1
# Copyright (C) 2005, 2006 Canonical Development Ltd
3
# This program is free software; you can redistribute it and/or modify
4
# it under the terms of the GNU General Public License as published by
5
# the Free Software Foundation; either version 2 of the License, or
6
# (at your option) any later version.
8
# This program is distributed in the hope that it will be useful,
9
# but WITHOUT ANY WARRANTY; without even the implied warranty of
10
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11
# GNU General Public License for more details.
13
# You should have received a copy of the GNU General Public License
14
# along with this program; if not, write to the Free Software
15
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
18
from cStringIO import StringIO
20
from tempfile import TemporaryFile
22
from bzrlib.diff import internal_diff, external_diff, show_diff_trees
23
from bzrlib.errors import BinaryFile, NoDiff
24
import bzrlib.patiencediff
25
from bzrlib.tests import (TestCase, TestCaseWithTransport,
26
TestCaseInTempDir, TestSkipped)
29
def udiff_lines(old, new, allow_binary=False):
31
internal_diff('old', old, 'new', new, output, allow_binary)
33
return output.readlines()
36
def external_udiff_lines(old, new, use_stringio=False):
38
# StringIO has no fileno, so it tests a different codepath
41
output = TemporaryFile()
43
external_diff('old', old, 'new', new, output, diff_opts=['-u'])
45
raise TestSkipped('external "diff" not present to test')
47
lines = output.readlines()
52
class TestDiff(TestCase):
54
def test_add_nl(self):
55
"""diff generates a valid diff for patches that add a newline"""
56
lines = udiff_lines(['boo'], ['boo\n'])
57
self.check_patch(lines)
58
self.assertEquals(lines[4], '\\ No newline at end of file\n')
59
## "expected no-nl, got %r" % lines[4]
61
def test_add_nl_2(self):
62
"""diff generates a valid diff for patches that change last line and
65
lines = udiff_lines(['boo'], ['goo\n'])
66
self.check_patch(lines)
67
self.assertEquals(lines[4], '\\ No newline at end of file\n')
68
## "expected no-nl, got %r" % lines[4]
70
def test_remove_nl(self):
71
"""diff generates a valid diff for patches that change last line and
74
lines = udiff_lines(['boo\n'], ['boo'])
75
self.check_patch(lines)
76
self.assertEquals(lines[5], '\\ No newline at end of file\n')
77
## "expected no-nl, got %r" % lines[5]
79
def check_patch(self, lines):
80
self.assert_(len(lines) > 1)
81
## "Not enough lines for a file header for patch:\n%s" % "".join(lines)
82
self.assert_(lines[0].startswith ('---'))
83
## 'No orig line for patch:\n%s' % "".join(lines)
84
self.assert_(lines[1].startswith ('+++'))
85
## 'No mod line for patch:\n%s' % "".join(lines)
86
self.assert_(len(lines) > 2)
87
## "No hunks for patch:\n%s" % "".join(lines)
88
self.assert_(lines[2].startswith('@@'))
89
## "No hunk header for patch:\n%s" % "".join(lines)
90
self.assert_('@@' in lines[2][2:])
91
## "Unterminated hunk header for patch:\n%s" % "".join(lines)
93
def test_binary_lines(self):
94
self.assertRaises(BinaryFile, udiff_lines, [1023 * 'a' + '\x00'], [])
95
self.assertRaises(BinaryFile, udiff_lines, [], [1023 * 'a' + '\x00'])
96
udiff_lines([1023 * 'a' + '\x00'], [], allow_binary=True)
97
udiff_lines([], [1023 * 'a' + '\x00'], allow_binary=True)
99
def test_external_diff(self):
100
lines = external_udiff_lines(['boo\n'], ['goo\n'])
101
self.check_patch(lines)
102
self.assertEqual('\n', lines[-1])
104
def test_external_diff_no_fileno(self):
105
# Make sure that we can handle not having a fileno, even
106
# if the diff is large
107
lines = external_udiff_lines(['boo\n']*10000,
110
self.check_patch(lines)
112
def test_external_diff_binary(self):
113
lines = external_udiff_lines(['\x00foobar\n'], ['foo\x00bar\n'])
114
self.assertEqual(['Binary files old and new differ\n', '\n'], lines)
116
def test_no_external_diff(self):
117
"""Check that NoDiff is raised when diff is not available"""
118
# Use os.environ['PATH'] to make sure no 'diff' command is available
119
orig_path = os.environ['PATH']
121
os.environ['PATH'] = ''
122
self.assertRaises(NoDiff, external_diff,
123
'old', ['boo\n'], 'new', ['goo\n'],
124
StringIO(), diff_opts=['-u'])
126
os.environ['PATH'] = orig_path
128
def test_internal_diff_default(self):
129
# Default internal diff encoding is utf8
131
internal_diff(u'old_\xb5', ['old_text\n'],
132
u'new_\xe5', ['new_text\n'], output)
133
lines = output.getvalue().splitlines(True)
134
self.check_patch(lines)
135
self.assertEquals(['--- old_\xc2\xb5\n',
136
'+++ new_\xc3\xa5\n',
144
def test_internal_diff_utf8(self):
146
internal_diff(u'old_\xb5', ['old_text\n'],
147
u'new_\xe5', ['new_text\n'], output,
148
path_encoding='utf8')
149
lines = output.getvalue().splitlines(True)
150
self.check_patch(lines)
151
self.assertEquals(['--- old_\xc2\xb5\n',
152
'+++ new_\xc3\xa5\n',
160
def test_internal_diff_iso_8859_1(self):
162
internal_diff(u'old_\xb5', ['old_text\n'],
163
u'new_\xe5', ['new_text\n'], output,
164
path_encoding='iso-8859-1')
165
lines = output.getvalue().splitlines(True)
166
self.check_patch(lines)
167
self.assertEquals(['--- old_\xb5\n',
176
def test_internal_diff_returns_bytes(self):
178
output = StringIO.StringIO()
179
internal_diff(u'old_\xb5', ['old_text\n'],
180
u'new_\xe5', ['new_text\n'], output)
181
self.failUnless(isinstance(output.getvalue(), str),
182
'internal_diff should return bytestrings')
185
class TestDiffDates(TestCaseWithTransport):
188
super(TestDiffDates, self).setUp()
189
self.wt = self.make_branch_and_tree('.')
190
self.b = self.wt.branch
191
self.build_tree_contents([
192
('file1', 'file1 contents at rev 1\n'),
193
('file2', 'file2 contents at rev 1\n')
195
self.wt.add(['file1', 'file2'])
197
message='Revision 1',
198
timestamp=1143849600, # 2006-04-01 00:00:00 UTC
201
self.build_tree_contents([('file1', 'file1 contents at rev 2\n')])
203
message='Revision 2',
204
timestamp=1143936000, # 2006-04-02 00:00:00 UTC
207
self.build_tree_contents([('file2', 'file2 contents at rev 3\n')])
209
message='Revision 3',
210
timestamp=1144022400, # 2006-04-03 00:00:00 UTC
213
self.wt.remove(['file2'])
215
message='Revision 4',
216
timestamp=1144108800, # 2006-04-04 00:00:00 UTC
219
self.build_tree_contents([
220
('file1', 'file1 contents in working tree\n')
222
# set the date stamps for files in the working tree to known values
223
os.utime('file1', (1144195200, 1144195200)) # 2006-04-05 00:00:00 UTC
225
def get_diff(self, tree1, tree2, specific_files=None, working_tree=None):
227
if working_tree is not None:
228
extra_trees = (working_tree,)
231
show_diff_trees(tree1, tree2, output, specific_files=specific_files,
232
extra_trees=extra_trees, old_label='old/',
234
return output.getvalue()
236
def test_diff_rev_tree_working_tree(self):
237
output = self.get_diff(self.wt.basis_tree(), self.wt)
238
# note that the date for old/file1 is from rev 2 rather than from
239
# the basis revision (rev 4)
240
self.assertEqualDiff(output, '''\
241
=== modified file 'file1'
242
--- old/file1\t2006-04-02 00:00:00 +0000
243
+++ new/file1\t2006-04-05 00:00:00 +0000
245
-file1 contents at rev 2
246
+file1 contents in working tree
250
def test_diff_rev_tree_rev_tree(self):
251
tree1 = self.b.repository.revision_tree('rev-2')
252
tree2 = self.b.repository.revision_tree('rev-3')
253
output = self.get_diff(tree1, tree2)
254
self.assertEqualDiff(output, '''\
255
=== modified file 'file2'
256
--- old/file2\t2006-04-01 00:00:00 +0000
257
+++ new/file2\t2006-04-03 00:00:00 +0000
259
-file2 contents at rev 1
260
+file2 contents at rev 3
264
def test_diff_add_files(self):
265
tree1 = self.b.repository.revision_tree(None)
266
tree2 = self.b.repository.revision_tree('rev-1')
267
output = self.get_diff(tree1, tree2)
268
# the files have the epoch time stamp for the tree in which
270
self.assertEqualDiff(output, '''\
271
=== added file 'file1'
272
--- old/file1\t1970-01-01 00:00:00 +0000
273
+++ new/file1\t2006-04-01 00:00:00 +0000
275
+file1 contents at rev 1
277
=== added file 'file2'
278
--- old/file2\t1970-01-01 00:00:00 +0000
279
+++ new/file2\t2006-04-01 00:00:00 +0000
281
+file2 contents at rev 1
285
def test_diff_remove_files(self):
286
tree1 = self.b.repository.revision_tree('rev-3')
287
tree2 = self.b.repository.revision_tree('rev-4')
288
output = self.get_diff(tree1, tree2)
289
# the file has the epoch time stamp for the tree in which
291
self.assertEqualDiff(output, '''\
292
=== removed file 'file2'
293
--- old/file2\t2006-04-03 00:00:00 +0000
294
+++ new/file2\t1970-01-01 00:00:00 +0000
296
-file2 contents at rev 3
300
def test_show_diff_specified(self):
301
"""A working tree filename can be used to identify a file"""
302
self.wt.rename_one('file1', 'file1b')
303
old_tree = self.b.repository.revision_tree('rev-1')
304
new_tree = self.b.repository.revision_tree('rev-4')
305
out = self.get_diff(old_tree, new_tree, specific_files=['file1b'],
306
working_tree=self.wt)
307
self.assertContainsRe(out, 'file1\t')
309
def test_recursive_diff(self):
310
"""Children of directories are matched"""
313
self.wt.add(['dir1', 'dir2'])
314
self.wt.rename_one('file1', 'dir1/file1')
315
old_tree = self.b.repository.revision_tree('rev-1')
316
new_tree = self.b.repository.revision_tree('rev-4')
317
out = self.get_diff(old_tree, new_tree, specific_files=['dir1'],
318
working_tree=self.wt)
319
self.assertContainsRe(out, 'file1\t')
320
out = self.get_diff(old_tree, new_tree, specific_files=['dir2'],
321
working_tree=self.wt)
322
self.assertNotContainsRe(out, 'file1\t')
325
class TestPatienceDiffLib(TestCase):
327
def test_unique_lcs(self):
328
unique_lcs = bzrlib.patiencediff.unique_lcs
329
self.assertEquals(unique_lcs('', ''), [])
330
self.assertEquals(unique_lcs('a', 'a'), [(0,0)])
331
self.assertEquals(unique_lcs('a', 'b'), [])
332
self.assertEquals(unique_lcs('ab', 'ab'), [(0,0), (1,1)])
333
self.assertEquals(unique_lcs('abcde', 'cdeab'), [(2,0), (3,1), (4,2)])
334
self.assertEquals(unique_lcs('cdeab', 'abcde'), [(0,2), (1,3), (2,4)])
335
self.assertEquals(unique_lcs('abXde', 'abYde'), [(0,0), (1,1),
337
self.assertEquals(unique_lcs('acbac', 'abc'), [(2,1)])
339
def test_recurse_matches(self):
340
def test_one(a, b, matches):
342
bzrlib.patiencediff.recurse_matches(a, b, 0, 0, len(a), len(b),
344
self.assertEquals(test_matches, matches)
346
test_one(['a', '', 'b', '', 'c'], ['a', 'a', 'b', 'c', 'c'],
347
[(0, 0), (2, 2), (4, 4)])
348
test_one(['a', 'c', 'b', 'a', 'c'], ['a', 'b', 'c'],
349
[(0, 0), (2, 1), (4, 2)])
351
# recurse_matches doesn't match non-unique
352
# lines surrounded by bogus text.
353
# The update has been done in patiencediff.SequenceMatcher instead
355
# This is what it could be
356
#test_one('aBccDe', 'abccde', [(0,0), (2,2), (3,3), (5,5)])
358
# This is what it currently gives:
359
test_one('aBccDe', 'abccde', [(0,0), (5,5)])
361
def test_matching_blocks(self):
362
def chk_blocks(a, b, expected_blocks):
363
# difflib always adds a signature of the total
364
# length, with no matching entries at the end
365
s = bzrlib.patiencediff.PatienceSequenceMatcher(None, a, b)
366
blocks = s.get_matching_blocks()
367
self.assertEquals((len(a), len(b), 0), blocks[-1])
368
self.assertEquals(expected_blocks, blocks[:-1])
370
# Some basic matching tests
371
chk_blocks('', '', [])
372
chk_blocks([], [], [])
373
chk_blocks('abcd', 'abcd', [(0, 0, 4)])
374
chk_blocks('abcd', 'abce', [(0, 0, 3)])
375
chk_blocks('eabc', 'abce', [(1, 0, 3)])
376
chk_blocks('eabce', 'abce', [(1, 0, 4)])
377
chk_blocks('abcde', 'abXde', [(0, 0, 2), (3, 3, 2)])
378
chk_blocks('abcde', 'abXYZde', [(0, 0, 2), (3, 5, 2)])
379
chk_blocks('abde', 'abXYZde', [(0, 0, 2), (2, 5, 2)])
380
# This may check too much, but it checks to see that
381
# a copied block stays attached to the previous section,
383
# difflib would tend to grab the trailing longest match
384
# which would make the diff not look right
385
chk_blocks('abcdefghijklmnop', 'abcdefxydefghijklmnop',
386
[(0, 0, 6), (6, 11, 10)])
388
# make sure it supports passing in lists
392
'how are you today?\n'],
394
'how are you today?\n'],
395
[(0, 0, 1), (2, 1, 1)])
397
# non unique lines surrounded by non-matching lines
399
chk_blocks('aBccDe', 'abccde', [(0,0,1), (5,5,1)])
401
# But they only need to be locally unique
402
chk_blocks('aBcDec', 'abcdec', [(0,0,1), (2,2,1), (4,4,2)])
404
# non unique blocks won't be matched
405
chk_blocks('aBcdEcdFg', 'abcdecdfg', [(0,0,1), (8,8,1)])
407
# but locally unique ones will
408
chk_blocks('aBcdEeXcdFg', 'abcdecdfg', [(0,0,1), (2,2,2),
409
(5,4,1), (7,5,2), (10,8,1)])
411
chk_blocks('abbabbXd', 'cabbabxd', [(7,7,1)])
412
chk_blocks('abbabbbb', 'cabbabbc', [])
413
chk_blocks('bbbbbbbb', 'cbbbbbbc', [])
415
def test_opcodes(self):
416
def chk_ops(a, b, expected_codes):
417
s = bzrlib.patiencediff.PatienceSequenceMatcher(None, a, b)
418
self.assertEquals(expected_codes, s.get_opcodes())
422
chk_ops('abcd', 'abcd', [('equal', 0,4, 0,4)])
423
chk_ops('abcd', 'abce', [('equal', 0,3, 0,3),
424
('replace', 3,4, 3,4)
426
chk_ops('eabc', 'abce', [('delete', 0,1, 0,0),
430
chk_ops('eabce', 'abce', [('delete', 0,1, 0,0),
433
chk_ops('abcde', 'abXde', [('equal', 0,2, 0,2),
434
('replace', 2,3, 2,3),
437
chk_ops('abcde', 'abXYZde', [('equal', 0,2, 0,2),
438
('replace', 2,3, 2,5),
441
chk_ops('abde', 'abXYZde', [('equal', 0,2, 0,2),
442
('insert', 2,2, 2,5),
445
chk_ops('abcdefghijklmnop', 'abcdefxydefghijklmnop',
446
[('equal', 0,6, 0,6),
447
('insert', 6,6, 6,11),
448
('equal', 6,16, 11,21)
453
, 'how are you today?\n'],
455
, 'how are you today?\n'],
456
[('equal', 0,1, 0,1),
457
('delete', 1,2, 1,1),
460
chk_ops('aBccDe', 'abccde',
461
[('equal', 0,1, 0,1),
462
('replace', 1,5, 1,5),
465
chk_ops('aBcDec', 'abcdec',
466
[('equal', 0,1, 0,1),
467
('replace', 1,2, 1,2),
469
('replace', 3,4, 3,4),
472
chk_ops('aBcdEcdFg', 'abcdecdfg',
473
[('equal', 0,1, 0,1),
474
('replace', 1,8, 1,8),
477
chk_ops('aBcdEeXcdFg', 'abcdecdfg',
478
[('equal', 0,1, 0,1),
479
('replace', 1,2, 1,2),
481
('delete', 4,5, 4,4),
483
('delete', 6,7, 5,5),
485
('replace', 9,10, 7,8),
486
('equal', 10,11, 8,9)
489
def test_multiple_ranges(self):
490
# There was an earlier bug where we used a bad set of ranges,
491
# this triggers that specific bug, to make sure it doesn't regress
492
def chk_blocks(a, b, expected_blocks):
493
# difflib always adds a signature of the total
494
# length, with no matching entries at the end
495
s = bzrlib.patiencediff.PatienceSequenceMatcher(None, a, b)
496
blocks = s.get_matching_blocks()
498
self.assertEquals(x, (len(a), len(b), 0))
499
self.assertEquals(expected_blocks, blocks)
501
chk_blocks('abcdefghijklmnop'
502
, 'abcXghiYZQRSTUVWXYZijklmnop'
503
, [(0, 0, 3), (6, 4, 3), (9, 20, 7)])
505
chk_blocks('ABCd efghIjk L'
506
, 'AxyzBCn mo pqrstuvwI1 2 L'
507
, [(0,0,1), (1, 4, 2), (9, 19, 1), (12, 23, 3)])
509
# These are rot13 code snippets.
511
trg nqqrq jura lbh nqq n svyr va gur qverpgbel.
513
gnxrf_netf = ['svyr*']
514
gnxrf_bcgvbaf = ['ab-erphefr']
516
qrs eha(frys, svyr_yvfg, ab_erphefr=Snyfr):
517
sebz omeyvo.nqq vzcbeg fzneg_nqq, nqq_ercbegre_cevag, nqq_ercbegre_ahyy
519
ercbegre = nqq_ercbegre_ahyy
521
ercbegre = nqq_ercbegre_cevag
522
fzneg_nqq(svyr_yvfg, abg ab_erphefr, ercbegre)
525
pynff pzq_zxqve(Pbzznaq):
526
'''.splitlines(True), '''\
527
trg nqqrq jura lbh nqq n svyr va gur qverpgbel.
529
--qel-eha jvyy fubj juvpu svyrf jbhyq or nqqrq, ohg abg npghnyyl
532
gnxrf_netf = ['svyr*']
533
gnxrf_bcgvbaf = ['ab-erphefr', 'qel-eha']
535
qrs eha(frys, svyr_yvfg, ab_erphefr=Snyfr, qel_eha=Snyfr):
540
# Guvf vf cbvagyrff, ohg V'q engure abg envfr na reebe
541
npgvba = omeyvo.nqq.nqq_npgvba_ahyy
543
npgvba = omeyvo.nqq.nqq_npgvba_cevag
545
npgvba = omeyvo.nqq.nqq_npgvba_nqq
547
npgvba = omeyvo.nqq.nqq_npgvba_nqq_naq_cevag
549
omeyvo.nqq.fzneg_nqq(svyr_yvfg, abg ab_erphefr, npgvba)
552
pynff pzq_zxqve(Pbzznaq):
554
, [(0,0,1), (1, 4, 2), (9, 19, 1), (12, 23, 3)])
556
def test_patience_unified_diff(self):
557
txt_a = ['hello there\n',
559
'how are you today?\n']
560
txt_b = ['hello there\n',
561
'how are you today?\n']
562
unified_diff = bzrlib.patiencediff.unified_diff
563
psm = bzrlib.patiencediff.PatienceSequenceMatcher
564
self.assertEquals([ '--- \n',
569
' how are you today?\n'
571
, list(unified_diff(txt_a, txt_b,
572
sequencematcher=psm)))
573
txt_a = map(lambda x: x+'\n', 'abcdefghijklmnop')
574
txt_b = map(lambda x: x+'\n', 'abcdefxydefghijklmnop')
575
# This is the result with LongestCommonSubstring matching
576
self.assertEquals(['--- \n',
578
'@@ -1,6 +1,11 @@\n',
590
, list(unified_diff(txt_a, txt_b)))
591
# And the patience diff
592
self.assertEquals(['--- \n',
594
'@@ -4,6 +4,11 @@\n',
607
, list(unified_diff(txt_a, txt_b,
608
sequencematcher=psm)))
611
class TestPatienceDiffLibFiles(TestCaseInTempDir):
613
def test_patience_unified_diff_files(self):
614
txt_a = ['hello there\n',
616
'how are you today?\n']
617
txt_b = ['hello there\n',
618
'how are you today?\n']
619
open('a1', 'wb').writelines(txt_a)
620
open('b1', 'wb').writelines(txt_b)
622
unified_diff_files = bzrlib.patiencediff.unified_diff_files
623
psm = bzrlib.patiencediff.PatienceSequenceMatcher
624
self.assertEquals(['--- a1 \n',
629
' how are you today?\n',
631
, list(unified_diff_files('a1', 'b1',
632
sequencematcher=psm)))
634
txt_a = map(lambda x: x+'\n', 'abcdefghijklmnop')
635
txt_b = map(lambda x: x+'\n', 'abcdefxydefghijklmnop')
636
open('a2', 'wb').writelines(txt_a)
637
open('b2', 'wb').writelines(txt_b)
639
# This is the result with LongestCommonSubstring matching
640
self.assertEquals(['--- a2 \n',
642
'@@ -1,6 +1,11 @@\n',
654
, list(unified_diff_files('a2', 'b2')))
656
# And the patience diff
657
self.assertEquals(['--- a2 \n',
659
'@@ -4,6 +4,11 @@\n',
672
, list(unified_diff_files('a2', 'b2',
673
sequencematcher=psm)))
added
removed
bzrlib/tests/test_diff.py
