[webkitpy] Refactor diff_parser.py

Tools/Scripts/webkitpy/common/checkout/diff_parser.py

 # Copyright (C) 2009 Google Inc. All rights reserved.
 #
 # Redistribution and use in source and binary forms, with or without
 # modification, are permitted provided that the following conditions are
 # met:
 #
 #    * Redistributions of source code must retain the above copyright
 # notice, this list of conditions and the following disclaimer.
 #    * Redistributions in binary form must reproduce the above
 # copyright notice, this list of conditions and the following disclaimer
@@ skipping 15 matching lines @@
 # (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 # OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
 """WebKit's Python module for interacting with patches."""
 
 import logging
 import re
 
 _log = logging.getLogger(__name__)
 
+conversion_patterns = (
+    (re.compile("^diff --git \w/(.+) \w/(?P<FilePath>.+)"), lambda matched: "Index: " + matched.group('FilePath') + "\n"),
+    (re.compile("^new file.*"), lambda matched: "\n"),
+    (re.compile("^index (([0-9a-f]{7}\.\.[0-9a-f]{7})|([0-9a-f]{40}\.\.[0-9a-f]{40})) [0-9]{6}"), lambda matched: ("=" * 67) + "\n"),
+    (re.compile("^--- \w/(?P<FilePath>.+)"), lambda matched: "--- " + matched.group('FilePath') + "\n"),
+    (re.compile("^\+\+\+ \w/(?P<FilePath>.+)"), lambda matched: "+++ " + matched.group('FilePath') + "\n"),
+)
 
-# FIXME: This is broken. We should compile our regexps up-front
-# instead of using a custom cache.
-_regexp_compile_cache = {}
+index_pattern = re.compile(r"^Index: (?P<FilePath>.+)")
+lines_changed_pattern = re.compile(r"^@@ -(?P<OldStartLine>\d+)(,\d+)? \+(?P<NewStartLine>\d+)(,\d+)? @@")
+diff_git_pattern = re.compile(r"^diff --git \w/")
 
 
-# FIXME: This function should be removed.
-def match(pattern, string):
-    """Matches the string with the pattern, caching the compiled regexp."""
-    if not pattern in _regexp_compile_cache:
-        _regexp_compile_cache[pattern] = re.compile(pattern)
-    return _regexp_compile_cache[pattern].match(string)
-
-
-# FIXME: This belongs on DiffParser (e.g. as to_svn_diff()).
 def git_diff_to_svn_diff(line):
     """Converts a git formatted diff line to a svn formatted line.
 
     Args:
       line: A string representing a line of the diff.
     """
-    # FIXME: This list should be a class member on DiffParser.
-    # These regexp patterns should be compiled once instead of every time.
-    conversion_patterns = (("^diff --git \w/(.+) \w/(?P<FilePath>.+)", lambda matched: "Index: " + matched.group('FilePath') + "\n"),
-                           ("^new file.*", lambda matched: "\n"),
-                           ("^index (([0-9a-f]{7}\.\.[0-9a-f]{7})|([0-9a-f]{40}\.\.[0-9a-f]{40})) [0-9]{6}", lambda matched: "===================================================================\n"),
-                           ("^--- \w/(?P<FilePath>.+)", lambda matched: "--- " + matched.group('FilePath') + "\n"),
-                           ("^\+\+\+ \w/(?P<FilePath>.+)", lambda matched: "+++ " + matched.group('FilePath') + "\n"))
-
     for pattern, conversion in conversion_patterns:
-        matched = match(pattern, line)
+        matched = pattern.match(line)
         if matched:
             return conversion(matched)
     return line
 
 
 # This function exists so we can unittest get_diff_converter function
 def svn_diff_to_svn_diff(line):
     return line
 
 
-# FIXME: This method belongs on DiffParser
 def get_diff_converter(lines):
     """Gets a converter function of diff lines.
 
     Args:
       lines: The lines of a diff file.
              If this line is git formatted, we'll return a
              converter from git to SVN.
     """
     for i, line in enumerate(lines[:-1]):
         # Stop when we find the first patch
         if line[:3] == "+++" and lines[i + 1] == "---":
             break
-        if match(r"^diff --git \w/", line):
+        if diff_git_pattern.match(line):
             return git_diff_to_svn_diff
     return svn_diff_to_svn_diff
 
 _INITIAL_STATE = 1
 _DECLARED_FILE_PATH = 2
 _PROCESSING_CHUNK = 3
 
 
 class DiffFile(object):
     """Contains the information for one file in a patch.
@@ skipping 47 matching lines @@
         files = {}
         state = _INITIAL_STATE
         current_file = None
         old_diff_line = None
         new_diff_line = None
         transform_line = get_diff_converter(diff_input)
         for line in diff_input:
             line = line.rstrip("\n")
             line = transform_line(line)
 
-            file_declaration = match(r"^Index: (?P<FilePath>.+)", line)
+            file_declaration = index_pattern.match(line)
             if file_declaration:
                 filename = file_declaration.group('FilePath')
                 current_file = DiffFile(filename)
                 files[filename] = current_file
                 state = _DECLARED_FILE_PATH
                 continue
 
-            lines_changed = match(r"^@@ -(?P<OldStartLine>\d+)(,\d+)? \+(?P<NewStartLine>\d+)(,\d+)? @@", line)
+            lines_changed = lines_changed_pattern.match(line)
             if lines_changed:
                 if state != _DECLARED_FILE_PATH and state != _PROCESSING_CHUNK:
                     _log.error('Unexpected line change without file path '
                                'declaration: %r' % line)
                 old_diff_line = int(lines_changed.group('OldStartLine'))
                 new_diff_line = int(lines_changed.group('NewStartLine'))
                 state = _PROCESSING_CHUNK
                 continue
 
             if state == _PROCESSING_CHUNK:
                 if line.startswith('+'):
                     current_file.add_new_line(new_diff_line, line[1:])
                     new_diff_line += 1
                 elif line.startswith('-'):
                     current_file.add_deleted_line(old_diff_line, line[1:])
                     old_diff_line += 1
                 elif line.startswith(' '):
                     current_file.add_unchanged_line(old_diff_line, new_diff_line, line[1:])
                     old_diff_line += 1
                     new_diff_line += 1
                 elif line == '\\ No newline at end of file':
                     # Nothing to do.  We may still have some added lines.
                     pass
                 else:
                     _log.error('Unexpected diff format when parsing a '
                                'chunk: %r' % line)
         return files