tools/diff.py

   1 #!/usr/bin/env python
   2 ##  Copyright (c) 2012 The WebM project authors. All Rights Reserved.
   3 ##
   4 ##  Use of this source code is governed by a BSD-style license
   5 ##  that can be found in the LICENSE file in the root of the source
   6 ##  tree. An additional intellectual property rights grant can be found
   7 ##  in the file PATENTS.  All contributing project authors may
   8 ##  be found in the AUTHORS file in the root of the source tree.
   9 ##
  10 """Classes for representing diff pieces."""
  11
  12 __author__ = "jkoleszar@google.com"
  13
  14 import re
  15
  16
  17 class DiffLines(object):
  18     """A container for one half of a diff."""
  19
  20     def __init__(self, filename, offset, length):
  21         self.filename = filename
  22         self.offset = offset
  23         self.length = length
  24         self.lines = []
  25         self.delta_line_nums = []
  26
  27     def Append(self, line):
  28         l = len(self.lines)
  29         if line[0] != " ":
  30             self.delta_line_nums.append(self.offset + l)
  31         self.lines.append(line[1:])
  32         assert l+1 <= self.length
  33
  34     def Complete(self):
  35         return len(self.lines) == self.length
  36
  37     def __contains__(self, item):
  38         return item >= self.offset and item <= self.offset + self.length - 1
  39
  40
  41 class DiffHunk(object):
  42     """A container for one diff hunk, consisting of two DiffLines."""
  43
  44     def __init__(self, header, file_a, file_b, start_a, len_a, start_b, len_b):
  45         self.header = header
  46         self.left = DiffLines(file_a, start_a, len_a)
  47         self.right = DiffLines(file_b, start_b, len_b)
  48         self.lines = []
  49
  50     def Append(self, line):
  51         """Adds a line to the DiffHunk and its DiffLines children."""
  52         if line[0] == "-":
  53             self.left.Append(line)
  54         elif line[0] == "+":
  55             self.right.Append(line)
  56         elif line[0] == " ":
  57             self.left.Append(line)
  58             self.right.Append(line)
  59         elif line[0] == "\\":
  60             # Ignore newline messages from git diff.
  61             pass
  62         else:
  63             assert False, ("Unrecognized character at start of diff line "
  64                            "%r" % line[0])
  65         self.lines.append(line)
  66
  67     def Complete(self):
  68         return self.left.Complete() and self.right.Complete()
  69
  70     def __repr__(self):
  71         return "DiffHunk(%s, %s, len %d)" % (
  72             self.left.filename, self.right.filename,
  73             max(self.left.length, self.right.length))
  74
  75
  76 def ParseDiffHunks(stream):
  77     """Walk a file-like object, yielding DiffHunks as they're parsed."""
  78
  79     file_regex = re.compile(r"(\+\+\+|---) (\S+)")
  80     range_regex = re.compile(r"@@ -(\d+)(,(\d+))? \+(\d+)(,(\d+))?")
  81     hunk = None
  82     while True:
  83         line = stream.readline()
  84         if not line:
  85             break
  86
  87         if hunk is None:
  88             # Parse file names
  89             diff_file = file_regex.match(line)
  90             if diff_file:
  91               if line.startswith("---"):
  92                   a_line = line
  93                   a = diff_file.group(2)
  94                   continue
  95               if line.startswith("+++"):
  96                   b_line = line
  97                   b = diff_file.group(2)
  98                   continue
  99
 100             # Parse offset/lengths
 101             diffrange = range_regex.match(line)
 102             if diffrange:
 103                 if diffrange.group(2):
 104                     start_a = int(diffrange.group(1))
 105                     len_a = int(diffrange.group(3))
 106                 else:
 107                     start_a = 1
 108                     len_a = int(diffrange.group(1))
 109
 110                 if diffrange.group(5):
 111                     start_b = int(diffrange.group(4))
 112                     len_b = int(diffrange.group(6))
 113                 else:
 114                     start_b = 1
 115                     len_b = int(diffrange.group(4))
 116
 117                 header = [a_line, b_line, line]
 118                 hunk = DiffHunk(header, a, b, start_a, len_a, start_b, len_b)
 119         else:
 120             # Add the current line to the hunk
 121             hunk.Append(line)
 122
 123             # See if the whole hunk has been parsed. If so, yield it and prepare
 124             # for the next hunk.
 125             if hunk.Complete():
 126                 yield hunk
 127                 hunk = None
 128
 129     # Partial hunks are a parse error
 130     assert hunk is None