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
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):
26
binary_files_re = b'Binary files (.*) and (.*) differ\n'
29
class PatchSyntax(BzrError):
30
"""Base class for patch syntax errors."""
33
class BinaryFiles(BzrError):
35
_fmt = 'Binary files section encountered.'
25
37
def __init__(self, orig_name, mod_name):
26
38
self.orig_name = orig_name
27
39
self.mod_name = mod_name
28
Exception.__init__(self, 'Binary files section encountered.')
31
class PatchSyntax(Exception):
32
def __init__(self, msg):
33
Exception.__init__(self, msg)
36
42
class MalformedPatchHeader(PatchSyntax):
37
def __init__(self, desc, line):
40
msg = "Malformed patch header. %s\n%r" % (self.desc, self.line)
41
PatchSyntax.__init__(self, msg)
44
class MalformedHunkHeader(PatchSyntax):
45
def __init__(self, desc, line):
48
msg = "Malformed hunk header. %s\n%r" % (self.desc, self.line)
49
PatchSyntax.__init__(self, msg)
44
_fmt = "Malformed patch header. %(desc)s\n%(line)r"
46
def __init__(self, desc, line):
52
51
class MalformedLine(PatchSyntax):
53
_fmt = "Malformed line. %(desc)s\n%(line)r"
53
55
def __init__(self, desc, line):
56
msg = "Malformed line. %s\n%s" % (self.desc, self.line)
57
PatchSyntax.__init__(self, msg)
60
class PatchConflict(Exception):
60
class PatchConflict(BzrError):
62
_fmt = ('Text contents mismatch at line %(line_no)d. Original has '
63
'"%(orig_line)s", but patch says it should be "%(patch_line)s"')
61
65
def __init__(self, line_no, orig_line, patch_line):
62
orig = orig_line.rstrip('\n')
63
patch = str(patch_line).rstrip('\n')
64
msg = 'Text contents mismatch at line %d. Original has "%s",'\
65
' but patch says it should be "%s"' % (line_no, orig, patch)
66
Exception.__init__(self, msg)
66
self.line_no = line_no
67
self.orig_line = orig_line.rstrip('\n')
68
self.patch_line = patch_line.rstrip('\n')
71
class MalformedHunkHeader(PatchSyntax):
73
_fmt = "Malformed hunk header. %(desc)s\n%(line)r"
75
def __init__(self, desc, line):
69
80
def get_patch_names(iter_lines):
81
line = next(iter_lines)
71
line = iter_lines.next()
72
83
match = re.match(binary_files_re, line)
73
84
if match is not None:
74
85
raise BinaryFiles(match.group(1), match.group(2))
75
if not line.startswith("--- "):
86
if not line.startswith(b"--- "):
76
87
raise MalformedPatchHeader("No orig name", line)
78
orig_name = line[4:].rstrip("\n")
89
orig_name = line[4:].rstrip(b"\n")
79
90
except StopIteration:
80
91
raise MalformedPatchHeader("No orig line", "")
82
line = iter_lines.next()
83
if not line.startswith("+++ "):
93
line = next(iter_lines)
94
if not line.startswith(b"+++ "):
84
95
raise PatchSyntax("No mod name")
86
mod_name = line[4:].rstrip("\n")
97
mod_name = line[4:].rstrip(b"\n")
87
98
except StopIteration:
88
99
raise MalformedPatchHeader("No mod line", "")
89
100
return (orig_name, mod_name)
111
122
def hunk_from_header(line):
113
matches = re.match(r'\@\@ ([^@]*) \@\@( (.*))?\n', line)
124
matches = re.match(br'\@\@ ([^@]*) \@\@( (.*))?\n', line)
114
125
if matches is None:
115
126
raise MalformedHunkHeader("Does not match format.", line)
117
(orig, mod) = matches.group(1).split(" ")
118
except (ValueError, IndexError), e:
128
(orig, mod) = matches.group(1).split(b" ")
129
except (ValueError, IndexError) as e:
119
130
raise MalformedHunkHeader(str(e), line)
120
if not orig.startswith('-') or not mod.startswith('+'):
131
if not orig.startswith(b'-') or not mod.startswith(b'+'):
121
132
raise MalformedHunkHeader("Positions don't start with + or -.", line)
123
134
(orig_pos, orig_range) = parse_range(orig[1:])
124
135
(mod_pos, mod_range) = parse_range(mod[1:])
125
except (ValueError, IndexError), e:
136
except (ValueError, IndexError) as e:
126
137
raise MalformedHunkHeader(str(e), line)
127
138
if mod_range < 0 or orig_range < 0:
128
139
raise MalformedHunkHeader("Hunk range is negative", line)
130
141
return Hunk(orig_pos, orig_range, mod_pos, mod_range, tail)
144
class HunkLine(object):
134
146
def __init__(self, contents):
135
147
self.contents = contents
137
149
def get_str(self, leadchar):
138
if self.contents == "\n" and leadchar == " " and False:
140
if not self.contents.endswith('\n'):
141
terminator = '\n' + NO_NL
150
if self.contents == b"\n" and leadchar == b" " and False:
152
if not self.contents.endswith(b'\n'):
153
terminator = b'\n' + NO_NL
144
156
return leadchar + self.contents + terminator
159
raise NotImplementedError
147
162
class ContextLine(HunkLine):
148
164
def __init__(self, contents):
149
165
HunkLine.__init__(self, contents)
152
return self.get_str(" ")
168
return self.get_str(b" ")
155
171
class InsertLine(HunkLine):
156
172
def __init__(self, contents):
157
173
HunkLine.__init__(self, contents)
160
return self.get_str("+")
176
return self.get_str(b"+")
163
179
class RemoveLine(HunkLine):
164
180
def __init__(self, contents):
165
181
HunkLine.__init__(self, contents)
168
return self.get_str("-")
170
NO_NL = '\\ No newline at end of file\n'
171
__pychecker__="no-returnvalues"
184
return self.get_str(b"-")
187
NO_NL = b'\\ No newline at end of file\n'
188
__pychecker__ = "no-returnvalues"
173
191
def parse_line(line):
174
if line.startswith("\n"):
192
if line.startswith(b"\n"):
175
193
return ContextLine(line)
176
elif line.startswith(" "):
194
elif line.startswith(b" "):
177
195
return ContextLine(line[1:])
178
elif line.startswith("+"):
196
elif line.startswith(b"+"):
179
197
return InsertLine(line[1:])
180
elif line.startswith("-"):
198
elif line.startswith(b"-"):
181
199
return RemoveLine(line[1:])
183
201
raise MalformedLine("Unknown line type", line)
188
209
def __init__(self, orig_pos, orig_range, mod_pos, mod_range, tail=None):
189
210
self.orig_pos = orig_pos
190
211
self.orig_range = orig_range
214
235
:return: a string in the format 1,4 except when range == pos == 1
219
return "%i,%i" % (pos, range)
240
return b"%i,%i" % (pos, range)
222
243
lines = [self.get_header()]
223
244
for line in self.lines:
224
lines.append(str(line))
225
return "".join(lines)
245
lines.append(line.as_bytes())
246
return b"".join(lines)
227
250
def shift_to_mod(self, pos):
228
if pos < self.orig_pos-1:
251
if pos < self.orig_pos - 1:
230
elif pos > self.orig_pos+self.orig_range:
253
elif pos > self.orig_pos + self.orig_range:
231
254
return self.mod_range - self.orig_range
233
256
return self.shift_to_mod_lines(pos)
235
258
def shift_to_mod_lines(self, pos):
236
position = self.orig_pos-1
259
position = self.orig_pos - 1
238
261
for line in self.lines:
239
262
if isinstance(line, InsertLine):
380
404
first patch are stripped away in iter_hunks() if it is also passed
381
405
allow_dirty=True. Default False.
383
### FIXME: Docstring is not quite true. We allow certain comments no
407
# FIXME: Docstring is not quite true. We allow certain comments no
384
408
# matter what, If they startwith '===', '***', or '#' Someone should
385
409
# reexamine this logic and decide if we should include those in
386
410
# allow_dirty or restrict those to only being before the patch is found
387
411
# (as allow_dirty does).
388
412
regex = re.compile(binary_files_re)
392
418
for line in iter_lines:
393
if line.startswith('=== ') or line.startswith('*** '):
395
if line.startswith('#'):
419
if line.startswith(b'=== '):
420
if allow_dirty and beginning:
421
# Patches can have "junk" at the beginning
422
# Stripping junk from the end of patches is handled when we
425
elif len(saved_lines) > 0:
426
if keep_dirty and len(dirty_head) > 0:
427
yield {'saved_lines': saved_lines,
428
'dirty_head': dirty_head}
433
dirty_head.append(line)
435
if line.startswith(b'*** '):
437
if line.startswith(b'#'):
397
439
elif orig_range > 0:
398
if line.startswith('-') or line.startswith(' '):
440
if line.startswith(b'-') or line.startswith(b' '):
400
elif line.startswith('--- ') or regex.match(line):
442
elif line.startswith(b'--- ') or regex.match(line):
401
443
if allow_dirty and beginning:
402
444
# Patches can have "junk" at the beginning
403
445
# Stripping junk from the end of patches is handled when we
404
446
# parse the patch
405
447
beginning = False
406
448
elif len(saved_lines) > 0:
449
if keep_dirty and len(dirty_head) > 0:
450
yield {'saved_lines': saved_lines,
451
'dirty_head': dirty_head}
409
elif line.startswith('@@'):
456
elif line.startswith(b'@@'):
410
457
hunk = hunk_from_header(line)
411
458
orig_range = hunk.orig_range
412
459
saved_lines.append(line)
413
460
if len(saved_lines) > 0:
461
if keep_dirty and len(dirty_head) > 0:
462
yield {'saved_lines': saved_lines,
463
'dirty_head': dirty_head}
417
468
def iter_lines_handle_nl(iter_lines):
438
def parse_patches(iter_lines, allow_dirty=False):
489
def parse_patches(iter_lines, allow_dirty=False, keep_dirty=False):
440
491
:arg iter_lines: iterable of lines to parse for patches
441
492
:kwarg allow_dirty: If True, allow text that's not part of the patch at
442
493
selected places. This includes comments before and after a patch
443
494
for instance. Default False.
495
:kwarg keep_dirty: If True, returns a dict of patches with dirty headers.
445
return [parse_patch(f.__iter__(), allow_dirty) for f in
446
iter_file_patch(iter_lines, allow_dirty)]
498
for patch_lines in iter_file_patch(iter_lines, allow_dirty, keep_dirty):
499
if 'dirty_head' in patch_lines:
500
yield ({'patch': parse_patch(patch_lines['saved_lines'], allow_dirty),
501
'dirty_head': patch_lines['dirty_head']})
503
yield parse_patch(patch_lines, allow_dirty)
449
506
def difference_index(atext, btext):
487
544
orig_lines = iter(orig_lines)
488
545
for hunk in hunks:
489
546
while line_no < hunk.orig_pos:
490
orig_line = orig_lines.next()
547
orig_line = next(orig_lines)
493
550
for hunk_line in hunk.lines:
494
seen_patch.append(str(hunk_line))
551
seen_patch.append(hunk_line.contents)
495
552
if isinstance(hunk_line, InsertLine):
496
553
yield hunk_line.contents
497
554
elif isinstance(hunk_line, (ContextLine, RemoveLine)):
498
orig_line = orig_lines.next()
555
orig_line = next(orig_lines)
499
556
if orig_line != hunk_line.contents:
500
raise PatchConflict(line_no, orig_line, "".join(seen_patch))
557
raise PatchConflict(line_no, orig_line,
558
b''.join(seen_patch))
501
559
if isinstance(hunk_line, ContextLine):
507
565
if orig_lines is not None:
508
566
for line in orig_lines:
570
def apply_patches(tt, patches, prefix=1):
571
"""Apply patches to a TreeTransform.
573
:param tt: TreeTransform instance
574
:param patches: List of patches
575
:param prefix: Number leading path segments to strip
578
return '/'.join(p.split('/')[1:])
580
from breezy.bzr.generate_ids import gen_file_id
581
# TODO(jelmer): Extract and set mode
582
for patch in patches:
583
if patch.oldname == b'/dev/null':
587
oldname = strip_prefix(patch.oldname.decode())
588
trans_id = tt.trans_id_tree_path(oldname)
589
orig_contents = tt._tree.get_file_text(oldname)
590
tt.delete_contents(trans_id)
592
if patch.newname != b'/dev/null':
593
newname = strip_prefix(patch.newname.decode())
594
new_contents = iter_patched_from_hunks(
595
orig_contents.splitlines(True), patch.hunks)
597
parts = os.path.split(newname)
599
for part in parts[1:-1]:
600
trans_id = tt.new_directory(part, trans_id)
602
parts[-1], trans_id, new_contents,
603
file_id=gen_file_id(newname))
605
tt.create_file(new_contents, trans_id)
608
class AppliedPatches(object):
609
"""Context that provides access to a tree with patches applied.
612
def __init__(self, tree, patches, prefix=1):
614
self.patches = patches
618
self._tt = self.tree.preview_transform()
619
apply_patches(self._tt, self.patches, prefix=self.prefix)
620
return self._tt.get_preview_tree()
622
def __exit__(self, exc_type, exc_value, exc_tb):