added the newline parameter to PatchSet::from_filename and a test cas…

.github/workflows/main.yml

@@ -0,0 +1,28 @@
+name: CI
+
+on:
+  push:
+    branches: [ master ]
+  pull_request:
+    branches: [ master ]
+
+  workflow_dispatch:
+
+jobs:
+  build:
+    runs-on: self-hosted
+    strategy:
+      matrix:
+        python-version: [ '2.7', '3.6', '3.7', '3.8', '3.9' ]
+      fail-fast: false  
+
+    steps:
+      - uses: actions/checkout@v2
+      - uses: actions/setup-python@v2
+        with:
+          python-version: ${{ matrix.python-version }}
+          architecture: 'x64' 
+      - name: 'Run Tests'
+        run: $GITHUB_WORKSPACE/run_tests.sh
+        shell: bash
+

tests/samples/git_cr.diff

@@ -0,0 +1,9 @@
+diff --git a/src/test/org/apache/commons/math/util/ExpandableDoubleArrayTest.java b/src/test/org/apache/commons/math/util/ExpandableDoubleArrayTest.java
+new file mode 100644
+index 000000000..2b38fa232
+--- /dev/null
++++ b/a.txt
+@@ -0,0 +1,3 @@
++ "This line is broken into two lines by CR. " +		"but it should be treated as one line in the text diff file"
++ "This has no CR"
++ "This line also has CR. " +              "but it should also be treated as one line in the text diff file". 

tests/test_parser.py

@@ -231,6 +231,18 @@ def test_parse_malformed_diff_shorter_than_expected(self):
         with open(utf8_file, 'r') as diff_file:
             self.assertRaises(UnidiffParseError, PatchSet, diff_file)
 
+    def test_from_filename_with_cr_in_diff_text_files(self):
+        """Parse git diff text files that contain CR"""
+        utf8_file = os.path.join(self.samples_dir, 'samples/git_cr.diff')
+        self.assertRaises(UnidiffParseError, PatchSet.from_filename, utf8_file)
+
+        ps1 = PatchSet.from_filename(utf8_file, newline='\n')
+        import io
+        with io.open(utf8_file, 'r', newline='\n') as diff_file:
+            ps2 = PatchSet(diff_file)
+
+        self.assertEqual(ps1, ps2)
+
     def test_parse_diff_with_new_and_modified_binary_files(self):
         """Parse git diff file with newly added and modified binaries files."""
         utf8_file = os.path.join(self.samples_dir, 'samples/sample8.diff')

unidiff/patch.py

@@ -52,8 +52,10 @@
 
 PY2 = sys.version_info[0] == 2
 if PY2:
+    import io
     from StringIO import StringIO
-    open_file = codecs.open
+    # open_file = codecs.open
+    open_file = io.open
     make_str = lambda x: x.encode(DEFAULT_ENCODING)
 
     def implements_to_string(cls):
@@ -557,10 +559,10 @@ def _parse(self, diff, encoding, metadata_only):
             patch_info.append(line)
 
     @classmethod
-    def from_filename(cls, filename, encoding=DEFAULT_ENCODING, errors=None):
+    def from_filename(cls, filename, encoding=DEFAULT_ENCODING, errors=None, newline=None):
         # type: (str, str, Optional[str]) -> PatchSet
         """Return a PatchSet instance given a diff filename."""
-        with open_file(filename, 'r', encoding=encoding, errors=errors) as f:
+        with open_file(filename, 'r', encoding=encoding, errors=errors, newline=newline) as f:
             instance = cls(f)
         return instance