Merge pull request #95 from asottile-sentry/filenames-with-spaces

fix filenames with spaces

Author: matiasb

tests/samples/git_filenames_with_spaces.diff

@@ -0,0 +1,7 @@
+diff --git a/has spaces/t.sql b/has spaces/t.sql
+new file mode 100644
+index 0000000..8a9b485
+--- /dev/null
++++ b/has spaces/t.sql
+@@ -0,0 +1 @@
++select * FROM t;

tests/test_parser.py

@@ -313,6 +313,18 @@ def test_parse_diff_git_no_prefix(self):
         self.assertTrue(res[2].is_added_file)
         self.assertEqual(res[2].path, 'file3')
 
+    def test_parse_filename_with_spaces(self):
+        filename = os.path.join(self.samples_dir, 'samples/git_filenames_with_spaces.diff')
+        with open(filename) as f:
+            res = PatchSet(f)
+
+        self.assertEqual(len(res), 1)
+
+        self.assertEqual(res[0].source_file, '/dev/null')
+        self.assertEqual(res[0].target_file, 'b/has spaces/t.sql')
+        self.assertTrue(res[0].is_added_file)
+        self.assertEqual(res[0].path, 'has spaces/t.sql')
+
     def test_diff_lines_linenos(self):
         with open(self.sample_file, 'rb') as diff_file:
             res = PatchSet(diff_file, encoding='utf-8')

unidiff/constants.py

@@ -37,7 +37,9 @@
 
 # check diff git line for git renamed files support
 RE_DIFF_GIT_HEADER = re.compile(
-    r'^diff --git (?P<source>(a/)?[^\t\n]+) (?P<target>(b/)?[^\t\n]+)')
+    r'^diff --git (?P<source>a/[^\t\n]+) (?P<target>b/[^\t\n]+)')
+RE_DIFF_GIT_HEADER_NO_PREFIX = re.compile(
+    r'^diff --git (?P<source>[^\t\n]+) (?P<target>[^\t\n]+)')
 
 # check diff git new file marker `deleted file mode 100644`
 RE_DIFF_GIT_DELETED_FILE = re.compile(r'^deleted file mode \d+\n$')

unidiff/patch.py

@@ -40,6 +40,7 @@
     LINE_VALUE_NO_NEWLINE,
     RE_DIFF_GIT_DELETED_FILE,
     RE_DIFF_GIT_HEADER,
+    RE_DIFF_GIT_HEADER_NO_PREFIX,
     RE_DIFF_GIT_NEW_FILE,
     RE_HUNK_BODY_LINE,
     RE_HUNK_EMPTY_BODY_LINE,
@@ -478,7 +479,7 @@ def _parse(self, diff, encoding, metadata_only):
                 line = line.decode(encoding)
 
             # check for a git file rename
-            is_diff_git_header = RE_DIFF_GIT_HEADER.match(line)
+            is_diff_git_header = RE_DIFF_GIT_HEADER.match(line) or RE_DIFF_GIT_HEADER_NO_PREFIX.match(line)
             if is_diff_git_header:
                 patch_info = PatchInfo()
                 source_file = is_diff_git_header.group('source')