe_unicode(self._commit.message) |
|
60 | return safe_unicode(self._commit.message) | |
61 |
|
61 | |||
62 | @LazyProperty |
|
62 | @LazyProperty | |
63 | def committer(self): |
|
63 | def committer(self): | |
64 | return safe_unicode(getattr(self._commit, self._committer_property)) |
|
64 | return safe_unicode(getattr(self._commit, self._committer_property)) | |
65 |
|
65 | |||
66 | @LazyProperty |
|
66 | @LazyProperty | |
67 | def author(self): |
|
67 | def author(self): | |
68 | return safe_unicode(getattr(self._commit, self._author_property)) |
|
68 | return safe_unicode(getattr(self._commit, self._author_property)) | |
69 |
|
69 | |||
70 | @LazyProperty |
|
70 | @LazyProperty | |
71 | def date(self): |
|
71 | def date(self): | |
72 | return date_fromtimestamp(getattr(self._commit, self._date_property), |
|
72 | return date_fromtimestamp(getattr(self._commit, self._date_property), | |
73 | getattr(self._commit, self._date_tz_property)) |
|
73 | getattr(self._commit, self._date_tz_property)) | |
74 |
|
74 | |||
75 | @LazyProperty |
|
75 | @LazyProperty | |
76 | def _timestamp(self): |
|
76 | def _timestamp(self): | |
77 | return getattr(self._commit, self._date_property) |
|
77 | return getattr(self._commit, self._date_property) | |
78 |
|
78 | |||
79 | @LazyProperty |
|
79 | @LazyProperty | |
80 | def status(self): |
|
80 | def status(self): | |
81 | """ |
|
81 | """ | |
82 | Returns modified, added, removed, deleted files for current changeset |
|
82 | Returns modified, added, removed, deleted files for current changeset | |
83 | """ |
|
83 | """ | |
84 | return self.changed, self.added, self.removed |
|
84 | return self.changed, self.added, self.removed | |
85 |
|
85 | |||
86 | @LazyProperty |
|
86 | @LazyProperty | |
87 | def tags(self): |
|
87 | def tags(self): | |
88 | _tags = [] |
|
88 | _tags = [] | |
89 | for tname, tsha in self.repository.tags.iteritems(): |
|
89 | for tname, tsha in self.repository.tags.iteritems(): | |
90 | if tsha == self.raw_id: |
|
90 | if tsha == self.raw_id: | |
91 | _tags.append(tname) |
|
91 | _tags.append(tname) | |
92 | return _tags |
|
92 | return _tags | |
93 |
|
93 | |||
94 | @LazyProperty |
|
94 | @LazyProperty | |
95 | def branch(self): |
|
95 | def branch(self): | |
96 | # Note: This function will return one branch name for the changeset - |
|
96 | # Note: This function will return one branch name for the changeset - | |
97 | # that might not make sense in Git where branches() is a better match |
|
97 | # that might not make sense in Git where branches() is a better match | |
98 | # for the basic model |
|
98 | # for the basic model | |
99 | heads = self.repository._heads(reverse=False) |
|
99 | heads = self.repository._heads(reverse=False) | |
100 | ref = heads.get(self.raw_id) |
|
100 | ref = heads.get(self.raw_id) | |
101 | if ref: |
|
101 | if ref: | |
102 | return safe_unicode(ref) |
|
102 | return safe_unicode(ref) | |
103 |
|
103 | |||
104 | @LazyProperty |
|
104 | @LazyProperty | |
105 | def branches(self): |
|
105 | def branches(self): | |
106 | heads = self.repository._heads(reverse=True) |
|
106 | heads = self.repository._heads(reverse=True) | |
107 | return [b for b in heads if heads[b] == self.raw_id] # FIXME: Inefficient ... and returning None! |
|
107 | return [b for b in heads if heads[b] == self.raw_id] # FIXME: Inefficient ... and returning None! | |
108 |
|
108 | |||
109 | def _fix_path(self, path): |
|
109 | def _fix_path(self, path): | |
110 | """ |
|
110 | """ | |
111 | Paths are stored without trailing slash so we need to get rid off it if |
|
111 | Paths are stored without trailing slash so we need to get rid off it if | |
112 | needed. |
|
112 | needed. | |
113 | """ |
|
113 | """ | |
114 | if path.endswith('/'): |
|
114 | if path.endswith('/'): | |
115 | path = path.rstrip('/') |
|
115 | path = path.rstrip('/') | |
116 | return path |
|
116 | return path | |
117 |
|
117 | |||
118 | def _get_id_for_path(self, path): |
|
118 | def _get_id_for_path(self, path): | |
119 | path = safe_str(path) |
|
119 | path = safe_str(path) | |
120 | # FIXME: Please, spare a couple of minutes and make those codes cleaner; |
|
120 | # FIXME: Please, spare a couple of minutes and make those codes cleaner; | |
121 | if path not in self._paths: |
|
121 | if path not in self._paths: | |
122 | path = path.strip('/') |
|
122 | path = path.strip('/') | |
123 | # set root tree |
|
123 | # set root tree | |
124 | tree = self.repository._repo[self._tree_id] |
|
124 | tree = self.repository._repo[self._tree_id] | |
125 | if path == '': |
|
125 | if path == '': | |
126 | self._paths[''] = tree.id |
|
126 | self._paths[''] = tree.id | |
127 | return tree.id |
|
127 | return tree.id | |
128 | splitted = path.split('/') |
|
128 | splitted = path.split('/') | |
129 | dirs, name = splitted[:-1], splitted[-1] |
|
129 | dirs, name = splitted[:-1], splitted[-1] | |
130 | curdir = '' |
|
130 | curdir = '' | |
131 |
|
131 | |||
132 | # initially extract things from root dir |
|
132 | # initially extract things from root dir | |
133 | for item, stat, id in tree.iteritems(): |
|
133 | for item, stat, id in tree.iteritems(): | |
134 | if curdir: |
|
134 | if curdir: | |
135 | name = '/'.join((curdir, item)) |
|
135 | name = '/'.join((curdir, item)) | |
136 | else: |
|
136 | else: | |
137 | name = item |
|
137 | name = item | |
138 | self._paths[name] = id |
|
138 | self._paths[name] = id | |
139 | self._stat_modes[name] = stat |
|
139 | self._stat_modes[name] = stat | |
140 |
|
140 | |||
141 | for dir in dirs: |
|
141 | for dir in dirs: | |
142 | if curdir: |
|
142 | if curdir: | |
143 | curdir = '/'.join((curdir, dir)) |
|
143 | curdir = '/'.join((curdir, dir)) | |
144 | else: |
|
144 | else: | |
145 | curdir = dir |
|
145 | curdir = dir | |
146 | dir_id = None |
|
146 | dir_id = None | |
147 | for item, stat, id in tree.iteritems(): |
|
147 | for item, stat, id in tree.iteritems(): | |
148 | if dir == item: |
|
148 | if dir == item: | |
149 | dir_id = id |
|
149 | dir_id = id | |
150 | if dir_id: |
|
150 | if dir_id: | |
151 | # Update tree |
|
151 | # Update tree | |
152 | tree = self.repository._repo[dir_id] |
|
152 | tree = self.repository._repo[dir_id] | |
153 | if not isinstance(tree, objects.Tree): |
|
153 | if not isinstance(tree, objects.Tree): | |
154 | raise ChangesetError('%s is not a directory' % curdir) |
|
154 | raise ChangesetError('%s is not a directory' % curdir) | |
155 | else: |
|
155 | else: | |
156 | raise ChangesetError('%s have not been found' % curdir) |
|
156 | raise ChangesetError('%s have not been found' % curdir) | |
157 |
|
157 | |||
158 | # cache all items from the given traversed tree |
|
158 | # cache all items from the given traversed tree | |
159 | for item, stat, id in tree.iteritems(): |
|
159 | for item, stat, id in tree.iteritems(): | |
160 | if curdir: |
|
160 | if curdir: | |
161 | name = '/'.join((curdir, item)) |
|
161 | name = '/'.join((curdir, item)) | |
162 | else: |
|
162 | else: | |
163 | name = item |
|
163 | name = item | |
164 | self._paths[name] = id |
|
164 | self._paths[name] = id | |
165 | self._stat_modes[name] = stat |
|
165 | self._stat_modes[name] = stat | |
166 | if path not in self._paths: |
|
166 | if path not in self._paths: | |
167 | raise NodeDoesNotExistError("There is no file nor directory " |
|
167 | raise NodeDoesNotExistError("There is no file nor directory " | |
168 | "at the given path '%s' at revision %s" |
|
168 | "at the given path '%s' at revision %s" | |
169 | % (path, safe_str(self.short_id))) |
|
169 | % (path, safe_str(self.short_id))) | |
170 | return self._paths[path] |
|
170 | return self._paths[path] | |
171 |
|
171 | |||
172 | def _get_kind(self, path): |
|
172 | def _get_kind(self, path): | |
173 | obj = self.repository._repo[self._get_id_for_path(path)] |
|
173 | obj = self.repository._repo[self._get_id_for_path(path)] | |
174 | if isinstance(obj, objects.Blob): |
|
174 | if isinstance(obj, objects.Blob): | |
175 | return NodeKind.FILE |
|
175 | return NodeKind.FILE | |
176 | elif isinstance(obj, objects.Tree): |
|
176 | elif isinstance(obj, objects.Tree): | |
177 | return NodeKind.DIR |
|
177 | return NodeKind.DIR | |
178 |
|
178 | |||
179 | def _get_filectx(self, path): |
|
179 | def _get_filectx(self, path): | |
180 | path = self._fix_path(path) |
|
180 | path = self._fix_path(path) | |
181 | if self._get_kind(path) != NodeKind.FILE: |
|
181 | if self._get_kind(path) != NodeKind.FILE: | |
182 | raise ChangesetError("File does not exist for revision %s at " |
|
182 | raise ChangesetError("File does not exist for revision %s at " | |
183 | " '%s'" % (self.raw_id, path)) |
|
183 | " '%s'" % (self.raw_id, path)) | |
184 | return path |
|
184 | return path | |
185 |
|
185 | |||
186 | def _get_file_nodes(self): |
|
186 | def _get_file_nodes(self): | |
187 | return chain(*(t[2] for t in self.walk())) |
|
187 | return chain(*(t[2] for t in self.walk())) | |
188 |
|
188 | |||
189 | @LazyProperty |
|
189 | @LazyProperty | |
190 | def parents(self): |
|
190 | def parents(self): | |
191 | """ |
|
191 | """ | |
192 | Returns list of parents changesets. |
|
192 | Returns list of parents changesets. | |
193 | """ |
|
193 | """ | |
194 | return [self.repository.get_changeset(parent) |
|
194 | return [self.repository.get_changeset(parent) | |
195 | for parent in self._commit.parents] |
|
195 | for parent in self._commit.parents] | |
196 |
|
196 | |||
197 | @LazyProperty |
|
197 | @LazyProperty | |
198 | def children(self): |
|
198 | def children(self): | |
199 | """ |
|
199 | """ | |
200 | Returns list of children changesets. |
|
200 | Returns list of children changesets. | |
201 | """ |
|
201 | """ | |
202 | rev_filter = settings.GIT_REV_FILTER |
|
202 | rev_filter = settings.GIT_REV_FILTER | |
203 | so, se = self.repository.run_git_command( |
|
203 | so, se = self.repository.run_git_command( | |
204 | ['rev-list', rev_filter, '--children'] |
|
204 | ['rev-list', rev_filter, '--children'] | |
205 | ) |
|
205 | ) | |
206 |
|
206 | |||
207 | children = [] |
|
207 | children = [] | |
208 | pat = re.compile(r'^%s' % self.raw_id) |
|
208 | pat = re.compile(r'^%s' % self.raw_id) | |
209 | for l in so.splitlines(): |
|
209 | for l in so.splitlines(): | |
210 | if pat.match(l): |
|
210 | if pat.match(l): | |
211 | childs = l.split(' ')[1:] |
|
211 | childs = l.split(' ')[1:] | |
212 | children.extend(childs) |
|
212 | children.extend(childs) | |
213 | return [self.repository.get_changeset(cs) for cs in children] |
|
213 | return [self.repository.get_changeset(cs) for cs in children] | |
214 |
|
214 | |||
215 | def next(self, branch=None): |
|
215 | def next(self, branch=None): | |
216 | if branch and self.branch != branch: |
|
216 | if branch and self.branch != branch: | |
217 | raise VCSError('Branch option used on changeset not belonging ' |
|
217 | raise VCSError('Branch option used on changeset not belonging ' | |
218 | 'to that branch') |
|
218 | 'to that branch') | |
219 |
|
219 | |||
220 | cs = self |
|
220 | cs = self | |
221 | while True: |
|
221 | while True: | |
222 | try: |
|
222 | try: | |
223 | next_ = cs.revision + 1 |
|
223 | next_ = cs.revision + 1 | |
224 | next_rev = cs.repository.revisions[next_] |
|
224 | next_rev = cs.repository.revisions[next_] | |
225 | except IndexError: |
|
225 | except IndexError: | |
226 | raise ChangesetDoesNotExistError |
|
226 | raise ChangesetDoesNotExistError | |
227 | cs = cs.repository.get_changeset(next_rev) |
|
227 | cs = cs.repository.get_changeset(next_rev) | |
228 |
|
228 | |||
229 | if not branch or branch == cs.branch: |
|
229 | if not branch or branch == cs.branch: | |
230 | return cs |
|
230 | return cs | |
231 |
|
231 | |||
232 | def prev(self, branch=None): |
|
232 | def prev(self, branch=None): | |
233 | if branch and self.branch != branch: |
|
233 | if branch and self.branch != branch: | |
234 | raise VCSError('Branch option used on changeset not belonging ' |
|
234 | raise VCSError('Branch option used on changeset not belonging ' | |
235 | 'to that branch') |
|
235 | 'to that branch') | |
236 |
|
236 | |||
237 | cs = self |
|
237 | cs = self | |
238 | while True: |
|
238 | while True: | |
239 | try: |
|
239 | try: | |
240 | prev_ = cs.revision - 1 |
|
240 | prev_ = cs.revision - 1 | |
241 | if prev_ < 0: |
|
241 | if prev_ < 0: | |
242 | raise IndexError |
|
242 | raise IndexError | |
243 | prev_rev = cs.repository.revisions[prev_] |
|
243 | prev_rev = cs.repository.revisions[prev_] | |
244 | except IndexError: |
|
244 | except IndexError: | |
245 | raise ChangesetDoesNotExistError |
|
245 | raise ChangesetDoesNotExistError | |
246 | cs = cs.repository.get_changeset(prev_rev) |
|
246 | cs = cs.repository.get_changeset(prev_rev) | |
247 |
|
247 | |||
248 | if not branch or branch == cs.branch: |
|
248 | if not branch or branch == cs.branch: | |
249 | return cs |
|
249 | return cs | |
250 |
|
250 | |||
251 | def diff(self, ignore_whitespace=True, context=3): |
|
251 | def diff(self, ignore_whitespace=True, context=3): | |
252 | rev1 = self.parents[0] if self.parents else self.repository.EMPTY_CHANGESET |
|
252 | rev1 = self.parents[0] if self.parents else self.repository.EMPTY_CHANGESET | |
253 | rev2 = self |
|
253 | rev2 = self | |
254 | return ''.join(self.repository.get_diff(rev1, rev2, |
|
254 | return ''.join(self.repository.get_diff(rev1, rev2, | |
255 | ignore_whitespace=ignore_whitespace, |
|
255 | ignore_whitespace=ignore_whitespace, | |
256 | context=context)) |
|
256 | context=context)) | |
257 |
|
257 | |||
258 | def get_file_mode(self, path): |
|
258 | def get_file_mode(self, path): | |
259 | """ |
|
259 | """ | |
260 | Returns stat mode of the file at the given ``path``. |
|
260 | Returns stat mode of the file at the given ``path``. | |
261 | """ |
|
261 | """ | |
262 | # ensure path is traversed |
|
262 | # ensure path is traversed | |
263 | path = safe_str(path) |
|
263 | path = safe_str(path) | |
264 | self._get_id_for_path(path) |
|
264 | self._get_id_for_path(path) | |
265 | return self._stat_modes[path] |
|
265 | return self._stat_modes[path] | |
266 |
|
266 | |||
267 | def get_file_content(self, path): |
|
267 | def get_file_content(self, path): | |
268 | """ |
|
268 | """ | |
269 | Returns content of the file at given ``path``. |
|
269 | Returns content of the file at given ``path``. | |
270 | """ |
|
270 | """ | |
271 | id = self._get_id_for_path(path) |
|
271 | id = self._get_id_for_path(path) | |
272 | blob = self.repository._repo[id] |
|
272 | blob = self.repository._repo[id] | |
273 | return blob.as_pretty_string() |
|
273 | return blob.as_pretty_string() | |
274 |
|
274 | |||
275 | def get_file_size(self, path): |
|
275 | def get_file_size(self, path): | |
276 | """ |
|
276 | """ | |
277 | Returns size of the file at given ``path``. |
|
277 | Returns size of the file at given ``path``. | |
278 | """ |
|
278 | """ | |
279 | id = self._get_id_for_path(path) |
|
279 | id = self._get_id_for_path(path) | |
280 | blob = self.repository._repo[id] |
|
280 | blob = self.repository._repo[id] | |
281 | return blob.raw_length() |
|
281 | return blob.raw_length() | |
282 |
|
282 | |||
283 | def get_file_changeset(self, path): |
|
283 | def get_file_changeset(self, path): | |
284 | """ |
|
284 | """ | |
285 | Returns last commit of the file at the given ``path``. |
|
285 | Returns last commit of the file at the given ``path``. | |
286 | """ |
|
286 | """ | |
287 | return self.get_file_history(path, limit=1)[0] |
|
287 | return self.get_file_history(path, limit=1)[0] | |
288 |
|
288 | |||
289 | def get_file_history(self, path, limit=None): |
|
289 | def get_file_history(self, path, limit=None): | |
290 | """ |
|
290 | """ | |
291 | Returns history of file as reversed list of ``Changeset`` objects for |
|
291 | Returns history of file as reversed list of ``Changeset`` objects for | |
292 | which file at given ``path`` has been modified. |
|
292 | which file at given ``path`` has been modified. | |
293 |
|
293 | |||
294 | TODO: This function now uses os underlying 'git' and 'grep' commands |
|
294 | TODO: This function now uses os underlying 'git' and 'grep' commands | |
295 | which is generally not good. Should be replaced with algorithm |
|
295 | which is generally not good. Should be replaced with algorithm | |
296 | iterating commits. |
|
296 | iterating commits. | |
297 | """ |
|
297 | """ | |
298 | self._get_filectx(path) |
|
298 | self._get_filectx(path) | |
299 | cs_id = safe_str(self.id) |
|
299 | cs_id = safe_str(self.id) | |
300 | f_path = safe_str(path) |
|
300 | f_path = safe_str(path) | |
301 |
|
301 | |||
302 | if limit is not None: |
|
302 | if limit is not None: | |
303 | cmd = ['log', '-n', str(safe_int(limit, 0)), |
|
303 | cmd = ['log', '-n', str(safe_int(limit, 0)), | |
304 | '--pretty=format:%H', '-s', cs_id, '--', f_path] |
|
304 | '--pretty=format:%H', '-s', cs_id, '--', f_path] | |
305 |
|
305 | |||
306 | else: |
|
306 | else: | |
307 | cmd = ['log', |
|
307 | cmd = ['log', | |
308 | '--pretty=format:%H', '-s', cs_id, '--', f_path] |
|
308 | '--pretty=format:%H', '-s', cs_id, '--', f_path] | |
309 | so, se = self.repository.run_git_command(cmd) |
|
309 | so, se = self.repository.run_git_command(cmd) | |
310 | ids = re.findall(r'[0-9a-fA-F]{40}', so) |
|
310 | ids = re.findall(r'[0-9a-fA-F]{40}', so) | |
311 | return [self.repository.get_changeset(sha) for sha in ids] |
|
311 | return [self.repository.get_changeset(sha) for sha in ids] | |
312 |
|
312 | |||
313 | def get_file_history_2(self, path): |
|
313 | def get_file_history_2(self, path): | |
314 | """ |
|
314 | """ | |
315 | Returns history of file as reversed list of ``Changeset`` objects for |
|
315 | Returns history of file as reversed list of ``Changeset`` objects for | |
316 | which file at given ``path`` has been modified. |
|
316 | which file at given ``path`` has been modified. | |
317 |
|
317 | |||
318 | """ |
|
318 | """ | |
319 | self._get_filectx(path) |
|
319 | self._get_filectx(path) | |
320 | from dulwich.walk import Walker |
|
320 | from dulwich.walk import Walker | |
321 | include = [self.id] |
|
321 | include = [self.id] | |
322 | walker = Walker(self.repository._repo.object_store, include, |
|
322 | walker = Walker(self.repository._repo.object_store, include, | |
323 | paths=[path], max_entries=1) |
|
323 | paths=[path], max_entries=1) | |
324 | return [self.repository.get_changeset(sha) |
|
324 | return [self.repository.get_changeset(sha) | |
325 | for sha in (x.commit.id for x in walker)] |
|
325 | for sha in (x.commit.id for x in walker)] | |
326 |
|
326 | |||
327 | def get_file_annotate(self, path): |
|
327 | def get_file_annotate(self, path): | |
328 | """ |
|
328 | """ | |
329 | Returns a generator of four element tuples with |
|
329 | Returns a generator of four element tuples with | |
330 | lineno, sha, changeset lazy loader and line |
|
330 | lineno, sha, changeset lazy loader and line | |
331 |
|
331 | |||
332 | TODO: This function now uses os underlying 'git' command which is |
|
332 | TODO: This function now uses os underlying 'git' command which is | |
333 | generally not good. Should be replaced with algorithm iterating |
|
333 | generally not good. Should be replaced with algorithm iterating | |
334 | commits. |
|
334 | commits. | |
335 | """ |
|
335 | """ | |
336 | cmd = ['blame', '-l', '--root', '-r', self.id, '--', path] |
|
336 | cmd = ['blame', '-l', '--root', '-r', self.id, '--', path] | |
337 | # -l ==> outputs long shas (and we need all 40 characters) |
|
337 | # -l ==> outputs long shas (and we need all 40 characters) | |
338 | # --root ==> doesn't put '^' character for boundaries |
|
338 | # --root ==> doesn't put '^' character for boundaries | |
339 | # -r sha ==> blames for the given revision |
|
339 | # -r sha ==> blames for the given revision | |
340 | so, se = self.repository.run_git_command(cmd) |
|
340 | so, se = self.repository.run_git_command(cmd) | |
341 |
|
341 | |||
342 | for i, blame_line in enumerate(so.split('\n')[:-1]): |
|
342 | for i, blame_line in enumerate(so.split('\n')[:-1]): | |
343 | ln_no = i + 1 |
|
343 | ln_no = i + 1 | |
344 | sha, line = re.split(r' ', blame_line, 1) |
|
344 | sha, line = re.split(r' ', blame_line, 1) | |
345 | yield (ln_no, sha, lambda: self.repository.get_changeset(sha), line) |
|
345 | yield (ln_no, sha, lambda: self.repository.get_changeset(sha), line) | |
346 |
|
346 | |||
347 | def fill_archive(self, stream=None, kind='tgz', prefix=None, |
|
347 | def fill_archive(self, stream=None, kind='tgz', prefix=None, | |
348 | subrepos=False): |
|
348 | subrepos=False): | |
349 | """ |
|
349 | """ | |
350 | Fills up given stream. |
|
350 | Fills up given stream. | |
351 |
|
351 | |||
352 | :param stream: file like object. |
|
352 | :param stream: file like object. | |
353 | :param kind: one of following: ``zip``, ``tgz`` or ``tbz2``. |
|
353 | :param kind: one of following: ``zip``, ``tgz`` or ``tbz2``. | |
354 | Default: ``tgz``. |
|
354 | Default: ``tgz``. | |
355 | :param prefix: name of root directory in archive. |
|
355 | :param prefix: name of root directory in archive. | |
356 | Default is repository name and changeset's raw_id joined with dash |
|
356 | Default is repository name and changeset's raw_id joined with dash | |
357 | (``repo-tip.<KIND>``). |
|
357 | (``repo-tip.<KIND>``). | |
358 | :param subrepos: include subrepos in this archive. |
|
358 | :param subrepos: include subrepos in this archive. | |
359 |
|
359 | |||
360 | :raise ImproperArchiveTypeError: If given kind is wrong. |
|
360 | :raise ImproperArchiveTypeError: If given kind is wrong. | |
361 | :raise VcsError: If given stream is None |
|
361 | :raise VcsError: If given stream is None | |
362 |
|
362 | |||
363 | """ |
|
363 | """ | |
364 | allowed_kinds = settings.ARCHIVE_SPECS.keys() |
|
364 | allowed_kinds = settings.ARCHIVE_SPECS.keys() | |
365 | if kind not in allowed_kinds: |
|
365 | if kind not in allowed_kinds: | |
366 | raise ImproperArchiveTypeError('Archive kind not supported use one' |
|
366 | raise ImproperArchiveTypeError('Archive kind not supported use one' | |
367 | 'of %s', allowed_kinds) |
|
367 | 'of %s', allowed_kinds) | |
368 |
|
368 | |||
369 | if prefix is None: |
|
369 | if prefix is None: | |
370 | prefix = '%s-%s' % (self.repository.name, self.short_id) |
|
370 | prefix = '%s-%s' % (self.repository.name, self.short_id) | |
371 | elif prefix.startswith('/'): |
|
371 | elif prefix.startswith('/'): | |
372 | raise VCSError("Prefix cannot start with leading slash") |
|
372 | raise VCSError("Prefix cannot start with leading slash") | |
373 | elif prefix.strip() == '': |
|
373 | elif prefix.strip() == '': | |
374 | raise VCSError("Prefix cannot be empty") |
|
374 | raise VCSError("Prefix cannot be empty") | |
375 |
|
375 | |||
376 | if kind == 'zip': |
|
376 | if kind == 'zip': | |
377 | frmt = 'zip' |
|
377 | frmt = 'zip' | |
378 | else: |
|
378 | else: | |
379 | frmt = 'tar' |
|
379 | frmt = 'tar' | |
380 | _git_path = settings.GIT_EXECUTABLE_PATH |
|
380 | _git_path = settings.GIT_EXECUTABLE_PATH | |
381 | cmd = '%s archive --format=%s --prefix=%s/ %s' % (_git_path, |
|
381 | cmd = '%s archive --format=%s --prefix=%s/ %s' % (_git_path, | |
382 | frmt, prefix, self.raw_id) |
|
382 | frmt, prefix, self.raw_id) | |
383 | if kind == 'tgz': |
|
383 | if kind == 'tgz': | |
384 | cmd += ' | gzip -9' |
|
384 | cmd += ' | gzip -9' | |
385 | elif kind == 'tbz2': |
|
385 | elif kind == 'tbz2': | |
386 | cmd += ' | bzip2 -9' |
|
386 | cmd += ' | bzip2 -9' | |
387 |
|
387 | |||
388 | if stream is None: |
|
388 | if stream is None: | |
389 | raise VCSError('You need to pass in a valid stream for filling' |
|
389 | raise VCSError('You need to pass in a valid stream for filling' | |
390 | ' with archival data') |
|
390 | ' with archival data') | |
391 | popen = Popen(cmd, stdout=PIPE, stderr=PIPE, shell=True, |
|
391 | popen = Popen(cmd, stdout=PIPE, stderr=PIPE, shell=True, | |
392 | cwd=self.repository.path) |
|
392 | cwd=self.repository.path) | |
393 |
|
393 | |||
394 | buffer_size = 1024 * 8 |
|
394 | buffer_size = 1024 * 8 | |
395 | chunk = popen.stdout.read(buffer_size) |
|
395 | chunk = popen.stdout.read(buffer_size) | |
396 | while chunk: |
|
396 | while chunk: | |
397 | stream.write(chunk) |
|
397 | stream.write(chunk) | |
398 | chunk = popen.stdout.read(buffer_size) |
|
398 | chunk = popen.stdout.read(buffer_size) | |
399 | # Make sure all descriptors would be read |
|
399 | # Make sure all descriptors would be read | |
400 | popen.communicate() |
|
400 | popen.communicate() | |
401 |
|
401 | |||
402 | def get_nodes(self, path): |
|
402 | def get_nodes(self, path): | |
403 | if self._get_kind(path) != NodeKind.DIR: |
|
403 | if self._get_kind(path) != NodeKind.DIR: | |
404 | raise ChangesetError("Directory does not exist for revision %s at " |
|
404 | raise ChangesetError("Directory does not exist for revision %s at " | |
405 | " '%s'" % (self.revision, path)) |
|
405 | " '%s'" % (self.revision, path)) | |
406 | path = self._fix_path(path) |
|
406 | path = self._fix_path(path) | |
407 | id = self._get_id_for_path(path) |
|
407 | id = self._get_id_for_path(path) | |
408 | tree = self.repository._repo[id] |
|
408 | tree = self.repository._repo[id] | |
409 | dirnodes = [] |
|
409 | dirnodes = [] | |
410 | filenodes = [] |
|
410 | filenodes = [] | |
411 | als = self.repository.alias |
|
411 | als = self.repository.alias | |
412 | for name, stat, id in tree.iteritems(): |
|
412 | for name, stat, id in tree.iteritems(): | |
|
413 | if path != '': | |||
|
414 | obj_path = '/'.join((path, name)) | |||
|
415 | else: | |||
|
416 | obj_path = name | |||
413 | if objects.S_ISGITLINK(stat): |
|
417 | if objects.S_ISGITLINK(stat): | |
414 | cf = ConfigFile.from_file(BytesIO(self.repository._repo.get_object(tree['.gitmodules'][1]).data)) |
|
418 | root_tree = self.repository._repo[self._tree_id] | |
415 | url = cf.get(('submodule', name), 'url') |
|
419 | cf = ConfigFile.from_file(BytesIO(self.repository._repo.get_object(root_tree['.gitmodules'][1]).data)) | |
416 | dirnodes.append(SubModuleNode(name, url=url, changeset=id, |
|
420 | url = cf.get(('submodule', obj_path), 'url') | |
|
421 | dirnodes.append(SubModuleNode(obj_path, url=url, changeset=id, | |||
417 | alias=als)) |
|
422 | alias=als)) | |
418 | continue |
|
423 | continue | |
419 |
|
424 | |||
420 | obj = self.repository._repo.get_object(id) |
|
425 | obj = self.repository._repo.get_object(id) | |
421 | if path != '': |
|
|||
422 | obj_path = '/'.join((path, name)) |
|
|||
423 | else: |
|
|||
424 | obj_path = name |
|
|||
425 | if obj_path not in self._stat_modes: |
|
426 | if obj_path not in self._stat_modes: | |
426 | self._stat_modes[obj_path] = stat |
|
427 | self._stat_modes[obj_path] = stat | |
427 | if isinstance(obj, objects.Tree): |
|
428 | if isinstance(obj, objects.Tree): | |
428 | dirnodes.append(DirNode(obj_path, changeset=self)) |
|
429 | dirnodes.append(DirNode(obj_path, changeset=self)) | |
429 | elif isinstance(obj, objects.Blob): |
|
430 | elif isinstance(obj, objects.Blob): | |
430 | filenodes.append(FileNode(obj_path, changeset=self, mode=stat)) |
|
431 | filenodes.append(FileNode(obj_path, changeset=self, mode=stat)) | |
431 | else: |
|
432 | else: | |
432 | raise ChangesetError("Requested object should be Tree " |
|
433 | raise ChangesetError("Requested object should be Tree " | |
433 | "or Blob, is %r" % type(obj)) |
|
434 | "or Blob, is %r" % type(obj)) | |
434 | nodes = dirnodes + filenodes |
|
435 | nodes = dirnodes + filenodes | |
435 | for node in nodes: |
|
436 | for node in nodes: | |
436 | if node.path not in self.nodes: |
|
437 | if node.path not in self.nodes: | |
437 | self.nodes[node.path] = node |
|
438 | self.nodes[node.path] = node | |
438 | nodes.sort() |
|
439 | nodes.sort() | |
439 | return nodes |
|
440 | return nodes | |
440 |
|
441 | |||
441 | def get_node(self, path): |
|
442 | def get_node(self, path): | |
442 | if isinstance(path, unicode): |
|
443 | if isinstance(path, unicode): | |
443 | path = path.encode('utf-8') |
|
444 | path = path.encode('utf-8') | |
444 | path = self._fix_path(path) |
|
445 | path = self._fix_path(path) | |
445 | if path not in self.nodes: |
|
446 | if path not in self.nodes: | |
446 | try: |
|
447 | try: | |
447 | id_ = self._get_id_for_path(path) |
|
448 | id_ = self._get_id_for_path(path) | |
448 | except ChangesetError: |
|
449 | except ChangesetError: | |
449 | raise NodeDoesNotExistError("Cannot find one of parents' " |
|
450 | raise NodeDoesNotExistError("Cannot find one of parents' " | |
450 | "directories for a given path: %s" % path) |
|
451 | "directories for a given path: %s" % path) | |
451 |
|
452 | |||
452 | _GL = lambda m: m and objects.S_ISGITLINK(m) |
|
453 | _GL = lambda m: m and objects.S_ISGITLINK(m) | |
453 | if _GL(self._stat_modes.get(path)): |
|
454 | if _GL(self._stat_modes.get(path)): | |
454 | tree = self.repository._repo[self._tree_id] |
|
455 | tree = self.repository._repo[self._tree_id] | |
455 | cf = ConfigFile.from_file(BytesIO(self.repository._repo.get_object(tree['.gitmodules'][1]).data)) |
|
456 | cf = ConfigFile.from_file(BytesIO(self.repository._repo.get_object(tree['.gitmodules'][1]).data)) | |
456 | url = cf.get(('submodule', path), 'url') |
|
457 | url = cf.get(('submodule', path), 'url') | |
457 | node = SubModuleNode(path, url=url, changeset=id_, |
|
458 | node = SubModuleNode(path, url=url, changeset=id_, | |
458 | alias=self.repository.alias) |
|
459 | alias=self.repository.alias) | |
459 | else: |
|
460 | else: | |
460 | obj = self.repository._repo.get_object(id_) |
|
461 | obj = self.repository._repo.get_object(id_) | |
461 |
|
462 | |||
462 | if isinstance(obj, objects.Tree): |
|
463 | if isinstance(obj, objects.Tree): | |
463 | if path == '': |
|
464 | if path == '': | |
464 | node = RootNode(changeset=self) |
|
465 | node = RootNode(changeset=self) | |
465 | else: |
|
466 | else: | |
466 | node = DirNode(path, changeset=self) |
|
467 | node = DirNode(path, changeset=self) | |
467 | node._tree = obj |
|
468 | node._tree = obj | |
468 | elif isinstance(obj, objects.Blob): |
|
469 | elif isinstance(obj, objects.Blob): | |
469 | node = FileNode(path, changeset=self) |
|
470 | node = FileNode(path, changeset=self) | |
470 | node._blob = obj |
|
471 | node._blob = obj | |
471 | else: |
|
472 | else: | |
472 | raise NodeDoesNotExistError("There is no file nor directory " |
|
473 | raise NodeDoesNotExistError("There is no file nor directory " | |
473 | "at the given path '%s' at revision %s" |
|
474 | "at the given path '%s' at revision %s" | |
474 | % (path, self.short_id)) |
|
475 | % (path, self.short_id)) | |
475 | # cache node |
|
476 | # cache node | |
476 | self.nodes[path] = node |
|
477 | self.nodes[path] = node | |
477 | return self.nodes[path] |
|
478 | return self.nodes[path] | |
478 |
|
479 | |||
479 | @LazyProperty |
|
480 | @LazyProperty | |
480 | def affected_files(self): |
|
481 | def affected_files(self): | |
481 | """ |
|
482 | """ | |
482 | Gets a fast accessible file changes for given changeset |
|
483 | Gets a fast accessible file changes for given changeset | |
483 | """ |
|
484 | """ | |
484 | added, modified, deleted = self._changes_cache |
|
485 | added, modified, deleted = self._changes_cache | |
485 | return list(added.union(modified).union(deleted)) |
|
486 | return list(added.union(modified).union(deleted)) | |
486 |
|
487 | |||
487 | @LazyProperty |
|
488 | @LazyProperty | |
488 | def _diff_name_status(self): |
|
489 | def _diff_name_status(self): | |
489 | output = [] |
|
490 | output = [] | |
490 | for parent in self.parents: |
|
491 | for parent in self.parents: | |
491 | cmd = ['diff', '--name-status', parent.raw_id, self.raw_id, |
|
492 | cmd = ['diff', '--name-status', parent.raw_id, self.raw_id, | |
492 | '--encoding=utf8'] |
|
493 | '--encoding=utf8'] | |
493 | so, se = self.repository.run_git_command(cmd) |
|
494 | so, se = self.repository.run_git_command(cmd) | |
494 | output.append(so.strip()) |
|
495 | output.append(so.strip()) | |
495 | return '\n'.join(output) |
|
496 | return '\n'.join(output) | |
496 |
|
497 | |||
497 | @LazyProperty |
|
498 | @LazyProperty | |
498 | def _changes_cache(self): |
|
499 | def _changes_cache(self): | |
499 | added = set() |
|
500 | added = set() | |
500 | modified = set() |
|
501 | modified = set() | |
501 | deleted = set() |
|
502 | deleted = set() | |
502 | _r = self.repository._repo |
|
503 | _r = self.repository._repo | |
503 |
|
504 | |||
504 | parents = self.parents |
|
505 | parents = self.parents | |
505 | if not self.parents: |
|
506 | if not self.parents: | |
506 | parents = [EmptyChangeset()] |
|
507 | parents = [EmptyChangeset()] | |
507 | for parent in parents: |
|
508 | for parent in parents: | |
508 | if isinstance(parent, EmptyChangeset): |
|
509 | if isinstance(parent, EmptyChangeset): | |
509 | oid = None |
|
510 | oid = None | |
510 | else: |
|
511 | else: | |
511 | oid = _r[parent.raw_id].tree |
|
512 | oid = _r[parent.raw_id].tree | |
512 | changes = _r.object_store.tree_changes(oid, _r[self.raw_id].tree) |
|
513 | changes = _r.object_store.tree_changes(oid, _r[self.raw_id].tree) | |
513 | for (oldpath, newpath), (_, _), (_, _) in changes: |
|
514 | for (oldpath, newpath), (_, _), (_, _) in changes: | |
514 | if newpath and oldpath: |
|
515 | if newpath and oldpath: | |
515 | modified.add(newpath) |
|
516 | modified.add(newpath) | |
516 | elif newpath and not oldpath: |
|
517 | elif newpath and not oldpath: | |
517 | added.add(newpath) |
|
518 | added.add(newpath) | |
518 | elif not newpath and oldpath: |
|
519 | elif not newpath and oldpath: | |
519 | deleted.add(oldpath) |
|
520 | deleted.add(oldpath) | |
520 | return added, modified, deleted |
|
521 | return added, modified, deleted | |
521 |
|
522 | |||
522 | def _get_paths_for_status(self, status): |
|
523 | def _get_paths_for_status(self, status): | |
523 | """ |
|
524 | """ | |
524 | Returns sorted list of paths for given ``status``. |
|
525 | Returns sorted list of paths for given ``status``. | |
525 |
|
526 | |||
526 | :param status: one of: *added*, *modified* or *deleted* |
|
527 | :param status: one of: *added*, *modified* or *deleted* | |
527 | """ |
|
528 | """ | |
528 | added, modified, deleted = self._changes_cache |
|
529 | added, modified, deleted = self._changes_cache | |
529 | return sorted({ |
|
530 | return sorted({ | |
530 | 'added': list(added), |
|
531 | 'added': list(added), | |
531 | 'modified': list(modified), |
|
532 | 'modified': list(modified), | |
532 | 'deleted': list(deleted)}[status] |
|
533 | 'deleted': list(deleted)}[status] | |
533 | ) |
|
534 | ) | |
534 |
|
535 | |||
535 | @LazyProperty |
|
536 | @LazyProperty | |
536 | def added(self): |
|
537 | def added(self): | |
537 | """ |
|
538 | """ | |
538 | Returns list of added ``FileNode`` objects. |
|
539 | Returns list of added ``FileNode`` objects. | |
539 | """ |
|
540 | """ | |
540 | if not self.parents: |
|
541 | if not self.parents: | |
541 | return list(self._get_file_nodes()) |
|
542 | return list(self._get_file_nodes()) | |
542 | return AddedFileNodesGenerator([n for n in |
|
543 | return AddedFileNodesGenerator([n for n in | |
543 | self._get_paths_for_status('added')], self) |
|
544 | self._get_paths_for_status('added')], self) | |
544 |
|
545 | |||
545 | @LazyProperty |
|
546 | @LazyProperty | |
546 | def changed(self): |
|
547 | def changed(self): | |
547 | """ |
|
548 | """ | |
548 | Returns list of modified ``FileNode`` objects. |
|
549 | Returns list of modified ``FileNode`` objects. | |
549 | """ |
|
550 | """ | |
550 | if not self.parents: |
|
551 | if not self.parents: | |
551 | return [] |
|
552 | return [] | |
552 | return ChangedFileNodesGenerator([n for n in |
|
553 | return ChangedFileNodesGenerator([n for n in | |
553 | self._get_paths_for_status('modified')], self) |
|
554 | self._get_paths_for_status('modified')], self) | |
554 |
|
555 | |||
555 | @LazyProperty |
|
556 | @LazyProperty | |
556 | def removed(self): |
|
557 | def removed(self): | |
557 | """ |
|
558 | """ | |
558 | Returns list of removed ``FileNode`` objects. |
|
559 | Returns list of removed ``FileNode`` objects. | |
559 | """ |
|
560 | """ | |
560 | if not self.parents: |
|
561 | if not self.parents: | |
561 | return [] |
|
562 | return [] | |
562 | return RemovedFileNodesGenerator([n for n in |
|
563 | return RemovedFileNodesGenerator([n for n in | |
563 | self._get_paths_for_status('deleted')], self) |
|
564 | self._get_paths_for_status('deleted')], self) | |
564 |
|
565 | |||
565 | extra = {} |
|
566 | extra = {} |
General Comments 0
You need to be logged in to leave comments.
Login now