1
# Copyright (C) 2004 - 2006 Aaron Bentley, Canonical Ltd
1
# Copyright (C) 2005-2010 Aaron Bentley, Canonical Ltd
2
2
# <aaron.bentley@utoronto.ca>
4
4
# This program is free software; you can redistribute it and/or modify
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., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
16
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
20
binary_files_re = 'Binary files (.*) and (.*) differ\n'
23
class BinaryFiles(Exception):
25
def __init__(self, orig_name, mod_name):
26
self.orig_name = orig_name
27
self.mod_name = mod_name
28
Exception.__init__(self, 'Binary files section encountered.')
20
31
class PatchSyntax(Exception):
21
32
def __init__(self, msg):
22
33
Exception.__init__(self, msg)
58
69
def get_patch_names(iter_lines):
60
71
line = iter_lines.next()
72
match = re.match(binary_files_re, line)
74
raise BinaryFiles(match.group(1), match.group(2))
61
75
if not line.startswith("--- "):
62
76
raise MalformedPatchHeader("No orig name", line)
164
179
return InsertLine(line[1:])
165
180
elif line.startswith("-"):
166
181
return RemoveLine(line[1:])
170
183
raise MalformedLine("Unknown line type", line)
240
def iter_hunks(iter_lines):
253
def iter_hunks(iter_lines, allow_dirty=False):
255
:arg iter_lines: iterable of lines to parse for hunks
256
:kwarg allow_dirty: If True, when we encounter something that is not
257
a hunk header when we're looking for one, assume the rest of the lines
258
are not part of the patch (comments or other junk). Default False
242
261
for line in iter_lines:
248
267
if hunk is not None:
250
hunk = hunk_from_header(line)
270
hunk = hunk_from_header(line)
271
except MalformedHunkHeader:
273
# If the line isn't a hunk header, then we've reached the end
274
# of this patch and there's "junk" at the end. Ignore the
275
# rest of this patch.
253
280
while orig_size < hunk.orig_range or mod_size < hunk.mod_range:
291
class BinaryPatch(object):
265
292
def __init__(self, oldname, newname):
266
293
self.oldname = oldname
267
294
self.newname = newname
297
return 'Binary files %s and %s differ\n' % (self.oldname, self.newname)
300
class Patch(BinaryPatch):
302
def __init__(self, oldname, newname):
303
BinaryPatch.__init__(self, oldname, newname)
270
306
def __str__(self):
271
ret = self.get_header()
307
ret = self.get_header()
272
308
ret += "".join([str(h) for h in self.hunks])
275
311
def get_header(self):
276
312
return "--- %s\n+++ %s\n" % (self.oldname, self.newname)
279
"""Return a string of patch statistics"""
314
def stats_values(self):
315
"""Calculate the number of inserts and removes."""
282
318
for hunk in self.hunks:
286
322
elif isinstance(line, RemoveLine):
324
return (inserts, removes, len(self.hunks))
327
"""Return a string of patch statistics"""
288
328
return "%i inserts, %i removes in %i hunks" % \
289
(inserts, removes, len(self.hunks))
291
331
def pos_in_mod(self, position):
292
332
newpos = position
316
def parse_patch(iter_lines):
317
(orig_name, mod_name) = get_patch_names(iter_lines)
318
patch = Patch(orig_name, mod_name)
319
for hunk in iter_hunks(iter_lines):
320
patch.hunks.append(hunk)
324
def iter_file_patch(iter_lines):
356
def parse_patch(iter_lines, allow_dirty=False):
358
:arg iter_lines: iterable of lines to parse
359
:kwarg allow_dirty: If True, allow the patch to have trailing junk.
362
iter_lines = iter_lines_handle_nl(iter_lines)
364
(orig_name, mod_name) = get_patch_names(iter_lines)
365
except BinaryFiles, e:
366
return BinaryPatch(e.orig_name, e.mod_name)
368
patch = Patch(orig_name, mod_name)
369
for hunk in iter_hunks(iter_lines, allow_dirty):
370
patch.hunks.append(hunk)
374
def iter_file_patch(iter_lines, allow_dirty=False):
376
:arg iter_lines: iterable of lines to parse for patches
377
:kwarg allow_dirty: If True, allow comments and other non-patch text
378
before the first patch. Note that the algorithm here can only find
379
such text before any patches have been found. Comments after the
380
first patch are stripped away in iter_hunks() if it is also passed
381
allow_dirty=True. Default False.
383
### FIXME: Docstring is not quite true. We allow certain comments no
384
# matter what, If they startwith '===', '***', or '#' Someone should
385
# reexamine this logic and decide if we should include those in
386
# allow_dirty or restrict those to only being before the patch is found
387
# (as allow_dirty does).
388
regex = re.compile(binary_files_re)
327
392
for line in iter_lines:
328
393
if line.startswith('=== ') or line.startswith('*** '):
332
397
elif orig_range > 0:
333
398
if line.startswith('-') or line.startswith(' '):
335
elif line.startswith('--- '):
336
if len(saved_lines) > 0:
400
elif line.startswith('--- ') or regex.match(line):
401
if allow_dirty and beginning:
402
# Patches can have "junk" at the beginning
403
# Stripping junk from the end of patches is handled when we
406
elif len(saved_lines) > 0:
337
407
yield saved_lines
339
409
elif line.startswith('@@'):
368
def parse_patches(iter_lines):
369
iter_lines = iter_lines_handle_nl(iter_lines)
370
return [parse_patch(f.__iter__()) for f in iter_file_patch(iter_lines)]
438
def parse_patches(iter_lines, allow_dirty=False):
440
:arg iter_lines: iterable of lines to parse for patches
441
:kwarg allow_dirty: If True, allow text that's not part of the patch at
442
selected places. This includes comments before and after a patch
443
for instance. Default False.
445
return [parse_patch(f.__iter__(), allow_dirty) for f in
446
iter_file_patch(iter_lines, allow_dirty)]
373
449
def difference_index(atext, btext):
393
469
"""Iterate through a series of lines with a patch applied.
394
470
This handles a single file, and does exact, not fuzzy patching.
396
if orig_lines is not None:
397
orig_lines = orig_lines.__iter__()
472
patch_lines = iter_lines_handle_nl(iter(patch_lines))
473
get_patch_names(patch_lines)
474
return iter_patched_from_hunks(orig_lines, iter_hunks(patch_lines))
477
def iter_patched_from_hunks(orig_lines, hunks):
478
"""Iterate through a series of lines with a patch applied.
479
This handles a single file, and does exact, not fuzzy patching.
481
:param orig_lines: The unpatched lines.
482
:param hunks: An iterable of Hunk instances.
399
patch_lines = iter_lines_handle_nl(patch_lines.__iter__())
400
get_patch_names(patch_lines)
402
for hunk in iter_hunks(patch_lines):
486
if orig_lines is not None:
487
orig_lines = iter(orig_lines)
403
489
while line_no < hunk.orig_pos:
404
490
orig_line = orig_lines.next()