patch-fixer 0.3.0__tar.gz → 0.3.2__tar.gz

{patch_fixer-0.3.0 → patch_fixer-0.3.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: patch-fixer
-Version: 0.3.0
+Version: 0.3.2
 Summary: Fixes erroneous git apply patches to the best of its ability.
 Maintainer-email: Alex Mueller <amueller474@gmail.com>
 License-Expression: Apache-2.0
@@ -22,6 +22,7 @@ Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: GitPython
 Provides-Extra: test
+Requires-Dist: hypothesis; extra == "test"
 Requires-Dist: pytest; extra == "test"
 Requires-Dist: requests; extra == "test"
 Dynamic: license-file
@@ -71,7 +72,7 @@ pytest
 ```
 From version `0.3.0` onward (at least until version `1.0`), some test failures are expected
 in bugfix versions as I like to use test-driven development to build out new features. 
-Please only report test failures if the same test passed in the most recent `0.x.0` version.
+Please only report test failures if the same test existed and passed in the most recent `0.x.0` version.
 
 ## License
 

{patch_fixer-0.3.0 → patch_fixer-0.3.2}/README.md

@@ -43,7 +43,7 @@ pytest
 ```
 From version `0.3.0` onward (at least until version `1.0`), some test failures are expected
 in bugfix versions as I like to use test-driven development to build out new features. 
-Please only report test failures if the same test passed in the most recent `0.x.0` version.
+Please only report test failures if the same test existed and passed in the most recent `0.x.0` version.
 
 ## License
 

patch_fixer-0.3.2/patch_fixer/__init__.py

@@ -0,0 +1,2 @@
+from .patch_fixer import fix_patch
+from .split import split_patch

{patch_fixer-0.3.0 → patch_fixer-0.3.2}/patch_fixer/patch_fixer.py

@@ -25,9 +25,40 @@ class MissingHunkError(Exception):
     pass
 
 
+class BadCarriageReturn(ValueError):
+    pass
+
+
 def normalize_line(line):
-    # preserve whitespace, only normalize line endings
-    return line.rstrip("\r\n") + "\n"
+    """Normalize line endings while preserving whitespace."""
+    if not isinstance(line, str):
+        raise TypeError(f"Cannot normalize non-string object {line}")
+
+    # edge case: empty string
+    if line == "":
+        return "\n"
+
+    # special malformed ending: ...\n\r
+    if line.endswith("\n\r"):
+        raise BadCarriageReturn(f"carriage return after line feed: {line}")
+
+    # handle CRLF and simple CR/LF endings
+    if line.endswith("\r\n"):
+        core = line[:-2]
+    elif line.endswith("\r"):
+        core = line[:-1]
+    elif line.endswith("\n"):
+        core = line[:-1]
+    else:
+        core = line
+
+    # check for interior CR/LF (anything before the final terminator)
+    if "\n" in core:
+        raise ValueError(f"line feed in middle of line: {line}")
+    if "\r" in core:
+        raise BadCarriageReturn(f"carriage return in middle of line: {line}")
+
+    return core + "\n"
 
 
 def find_hunk_start(context_lines, original_lines):
@@ -173,7 +204,7 @@ def fix_patch(patch_lines, original, remove_binary=False):
                         ) = capture_hunk(current_hunk, original_lines, offset, last_hunk, hunk_context)
                     except MissingHunkError:
                         raise NotImplementedError(f"Could not find hunk in {current_file}:"
-                                                  f"\n\n{"".join(current_hunk)}")
+                                                  f"\n\n{''.join(current_hunk)}")
                     fixed_lines.append(fixed_header)
                     fixed_lines.extend(current_hunk)
                     current_hunk = []
@@ -384,7 +415,7 @@ def fix_patch(patch_lines, original, remove_binary=False):
                     ) = capture_hunk(current_hunk, original_lines, offset, last_hunk, hunk_context)
                 except MissingHunkError:
                     raise NotImplementedError(f"Could not find hunk in {current_file}:"
-                                              f"\n\n{"".join(current_hunk)}")
+                                              f"\n\n{''.join(current_hunk)}")
                 fixed_lines.append(fixed_header)
                 fixed_lines.extend(current_hunk)
                 current_hunk = []
@@ -406,7 +437,7 @@ def fix_patch(patch_lines, original, remove_binary=False):
         ) = capture_hunk(current_hunk, original_lines, offset, last_hunk, hunk_context)
     except MissingHunkError:
         raise NotImplementedError(f"Could not find hunk in {current_file}:"
-                                  f"\n\n{"".join(current_hunk)}")
+                                  f"\n\n{''.join(current_hunk)}")
     fixed_lines.append(fixed_header)
     fixed_lines.extend(current_hunk)
 

patch_fixer-0.3.2/patch_fixer/split.py

@@ -0,0 +1,119 @@
+"""
+Idea:
+
+1. main function takes in:
+    a. patch file
+    b. list of files to split out
+2. reads patch file, splits based on file headers (assumed to be valid)
+3. for each file being patched:
+    a. if the file is in the list, send its hunks to output 1
+    b. otherwise send its hunks to output 2
+    c. hunks include all header lines so each output is a valid diff
+
+Could share some functionality with refactored, modular version of fix_patch
+"""
+
+import re
+from typing import List, Tuple
+
+from .patch_fixer import match_line, normalize_line, split_ab
+
+
+def get_file_path_from_diff(line: str) -> str:
+    """Extract the file path from a diff line."""
+    match_groups, line_type = match_line(line)
+    if line_type != "DIFF_LINE":
+        raise ValueError(f"Expected DIFF_LINE but got {line_type}")
+
+    # get the 'a' path (source file)
+    a_path, _ = split_ab(match_groups)
+    return a_path
+
+
+def split_patch(patch_lines: List[str], files_to_include: List[str]) -> Tuple[List[str], List[str]]:
+    """
+    Split a patch into two parts based on a list of files to include.
+
+    Parameters
+    ----------
+    patch_lines : List[str]
+        Lines of the patch file to split.
+    files_to_include : List[str]
+        List of file paths (relative, starting with ./) to include in the first output.
+        Files not in this list go to the second output.
+
+    Returns
+    -------
+    included_lines : List[str]
+        Lines for the patch containing only the included files.
+    excluded_lines : List[str]
+        Lines for the patch containing all other files.
+
+    Notes
+    -----
+    The function preserves all header information for each file's hunks
+    to ensure both output patches are valid. File paths are normalized
+    to start with './' for comparison purposes.
+
+    Raises
+    ------
+    ValueError
+        If the patch format is invalid or cannot be parsed.
+    """
+    if not patch_lines:
+        raise ValueError("Empty patch provided")
+
+    # normalize file paths to include
+    normalized_include = set()
+    for path in files_to_include:
+        if not path.startswith("./"):
+            path = f"./{path}"
+        normalized_include.add(path)
+
+    included_lines = []
+    excluded_lines = []
+    current_file_lines = []
+    current_file_path = None
+    in_file_block = False
+
+    for line in patch_lines:
+        match_groups, line_type = match_line(line)
+
+        if line_type == "DIFF_LINE":
+            # start of a new file block
+            if in_file_block and current_file_lines:
+                # output the previous file block
+                if current_file_path in normalized_include:
+                    included_lines.extend(current_file_lines)
+                else:
+                    excluded_lines.extend(current_file_lines)
+
+            # start collecting new file block
+            current_file_lines = [normalize_line(line)]
+            current_file_path = get_file_path_from_diff(line)
+            in_file_block = True
+
+        elif in_file_block:
+            # continue collecting lines for current file
+            current_file_lines.append(normalize_line(line))
+
+        else:
+            # lines before any diff (shouldn't happen in well-formed patches)
+            # add to both outputs to preserve any global headers
+            normalized = normalize_line(line)
+            included_lines.append(normalized)
+            excluded_lines.append(normalized)
+
+    # don't forget the last file block
+    if in_file_block and current_file_lines:
+        if current_file_path in normalized_include:
+            included_lines.extend(current_file_lines)
+        else:
+            excluded_lines.extend(current_file_lines)
+
+    # handle edge case where no files were split (no diff lines)
+    if not in_file_block:
+        # patch had no diff lines at all
+        return patch_lines, []
+
+    return included_lines, excluded_lines

{patch_fixer-0.3.0 → patch_fixer-0.3.2}/patch_fixer.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: patch-fixer
-Version: 0.3.0
+Version: 0.3.2
 Summary: Fixes erroneous git apply patches to the best of its ability.
 Maintainer-email: Alex Mueller <amueller474@gmail.com>
 License-Expression: Apache-2.0
@@ -22,6 +22,7 @@ Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: GitPython
 Provides-Extra: test
+Requires-Dist: hypothesis; extra == "test"
 Requires-Dist: pytest; extra == "test"
 Requires-Dist: requests; extra == "test"
 Dynamic: license-file
@@ -71,7 +72,7 @@ pytest
 ```
 From version `0.3.0` onward (at least until version `1.0`), some test failures are expected
 in bugfix versions as I like to use test-driven development to build out new features. 
-Please only report test failures if the same test passed in the most recent `0.x.0` version.
+Please only report test failures if the same test existed and passed in the most recent `0.x.0` version.
 
 ## License
 

{patch_fixer-0.3.0 → patch_fixer-0.3.2}/patch_fixer.egg-info/SOURCES.txt

@@ -3,9 +3,12 @@ README.md
 pyproject.toml
 patch_fixer/__init__.py
 patch_fixer/patch_fixer.py
+patch_fixer/split.py
 patch_fixer.egg-info/PKG-INFO
 patch_fixer.egg-info/SOURCES.txt
 patch_fixer.egg-info/dependency_links.txt
 patch_fixer.egg-info/requires.txt
 patch_fixer.egg-info/top_level.txt
-tests/test_repos.py
+tests/test_norm.py
+tests/test_repos.py
+tests/test_split.py

{patch_fixer-0.3.0 → patch_fixer-0.3.2}/patch_fixer.egg-info/requires.txt

@@ -1,5 +1,6 @@
 GitPython
 
 [test]
+hypothesis
 pytest
 requests

{patch_fixer-0.3.0 → patch_fixer-0.3.2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "patch-fixer"
-version = "0.3.0"
+version = "0.3.2"
 description = "Fixes erroneous git apply patches to the best of its ability."
 maintainers = [
     {name = "Alex Mueller", email="amueller474@gmail.com"},
@@ -34,6 +34,7 @@ license-files = [
 
 [project.optional-dependencies]
 test = [
+    "hypothesis",
     "pytest",
     "requests"
 ]

patch_fixer-0.3.2/tests/test_norm.py

@@ -0,0 +1,87 @@
+import pytest
+from hypothesis import given, strategies as st
+
+from patch_fixer.patch_fixer import normalize_line, BadCarriageReturn
+
+# --- Good cases --------------------------------------------------
+
+@pytest.mark.parametrize("line, expected", [
+    ("", "\n"),                # empty string -> newline
+    ("foo", "foo\n"),          # no terminator
+    ("foo\r", "foo\n"),        # CR terminator normalized
+    ("foo\n", "foo\n"),        # LF terminator unchanged
+    ("foo\r\n", "foo\n"),      # CRLF normalized
+])
+def test_normalize_good(line, expected):
+    assert normalize_line(line) == expected
+
+
+# --- Type errors -------------------------------------------------
+
+@pytest.mark.parametrize("bad", [
+    123,
+    4.56,
+    None,
+    True,
+    ["list"],
+    {"set"},
+    {"dict": "val"},
+    ("tuple",),
+])
+def test_normalize_type_error(bad):
+    with pytest.raises(TypeError):
+        normalize_line(bad)
+
+
+# --- Bad endings -------------------------------------------------
+
+@pytest.mark.parametrize("line", [
+    "foo\n\r",   # LF then CR
+    "foo\rx",    # CR not followed by LF at end
+])
+def test_normalize_bad_endings(line):
+    with pytest.raises(BadCarriageReturn):
+        normalize_line(line)
+
+
+# --- Interior CR/LF ----------------------------------------------
+
+def test_interior_lf_raises():
+    line = "bad\nline\n"
+    with pytest.raises(ValueError):
+        normalize_line(line)
+
+def test_interior_cr_raises():
+    line = "bad\rcarriage\n"
+    with pytest.raises(BadCarriageReturn):
+        normalize_line(line)
+
+# --- Hypothesis testing ------------------------------------------
+
+# generate arbitrary strings including \r and \n
+line_strategy = st.text(alphabet=st.characters(), min_size=0, max_size=100)
+
+@given(line=line_strategy)
+def test_normalize_line_hypothesis(line):
+    # we want to see that normalize_line either:
+    # 1. returns a string ending with exactly one "\n", or
+    # 2. raises ValueError for interior LF, or
+    # 3. raises BadCarriageReturn for interior CR or malformed endings
+    try:
+        result = normalize_line(line)
+    except BadCarriageReturn:
+        # must have an interior CR somewhere, or malformed ending
+        cr_condition = (("\r" in line[:-2])
+                        or (line.endswith("\r") and not line.endswith("\r\n"))
+                        or line.endswith("\n\r"))
+        assert cr_condition, f"BadCarriageReturn raised unexpectedly for line: {line!r}"
+    except ValueError:
+        # must have an interior LF somewhere
+        assert "\n" in line[:-1], f"ValueError raised unexpectedly for line: {line!r}"
+    else:
+        # function returned normally
+        assert result.endswith("\n"), f"Returned line does not end with \\n: {result!r}"
+
+        core = result[:-1]
+        assert "\n" not in core
+        assert "\r" not in core

{patch_fixer-0.3.0 → patch_fixer-0.3.2}/tests/test_repos.py

@@ -27,11 +27,20 @@ import pytest
 from patch_fixer import fix_patch
 
 REPOS = {
+    ("apache", "airflow"): ("26f6e54","2136f56"),   # big repo
     ("asottile", "astpretty"): ("5b68c7e", "5a8296f"),
+    ("astral-sh", "ruff"): ("7fee877", "11dae2c"),
+    ("gabrielecirulli", "2048"): ("878098f", "478b6ec"),    # adds binary files
+    ("mrdoob", "three.js"): ("5f3a718", "b97f111"),         # replaces images
+    ("myriadrf", "LimeSDR-Mini"): ("0bb75e7", "fb012c8"),   # gigantic diffs
     ("numpy", "numpy"): ("dca33b3", "5f82966"),
     ("pallets", "click"): ("93c6966", "e11a1ef"),
+    ("psf", "black"): ("8d9d18c", "903bef5"),   # whole year's worth of changes
+    ("PyCQA", "flake8"): ("8bdec0b", "d45bdc0"),    # two years of changes
     ("scipy", "scipy"): ("c2220c0", "4ca6dd9"),
+    ("tox-dev", "tox"): ("fb3fe66", "01442da"),     # four years
     ("yaml", "pyyaml"): ("48838a3", "a2d19c0"),
+    ("zertovitch", "hac"): ("c563d18", "17207ee")   # renamed binary files
 }
 
 CACHE_DIR = Path.home() / ".patch-testing"

patch_fixer-0.3.2/tests/test_split.py

@@ -0,0 +1,386 @@
+"""Tests for the split_patch functionality."""
+
+import pytest
+
+from patch_fixer.split import split_patch
+
+
+class TestSplitPatch:
+    """Test cases for split_patch function."""
+
+    def test_simple_split(self):
+        """Test basic splitting with two files."""
+        patch = [
+            "diff --git a/file1.txt b/file1.txt\n",
+            "index 1234567..abcdefg 100644\n",
+            "--- a/file1.txt\n",
+            "+++ b/file1.txt\n",
+            "@@ -1,3 +1,3 @@\n",
+            " line1\n",
+            "-old line\n",
+            "+new line\n",
+            " line3\n",
+            "diff --git a/file2.txt b/file2.txt\n",
+            "index 2234567..bbcdefg 100644\n",
+            "--- a/file2.txt\n",
+            "+++ b/file2.txt\n",
+            "@@ -1,2 +1,2 @@\n",
+            "-removed\n",
+            "+added\n",
+        ]
+
+        included, excluded = split_patch(patch, ["./file1.txt"])
+
+        # check that file1 is in included
+        assert "diff --git a/file1.txt b/file1.txt\n" in included
+        assert "+new line\n" in included
+
+        # check that file2 is in excluded
+        assert "diff --git a/file2.txt b/file2.txt\n" in excluded
+        assert "+added\n" in excluded
+
+        # check that files are not mixed
+        assert "file2.txt" not in "".join(included)
+        assert "file1.txt" not in "".join(excluded)
+
+    def test_split_with_multiple_includes(self):
+        """Test splitting with multiple files to include."""
+        patch = [
+            "diff --git a/file1.txt b/file1.txt\n",
+            "--- a/file1.txt\n",
+            "+++ b/file1.txt\n",
+            "@@ -1,1 +1,1 @@\n",
+            "-old1\n",
+            "+new1\n",
+            "diff --git a/file2.txt b/file2.txt\n",
+            "--- a/file2.txt\n",
+            "+++ b/file2.txt\n",
+            "@@ -1,1 +1,1 @@\n",
+            "-old2\n",
+            "+new2\n",
+            "diff --git a/file3.txt b/file3.txt\n",
+            "--- a/file3.txt\n",
+            "+++ b/file3.txt\n",
+            "@@ -1,1 +1,1 @@\n",
+            "-old3\n",
+            "+new3\n",
+        ]
+
+        included, excluded = split_patch(patch, ["./file1.txt", "./file3.txt"])
+
+        # verify file1 and file3 are included
+        assert "file1.txt" in "".join(included)
+        assert "file3.txt" in "".join(included)
+        assert "+new1\n" in included
+        assert "+new3\n" in included
+
+        # verify file2 is excluded
+        assert "file2.txt" in "".join(excluded)
+        assert "+new2\n" in excluded
+        assert "file1.txt" not in "".join(excluded)
+        assert "file3.txt" not in "".join(excluded)
+
+    def test_split_with_no_includes(self):
+        """Test when no files match the include list."""
+        patch = [
+            "diff --git a/file1.txt b/file1.txt\n",
+            "--- a/file1.txt\n",
+            "+++ b/file1.txt\n",
+            "@@ -1,1 +1,1 @@\n",
+            "-old\n",
+            "+new\n",
+        ]
+
+        included, excluded = split_patch(patch, ["./nonexistent.txt"])
+
+        # all content should be in excluded
+        assert len(included) == 0
+        assert "file1.txt" in "".join(excluded)
+        assert "+new\n" in excluded
+
+    def test_split_with_all_includes(self):
+        """Test when all files match the include list."""
+        patch = [
+            "diff --git a/file1.txt b/file1.txt\n",
+            "--- a/file1.txt\n",
+            "+++ b/file1.txt\n",
+            "@@ -1,1 +1,1 @@\n",
+            "-old\n",
+            "+new\n",
+        ]
+
+        included, excluded = split_patch(patch, ["./file1.txt"])
+
+        # all content should be in included
+        assert "file1.txt" in "".join(included)
+        assert "+new\n" in included
+        assert len(excluded) == 0
+
+    def test_split_preserves_headers(self):
+        """Test that all necessary headers are preserved."""
+        patch = [
+            "diff --git a/file.txt b/file.txt\n",
+            "index 1234567..abcdefg 100644\n",
+            "new file mode 100644\n",
+            "--- /dev/null\n",
+            "+++ b/file.txt\n",
+            "@@ -0,0 +1,2 @@\n",
+            "+new file\n",
+            "+content\n",
+        ]
+
+        included, excluded = split_patch(patch, ["./file.txt"])
+
+        # check all headers are preserved
+        assert "diff --git a/file.txt b/file.txt\n" in included
+        assert "index 1234567..abcdefg 100644\n" in included
+        assert "new file mode 100644\n" in included
+        assert "--- /dev/null\n" in included
+        assert "+++ b/file.txt\n" in included
+        assert "@@ -0,0 +1,2 @@\n" in included
+
+    def test_split_with_rename(self):
+        """Test splitting patches with file renames."""
+        patch = [
+            "diff --git a/old.txt b/new.txt\n",
+            "similarity index 95%\n",
+            "rename from old.txt\n",
+            "rename to new.txt\n",
+            "index 1234567..abcdefg 100644\n",
+            "--- a/old.txt\n",
+            "+++ b/new.txt\n",
+            "@@ -1,3 +1,3 @@\n",
+            " same line\n",
+            "-old content\n",
+            "+new content\n",
+        ]
+
+        # include based on old name (source file)
+        included, excluded = split_patch(patch, ["./old.txt"])
+
+        assert "rename from old.txt\n" in included
+        assert "rename to new.txt\n" in included
+        assert len(excluded) == 0
+
+    def test_split_with_binary_files(self):
+        """Test splitting patches containing binary files."""
+        patch = [
+            "diff --git a/image.png b/image.png\n",
+            "index 1234567..abcdefg 100644\n",
+            "Binary files a/image.png and b/image.png differ\n",
+            "diff --git a/text.txt b/text.txt\n",
+            "--- a/text.txt\n",
+            "+++ b/text.txt\n",
+            "@@ -1,1 +1,1 @@\n",
+            "-old\n",
+            "+new\n",
+        ]
+
+        included, excluded = split_patch(patch, ["./image.png"])
+
+        assert "Binary files a/image.png and b/image.png differ\n" in included
+        assert "text.txt" in "".join(excluded)
+        assert "image.png" not in "".join(excluded)
+
+    def test_normalization_of_paths(self):
+        """Test that file paths are normalized correctly."""
+        patch = [
+            "diff --git a/file.txt b/file.txt\n",
+            "--- a/file.txt\n",
+            "+++ b/file.txt\n",
+            "@@ -1,1 +1,1 @@\n",
+            "-old\n",
+            "+new\n",
+        ]
+
+        # test without ./ prefix
+        included1, excluded1 = split_patch(patch, ["file.txt"])
+        assert "file.txt" in "".join(included1)
+        assert len(excluded1) == 0
+
+        # test with ./ prefix
+        included2, excluded2 = split_patch(patch, ["./file.txt"])
+        assert "file.txt" in "".join(included2)
+        assert len(excluded2) == 0
+
+        # both should produce same result
+        assert included1 == included2
+        assert excluded1 == excluded2
+
+    def test_empty_patch(self):
+        """Test behavior with empty patch."""
+        with pytest.raises(ValueError, match="Empty patch provided"):
+            split_patch([], ["./file.txt"])
+
+    def test_invalid_diff_format(self):
+        """Test behavior with invalid diff format."""
+        invalid_patch = [
+            "not a valid diff line\n",
+            "diff --git a/file.txt b/file.txt\n",
+            "--- a/file.txt\n",
+            "+++ b/file.txt\n",
+            "@@ -1,1 +1,1 @@\n",
+            "-old\n",
+            "+new\n",
+        ]
+
+        # should handle gracefully - non-diff lines before first diff
+        included, excluded = split_patch(invalid_patch, ["./file.txt"])
+
+        # the invalid line should be in both outputs (global header behavior)
+        assert "not a valid diff line\n" in included
+        assert "not a valid diff line\n" in excluded
+
+    def test_no_files_in_include_list(self):
+        """Test when include list is empty."""
+        patch = [
+            "diff --git a/file.txt b/file.txt\n",
+            "--- a/file.txt\n",
+            "+++ b/file.txt\n",
+            "@@ -1,1 +1,1 @@\n",
+            "-old\n",
+            "+new\n",
+        ]
+
+        included, excluded = split_patch(patch, [])
+
+        # everything should go to excluded
+        assert len(included) == 0
+        assert "file.txt" in "".join(excluded)
+
+    def test_patch_with_no_diff_lines(self):
+        """Test patch that has no actual diff lines."""
+        patch = [
+            "This is a comment\n",
+            "Another comment\n",
+        ]
+
+        included, excluded = split_patch(patch, ["./file.txt"])
+
+        # non-diff lines should appear in original form
+        assert patch == included
+        assert len(excluded) == 0
+
+    def test_multiple_hunks_same_file(self):
+        """Test that multiple hunks for the same file stay together."""
+        patch = [
+            "diff --git a/file.txt b/file.txt\n",
+            "--- a/file.txt\n",
+            "+++ b/file.txt\n",
+            "@@ -1,1 +1,1 @@\n",
+            "-old1\n",
+            "+new1\n",
+            "@@ -10,1 +10,1 @@\n",
+            "-old2\n",
+            "+new2\n",
+            "@@ -20,1 +20,1 @@\n",
+            "-old3\n",
+            "+new3\n",
+        ]
+
+        included, excluded = split_patch(patch, ["./file.txt"])
+
+        # all hunks should be in included
+        assert "+new1\n" in included
+        assert "+new2\n" in included
+        assert "+new3\n" in included
+        assert len(excluded) == 0
+
+    def test_file_deletion(self):
+        """Test splitting patches with file deletions."""
+        patch = [
+            "diff --git a/deleted.txt b/deleted.txt\n",
+            "deleted file mode 100644\n",
+            "index 1234567..0000000\n",
+            "--- a/deleted.txt\n",
+            "+++ /dev/null\n",
+            "@@ -1,3 +0,0 @@\n",
+            "-line1\n",
+            "-line2\n",
+            "-line3\n",
+        ]
+
+        included, excluded = split_patch(patch, ["./deleted.txt"])
+
+        assert "deleted file mode 100644\n" in included
+        assert "+++ /dev/null\n" in included
+        assert "-line1\n" in included
+        assert len(excluded) == 0
+
+    def test_file_creation(self):
+        """Test splitting patches with new file creation."""
+        patch = [
+            "diff --git a/new.txt b/new.txt\n",
+            "new file mode 100644\n",
+            "index 0000000..1234567\n",
+            "--- /dev/null\n",
+            "+++ b/new.txt\n",
+            "@@ -0,0 +1,3 @@\n",
+            "+line1\n",
+            "+line2\n",
+            "+line3\n",
+        ]
+
+        included, excluded = split_patch(patch, ["./new.txt"])
+
+        assert "new file mode 100644\n" in included
+        assert "--- /dev/null\n" in included
+        assert "+line1\n" in included
+        assert len(excluded) == 0
+
+    def test_complex_patch(self):
+        """Test a complex patch with various file operations."""
+        patch = [
+            "diff --git a/modified.txt b/modified.txt\n",
+            "index 1234567..abcdefg 100644\n",
+            "--- a/modified.txt\n",
+            "+++ b/modified.txt\n",
+            "@@ -1,1 +1,1 @@\n",
+            "-old\n",
+            "+new\n",
+            "diff --git a/created.txt b/created.txt\n",
+            "new file mode 100644\n",
+            "index 0000000..2234567\n",
+            "--- /dev/null\n",
+            "+++ b/created.txt\n",
+            "@@ -0,0 +1,1 @@\n",
+            "+created content\n",
+            "diff --git a/deleted.txt b/deleted.txt\n",
+            "deleted file mode 100644\n",
+            "index 3234567..0000000\n",
+            "--- a/deleted.txt\n",
+            "+++ /dev/null\n",
+            "@@ -1,1 +0,0 @@\n",
+            "-deleted content\n",
+            "diff --git a/renamed_old.txt b/renamed_new.txt\n",
+            "similarity index 90%\n",
+            "rename from renamed_old.txt\n",
+            "rename to renamed_new.txt\n",
+            "index 4234567..5234567\n",
+            "--- a/renamed_old.txt\n",
+            "+++ b/renamed_new.txt\n",
+            "@@ -1,1 +1,1 @@\n",
+            "-before rename\n",
+            "+after rename\n",
+        ]
+
+        # include modified and renamed files
+        included, excluded = split_patch(patch, ["./modified.txt", "./renamed_old.txt"])
+
+        # check included has modified and renamed
+        assert "modified.txt" in "".join(included)
+        assert "renamed_old.txt" in "".join(included)
+        assert "+new\n" in included
+        assert "+after rename\n" in included
+
+        # check excluded has created and deleted
+        assert "created.txt" in "".join(excluded)
+        assert "deleted.txt" in "".join(excluded)
+        assert "+created content\n" in excluded
+        assert "-deleted content\n" in excluded
+
+        # check no cross-contamination
+        assert "created.txt" not in "".join(included)
+        assert "deleted.txt" not in "".join(included)
+        assert "modified.txt" not in "".join(excluded)
+        assert "renamed" not in "".join(excluded)

patch_fixer-0.3.0/patch_fixer/__init__.py

@@ -1 +0,0 @@
-from .patch_fixer import fix_patch