Show More
@@ -1,332 +1,345 | |||
|
1 | 1 | # archival.py - revision archival for mercurial |
|
2 | 2 | # |
|
3 | 3 | # Copyright 2006 Vadim Gelfer <vadim.gelfer@gmail.com> |
|
4 | 4 | # |
|
5 | 5 | # This software may be used and distributed according to the terms of the |
|
6 | 6 | # GNU General Public License version 2 or any later version. |
|
7 | 7 | |
|
8 | from i18n import _ | |
|
9 | import match as matchmod | |
|
10 |
import c |
|
|
11 | import scmutil, util, encoding | |
|
12 | import cStringIO, os, tarfile, time, zipfile | |
|
13 | import zlib, gzip | |
|
8 | from __future__ import absolute_import | |
|
9 | ||
|
10 | import cStringIO | |
|
11 | import gzip | |
|
12 | import os | |
|
14 | 13 | import struct |
|
15 |
import |
|
|
14 | import tarfile | |
|
15 | import time | |
|
16 | import zipfile | |
|
17 | import zlib | |
|
18 | ||
|
19 | from .i18n import _ | |
|
20 | ||
|
21 | from . import ( | |
|
22 | cmdutil, | |
|
23 | encoding, | |
|
24 | error, | |
|
25 | match as matchmod, | |
|
26 | scmutil, | |
|
27 | util, | |
|
28 | ) | |
|
16 | 29 | |
|
17 | 30 | # from unzip source code: |
|
18 | 31 | _UNX_IFREG = 0x8000 |
|
19 | 32 | _UNX_IFLNK = 0xa000 |
|
20 | 33 | |
|
21 | 34 | def tidyprefix(dest, kind, prefix): |
|
22 | 35 | '''choose prefix to use for names in archive. make sure prefix is |
|
23 | 36 | safe for consumers.''' |
|
24 | 37 | |
|
25 | 38 | if prefix: |
|
26 | 39 | prefix = util.normpath(prefix) |
|
27 | 40 | else: |
|
28 | 41 | if not isinstance(dest, str): |
|
29 | 42 | raise ValueError('dest must be string if no prefix') |
|
30 | 43 | prefix = os.path.basename(dest) |
|
31 | 44 | lower = prefix.lower() |
|
32 | 45 | for sfx in exts.get(kind, []): |
|
33 | 46 | if lower.endswith(sfx): |
|
34 | 47 | prefix = prefix[:-len(sfx)] |
|
35 | 48 | break |
|
36 | 49 | lpfx = os.path.normpath(util.localpath(prefix)) |
|
37 | 50 | prefix = util.pconvert(lpfx) |
|
38 | 51 | if not prefix.endswith('/'): |
|
39 | 52 | prefix += '/' |
|
40 | 53 | # Drop the leading '.' path component if present, so Windows can read the |
|
41 | 54 | # zip files (issue4634) |
|
42 | 55 | if prefix.startswith('./'): |
|
43 | 56 | prefix = prefix[2:] |
|
44 | 57 | if prefix.startswith('../') or os.path.isabs(lpfx) or '/../' in prefix: |
|
45 | 58 | raise util.Abort(_('archive prefix contains illegal components')) |
|
46 | 59 | return prefix |
|
47 | 60 | |
|
48 | 61 | exts = { |
|
49 | 62 | 'tar': ['.tar'], |
|
50 | 63 | 'tbz2': ['.tbz2', '.tar.bz2'], |
|
51 | 64 | 'tgz': ['.tgz', '.tar.gz'], |
|
52 | 65 | 'zip': ['.zip'], |
|
53 | 66 | } |
|
54 | 67 | |
|
55 | 68 | def guesskind(dest): |
|
56 | 69 | for kind, extensions in exts.iteritems(): |
|
57 | 70 | if any(dest.endswith(ext) for ext in extensions): |
|
58 | 71 | return kind |
|
59 | 72 | return None |
|
60 | 73 | |
|
61 | 74 | def _rootctx(repo): |
|
62 | 75 | # repo[0] may be hidden |
|
63 | 76 | for rev in repo: |
|
64 | 77 | return repo[rev] |
|
65 | 78 | return repo['null'] |
|
66 | 79 | |
|
67 | 80 | def buildmetadata(ctx): |
|
68 | 81 | '''build content of .hg_archival.txt''' |
|
69 | 82 | repo = ctx.repo() |
|
70 | 83 | hex = ctx.hex() |
|
71 | 84 | if ctx.rev() is None: |
|
72 | 85 | hex = ctx.p1().hex() |
|
73 | 86 | if ctx.dirty(): |
|
74 | 87 | hex += '+' |
|
75 | 88 | |
|
76 | 89 | base = 'repo: %s\nnode: %s\nbranch: %s\n' % ( |
|
77 | 90 | _rootctx(repo).hex(), hex, encoding.fromlocal(ctx.branch())) |
|
78 | 91 | |
|
79 | 92 | tags = ''.join('tag: %s\n' % t for t in ctx.tags() |
|
80 | 93 | if repo.tagtype(t) == 'global') |
|
81 | 94 | if not tags: |
|
82 | 95 | repo.ui.pushbuffer() |
|
83 | 96 | opts = {'template': '{latesttag}\n{latesttagdistance}\n' |
|
84 | 97 | '{changessincelatesttag}', |
|
85 | 98 | 'style': '', 'patch': None, 'git': None} |
|
86 | 99 | cmdutil.show_changeset(repo.ui, repo, opts).show(ctx) |
|
87 | 100 | ltags, dist, changessince = repo.ui.popbuffer().split('\n') |
|
88 | 101 | ltags = ltags.split(':') |
|
89 | 102 | tags = ''.join('latesttag: %s\n' % t for t in ltags) |
|
90 | 103 | tags += 'latesttagdistance: %s\n' % dist |
|
91 | 104 | tags += 'changessincelatesttag: %s\n' % changessince |
|
92 | 105 | |
|
93 | 106 | return base + tags |
|
94 | 107 | |
|
95 | 108 | class tarit(object): |
|
96 | 109 | '''write archive to tar file or stream. can write uncompressed, |
|
97 | 110 | or compress with gzip or bzip2.''' |
|
98 | 111 | |
|
99 | 112 | class GzipFileWithTime(gzip.GzipFile): |
|
100 | 113 | |
|
101 | 114 | def __init__(self, *args, **kw): |
|
102 | 115 | timestamp = None |
|
103 | 116 | if 'timestamp' in kw: |
|
104 | 117 | timestamp = kw.pop('timestamp') |
|
105 | 118 | if timestamp is None: |
|
106 | 119 | self.timestamp = time.time() |
|
107 | 120 | else: |
|
108 | 121 | self.timestamp = timestamp |
|
109 | 122 | gzip.GzipFile.__init__(self, *args, **kw) |
|
110 | 123 | |
|
111 | 124 | def _write_gzip_header(self): |
|
112 | 125 | self.fileobj.write('\037\213') # magic header |
|
113 | 126 | self.fileobj.write('\010') # compression method |
|
114 | 127 | # Python 2.6 introduced self.name and deprecated self.filename |
|
115 | 128 | try: |
|
116 | 129 | fname = self.name |
|
117 | 130 | except AttributeError: |
|
118 | 131 | fname = self.filename |
|
119 | 132 | if fname and fname.endswith('.gz'): |
|
120 | 133 | fname = fname[:-3] |
|
121 | 134 | flags = 0 |
|
122 | 135 | if fname: |
|
123 | 136 | flags = gzip.FNAME |
|
124 | 137 | self.fileobj.write(chr(flags)) |
|
125 | 138 | gzip.write32u(self.fileobj, long(self.timestamp)) |
|
126 | 139 | self.fileobj.write('\002') |
|
127 | 140 | self.fileobj.write('\377') |
|
128 | 141 | if fname: |
|
129 | 142 | self.fileobj.write(fname + '\000') |
|
130 | 143 | |
|
131 | 144 | def __init__(self, dest, mtime, kind=''): |
|
132 | 145 | self.mtime = mtime |
|
133 | 146 | self.fileobj = None |
|
134 | 147 | |
|
135 | 148 | def taropen(name, mode, fileobj=None): |
|
136 | 149 | if kind == 'gz': |
|
137 | 150 | mode = mode[0] |
|
138 | 151 | if not fileobj: |
|
139 | 152 | fileobj = open(name, mode + 'b') |
|
140 | 153 | gzfileobj = self.GzipFileWithTime(name, mode + 'b', |
|
141 | 154 | zlib.Z_BEST_COMPRESSION, |
|
142 | 155 | fileobj, timestamp=mtime) |
|
143 | 156 | self.fileobj = gzfileobj |
|
144 | 157 | return tarfile.TarFile.taropen(name, mode, gzfileobj) |
|
145 | 158 | else: |
|
146 | 159 | return tarfile.open(name, mode + kind, fileobj) |
|
147 | 160 | |
|
148 | 161 | if isinstance(dest, str): |
|
149 | 162 | self.z = taropen(dest, mode='w:') |
|
150 | 163 | else: |
|
151 | 164 | # Python 2.5-2.5.1 have a regression that requires a name arg |
|
152 | 165 | self.z = taropen(name='', mode='w|', fileobj=dest) |
|
153 | 166 | |
|
154 | 167 | def addfile(self, name, mode, islink, data): |
|
155 | 168 | i = tarfile.TarInfo(name) |
|
156 | 169 | i.mtime = self.mtime |
|
157 | 170 | i.size = len(data) |
|
158 | 171 | if islink: |
|
159 | 172 | i.type = tarfile.SYMTYPE |
|
160 | 173 | i.mode = 0o777 |
|
161 | 174 | i.linkname = data |
|
162 | 175 | data = None |
|
163 | 176 | i.size = 0 |
|
164 | 177 | else: |
|
165 | 178 | i.mode = mode |
|
166 | 179 | data = cStringIO.StringIO(data) |
|
167 | 180 | self.z.addfile(i, data) |
|
168 | 181 | |
|
169 | 182 | def done(self): |
|
170 | 183 | self.z.close() |
|
171 | 184 | if self.fileobj: |
|
172 | 185 | self.fileobj.close() |
|
173 | 186 | |
|
174 | 187 | class tellable(object): |
|
175 | 188 | '''provide tell method for zipfile.ZipFile when writing to http |
|
176 | 189 | response file object.''' |
|
177 | 190 | |
|
178 | 191 | def __init__(self, fp): |
|
179 | 192 | self.fp = fp |
|
180 | 193 | self.offset = 0 |
|
181 | 194 | |
|
182 | 195 | def __getattr__(self, key): |
|
183 | 196 | return getattr(self.fp, key) |
|
184 | 197 | |
|
185 | 198 | def write(self, s): |
|
186 | 199 | self.fp.write(s) |
|
187 | 200 | self.offset += len(s) |
|
188 | 201 | |
|
189 | 202 | def tell(self): |
|
190 | 203 | return self.offset |
|
191 | 204 | |
|
192 | 205 | class zipit(object): |
|
193 | 206 | '''write archive to zip file or stream. can write uncompressed, |
|
194 | 207 | or compressed with deflate.''' |
|
195 | 208 | |
|
196 | 209 | def __init__(self, dest, mtime, compress=True): |
|
197 | 210 | if not isinstance(dest, str): |
|
198 | 211 | try: |
|
199 | 212 | dest.tell() |
|
200 | 213 | except (AttributeError, IOError): |
|
201 | 214 | dest = tellable(dest) |
|
202 | 215 | self.z = zipfile.ZipFile(dest, 'w', |
|
203 | 216 | compress and zipfile.ZIP_DEFLATED or |
|
204 | 217 | zipfile.ZIP_STORED) |
|
205 | 218 | |
|
206 | 219 | # Python's zipfile module emits deprecation warnings if we try |
|
207 | 220 | # to store files with a date before 1980. |
|
208 | 221 | epoch = 315532800 # calendar.timegm((1980, 1, 1, 0, 0, 0, 1, 1, 0)) |
|
209 | 222 | if mtime < epoch: |
|
210 | 223 | mtime = epoch |
|
211 | 224 | |
|
212 | 225 | self.mtime = mtime |
|
213 | 226 | self.date_time = time.gmtime(mtime)[:6] |
|
214 | 227 | |
|
215 | 228 | def addfile(self, name, mode, islink, data): |
|
216 | 229 | i = zipfile.ZipInfo(name, self.date_time) |
|
217 | 230 | i.compress_type = self.z.compression |
|
218 | 231 | # unzip will not honor unix file modes unless file creator is |
|
219 | 232 | # set to unix (id 3). |
|
220 | 233 | i.create_system = 3 |
|
221 | 234 | ftype = _UNX_IFREG |
|
222 | 235 | if islink: |
|
223 | 236 | mode = 0o777 |
|
224 | 237 | ftype = _UNX_IFLNK |
|
225 | 238 | i.external_attr = (mode | ftype) << 16L |
|
226 | 239 | # add "extended-timestamp" extra block, because zip archives |
|
227 | 240 | # without this will be extracted with unexpected timestamp, |
|
228 | 241 | # if TZ is not configured as GMT |
|
229 | 242 | i.extra += struct.pack('<hhBl', |
|
230 | 243 | 0x5455, # block type: "extended-timestamp" |
|
231 | 244 | 1 + 4, # size of this block |
|
232 | 245 | 1, # "modification time is present" |
|
233 | 246 | int(self.mtime)) # last modification (UTC) |
|
234 | 247 | self.z.writestr(i, data) |
|
235 | 248 | |
|
236 | 249 | def done(self): |
|
237 | 250 | self.z.close() |
|
238 | 251 | |
|
239 | 252 | class fileit(object): |
|
240 | 253 | '''write archive as files in directory.''' |
|
241 | 254 | |
|
242 | 255 | def __init__(self, name, mtime): |
|
243 | 256 | self.basedir = name |
|
244 | 257 | self.opener = scmutil.opener(self.basedir) |
|
245 | 258 | |
|
246 | 259 | def addfile(self, name, mode, islink, data): |
|
247 | 260 | if islink: |
|
248 | 261 | self.opener.symlink(data, name) |
|
249 | 262 | return |
|
250 | 263 | f = self.opener(name, "w", atomictemp=True) |
|
251 | 264 | f.write(data) |
|
252 | 265 | f.close() |
|
253 | 266 | destfile = os.path.join(self.basedir, name) |
|
254 | 267 | os.chmod(destfile, mode) |
|
255 | 268 | |
|
256 | 269 | def done(self): |
|
257 | 270 | pass |
|
258 | 271 | |
|
259 | 272 | archivers = { |
|
260 | 273 | 'files': fileit, |
|
261 | 274 | 'tar': tarit, |
|
262 | 275 | 'tbz2': lambda name, mtime: tarit(name, mtime, 'bz2'), |
|
263 | 276 | 'tgz': lambda name, mtime: tarit(name, mtime, 'gz'), |
|
264 | 277 | 'uzip': lambda name, mtime: zipit(name, mtime, False), |
|
265 | 278 | 'zip': zipit, |
|
266 | 279 | } |
|
267 | 280 | |
|
268 | 281 | def archive(repo, dest, node, kind, decode=True, matchfn=None, |
|
269 | 282 | prefix='', mtime=None, subrepos=False): |
|
270 | 283 | '''create archive of repo as it was at node. |
|
271 | 284 | |
|
272 | 285 | dest can be name of directory, name of archive file, or file |
|
273 | 286 | object to write archive to. |
|
274 | 287 | |
|
275 | 288 | kind is type of archive to create. |
|
276 | 289 | |
|
277 | 290 | decode tells whether to put files through decode filters from |
|
278 | 291 | hgrc. |
|
279 | 292 | |
|
280 | 293 | matchfn is function to filter names of files to write to archive. |
|
281 | 294 | |
|
282 | 295 | prefix is name of path to put before every archive member.''' |
|
283 | 296 | |
|
284 | 297 | if kind == 'files': |
|
285 | 298 | if prefix: |
|
286 | 299 | raise util.Abort(_('cannot give prefix when archiving to files')) |
|
287 | 300 | else: |
|
288 | 301 | prefix = tidyprefix(dest, kind, prefix) |
|
289 | 302 | |
|
290 | 303 | def write(name, mode, islink, getdata): |
|
291 | 304 | data = getdata() |
|
292 | 305 | if decode: |
|
293 | 306 | data = repo.wwritedata(name, data) |
|
294 | 307 | archiver.addfile(prefix + name, mode, islink, data) |
|
295 | 308 | |
|
296 | 309 | if kind not in archivers: |
|
297 | 310 | raise util.Abort(_("unknown archive type '%s'") % kind) |
|
298 | 311 | |
|
299 | 312 | ctx = repo[node] |
|
300 | 313 | archiver = archivers[kind](dest, mtime or ctx.date()[0]) |
|
301 | 314 | |
|
302 | 315 | if repo.ui.configbool("ui", "archivemeta", True): |
|
303 | 316 | name = '.hg_archival.txt' |
|
304 | 317 | if not matchfn or matchfn(name): |
|
305 | 318 | write(name, 0o644, False, lambda: buildmetadata(ctx)) |
|
306 | 319 | |
|
307 | 320 | if matchfn: |
|
308 | 321 | files = [f for f in ctx.manifest().keys() if matchfn(f)] |
|
309 | 322 | else: |
|
310 | 323 | files = ctx.manifest().keys() |
|
311 | 324 | total = len(files) |
|
312 | 325 | if total: |
|
313 | 326 | files.sort() |
|
314 | 327 | repo.ui.progress(_('archiving'), 0, unit=_('files'), total=total) |
|
315 | 328 | for i, f in enumerate(files): |
|
316 | 329 | ff = ctx.flags(f) |
|
317 | 330 | write(f, 'x' in ff and 0o755 or 0o644, 'l' in ff, ctx[f].data) |
|
318 | 331 | repo.ui.progress(_('archiving'), i + 1, item=f, |
|
319 | 332 | unit=_('files'), total=total) |
|
320 | 333 | repo.ui.progress(_('archiving'), None) |
|
321 | 334 | |
|
322 | 335 | if subrepos: |
|
323 | 336 | for subpath in sorted(ctx.substate): |
|
324 | 337 | sub = ctx.workingsub(subpath) |
|
325 | 338 | submatch = matchmod.narrowmatcher(subpath, matchfn) |
|
326 | 339 | total += sub.archive(archiver, prefix, submatch) |
|
327 | 340 | |
|
328 | 341 | if total == 0: |
|
329 | 342 | raise error.Abort(_('no files match the archive pattern')) |
|
330 | 343 | |
|
331 | 344 | archiver.done() |
|
332 | 345 | return total |
General Comments 0
You need to be logged in to leave comments.
Login now