« back to all changes in this revision
Viewing changes to bzrlib/tests/blackbox/test_non_ascii.py
- browse files at revision 2978.8.5
- compare with another revision
- compare with revision 2978.8.5
- download diff
- download tarball
- view history from revision 2978.8.5
- Committer: Alexander Belchenko
- Date: 2007-11-19 22:54:30 UTC
- mfrom: (3006 +trunk)
- mto: This revision was merged to the branch mainline in revision 3008.
- Revision ID: bialix@ukr.net-20071119225430-x0ewosrsagis0yno
merge bzr.dev
- files added:
- bzrlib/lru_cache.py
- files modified:
- NEWS
added
removed
bzrlib/tests/blackbox/test_non_ascii.py
47
47
bzrlib.user_encoding = self._orig_encoding
48
48
super(TestNonAscii, self).tearDown()
49
49
50
def run_bzr_decode(self, args, encoding=None, fail=False, retcode=None):
50
def run_bzr_decode(self, args, encoding=None, fail=False, retcode=None,
51
working_dir=None):
51
52
"""Run bzr and decode the output into a particular encoding.
52
53
53
54
Returns a string containing the stdout output from bzr.
59
60
encoding = bzrlib.user_encoding
60
61
try:
61
62
out = self.run_bzr(args, output_encoding=encoding, encoding=encoding,
62
retcode=retcode)[0]
63
retcode=retcode, working_dir=working_dir)[0]
63
64
return out.decode(encoding)
64
65
except UnicodeError, e:
65
66
if not fail:
94
95
% (thing, terminal_enc, fs_enc))
95
96
96
97
wt = self.make_branch_and_tree('.')
97
open('a', 'wb').write('foo\n')
98
self.build_tree_contents([('a', 'foo\n')])
98
99
wt.add('a')
99
100
wt.commit('adding a')
100
101
101
open('b', 'wb').write('non-ascii \xFF\xFF\xFC\xFB\x00 in b\n')
102
self.build_tree_contents(
103
[('b', 'non-ascii \xFF\xFF\xFC\xFB\x00 in b\n')])
102
104
wt.add('b')
103
105
wt.commit(self.info['message'])
104
106
105
open(fname, 'wb').write('unicode filename\n')
107
self.build_tree_contents([(fname, 'unicode filename\n')])
106
108
wt.add(fname)
107
109
wt.commit(u'And a unicode file\n')
108
110
self.wt = wt
109
111
110
112
def test_status(self):
111
open(self.info['filename'], 'ab').write('added something\n')
113
self.build_tree_contents(
114
[(self.info['filename'], 'changed something\n')])
112
115
txt = self.run_bzr_decode('status')
113
116
self.assertEqual(u'modified:\n %s\n' % (self.info['filename'],), txt)
114
117
127
130
self.assertEqual('unicode filename\n', txt)
128
131
129
132
def test_cat_revision(self):
130
bzr = self.run_bzr_decode
131
132
133
committer = self.info['committer']
133
txt = bzr('cat-revision -r 1')
134
txt = self.run_bzr_decode('cat-revision -r 1')
134
135
self.failUnless(committer in txt,
135
136
'failed to find %r in %r' % (committer, txt))
136
137
137
138
msg = self.info['message']
138
txt = bzr('cat-revision -r 2')
139
txt = self.run_bzr_decode('cat-revision -r 2')
139
140
self.failUnless(msg in txt, 'failed to find %r in %r' % (msg, txt))
140
141
141
142
def test_mkdir(self):
142
bzr = self.run_bzr_decode
143
144
txt = bzr(['mkdir', self.info['directory']])
143
txt = self.run_bzr_decode(['mkdir', self.info['directory']])
145
144
self.assertEqual(u'added %s\n' % self.info['directory'], txt)
146
145
147
146
# The text should be garbled, but the command should succeed
148
txt = bzr(['mkdir', self.info['directory'] + '2'], encoding='ascii')
147
txt = self.run_bzr_decode(['mkdir', self.info['directory'] + '2'],
148
encoding='ascii')
149
149
expected = u'added %s2\n' % (self.info['directory'],)
150
150
expected = expected.encode('ascii', 'replace')
151
151
self.assertEqual(expected, txt)
152
152
153
153
def test_relpath(self):
154
bzr = self.run_bzr_decode
155
156
txt = bzr(['relpath', self.info['filename']])
154
txt = self.run_bzr_decode(['relpath', self.info['filename']])
157
155
self.assertEqual(self.info['filename'] + '\n', txt)
158
156
159
bzr(['relpath', self.info['filename']], encoding='ascii', fail=True)
157
self.run_bzr_decode(['relpath', self.info['filename']],
158
encoding='ascii', fail=True)
160
159
161
160
def test_inventory(self):
162
bzr = self.run_bzr_decode
163
164
txt = bzr('inventory')
161
txt = self.run_bzr_decode('inventory')
165
162
self.assertEqual(['a', 'b', self.info['filename']],
166
163
txt.splitlines())
167
164
168
165
# inventory should fail if unable to encode
169
bzr('inventory', encoding='ascii', fail=True)
166
self.run_bzr_decode('inventory', encoding='ascii', fail=True)
170
167
171
168
# We don't really care about the ids themselves,
172
169
# but the command shouldn't fail
173
txt = bzr('inventory --show-ids')
170
txt = self.run_bzr_decode('inventory --show-ids')
174
171
175
172
def test_revno(self):
176
173
# There isn't a lot to test here, since revno should always
177
174
# be an integer
178
bzr = self.run_bzr_decode
179
180
self.assertEqual('3\n', bzr('revno'))
181
self.assertEqual('3\n', bzr('revno', encoding='ascii'))
175
self.assertEqual('3\n', self.run_bzr_decode('revno'))
176
self.assertEqual('3\n', self.run_bzr_decode('revno', encoding='ascii'))
182
177
183
178
def test_revision_info(self):
184
bzr = self.run_bzr_decode
185
186
bzr('revision-info -r 1')
179
self.run_bzr_decode('revision-info -r 1')
187
180
188
181
# TODO: jam 20060105 If we support revisions with non-ascii characters,
189
182
# this should be strict and fail.
190
bzr('revision-info -r 1', encoding='ascii')
183
self.run_bzr_decode('revision-info -r 1', encoding='ascii')
191
184
192
185
def test_mv(self):
193
bzr = self.run_bzr_decode
194
195
186
fname1 = self.info['filename']
196
187
fname2 = self.info['filename'] + '2'
197
188
dirname = self.info['directory']
198
189
199
190
# fname1 already exists
200
bzr(['mv', 'a', fname1], fail=True)
191
self.run_bzr_decode(['mv', 'a', fname1], fail=True)
201
192
202
txt = bzr(['mv', 'a', fname2])
193
txt = self.run_bzr_decode(['mv', 'a', fname2])
203
194
self.assertEqual(u'a => %s\n' % fname2, txt)
204
195
self.failIfExists('a')
205
196
self.failUnlessExists(fname2)
210
201
211
202
os.mkdir(dirname)
212
203
self.wt.add(dirname)
213
txt = bzr(['mv', fname1, fname2, dirname])
204
txt = self.run_bzr_decode(['mv', fname1, fname2, dirname])
214
205
self.assertEqual([u'%s => %s/%s' % (fname1, dirname, fname1),
215
206
u'%s => %s/%s' % (fname2, dirname, fname2)]
216
207
, txt.splitlines())
217
208
218
209
# The rename should still succeed
219
210
newpath = u'%s/%s' % (dirname, fname2)
220
txt = bzr(['mv', newpath, 'a'], encoding='ascii')
211
txt = self.run_bzr_decode(['mv', newpath, 'a'], encoding='ascii')
221
212
self.failUnlessExists('a')
222
213
self.assertEqual(newpath.encode('ascii', 'replace') + ' => a\n', txt)
223
214
224
215
def test_branch(self):
225
216
# We should be able to branch into a directory that
226
217
# has a unicode name, even if we can't display the name
227
bzr = self.run_bzr_decode
228
bzr(['branch', u'.', self.info['directory']])
229
bzr(['branch', u'.', self.info['directory'] + '2'], encoding='ascii')
218
self.run_bzr_decode(['branch', u'.', self.info['directory']])
219
self.run_bzr_decode(['branch', u'.', self.info['directory'] + '2'],
220
encoding='ascii')
230
221
231
222
def test_pull(self):
232
223
# Make sure we can pull from paths that can't be encoded
233
bzr = self.run_bzr_decode
234
235
224
dirname1 = self.info['directory']
236
225
dirname2 = self.info['directory'] + '2'
237
226
url1 = urlutils.local_path_to_url(dirname1)
239
228
out_bzrdir = self.wt.bzrdir.sprout(url1)
240
229
out_bzrdir.sprout(url2)
241
230
242
os.chdir(dirname1)
243
open('a', 'ab').write('more text\n')
231
self.build_tree_contents(
232
[(osutils.pathjoin(dirname1, "a"), 'different text\n')])
244
233
self.wt.commit('mod a')
245
234
246
pwd = osutils.getcwd()
247
248
os.chdir(u'../' + dirname2)
249
txt = bzr('pull')
250
235
txt = self.run_bzr_decode('pull', working_dir=dirname2)
236
237
expected = osutils.pathjoin(osutils.getcwd(), dirname1)
251
238
self.assertEqual(u'Using saved location: %s/\n'
252
'No revisions to pull.\n' % (pwd,), txt)
239
'No revisions to pull.\n' % (expected,), txt)
253
240
254
os.chdir('../' + dirname1)
255
open('a', 'ab').write('and yet more\n')
241
self.build_tree_contents(
242
[(osutils.pathjoin(dirname1, 'a'), 'and yet more\n')])
256
243
self.wt.commit(u'modifying a by ' + self.info['committer'])
257
244
258
os.chdir('../' + dirname2)
259
245
# We should be able to pull, even if our encoding is bad
260
bzr('pull --verbose', encoding='ascii')
246
self.run_bzr_decode('pull --verbose', encoding='ascii',
247
working_dir=dirname2)
261
248
262
249
def test_push(self):
263
250
# TODO: Test push to an SFTP location
264
251
# Make sure we can pull from paths that can't be encoded
265
bzr = self.run_bzr_decode
266
267
252
# TODO: jam 20060427 For drastically improving performance, we probably
268
253
# could create a local repository, so it wouldn't have to copy
269
254
# the files around as much.
270
255
271
256
dirname = self.info['directory']
272
bzr(['push', dirname])
257
self.run_bzr_decode(['push', dirname])
273
258
274
open('a', 'ab').write('adding more text\n')
259
self.build_tree_contents([('a', 'adding more text\n')])
275
260
self.wt.commit('added some stuff')
276
261
277
262
# TODO: check the output text is properly encoded
278
bzr('push')
263
self.run_bzr_decode('push')
279
264
280
f = open('a', 'ab')
281
try:
282
f.write('and a bit more: ')
283
f.write(dirname.encode('utf-8'))
284
f.write('\n')
285
finally:
286
f.close()
265
self.build_tree_contents(
266
[('a', 'and a bit more: \n%s\n' % (dirname.encode('utf-8'),))])
287
267
288
268
self.wt.commit('Added some ' + dirname)
289
bzr('push --verbose', encoding='ascii')
290
291
bzr(['push', '--verbose', dirname + '2'])
292
293
bzr(['push', '--verbose', dirname + '3'], encoding='ascii')
294
295
bzr(['push', '--verbose', '--create-prefix',
296
dirname + '4/' + dirname + '5'])
297
bzr(['push', '--verbose', '--create-prefix',
298
dirname + '6/' + dirname + '7'], encoding='ascii')
269
self.run_bzr_decode('push --verbose', encoding='ascii')
270
271
self.run_bzr_decode(['push', '--verbose', dirname + '2'])
272
273
self.run_bzr_decode(['push', '--verbose', dirname + '3'],
274
encoding='ascii')
275
276
self.run_bzr_decode(['push', '--verbose', '--create-prefix',
277
dirname + '4/' + dirname + '5'])
278
self.run_bzr_decode(['push', '--verbose', '--create-prefix',
279
dirname + '6/' + dirname + '7'], encoding='ascii')
299
280
300
281
def test_renames(self):
301
bzr = self.run_bzr_decode
302
303
282
fname = self.info['filename'] + '2'
304
bzr(['mv', 'a', fname])
305
txt = bzr('renames')
283
self.wt.rename_one('a', fname)
284
txt = self.run_bzr_decode('renames')
306
285
self.assertEqual(u'a => %s\n' % fname, txt)
307
286
308
bzr('renames', fail=True, encoding='ascii')
287
self.run_bzr_decode('renames', fail=True, encoding='ascii')
309
288
310
289
def test_remove(self):
311
bzr = self.run_bzr_decode
312
313
290
fname = self.info['filename']
314
txt = bzr(['remove', fname], encoding='ascii')
291
txt = self.run_bzr_decode(['remove', fname], encoding='ascii')
315
292
316
293
def test_remove_verbose(self):
317
bzr = self.run_bzr_decode
318
319
294
fname = self.info['filename']
320
txt = bzr(['remove', '--verbose', fname], encoding='ascii')
295
txt = self.run_bzr_decode(['remove', '--verbose', fname],
296
encoding='ascii')
321
297
322
298
def test_file_id(self):
323
bzr = self.run_bzr_decode
324
325
299
fname = self.info['filename']
326
txt = bzr(['file-id', fname])
300
txt = self.run_bzr_decode(['file-id', fname])
327
301
328
302
# TODO: jam 20060106 We don't support non-ascii file ids yet,
329
303
# so there is nothing which would fail in ascii encoding
330
304
# This *should* be retcode=3
331
txt = bzr(['file-id', fname], encoding='ascii')
305
txt = self.run_bzr_decode(['file-id', fname], encoding='ascii')
332
306
333
307
def test_file_path(self):
334
bzr = self.run_bzr_decode
335
336
308
# Create a directory structure
337
309
fname = self.info['filename']
338
310
dirname = self.info['directory']
339
os.mkdir('base')
340
os.mkdir('base/' + dirname)
311
self.build_tree_contents([
312
('base/', ),
313
(osutils.pathjoin('base', '%s/' % (dirname,)), )])
341
314
self.wt.add('base')
342
315
self.wt.add('base/'+dirname)
343
path = '/'.join(['base', dirname, fname])
316
path = osutils.pathjoin('base', dirname, fname)
344
317
self.wt.rename_one(fname, path)
345
318
self.wt.commit('moving things around')
346
319
347
txt = bzr(['file-path', path])
320
txt = self.run_bzr_decode(['file-path', path])
348
321
349
322
# TODO: jam 20060106 We don't support non-ascii file ids yet,
350
323
# so there is nothing which would fail in ascii encoding
351
324
# This *should* be retcode=3
352
txt = bzr(['file-path', path], encoding='ascii')
325
txt = self.run_bzr_decode(['file-path', path], encoding='ascii')
353
326
354
327
def test_revision_history(self):
355
bzr = self.run_bzr_decode
356
357
328
# TODO: jam 20060106 We don't support non-ascii revision ids yet,
358
329
# so there is nothing which would fail in ascii encoding
359
txt = bzr('revision-history')
330
txt = self.run_bzr_decode('revision-history')
360
331
361
332
def test_ancestry(self):
362
bzr = self.run_bzr_decode
363
364
333
# TODO: jam 20060106 We don't support non-ascii revision ids yet,
365
334
# so there is nothing which would fail in ascii encoding
366
txt = bzr('ancestry')
335
txt = self.run_bzr_decode('ancestry')
367
336
368
337
def test_diff(self):
369
338
# TODO: jam 20060106 diff is a difficult one to test, because it
370
339
# shouldn't encode the file contents, but it needs some sort
371
340
# of encoding for the paths, etc which are displayed.
372
open(self.info['filename'], 'ab').write('newline\n')
341
self.build_tree_contents([(self.info['filename'], 'newline\n')])
373
342
txt = self.run_bzr('diff', retcode=1)[0]
374
343
375
344
def test_deleted(self):
376
bzr = self.run_bzr_decode
377
378
345
fname = self.info['filename']
379
346
os.remove(fname)
380
347
self.wt.remove(fname)
381
348
382
txt = bzr('deleted')
349
txt = self.run_bzr_decode('deleted')
383
350
self.assertEqual(fname+'\n', txt)
384
351
385
txt = bzr('deleted --show-ids')
352
txt = self.run_bzr_decode('deleted --show-ids')
386
353
self.failUnless(txt.startswith(fname))
387
354
388
355
# Deleted should fail if cannot decode
389
356
# Because it is giving the exact paths
390
357
# which might be used by a front end
391
bzr('deleted', encoding='ascii', fail=True)
358
self.run_bzr_decode('deleted', encoding='ascii', fail=True)
392
359
393
360
def test_modified(self):
394
bzr = self.run_bzr_decode
395
396
361
fname = self.info['filename']
397
open(fname, 'ab').write('modified\n')
362
self.build_tree_contents([(fname, 'modified\n')])
398
363
399
txt = bzr('modified')
364
txt = self.run_bzr_decode('modified')
400
365
self.assertEqual(fname+'\n', txt)
401
366
402
bzr('modified', encoding='ascii', fail=True)
367
self.run_bzr_decode('modified', encoding='ascii', fail=True)
403
368
404
369
def test_added(self):
405
bzr = self.run_bzr_decode
406
407
370
fname = self.info['filename'] + '2'
408
open(fname, 'wb').write('added\n')
371
self.build_tree_contents([(fname, 'added\n')])
409
372
self.wt.add(fname)
410
373
411
txt = bzr('added')
374
txt = self.run_bzr_decode('added')
412
375
self.assertEqual(fname+'\n', txt)
413
376
414
bzr('added', encoding='ascii', fail=True)
377
self.run_bzr_decode('added', encoding='ascii', fail=True)
415
378
416
379
def test_root(self):
417
bzr = self.run_bzr_decode
418
419
380
dirname = self.info['directory']
420
381
url = urlutils.local_path_to_url(dirname)
421
bzr('root')
382
self.run_bzr_decode('root')
422
383
423
384
self.wt.bzrdir.sprout(url)
424
385
425
os.chdir(dirname)
426
427
txt = bzr('root')
386
txt = self.run_bzr_decode('root', working_dir=dirname)
428
387
self.failUnless(txt.endswith(dirname+'\n'))
429
388
430
txt = bzr('root', encoding='ascii', fail=True)
389
txt = self.run_bzr_decode('root', encoding='ascii', fail=True,
390
working_dir=dirname)
431
391
432
392
def test_log(self):
433
bzr = self.run_bzr_decode
434
435
393
fname = self.info['filename']
436
394
437
txt = bzr('log')
395
txt = self.run_bzr_decode('log')
438
396
self.assertNotEqual(-1, txt.find(self.info['committer']))
439
397
self.assertNotEqual(-1, txt.find(self.info['message']))
440
398
441
txt = bzr('log --verbose')
399
txt = self.run_bzr_decode('log --verbose')
442
400
self.assertNotEqual(-1, txt.find(fname))
443
401
444
402
# Make sure log doesn't fail even if we can't write out
445
txt = bzr('log --verbose', encoding='ascii')
403
txt = self.run_bzr_decode('log --verbose', encoding='ascii')
446
404
self.assertEqual(-1, txt.find(fname))
447
405
self.assertNotEqual(-1, txt.find(fname.encode('ascii', 'replace')))
448
406
449
407
def test_touching_revisions(self):
450
bzr = self.run_bzr_decode
451
452
408
fname = self.info['filename']
453
txt = bzr(['touching-revisions', fname])
409
txt = self.run_bzr_decode(['touching-revisions', fname])
454
410
self.assertEqual(u' 3 added %s\n' % (fname,), txt)
455
411
456
412
fname2 = self.info['filename'] + '2'
457
413
self.wt.rename_one(fname, fname2)
458
414
self.wt.commit(u'Renamed %s => %s' % (fname, fname2))
459
415
460
txt = bzr(['touching-revisions', fname2])
416
txt = self.run_bzr_decode(['touching-revisions', fname2])
461
417
expected_txt = (u' 3 added %s\n'
462
418
u' 4 renamed %s => %s\n'
463
419
% (fname, fname, fname2))
464
420
self.assertEqual(expected_txt, txt)
465
421
466
bzr(['touching-revisions', fname2], encoding='ascii', fail=True)
422
self.run_bzr_decode(['touching-revisions', fname2], encoding='ascii',
423
fail=True)
467
424
468
425
def test_ls(self):
469
bzr = self.run_bzr_decode
470
471
txt = bzr('ls')
426
txt = self.run_bzr_decode('ls')
472
427
self.assertEqual(sorted(['a', 'b', self.info['filename']]),
473
428
sorted(txt.splitlines()))
474
txt = bzr('ls --null')
429
txt = self.run_bzr_decode('ls --null')
475
430
self.assertEqual(sorted(['', 'a', 'b', self.info['filename']]),
476
431
sorted(txt.split('\0')))
477
432
478
txt = bzr('ls', encoding='ascii', fail=True)
479
txt = bzr('ls --null', encoding='ascii', fail=True)
433
txt = self.run_bzr_decode('ls', encoding='ascii', fail=True)
434
txt = self.run_bzr_decode('ls --null', encoding='ascii', fail=True)
480
435
481
436
def test_unknowns(self):
482
bzr = self.run_bzr_decode
483
484
437
fname = self.info['filename'] + '2'
485
open(fname, 'wb').write('unknown\n')
438
self.build_tree_contents([(fname, 'unknown\n')])
486
439
487
440
# TODO: jam 20060112 bzr unknowns is the only one which
488
441
# quotes paths do we really want it to?
489
txt = bzr('unknowns')
442
txt = self.run_bzr_decode('unknowns')
490
443
self.assertEqual(u'"%s"\n' % (fname,), txt)
491
444
492
bzr('unknowns', encoding='ascii', fail=True)
445
self.run_bzr_decode('unknowns', encoding='ascii', fail=True)
493
446
494
447
def test_ignore(self):
495
bzr = self.run_bzr_decode
496
497
448
fname2 = self.info['filename'] + '2.txt'
498
open(fname2, 'wb').write('ignored\n')
449
self.build_tree_contents([(fname2, 'ignored\n')])
499
450
500
451
def check_unknowns(expected):
501
452
self.assertEqual(expected, list(self.wt.unknowns()))
502
453
503
454
check_unknowns([fname2])
504
455
505
bzr(['ignore', './' + fname2])
506
# After 'ignore' you must re-open the working tree
507
self.wt = self.wt.bzrdir.open_workingtree()
456
self.run_bzr_decode(['ignore', './' + fname2])
508
457
check_unknowns([])
509
458
510
459
fname3 = self.info['filename'] + '3.txt'
511
open(fname3, 'wb').write('unknown 3\n')
460
self.build_tree_contents([(fname3, 'unknown 3\n')])
512
461
check_unknowns([fname3])
513
462
514
463
# Ignore should not care what the encoding is
515
464
# (right now it doesn't print anything)
516
bzr(['ignore', fname3], encoding='ascii')
517
self.wt = self.wt.bzrdir.open_workingtree()
465
self.run_bzr_decode(['ignore', fname3], encoding='ascii')
518
466
check_unknowns([])
519
467
520
468
# Now try a wildcard match
521
469
fname4 = self.info['filename'] + '4.txt'
522
open(fname4, 'wb').write('unknown 4\n')
523
bzr('ignore *.txt')
524
self.wt = self.wt.bzrdir.open_workingtree()
470
self.build_tree_contents([(fname4, 'unknown 4\n')])
471
self.run_bzr_decode('ignore *.txt')
525
472
check_unknowns([])
526
473
527
474
# and a different wildcard that matches everything
528
475
os.remove('.bzrignore')
529
bzr(['ignore', self.info['filename'] + '*'])
530
self.wt = self.wt.bzrdir.open_workingtree()
476
self.run_bzr_decode(['ignore', self.info['filename'] + '*'])
531
477
check_unknowns([])
532
478
533
479
def test_missing(self):
534
bzr = self.run_bzr_decode
535
536
480
# create empty tree as reference for missing
537
self.run_bzr('init empty-tree')
481
self.make_branch_and_tree('empty-tree')
538
482
539
483
msg = self.info['message']
540
484
541
txt = bzr('missing empty-tree', retcode=1)
485
txt = self.run_bzr_decode('missing empty-tree')
542
486
self.assertNotEqual(-1, txt.find(self.info['committer']))
543
487
self.assertNotEqual(-1, txt.find(msg))
544
488
545
489
# Make sure missing doesn't fail even if we can't write out
546
txt = bzr('missing empty-tree', encoding='ascii', retcode=1)
490
txt = self.run_bzr_decode('missing empty-tree', encoding='ascii')
547
491
self.assertEqual(-1, txt.find(msg))
548
492
self.assertNotEqual(-1, txt.find(msg.encode('ascii', 'replace')))
549
493
Loggerhead is a web-based interface for Breezy
Version: 2.0.1