def _parse(self, diff, encoding):
current_file = None
diff = enumerate(diff, 1)
for unused_diff_line_no, line in diff:
if encoding is not None:
line = line.decode(encoding)
# check for source file header
is_source_filename = RE_SOURCE_FILENAME.match(line)
if is_source_filename:
source_file = is_source_filename.group('filename')
source_timestamp = is_source_filename.group('timestamp')
# reset current file
current_file = None
continue
# check for target file header
is_target_filename = RE_TARGET_FILENAME.match(line)
if is_target_filename:
if current_file is not None:
raise UnidiffParseError('Target without source: %s' % line)
target_file = is_target_filename.group('filename')
target_timestamp = is_target_filename.group('timestamp')
# add current file to PatchSet
current_file = PatchedFile(source_file, target_file,
source_timestamp, target_timestamp)
self.append(current_file)
continue
# check for hunk header
is_hunk_header = RE_HUNK_HEADER.match(line)
if is_hunk_header:
if current_file is None:
raise UnidiffParseError('Unexpected hunk found: %s' % line)
current_file._parse_hunk(line, diff, encoding)
def _parse(self, diff, encoding):
current_file = None
patch_info = None
diff = enumerate(diff, 1)
for unused_diff_line_no, line in diff:
if encoding is not None:
line = line.decode(encoding)
# check for source file header
is_source_filename = RE_SOURCE_FILENAME.match(line)
if is_source_filename:
source_file = is_source_filename.group('filename')
source_timestamp = is_source_filename.group('timestamp')
# reset current file
current_file = None
continue
# check for target file header
is_target_filename = RE_TARGET_FILENAME.match(line)
if is_target_filename:
if current_file is not None:
raise UnidiffParseError('Target without source: %s' % line)
target_file = is_target_filename.group('filename')
target_timestamp = is_target_filename.group('timestamp')
# add current file to PatchSet
current_file = PatchedFile(patch_info, source_file,
target_file, source_timestamp,
target_timestamp)
self.append(current_file)
patch_info = None
continue
# check for hunk header
is_hunk_header = RE_HUNK_HEADER.match(line)
if is_hunk_header:
if current_file is None:
raise UnidiffParseError('Unexpected hunk found: %s' % line)
current_file._parse_hunk(line, diff, encoding)
continue
# check for no newline marker
is_no_newline = RE_NO_NEWLINE_MARKER.match(line)
if is_no_newline:
if current_file is None:
raise UnidiffParseError('Unexpected marker: %s' % line)
current_file._add_no_newline_marker_to_last_hunk()
continue
# sometimes hunks can be followed by empty lines
if line == '\n' and current_file is not None:
current_file._append_trailing_empty_line()
continue
# if nothing has matched above then this line is a patch info
if patch_info is None:
current_file = None
patch_info = PatchInfo()
patch_info.append(line)
def _parse(self, diff, encoding):
current_file = None
diff = enumerate(diff, 1)
for unused_diff_line_no, line in diff:
if encoding is not None:#don't care - stuff about non utf-8 files
line = line.decode(encoding)
# check for source file header
is_source_filename = RE_SOURCE_FILENAME.match(line)
if is_source_filename:
source_file = is_source_filename.group('filename')
source_timestamp = is_source_filename.group('timestamp')
if source_file is None:#must be the other side of regex triggered
source_file = is_source_filename.group('renamefile')
# reset current file
current_file = None
continue
# check for target file header
is_target_filename = RE_TARGET_FILENAME.match(line)
if is_target_filename:
is_renamed = False
if current_file is not None:
raise UnidiffParseError('Target without source: %s' % line)
target_file = is_target_filename.group('filename')
if target_file is None:
target_file = is_target_filename.group('renamefile')
is_renamed = True
target_timestamp = is_target_filename.group('timestamp')
# add current file to PatchSet
current_file = PatchedFile(source_file, target_file,
source_timestamp, target_timestamp, is_renamed)
self.append(current_file)
continue
# check for hunk header
is_hunk_header = RE_HUNK_HEADER.match(line)
if is_hunk_header:
if current_file is None:
raise UnidiffParseError('Unexpected hunk found: %s' % line)
current_file._parse_hunk(line, diff, encoding)
# check for no newline marker
is_no_newline = RE_NO_NEWLINE_MARKER.match(line)
if is_no_newline:
if current_file is None:
raise UnidiffParseError('Unexpected marker: %s' % line)
current_file._add_no_newline_marker_to_last_hunk()
def _add_no_newline_marker_to_last_hunk(self):
if not self:
raise UnidiffParseError(
'Unexpected marker:' + LINE_VALUE_NO_NEWLINE)
last_hunk = self[-1]
last_hunk.append(
Line(LINE_VALUE_NO_NEWLINE + '\n', line_type=LINE_TYPE_NO_NEWLINE))
def _parse_hunk(self, header, diff, encoding):
"""Parse hunk details."""
header_info = RE_HUNK_HEADER.match(header)
hunk_info = header_info.groups()
hunk = Hunk(*hunk_info)
source_line_no = hunk.source_start
target_line_no = hunk.target_start
expected_source_end = source_line_no + hunk.source_length
expected_target_end = target_line_no + hunk.target_length
for diff_line_no, line in diff:
if encoding is not None:
line = line.decode(encoding)
valid_line = RE_HUNK_BODY_LINE.match(line)
if not valid_line:
raise UnidiffParseError('Hunk diff line expected: %s' % line)
line_type = valid_line.group('line_type')
if line_type == LINE_TYPE_EMPTY:
line_type = LINE_TYPE_CONTEXT
value = valid_line.group('value')
original_line = Line(value, line_type=line_type)
if line_type == LINE_TYPE_ADDED:
original_line.target_line_no = target_line_no
target_line_no += 1
self.added_lines_content = "{0}\n{1}".format(
self.added_lines_content, value)
elif line_type == LINE_TYPE_REMOVED:
original_line.source_line_no = source_line_no
source_line_no += 1
self.removed_lines = "{0}\n{1}".format(
self.removed_lines_content, value)
elif line_type == LINE_TYPE_CONTEXT:
original_line.target_line_no = target_line_no
target_line_no += 1
original_line.source_line_no = source_line_no
source_line_no += 1
else:
original_line = None
if original_line:
original_line.diff_line_no = diff_line_no
hunk.append(original_line)
# if hunk source/target lengths are ok, hunk is complete
if (source_line_no == expected_source_end
and target_line_no == expected_target_end):
break
self.append(hunk)
def _parse_hunk(self, header, diff, encoding):
"""Parse hunk details."""
header_info = RE_HUNK_HEADER.match(header)
hunk_info = header_info.groups()
hunk = Hunk(*hunk_info)
source_line_no = hunk.source_start
target_line_no = hunk.target_start
for line in diff:
if encoding is not None:
line = line.decode(encoding)
valid_line = RE_HUNK_BODY_LINE.match(line)
if not valid_line:
raise UnidiffParseError('Hunk diff line expected: %s' % line)
line_type = valid_line.group('line_type')
value = valid_line.group('value')
original_line = Line(value, line_type=line_type)
if line_type == LINE_TYPE_ADDED:
original_line.target_line_no = target_line_no
target_line_no += 1
elif line_type == LINE_TYPE_REMOVED:
original_line.source_line_no = source_line_no
source_line_no += 1
elif line_type == LINE_TYPE_CONTEXT:
original_line.target_line_no = target_line_no
target_line_no += 1
original_line.source_line_no = source_line_no
source_line_no += 1
else:
original_line = None
if original_line:
hunk.append(original_line)
# check hunk len(old_lines) and len(new_lines) are ok
if hunk.is_valid():
break
self.append(hunk)
def _parse(self, diff, encoding, metadata_only):
current_file = None
patch_info = None
diff = enumerate(diff, 1)
for unused_diff_line_no, line in diff:
if encoding is not None:
line = line.decode(encoding)
# check for a git rename, source file
is_rename_source_filename = RE_RENAME_SOURCE_FILENAME.match(line)
if is_rename_source_filename:
# prefix with 'a/' to match expected git source format
source_file = ('a/' +
is_rename_source_filename.group('filename'))
# keep line as patch_info
patch_info.append(line)
# reset current file
current_file = None
continue
# check for a git rename, target file
is_rename_target_filename = RE_RENAME_TARGET_FILENAME.match(line)
if is_rename_target_filename:
if current_file is not None:
raise UnidiffParseError('Target without source: %s' % line)
# prefix with 'b/' to match expected git source format
target_file = ('b/' +
is_rename_target_filename.group('filename'))
# keep line as patch_info
patch_info.append(line)
# add current file to PatchSet
current_file = PatchedFile(patch_info,
source_file,
target_file,
None,
None,
is_rename=True)
self.append(current_file)
continue
# check for source file header
is_source_filename = RE_SOURCE_FILENAME.match(line)
if is_source_filename:
source_file = is_source_filename.group('filename')
source_timestamp = is_source_filename.group('timestamp')
# reset current file, unless we are processing a rename
# (in that case, source files should match)
if current_file is not None and not (
current_file.is_rename
and current_file.source_file == source_file):
current_file = None
continue
# check for target file header
is_target_filename = RE_TARGET_FILENAME.match(line)
if is_target_filename:
if current_file is not None and not current_file.is_rename:
raise UnidiffParseError('Target without source: %s' % line)
target_file = is_target_filename.group('filename')
target_timestamp = is_target_filename.group('timestamp')
if current_file is None:
# add current file to PatchSet
current_file = PatchedFile(patch_info, source_file,
target_file, source_timestamp,
target_timestamp)
self.append(current_file)
patch_info = None
continue
# check for hunk header
is_hunk_header = RE_HUNK_HEADER.match(line)
if is_hunk_header:
if current_file is None:
raise UnidiffParseError('Unexpected hunk found: %s' % line)
current_file._parse_hunk(line, diff, encoding, metadata_only)
continue
# check for no newline marker
is_no_newline = RE_NO_NEWLINE_MARKER.match(line)
if is_no_newline:
if current_file is None:
raise UnidiffParseError('Unexpected marker: %s' % line)
current_file._add_no_newline_marker_to_last_hunk()
continue
# sometimes hunks can be followed by empty lines
if line == '\n' and current_file is not None:
current_file._append_trailing_empty_line()
continue
is_binary_diff = RE_BINARY_DIFF.match(line)
if is_binary_diff:
source_file = is_binary_diff.group('source_filename')
target_file = is_binary_diff.group('target_filename')
patch_info.append(line)
current_file = PatchedFile(patch_info,
source_file,
target_file,
is_binary_file=True)
self.append(current_file)
patch_info = None
current_file = None
continue
# if nothing has matched above then this line is a patch info
if patch_info is None:
current_file = None
patch_info = PatchInfo()
patch_info.append(line)
def _append_trailing_empty_line(self):
if not self:
raise UnidiffParseError('Unexpected trailing newline character')
last_hunk = self[-1]
last_hunk.append(Line('\n', line_type=LINE_TYPE_EMPTY))
def _parse_hunk(self, header, diff, encoding, metadata_only):
"""Parse hunk details."""
header_info = RE_HUNK_HEADER.match(header)
hunk_info = header_info.groups()
hunk = Hunk(*hunk_info)
source_line_no = hunk.source_start
target_line_no = hunk.target_start
expected_source_end = source_line_no + hunk.source_length
expected_target_end = target_line_no + hunk.target_length
added = 0
removed = 0
for diff_line_no, line in diff:
if encoding is not None:
line = line.decode(encoding)
if metadata_only:
# quick line type detection, no regex required
line_type = line[0] if line else LINE_TYPE_CONTEXT
if line_type not in (LINE_TYPE_ADDED, LINE_TYPE_REMOVED,
LINE_TYPE_CONTEXT, LINE_TYPE_NO_NEWLINE):
raise UnidiffParseError('Hunk diff line expected: %s' %
line)
if line_type == LINE_TYPE_ADDED:
target_line_no += 1
added += 1
elif line_type == LINE_TYPE_REMOVED:
source_line_no += 1
removed += 1
elif line_type == LINE_TYPE_CONTEXT:
target_line_no += 1
source_line_no += 1
# no file content tracking
original_line = None
else:
# parse diff line content
valid_line = RE_HUNK_BODY_LINE.match(line)
if not valid_line:
valid_line = RE_HUNK_EMPTY_BODY_LINE.match(line)
if not valid_line:
raise UnidiffParseError('Hunk diff line expected: %s' %
line)
line_type = valid_line.group('line_type')
if line_type == LINE_TYPE_EMPTY:
line_type = LINE_TYPE_CONTEXT
value = valid_line.group('value')
original_line = Line(value, line_type=line_type)
if line_type == LINE_TYPE_ADDED:
original_line.target_line_no = target_line_no
target_line_no += 1
elif line_type == LINE_TYPE_REMOVED:
original_line.source_line_no = source_line_no
source_line_no += 1
elif line_type == LINE_TYPE_CONTEXT:
original_line.target_line_no = target_line_no
original_line.source_line_no = source_line_no
target_line_no += 1
source_line_no += 1
elif line_type == LINE_TYPE_NO_NEWLINE:
pass
else:
original_line = None
# stop parsing if we got past expected number of lines
if (source_line_no > expected_source_end
or target_line_no > expected_target_end):
raise UnidiffParseError('Hunk is longer than expected')
if original_line:
original_line.diff_line_no = diff_line_no
hunk.append(original_line)
# if hunk source/target lengths are ok, hunk is complete
if (source_line_no == expected_source_end
and target_line_no == expected_target_end):
break
# report an error if we haven't got expected number of lines
if (source_line_no < expected_source_end
or target_line_no < expected_target_end):
raise UnidiffParseError('Hunk is shorter than expected')
if metadata_only:
# HACK: set fixed calculated values when metadata_only is enabled
hunk._added = added
hunk._removed = removed
self.append(hunk)
def _parse_hunk(self, header, diff, encoding):
"""Parse hunk details."""
header_info = RE_HUNK_HEADER.match(header)
hunk_info = header_info.groups()
hunk = Hunk(*hunk_info)
source_line_no = hunk.source_start
target_line_no = hunk.target_start
expected_source_end = source_line_no + hunk.source_length
expected_target_end = target_line_no + hunk.target_length
for diff_line_no, line in diff:
if encoding is not None:
line = line.decode(encoding)
valid_line = RE_HUNK_EMPTY_BODY_LINE.match(line)
if not valid_line:
valid_line = RE_HUNK_BODY_LINE.match(line)
if not valid_line:
raise UnidiffParseError('Hunk diff line expected: %s' % line)
line_type = valid_line.group('line_type')
if line_type == LINE_TYPE_EMPTY:
line_type = LINE_TYPE_CONTEXT
value = valid_line.group('value')
original_line = Line(value, line_type=line_type)
if line_type == LINE_TYPE_ADDED:
original_line.target_line_no = target_line_no
target_line_no += 1
elif line_type == LINE_TYPE_REMOVED:
original_line.source_line_no = source_line_no
source_line_no += 1
elif line_type == LINE_TYPE_CONTEXT:
original_line.target_line_no = target_line_no
target_line_no += 1
original_line.source_line_no = source_line_no
source_line_no += 1
elif line_type == LINE_TYPE_NO_NEWLINE:
pass
else:
original_line = None
# stop parsing if we got past expected number of lines
if (source_line_no > expected_source_end
or target_line_no > expected_target_end):
raise UnidiffParseError('Hunk is longer than expected')
if original_line:
original_line.diff_line_no = diff_line_no
hunk.append(original_line)
# if hunk source/target lengths are ok, hunk is complete
if (source_line_no == expected_source_end
and target_line_no == expected_target_end):
break
# report an error if we haven't got expected number of lines
if (source_line_no < expected_source_end
or target_line_no < expected_target_end):
raise UnidiffParseError('Hunk is shorter than expected')
self.append(hunk)