14
14
# You should have received a copy of the GNU General Public License
15
15
# along with this program; if not, write to the Free Software
16
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
18
from __future__ import absolute_import
20
from bzrlib.errors import (
32
binary_files_re = 'Binary files (.*) and (.*) differ\n'
16
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
19
class PatchSyntax(Exception):
20
def __init__(self, msg):
21
Exception.__init__(self, msg)
24
class MalformedPatchHeader(PatchSyntax):
25
def __init__(self, desc, line):
28
msg = "Malformed patch header. %s\n%r" % (self.desc, self.line)
29
PatchSyntax.__init__(self, msg)
32
class MalformedHunkHeader(PatchSyntax):
33
def __init__(self, desc, line):
36
msg = "Malformed hunk header. %s\n%r" % (self.desc, self.line)
37
PatchSyntax.__init__(self, msg)
40
class MalformedLine(PatchSyntax):
41
def __init__(self, desc, line):
44
msg = "Malformed line. %s\n%s" % (self.desc, self.line)
45
PatchSyntax.__init__(self, msg)
48
class PatchConflict(Exception):
49
def __init__(self, line_no, orig_line, patch_line):
50
orig = orig_line.rstrip('\n')
51
patch = str(patch_line).rstrip('\n')
52
msg = 'Text contents mismatch at line %d. Original has "%s",'\
53
' but patch says it should be "%s"' % (line_no, orig, patch)
54
Exception.__init__(self, msg)
35
57
def get_patch_names(iter_lines):
36
line = iter_lines.next()
38
match = re.match(binary_files_re, line)
40
raise BinaryFiles(match.group(1), match.group(2))
59
line = iter_lines.next()
41
60
if not line.startswith("--- "):
42
61
raise MalformedPatchHeader("No orig name", line)
74
93
return (pos, range)
77
96
def hunk_from_header(line):
79
matches = re.match(r'\@\@ ([^@]*) \@\@( (.*))?\n', line)
81
raise MalformedHunkHeader("Does not match format.", line)
97
if not line.startswith("@@") or not line.endswith("@@\n") \
99
raise MalformedHunkHeader("Does not start and end with @@.", line)
83
(orig, mod) = matches.group(1).split(" ")
84
except (ValueError, IndexError), e:
101
(orig, mod) = line[3:-4].split(" ")
85
103
raise MalformedHunkHeader(str(e), line)
86
104
if not orig.startswith('-') or not mod.startswith('+'):
87
105
raise MalformedHunkHeader("Positions don't start with + or -.", line)
89
107
(orig_pos, orig_range) = parse_range(orig[1:])
90
108
(mod_pos, mod_range) = parse_range(mod[1:])
91
except (ValueError, IndexError), e:
92
110
raise MalformedHunkHeader(str(e), line)
93
111
if mod_range < 0 or orig_range < 0:
94
112
raise MalformedHunkHeader("Hunk range is negative", line)
95
tail = matches.group(3)
96
return Hunk(orig_pos, orig_range, mod_pos, mod_range, tail)
113
return Hunk(orig_pos, orig_range, mod_pos, mod_range)
145
162
return InsertLine(line[1:])
146
163
elif line.startswith("-"):
147
164
return RemoveLine(line[1:])
149
168
raise MalformedLine("Unknown line type", line)
154
def __init__(self, orig_pos, orig_range, mod_pos, mod_range, tail=None):
173
def __init__(self, orig_pos, orig_range, mod_pos, mod_range):
155
174
self.orig_pos = orig_pos
156
175
self.orig_range = orig_range
157
176
self.mod_pos = mod_pos
158
177
self.mod_range = mod_range
162
180
def get_header(self):
163
if self.tail is None:
166
tail_str = ' ' + self.tail
167
return "@@ -%s +%s @@%s\n" % (self.range_str(self.orig_pos,
169
self.range_str(self.mod_pos,
181
return "@@ -%s +%s @@\n" % (self.range_str(self.orig_pos,
183
self.range_str(self.mod_pos,
173
186
def range_str(self, pos, range):
174
187
"""Return a file range, special-casing for 1-line files.
257
class BinaryPatch(object):
258
258
def __init__(self, oldname, newname):
259
259
self.oldname = oldname
260
260
self.newname = newname
263
return 'Binary files %s and %s differ\n' % (self.oldname, self.newname)
266
class Patch(BinaryPatch):
268
def __init__(self, oldname, newname):
269
BinaryPatch.__init__(self, oldname, newname)
272
263
def __str__(self):
273
ret = self.get_header()
264
ret = self.get_header()
274
265
ret += "".join([str(h) for h in self.hunks])
277
268
def get_header(self):
278
269
return "--- %s\n+++ %s\n" % (self.oldname, self.newname)
280
def stats_values(self):
281
"""Calculate the number of inserts and removes."""
272
"""Return a string of patch statistics"""
284
275
for hunk in self.hunks:
288
279
elif isinstance(line, RemoveLine):
290
return (inserts, removes, len(self.hunks))
293
"""Return a string of patch statistics"""
294
281
return "%i inserts, %i removes in %i hunks" % \
282
(inserts, removes, len(self.hunks))
297
284
def pos_in_mod(self, position):
298
285
newpos = position
318
305
if isinstance(line, ContextLine):
321
def parse_patch(iter_lines, allow_dirty=False):
323
:arg iter_lines: iterable of lines to parse
324
:kwarg allow_dirty: If True, allow the patch to have trailing junk.
327
iter_lines = iter_lines_handle_nl(iter_lines)
329
(orig_name, mod_name) = get_patch_names(iter_lines)
330
except BinaryFiles, e:
331
return BinaryPatch(e.orig_name, e.mod_name)
333
patch = Patch(orig_name, mod_name)
334
for hunk in iter_hunks(iter_lines, allow_dirty):
335
patch.hunks.append(hunk)
339
def iter_file_patch(iter_lines, allow_dirty=False, keep_dirty=False):
341
:arg iter_lines: iterable of lines to parse for patches
342
:kwarg allow_dirty: If True, allow comments and other non-patch text
343
before the first patch. Note that the algorithm here can only find
344
such text before any patches have been found. Comments after the
345
first patch are stripped away in iter_hunks() if it is also passed
346
allow_dirty=True. Default False.
348
### FIXME: Docstring is not quite true. We allow certain comments no
349
# matter what, If they startwith '===', '***', or '#' Someone should
350
# reexamine this logic and decide if we should include those in
351
# allow_dirty or restrict those to only being before the patch is found
352
# (as allow_dirty does).
353
regex = re.compile(binary_files_re)
309
def parse_patch(iter_lines):
310
(orig_name, mod_name) = get_patch_names(iter_lines)
311
patch = Patch(orig_name, mod_name)
312
for hunk in iter_hunks(iter_lines):
313
patch.hunks.append(hunk)
317
def iter_file_patch(iter_lines):
359
320
for line in iter_lines:
360
if line.startswith('=== '):
361
if len(saved_lines) > 0:
362
if keep_dirty and len(dirty_head) > 0:
363
yield {'saved_lines': saved_lines,
364
'dirty_head': dirty_head}
369
dirty_head.append(line)
371
if line.startswith('*** '):
321
if line.startswith('=== ') or line.startswith('*** '):
373
323
if line.startswith('#'):
375
325
elif orig_range > 0:
376
326
if line.startswith('-') or line.startswith(' '):
378
elif line.startswith('--- ') or regex.match(line):
379
if allow_dirty and beginning:
380
# Patches can have "junk" at the beginning
381
# Stripping junk from the end of patches is handled when we
384
elif len(saved_lines) > 0:
385
if keep_dirty and len(dirty_head) > 0:
386
yield {'saved_lines': saved_lines,
387
'dirty_head': dirty_head}
328
elif line.startswith('--- '):
329
if len(saved_lines) > 0:
392
332
elif line.startswith('@@'):
393
333
hunk = hunk_from_header(line)
394
334
orig_range = hunk.orig_range
395
335
saved_lines.append(line)
396
336
if len(saved_lines) > 0:
397
if keep_dirty and len(dirty_head) > 0:
398
yield {'saved_lines': saved_lines,
399
'dirty_head': dirty_head}
404
340
def iter_lines_handle_nl(iter_lines):
425
def parse_patches(iter_lines, allow_dirty=False, keep_dirty=False):
427
:arg iter_lines: iterable of lines to parse for patches
428
:kwarg allow_dirty: If True, allow text that's not part of the patch at
429
selected places. This includes comments before and after a patch
430
for instance. Default False.
431
:kwarg keep_dirty: If True, returns a dict of patches with dirty headers.
435
for patch_lines in iter_file_patch(iter_lines, allow_dirty, keep_dirty):
436
if 'dirty_head' in patch_lines:
437
patches.append({'patch': parse_patch(
438
patch_lines['saved_lines'], allow_dirty),
439
'dirty_head': patch_lines['dirty_head']})
441
patches.append(parse_patch(patch_lines, allow_dirty))
360
def parse_patches(iter_lines):
361
iter_lines = iter_lines_handle_nl(iter_lines)
362
return [parse_patch(f.__iter__()) for f in iter_file_patch(iter_lines)]
445
365
def difference_index(atext, btext):
465
385
"""Iterate through a series of lines with a patch applied.
466
386
This handles a single file, and does exact, not fuzzy patching.
468
patch_lines = iter_lines_handle_nl(iter(patch_lines))
388
if orig_lines is not None:
389
orig_lines = orig_lines.__iter__()
391
patch_lines = iter_lines_handle_nl(patch_lines.__iter__())
469
392
get_patch_names(patch_lines)
470
return iter_patched_from_hunks(orig_lines, iter_hunks(patch_lines))
473
def iter_patched_from_hunks(orig_lines, hunks):
474
"""Iterate through a series of lines with a patch applied.
475
This handles a single file, and does exact, not fuzzy patching.
477
:param orig_lines: The unpatched lines.
478
:param hunks: An iterable of Hunk instances.
482
if orig_lines is not None:
483
orig_lines = iter(orig_lines)
394
for hunk in iter_hunks(patch_lines):
485
395
while line_no < hunk.orig_pos:
486
396
orig_line = orig_lines.next()