diffpdf 0.2.0__tar.gz → 0.3.0__tar.gz

{diffpdf-0.2.0 → diffpdf-0.3.0}/.github/workflows/pypi-publish.yml

@@ -11,10 +11,10 @@ jobs:
       id-token: write
 
     steps:
-    - uses: actions/checkout@v4
+    - uses: actions/checkout@v6
 
     - name: Set up Python
-      uses: actions/setup-python@v5
+      uses: actions/setup-python@v6
       with:
         python-version: '3.10'
 

{diffpdf-0.2.0 → diffpdf-0.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: diffpdf
-Version: 0.2.0
+Version: 0.3.0
 Summary: A tool for comparing PDF files
 Project-URL: Homepage, https://github.com/JustusRijke/DiffPDF
 Project-URL: Issues, https://github.com/JustusRijke/DiffPDF/issues
@@ -15,7 +15,7 @@ Requires-Python: >=3.10
 Requires-Dist: click
 Requires-Dist: colorlog
 Requires-Dist: pillow>=10.0.0
-Requires-Dist: pixelmatch>=0.3.0
+Requires-Dist: pixelmatch-fast>=1.1.0
 Requires-Dist: pymupdf>=1.23.0
 Provides-Extra: dev
 Requires-Dist: pytest; extra == 'dev'
@@ -39,12 +39,10 @@ DiffPDF uses a fail-fast sequential pipeline to compare PDFs:
 1. **Hash Check** - SHA-256 comparison. If identical, exit immediately with pass.
 2. **Page Count** - Verify both PDFs have the same number of pages.
 3. **Text Content** - Extract and compare text from all pages (ignoring whitespace).
-4. **Visual Check** - Render pages to images and compare using pixelmatch.
+4. **Visual Check** - Render pages to images and compare using [pixelmatch-fast](https://pypi.org/project/pixelmatch-fast/).
 
 Each stage only runs if all previous stages pass.
 
-**⚠️ Performance Warning:** The Python port of pixelmatch is extremely slow.
-
 ## Installation
 
 ```bash
@@ -60,7 +58,7 @@ Usage: diffpdf [OPTIONS] REFERENCE ACTUAL
 Options:
   --threshold FLOAT       Pixelmatch threshold (0.0-1.0)
   --dpi INTEGER           Render resolution
-  --output-dir DIRECTORY  Diff image output directory
+  --output-dir DIRECTORY  Diff image output directory (optional, if not specified no diff images are saved)
   -v, --verbose           Increase verbosity (-v for INFO, -vv for DEBUG)
   --save-log              Write log output to log.txt
   --version               Show the version and exit.
@@ -75,10 +73,14 @@ Options:
 
 ## Library Usage
 
-Call the CLI from Python:
 ```python
-from diffpdf import main
-main(["-vv","foo.pdf", "bar.pdf"])
+from diffpdf import diffpdf
+
+# Basic usage (no diff images saved)
+diffpdf("reference.pdf", "actual.pdf")
+
+# With options (save diff images to ./output directory)
+diffpdf("reference.pdf", "actual.pdf", output_dir="./output", threshold=0.2, dpi=150, verbosity=2)
 ```
 
 ## Development
@@ -91,4 +93,4 @@ ruff check .
 
 ## Acknowledgements
 
-Built with [PyMuPDF](https://pymupdf.readthedocs.io/) for PDF parsing and [pixelmatch-py](https://github.com/whtsky/pixelmatch-py) (Python port of [pixelmatch](https://github.com/mapbox/pixelmatch)) for visual comparison.
+Built with [PyMuPDF](https://pymupdf.readthedocs.io/) for PDF parsing and [pixelmatch-fast](https://pypi.org/project/pixelmatch-fast/) (Python port of [pixelmatch](https://github.com/mapbox/pixelmatch)) for visual comparison.

{diffpdf-0.2.0 → diffpdf-0.3.0}/README.md

@@ -14,12 +14,10 @@ DiffPDF uses a fail-fast sequential pipeline to compare PDFs:
 1. **Hash Check** - SHA-256 comparison. If identical, exit immediately with pass.
 2. **Page Count** - Verify both PDFs have the same number of pages.
 3. **Text Content** - Extract and compare text from all pages (ignoring whitespace).
-4. **Visual Check** - Render pages to images and compare using pixelmatch.
+4. **Visual Check** - Render pages to images and compare using [pixelmatch-fast](https://pypi.org/project/pixelmatch-fast/).
 
 Each stage only runs if all previous stages pass.
 
-**⚠️ Performance Warning:** The Python port of pixelmatch is extremely slow.
-
 ## Installation
 
 ```bash
@@ -35,7 +33,7 @@ Usage: diffpdf [OPTIONS] REFERENCE ACTUAL
 Options:
   --threshold FLOAT       Pixelmatch threshold (0.0-1.0)
   --dpi INTEGER           Render resolution
-  --output-dir DIRECTORY  Diff image output directory
+  --output-dir DIRECTORY  Diff image output directory (optional, if not specified no diff images are saved)
   -v, --verbose           Increase verbosity (-v for INFO, -vv for DEBUG)
   --save-log              Write log output to log.txt
   --version               Show the version and exit.
@@ -50,10 +48,14 @@ Options:
 
 ## Library Usage
 
-Call the CLI from Python:
 ```python
-from diffpdf import main
-main(["-vv","foo.pdf", "bar.pdf"])
+from diffpdf import diffpdf
+
+# Basic usage (no diff images saved)
+diffpdf("reference.pdf", "actual.pdf")
+
+# With options (save diff images to ./output directory)
+diffpdf("reference.pdf", "actual.pdf", output_dir="./output", threshold=0.2, dpi=150, verbosity=2)
 ```
 
 ## Development
@@ -66,4 +68,4 @@ ruff check .
 
 ## Acknowledgements
 
-Built with [PyMuPDF](https://pymupdf.readthedocs.io/) for PDF parsing and [pixelmatch-py](https://github.com/whtsky/pixelmatch-py) (Python port of [pixelmatch](https://github.com/mapbox/pixelmatch)) for visual comparison.
+Built with [PyMuPDF](https://pymupdf.readthedocs.io/) for PDF parsing and [pixelmatch-fast](https://pypi.org/project/pixelmatch-fast/) (Python port of [pixelmatch](https://github.com/mapbox/pixelmatch)) for visual comparison.

{diffpdf-0.2.0 → diffpdf-0.3.0}/pyproject.toml

@@ -24,7 +24,7 @@ dependencies = [
     "click",
     "colorlog",
     "pymupdf>=1.23.0",
-    "pixelmatch>=0.3.0",
+    "pixelmatch-fast>=1.1.0",
     "Pillow>=10.0.0",
 ]
 
@@ -40,7 +40,7 @@ dev = [
 ]
 
 [project.scripts]
-diffpdf = "diffpdf:main"
+diffpdf = "diffpdf.cli:cli"
 
 [tool.hatch.version]
 source = "vcs"

diffpdf-0.3.0/src/diffpdf/__init__.py

@@ -0,0 +1,27 @@
+from importlib.metadata import version
+from pathlib import Path
+
+from .comparators import compare_pdfs
+from .logger import setup_logging
+
+__version__ = version("diffpdf")
+
+
+def diffpdf(
+    reference: str | Path,
+    actual: str | Path,
+    threshold: float = 0.1,
+    dpi: int = 96,
+    output_dir: str | Path | None = None,
+    verbosity: int = 0,
+    save_log: bool = False,
+) -> None:
+    ref_path = Path(reference) if isinstance(reference, str) else reference
+    actual_path = Path(actual) if isinstance(actual, str) else actual
+    out_path = Path(output_dir) if isinstance(output_dir, str) else output_dir
+
+    logger = setup_logging(verbosity, save_log)
+    compare_pdfs(ref_path, actual_path, threshold, dpi, out_path, logger)
+
+
+__all__ = ["diffpdf", "__version__"]

{diffpdf-0.2.0 → diffpdf-0.3.0}/src/diffpdf/cli.py

@@ -19,8 +19,8 @@ from .logger import setup_logging
 @click.option(
     "--output-dir",
     type=click.Path(file_okay=False, path_type=Path),
-    default="./",
-    help="Diff image output directory",
+    default=None,
+    help="Diff image output directory (if not specified, no diff images are saved)",
 )
 @click.option(
     "-v",

{diffpdf-0.2.0 → diffpdf-0.3.0}/src/diffpdf/comparators.py

@@ -8,7 +8,7 @@ from .visual_check import check_visual_content
 
 
 def compare_pdfs(
-    ref: Path, actual: Path, threshold: float, dpi: int, output_dir: Path, logger
+    ref: Path, actual: Path, threshold: float, dpi: int, output_dir: Path | None, logger
 ) -> None:
     check_hash(ref, actual, logger)
 

{diffpdf-0.2.0 → diffpdf-0.3.0}/src/diffpdf/visual_check.py

@@ -3,7 +3,7 @@ from pathlib import Path
 
 import fitz
 from PIL import Image
-from pixelmatch.contrib.PIL import pixelmatch
+from pixelmatch import pixelmatch
 
 
 def render_page_to_image(pdf_path: Path, page_num: int, dpi: int) -> Image.Image:
@@ -16,24 +16,28 @@ def render_page_to_image(pdf_path: Path, page_num: int, dpi: int) -> Image.Image
 
 
 def compare_images(
-    ref_img: Image.Image, actual_img: Image.Image, threshold: float, output_path: Path
+    ref_img: Image.Image,
+    actual_img: Image.Image,
+    threshold: float,
+    output_path: Path | None,
 ) -> bool:
-    diff_img = Image.new("RGB", ref_img.size)
-    mismatch_count = pixelmatch(ref_img, actual_img, diff_img, threshold=threshold)
+    mismatch_count = pixelmatch(
+        ref_img, actual_img, diff_path=output_path, threshold=threshold
+    )
 
     if mismatch_count > 0:
-        diff_img.save(output_path)
         return False
 
     return True
 
 
 def check_visual_content(
-    ref: Path, actual: Path, threshold: float, dpi: int, output_dir: Path, logger
+    ref: Path, actual: Path, threshold: float, dpi: int, output_dir: Path | None, logger
 ) -> None:
     logger.info("[4/4] Checking visual content...")
 
-    output_dir.mkdir(parents=True, exist_ok=True)
+    if output_dir is not None:
+        output_dir.mkdir(parents=True, exist_ok=True)
 
     ref_doc = fitz.open(ref)
     page_count = len(ref_doc)
@@ -45,11 +49,13 @@ def check_visual_content(
         ref_img = render_page_to_image(ref, page_num, dpi)
         actual_img = render_page_to_image(actual, page_num, dpi)
 
-        ref_name = ref.stem
-        actual_name = actual.stem
-        output_path = (
-            output_dir / f"{ref_name}_vs_{actual_name}_page{page_num + 1}_diff.png"
-        )
+        output_path = None
+        if output_dir is not None:
+            ref_name = ref.stem
+            actual_name = actual.stem
+            output_path = (
+                output_dir / f"{ref_name}_vs_{actual_name}_page{page_num + 1}_diff.png"
+            )
 
         passed = compare_images(ref_img, actual_img, threshold, output_path)
 

diffpdf-0.3.0/tests/test_api.py

@@ -0,0 +1,16 @@
+from pathlib import Path
+
+import pytest
+
+from diffpdf import diffpdf
+
+TEST_ASSETS_DIR = Path(__file__).parent / "assets"
+
+
+def test_diffpdf():
+    with pytest.raises(SystemExit) as exc_info:
+        diffpdf(
+            TEST_ASSETS_DIR / "pass/identical-A.pdf",
+            TEST_ASSETS_DIR / "pass/identical-B.pdf",
+        )
+    assert exc_info.value.code == 0

diffpdf-0.3.0/tests/test_cli.py

@@ -0,0 +1,36 @@
+from pathlib import Path
+
+from click.testing import CliRunner
+
+from diffpdf.cli import cli
+
+TEST_ASSETS_DIR = Path(__file__).parent / "assets"
+
+
+def test_verbose_flag():
+    runner = CliRunner()
+    result = runner.invoke(
+        cli,
+        [
+            str(TEST_ASSETS_DIR / "pass/identical-A.pdf"),
+            str(TEST_ASSETS_DIR / "pass/identical-B.pdf"),
+            "-v",
+        ],
+    )
+    assert result.exit_code == 0
+    assert "INFO" in result.output
+    assert "DEBUG" not in result.output
+
+
+def test_double_verbose_flag():
+    runner = CliRunner()
+    result = runner.invoke(
+        cli,
+        [
+            str(TEST_ASSETS_DIR / "pass/identical-A.pdf"),
+            str(TEST_ASSETS_DIR / "pass/identical-B.pdf"),
+            "-vv",
+        ],
+    )
+    assert result.exit_code == 0
+    assert "DEBUG" in result.output

diffpdf-0.2.0/tests/test_cli.py → diffpdf-0.3.0/tests/test_comparators.py

@@ -8,35 +8,6 @@ from diffpdf.cli import cli
 TEST_ASSETS_DIR = Path(__file__).parent / "assets"
 
 
-def test_verbose_flag():
-    runner = CliRunner()
-    result = runner.invoke(
-        cli,
-        [
-            str(TEST_ASSETS_DIR / "pass/identical-A.pdf"),
-            str(TEST_ASSETS_DIR / "pass/identical-B.pdf"),
-            "-v",
-        ],
-    )
-    assert result.exit_code == 0
-    assert "INFO" in result.output
-    assert "DEBUG" not in result.output
-
-
-def test_double_verbose_flag():
-    runner = CliRunner()
-    result = runner.invoke(
-        cli,
-        [
-            str(TEST_ASSETS_DIR / "pass/identical-A.pdf"),
-            str(TEST_ASSETS_DIR / "pass/identical-B.pdf"),
-            "-vv",
-        ],
-    )
-    assert result.exit_code == 0
-    assert "DEBUG" in result.output
-
-
 @pytest.mark.parametrize(
     "ref_pdf_rel,actual_pdf_rel,expected_exit_code",
     [
@@ -54,7 +25,6 @@ def test_double_verbose_flag():
     ],
 )
 def test_comparators(ref_pdf_rel, actual_pdf_rel, expected_exit_code):
-    """Parametric integration test: CLI should exit with correct code for various PDF pairs."""
     runner = CliRunner()
 
     ref_pdf = str(TEST_ASSETS_DIR / ref_pdf_rel)
@@ -63,3 +33,16 @@ def test_comparators(ref_pdf_rel, actual_pdf_rel, expected_exit_code):
     result = runner.invoke(cli, [ref_pdf, actual_pdf])
 
     assert result.exit_code == expected_exit_code
+
+
+def test_comparators_with_output_dir():
+    runner = CliRunner()
+
+    with runner.isolated_filesystem():
+        ref_pdf = str(TEST_ASSETS_DIR / "fail/major-color-diff-A.pdf")
+        actual_pdf = str(TEST_ASSETS_DIR / "fail/major-color-diff-B.pdf")
+
+        result = runner.invoke(cli, [ref_pdf, actual_pdf, "--output-dir", "./diff"])
+
+        assert result.exit_code == 1
+        assert Path("./diff").exists()

diffpdf-0.2.0/src/diffpdf/__init__.py

@@ -1,15 +0,0 @@
-from importlib.metadata import version
-
-from .cli import cli
-
-__version__ = version("diffpdf")
-
-
-def main(args=None):  # pragma: no cover
-    if args is None:
-        cli()
-    else:
-        cli(args, standalone_mode=False)
-
-
-__all__ = ["main", "__version__"]