« back to all changes in this revision
Viewing changes to bzrlib/tests/blackbox/test_non_ascii.py
- browse files at revision 1685.1.1
- compare with another revision
- compare with revision 5370
- download diff
- download tarball
- view history from revision 1685.1.1
- Committer: John Arbash Meinel
- Date: 2006-04-25 15:05:42 UTC
- mfrom: (1185.85.85 bzr-encoding)
- mto: This revision was merged to the branch mainline in revision 1752.
- Revision ID: john@arbash-meinel.com-20060425150542-c7b518dca9928691
[merge] the old bzr-encoding changes, reparenting them on bzr.dev
- files added:
- bzrlib/tests/EncodingAdapter.py
- files modified:
- bzrlib/add.py
added
removed
bzrlib/tests/blackbox/test_non_ascii.py
1
# Copyright (C) 2006 by Canonical Ltd
2
# -*- coding: utf-8 -*-
3
4
# This program is free software; you can redistribute it and/or modify
5
# it under the terms of the GNU General Public License as published by
6
# the Free Software Foundation; either version 2 of the License, or
7
# (at your option) any later version.
8
9
# This program is distributed in the hope that it will be useful,
10
# but WITHOUT ANY WARRANTY; without even the implied warranty of
11
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12
# GNU General Public License for more details.
13
14
# You should have received a copy of the GNU General Public License
15
# along with this program; if not, write to the Free Software
16
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
17
18
"""\
19
Black-box tests for bzr handling non-ascii characters.
20
"""
21
22
import sys
23
import os
24
import bzrlib
25
from bzrlib.tests import TestCaseInTempDir, TestSkipped
26
from bzrlib.trace import mutter, note
27
28
29
class TestNonAscii(TestCaseInTempDir):
30
"""Test that bzr handles files/committers/etc which are non-ascii."""
31
32
def setUp(self):
33
super(TestNonAscii, self).setUp()
34
raise TestSkipped('We cannot handle unicode filenames (yet)')
35
# TODO: jam 20060425 because of the confusion of URLs versus Unicode
36
# 'bzr add f' will not add a unicode filename
37
self._orig_email = os.environ.get('BZREMAIL', None)
38
self._orig_encoding = bzrlib.user_encoding
39
40
bzrlib.user_encoding = self.encoding
41
email = self.info['committer'] + ' <joe@foo.com>'
42
os.environ['BZREMAIL'] = email.encode(bzrlib.user_encoding)
43
self.create_base()
44
45
def tearDown(self):
46
if self._orig_email is not None:
47
os.environ['BZREMAIL'] = self._orig_email
48
else:
49
if os.environ.get('BZREMAIL', None) is not None:
50
del os.environ['BZREMAIL']
51
bzrlib.user_encoding = self._orig_encoding
52
super(TestNonAscii, self).tearDown()
53
54
def create_base(self):
55
bzr = self.run_bzr
56
57
bzr('init')
58
open('a', 'wb').write('foo\n')
59
bzr('add', 'a')
60
bzr('commit', '-m', 'adding a')
61
62
open('b', 'wb').write('non-ascii \xFF\xFF\xFC\xFB\x00 in b\n')
63
bzr('add', 'b')
64
bzr('commit', '-m', self.info['message'])
65
66
fname = self.info['filename']
67
try:
68
open(fname, 'wb').write('unicode filename\n')
69
except UnicodeEncodeError:
70
raise TestSkipped(('Unable to represent filename %r'
71
' in filesystem encoding %s')
72
% (fname, sys.getfilesystemencoding()))
73
74
bzr('add', fname)
75
bzr('commit', '-m', u'And an unicode file\n')
76
77
def test_status(self):
78
bzr = self.run_bzr_decode
79
80
open(self.info['filename'], 'ab').write('added something\n')
81
txt = bzr('status')
82
self.assertEqual(u'modified:\n %s\n' % (self.info['filename'],), txt)
83
84
def test_cat(self):
85
# bzr cat shouldn't change the contents
86
# using run_bzr since that doesn't decode
87
txt = self.run_bzr('cat', 'b')[0]
88
self.assertEqual('non-ascii \xFF\xFF\xFC\xFB\x00 in b\n', txt)
89
90
txt = self.run_bzr('cat', self.info['filename'])[0]
91
self.assertEqual('unicode filename\n', txt)
92
93
def test_cat_revision(self):
94
bzr = self.run_bzr_decode
95
96
# TODO: jam 20060112 should cat-revision always output utf-8?
97
committer = self.info['committer']
98
txt = bzr('cat-revision', '-r', '1')
99
self.failUnless(committer in txt,
100
'failed to find %r in %r' % (committer, txt))
101
102
msg = self.info['message']
103
txt = bzr('cat-revision', '-r', '2')
104
self.failUnless(msg in txt, 'failed to find %r in %r' % (msg, txt))
105
106
def test_mkdir(self):
107
bzr = self.run_bzr_decode
108
109
txt = bzr('mkdir', self.info['directory'])
110
self.assertEqual(u'added %s\n' % self.info['directory'], txt)
111
112
# The text should be garbled, but the command should succeed
113
txt = bzr('mkdir', self.info['directory'] + '2', encoding='ascii')
114
expected = u'added %s2\n' % (self.info['directory'],)
115
expected = expected.encode('ascii', 'replace')
116
self.assertEqual(expected, txt)
117
118
def test_relpath(self):
119
bzr = self.run_bzr_decode
120
121
txt = bzr('relpath', self.info['filename'])
122
self.assertEqual(self.info['filename'] + '\n', txt)
123
124
# TODO: jam 20060106 if relpath can return a munged string
125
# this text needs to be fixed
126
bzr('relpath', self.info['filename'], encoding='ascii', retcode=3)
127
128
def test_inventory(self):
129
bzr = self.run_bzr_decode
130
131
txt = bzr('inventory')
132
self.assertEqual(['a', 'b', self.info['filename']],
133
txt.splitlines())
134
135
# inventory should fail if unable to encode
136
bzr('inventory', encoding='ascii', retcode=3)
137
138
# We don't really care about the ids themselves,
139
# but the command shouldn't fail
140
txt = bzr('inventory', '--show-ids')
141
142
def test_revno(self):
143
# There isn't a lot to test here, since revno should always
144
# be an integer
145
bzr = self.run_bzr_decode
146
147
self.assertEqual('3\n', bzr('revno'))
148
self.assertEqual('3\n', bzr('revno', encoding='ascii'))
149
150
def test_revision_info(self):
151
bzr = self.run_bzr_decode
152
153
bzr('revision-info', '-r', '1')
154
155
# TODO: jam 20060105 We have no revisions with non-ascii characters.
156
bzr('revision-info', '-r', '1', encoding='ascii')
157
158
def test_mv(self):
159
bzr = self.run_bzr_decode
160
161
fname1 = self.info['filename']
162
fname2 = self.info['filename'] + '2'
163
dirname = self.info['directory']
164
165
bzr('mv', 'a', fname1, retcode=3)
166
167
txt = bzr('mv', 'a', fname2)
168
self.assertEqual(u'a => %s\n' % fname2, txt)
169
self.failIfExists('a')
170
self.failUnlessExists(fname2)
171
172
bzr('commit', '-m', 'renamed to non-ascii')
173
174
bzr('mkdir', dirname)
175
txt = bzr('mv', fname1, fname2, dirname)
176
self.assertEqual([u'%s => %s/%s' % (fname1, dirname, fname1),
177
u'%s => %s/%s' % (fname2, dirname, fname2)]
178
, txt.splitlines())
179
180
# The rename should still succeed
181
newpath = u'%s/%s' % (dirname, fname2)
182
txt = bzr('mv', newpath, 'a', encoding='ascii')
183
self.failUnlessExists('a')
184
self.assertEqual(newpath.encode('ascii', 'replace'), txt)
185
186
def test_branch(self):
187
# We should be able to branch into a directory that
188
# has a unicode name, even if we can't display the name
189
bzr = self.run_bzr_decode
190
191
bzr('branch', u'.', self.info['directory'])
192
193
bzr('branch', u'.', self.info['directory'] + '2', encoding='ascii')
194
195
def test_pull(self):
196
# Make sure we can pull from paths that can't be encoded
197
bzr = self.run_bzr_decode
198
199
dirname1 = self.info['directory']
200
dirname2 = self.info['directory'] + '2'
201
bzr('branch', '.', dirname1)
202
bzr('branch', dirname1, dirname2)
203
204
os.chdir(dirname1)
205
open('a', 'ab').write('more text\n')
206
bzr('commit', '-m', 'mod a')
207
208
pwd = os.getcwdu()
209
210
os.chdir(u'../' + dirname2)
211
txt = bzr('pull')
212
213
self.assertEqual(u'Using saved location: %s\n' % (pwd,), txt)
214
215
os.chdir('../' + dirname1)
216
open('a', 'ab').write('and yet more\n')
217
bzr('commit', '-m', 'modifying a by ' + self.info['committer'])
218
219
os.chdir('../' + dirname2)
220
# We should be able to pull, even if our encoding is bad
221
bzr('pull', '--verbose', encoding='ascii')
222
223
def test_push(self):
224
# TODO: Test push to an SFTP location
225
# Make sure we can pull from paths that can't be encoded
226
bzr = self.run_bzr_decode
227
228
dirname = self.info['directory']
229
bzr('push', dirname)
230
231
open('a', 'ab').write('adding more text\n')
232
bzr('commit', '-m', 'added some stuff')
233
234
bzr('push')
235
236
f = open('a', 'ab')
237
f.write('and a bit more: ')
238
f.write(dirname.encode('utf-8'))
239
f.write('\n')
240
f.close()
241
242
bzr('commit', '-m', u'Added some ' + dirname)
243
bzr('push', '--verbose', encoding='ascii')
244
245
bzr('push', '--verbose', dirname + '2')
246
247
bzr('push', '--verbose', dirname + '3', encoding='ascii')
248
249
def test_renames(self):
250
bzr = self.run_bzr_decode
251
252
fname = self.info['filename'] + '2'
253
bzr('mv', 'a', fname)
254
txt = bzr('renames')
255
self.assertEqual(u'a => %s\n' % fname, txt)
256
257
bzr('renames', retcode=3, encoding='ascii')
258
259
def test_remove(self):
260
bzr = self.run_bzr_decode
261
262
fname = self.info['filename']
263
txt = bzr('remove', fname, encoding='ascii')
264
265
def test_remove_verbose(self):
266
bzr = self.run_bzr_decode
267
268
raise TestSkipped('bzr remove --verbose uses tree.remove, which calls print directly.')
269
fname = self.info['filename']
270
txt = bzr('remove', '--verbose', fname, encoding='ascii')
271
272
def test_file_id(self):
273
bzr = self.run_bzr_decode
274
275
fname = self.info['filename']
276
txt = bzr('file-id', fname)
277
278
# TODO: jam 20060106 We don't support non-ascii file ids yet,
279
# so there is nothing which would fail in ascii encoding
280
# This *should* be retcode=3
281
txt = bzr('file-id', fname, encoding='ascii')
282
283
def test_file_path(self):
284
bzr = self.run_bzr_decode
285
286
# Create a directory structure
287
fname = self.info['filename']
288
dirname = self.info['directory']
289
bzr('mkdir', 'base')
290
bzr('mkdir', 'base/' + dirname)
291
path = '/'.join(['base', dirname, fname])
292
bzr('mv', fname, path)
293
bzr('commit', '-m', 'moving things around')
294
295
txt = bzr('file-path', path)
296
297
# TODO: jam 20060106 We don't support non-ascii file ids yet,
298
# so there is nothing which would fail in ascii encoding
299
# This *should* be retcode=3
300
txt = bzr('file-path', path, encoding='ascii')
301
302
def test_revision_history(self):
303
bzr = self.run_bzr_decode
304
305
# TODO: jam 20060106 We don't support non-ascii revision ids yet,
306
# so there is nothing which would fail in ascii encoding
307
txt = bzr('revision-history')
308
309
def test_ancestry(self):
310
bzr = self.run_bzr_decode
311
312
# TODO: jam 20060106 We don't support non-ascii revision ids yet,
313
# so there is nothing which would fail in ascii encoding
314
txt = bzr('ancestry')
315
316
def test_diff(self):
317
# TODO: jam 20060106 diff is a difficult one to test, because it
318
# shouldn't encode the file contents, but it needs some sort
319
# of encoding for the paths, etc which are displayed.
320
pass
321
322
def test_deleted(self):
323
bzr = self.run_bzr_decode
324
325
fname = self.info['filename']
326
os.remove(fname)
327
bzr('rm', fname)
328
329
txt = bzr('deleted')
330
self.assertEqual(fname+'\n', txt)
331
332
txt = bzr('deleted', '--show-ids')
333
self.failUnless(txt.startswith(fname))
334
335
# Deleted should fail if cannot decode
336
# Because it is giving the exact paths
337
# which might be used by a front end
338
bzr('deleted', encoding='ascii', retcode=3)
339
340
def test_modified(self):
341
bzr = self.run_bzr_decode
342
343
fname = self.info['filename']
344
open(fname, 'ab').write('modified\n')
345
346
txt = bzr('modified')
347
self.assertEqual(fname+'\n', txt)
348
349
bzr('modified', encoding='ascii', retcode=3)
350
351
def test_added(self):
352
bzr = self.run_bzr_decode
353
354
fname = self.info['filename'] + '2'
355
open(fname, 'wb').write('added\n')
356
bzr('add', fname)
357
358
txt = bzr('added')
359
self.assertEqual(fname+'\n', txt)
360
361
bzr('added', encoding='ascii', retcode=3)
362
363
def test_root(self):
364
bzr = self.run_bzr_decode
365
366
dirname = self.info['directory']
367
bzr('root')
368
369
bzr('branch', u'.', dirname)
370
371
os.chdir(dirname)
372
373
txt = bzr('root')
374
self.failUnless(txt.endswith(dirname+'\n'))
375
376
txt = bzr('root', encoding='ascii', retcode=3)
377
378
def test_log(self):
379
bzr = self.run_bzr_decode
380
381
fname = self.info['filename']
382
383
txt = bzr('log')
384
self.assertNotEqual(-1, txt.find(self.info['committer']))
385
self.assertNotEqual(-1, txt.find(self.info['message']))
386
387
txt = bzr('log', '--verbose')
388
self.assertNotEqual(-1, txt.find(fname))
389
390
# Make sure log doesn't fail even if we can't write out
391
txt = bzr('log', '--verbose', encoding='ascii')
392
self.assertEqual(-1, txt.find(fname))
393
self.assertNotEqual(-1, txt.find(fname.encode('ascii', 'replace')))
394
395
def test_touching_revisions(self):
396
bzr = self.run_bzr_decode
397
398
fname = self.info['filename']
399
txt = bzr('touching-revisions', fname)
400
self.assertEqual(u' 3 added %s\n' % (fname,), txt)
401
402
fname2 = self.info['filename'] + '2'
403
bzr('mv', fname, fname2)
404
bzr('commit', '-m', u'Renamed %s => %s' % (fname, fname2))
405
406
txt = bzr('touching-revisions', fname2)
407
expected_txt = (u' 3 added %s\n'
408
u' 4 renamed %s => %s\n'
409
% (fname, fname, fname2))
410
self.assertEqual(expected_txt, txt)
411
412
txt = bzr('touching-revisions', fname2, encoding='ascii')
413
expected_ascii = expected_txt.encode('ascii', 'replace')
414
self.assertEqual(expected_ascii, txt)
415
416
def test_ls(self):
417
bzr = self.run_bzr_decode
418
419
txt = bzr('ls')
420
self.assertEqual(['a', 'b', self.info['filename']],
421
txt.splitlines())
422
txt = bzr('ls', '--null')
423
self.assertEqual(['a', 'b', self.info['filename'], ''],
424
txt.split('\0'))
425
426
txt = bzr('ls', encoding='ascii', retcode=3)
427
txt = bzr('ls', '--null', encoding='ascii', retcode=3)
428
429
def test_unknowns(self):
430
bzr = self.run_bzr_decode
431
432
fname = self.info['filename'] + '2'
433
open(fname, 'wb').write('unknown\n')
434
435
# TODO: jam 20060112 bzr unknowns is the only one which
436
# quotes paths do we really want it to?
437
txt = bzr('unknowns')
438
self.assertEqual(u'"%s"\n' % (fname,), txt)
439
440
bzr('unknowns', encoding='ascii', retcode=3)
441
442
def test_ignore(self):
443
bzr = self.run_bzr_decode
444
445
fname2 = self.info['filename'] + '2.txt'
446
open(fname2, 'wb').write('ignored\n')
447
448
txt = bzr('unknowns')
449
self.assertEqual(u'"%s"\n' % (fname2,), txt)
450
451
bzr('ignore', './' + fname2)
452
txt = bzr('unknowns')
453
# TODO: jam 20060107 This is the correct output
454
# self.assertEqual('', txt)
455
# This is the incorrect output
456
self.assertEqual(u'"%s"\n' % (fname2,), txt)
457
458
fname3 = self.info['filename'] + '3.txt'
459
open(fname3, 'wb').write('unknown 3\n')
460
txt = bzr('unknowns')
461
# TODO: jam 20060107 This is the correct output
462
# self.assertEqual(u'"%s"\n' % (fname3,), txt)
463
# This is the incorrect output
464
self.assertEqual(u'"%s"\n"%s"\n' % (fname2, fname3,), txt)
465
466
# Ignore should not care what the encoding is
467
# (right now it doesn't print anything)
468
bzr('ignore', fname3, encoding='ascii')
469
txt = bzr('unknowns')
470
# TODO: jam 20060107 This is the correct output
471
# self.assertEqual('', txt)
472
# This is the incorrect output
473
self.assertEqual(u'"%s"\n"%s"\n' % (fname2, fname3), txt)
474
475
# Now try a wildcard match
476
fname4 = self.info['filename'] + '4.txt'
477
open(fname4, 'wb').write('unknown 4\n')
478
bzr('ignore', '*.txt')
479
txt = bzr('unknowns')
480
self.assertEqual('', txt)
481
482
os.remove('.bzrignore')
483
bzr('ignore', self.info['filename'] + '*')
484
txt = bzr('unknowns')
485
# TODO: jam 20060107 This is the correct output
486
# self.assertEqual('', txt)
487
# This is the incorrect output
488
self.assertEqual(u'"%s"\n"%s"\n"%s"\n' % (fname2, fname3, fname4), txt)
489
490
# TODO: jam 20060107 The best error we have right now is TestSkipped
491
# to indicate that this test is known to fail
492
raise TestSkipped("WorkingTree.is_ignored doesn't match unicode filenames (yet)")
493
494
Loggerhead is a web-based interface for Breezy
Version: 2.0.1