Python class for accessing Rietveld reviews.

reviewbot/patching.py

Index: reviewbot/patching.py
===================================================================
--- reviewbot/patching.py	(revision 0)
+++ reviewbot/patching.py	(revision 0)
@@ -0,0 +1,78 @@
+# Copyright (c) 2013 The Chromium Authors. All rights reserved.
+# Use of this source code is governed by a BSD-style license that can be
+# found in the LICENSE file.
+
+# Copyright 2008 Google Inc. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Utility for parsing patches, shamelessly stolen from rietveld."""
+
+import logging
+import re
+
+
+_CHUNK_RE = re.compile(r"""
+  @@
+  \s+
+  -
+  (?: (\d+) (?: , (\d+) )?)
+  \s+
+  \+
+  (?: (\d+) (?: , (\d+) )?)
+  \s+
+  @@
+""", re.VERBOSE)

[iannucci, 2013/08/08 10:35:20]

Scary :( I really wish we had a real, robust, tested diff parsing library.

May also want to provide some examples (or better, tests) and/or comments on
this guy.

[Mattias Nissler (ping if slow), 2013/08/08 13:04:13]

I'm with you on this, but the python standard library doesn't seem to have patch
parsing, and the few hits on the web that I found weren't much better than this.
The code here is stolen from rietveld and simplified to the bare essentials.
I've reworked the code a bit and added some tests.

+
+
+_NO_NEWLINE_MESSAGE = "\\ No newline at end of file"
+
+
+def ParsePatchToLines(lines):
+  """Parses a patch from a list of lines.
+
+  Args:
+    lines: The lines to parse.
+
+  Returns:
+    None on error, otherwise a list of 3-tuples:
+    (old_line_no, new_line_no, line)
+
+    A line number can be 0 if it doesn't exist in the old/new file.

[iannucci, 2013/08/08 10:35:20]

why not None? Unless lines start at 1?

[Mattias Nissler (ping if slow), 2013/08/08 13:04:13]

None is indeed better. Changed.

+  """
+  result = []
+  in_prelude = True
+  for line in lines:
+    if in_prelude:
+      # Skip leading lines until after we've seen one starting with '+++'
+      if line.startswith("+++"):
+        in_prelude = False

[iannucci, 2013/08/08 10:35:20]

how do we exit the prelude for a multi-chunk patch?

[Mattias Nissler (ping if slow), 2013/08/08 13:04:13]

Hunks start with @@, so we reset.

Multi-file patches are not supported by this code. They're just concatenated, so
at some point, we'll see another prelude. I've now added code that raises an
error when this happens.

+    elif line.startswith("@"):
+      match = _CHUNK_RE.match(line)
+      if not match:
+        logging.warn("ParsePatchToLines match failed on %s", line)
+        return None
+      old_ln = int(match.groups()[0])
+      new_ln = int(match.groups()[2])
+    else:
+      if line[0] == "-":
+        result.append((old_ln, 0, line[1:]))
+        old_ln += 1
+      elif line[0] == "+":
+        result.append((0, new_ln, line[1:]))
+        new_ln += 1
+      elif line[0] == " ":
+        result.append((old_ln, new_ln, line[1:]))
+        old_ln += 1
+        new_ln += 1
+  return result
Property changes on: reviewbot/patching.py
___________________________________________________________________
Added: svn:eol-style
   + LF