sigdetect 0.3.0__tar.gz → 0.3.1__tar.gz

{sigdetect-0.3.0 → sigdetect-0.3.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sigdetect
-Version: 0.3.0
+Version: 0.3.1
 Summary: Signature detection and role attribution for PDFs
 Author-email: BT Asmamaw <basmamaw@angeiongroup.com>
 License: MIT
@@ -198,6 +198,7 @@ crops = CropSignatureImages(
     outputDirectory="./signature_crops",
     dpi=200,
     returnBytes=True,  # also returns in-memory PNG bytes for each crop
+    # saveToDisk=False,  # optional: skip writing PNGs to disk
 )
 
 first_crop = crops[0]
@@ -206,6 +207,7 @@ print(first_crop.path, len(first_crop.image_bytes))
 
 When ``returnBytes=True`` the helper returns ``SignatureCrop`` objects containing the saved path,
 PNG bytes, and the originating signature metadata.
+Pass ``saveToDisk=False`` if you only want in-memory PNG bytes (no files on disk or ``crop_path`` updates).
 
 
 ## Result schema

{sigdetect-0.3.0 → sigdetect-0.3.1}/README.md

@@ -182,6 +182,7 @@ crops = CropSignatureImages(
     outputDirectory="./signature_crops",
     dpi=200,
     returnBytes=True,  # also returns in-memory PNG bytes for each crop
+    # saveToDisk=False,  # optional: skip writing PNGs to disk
 )
 
 first_crop = crops[0]
@@ -190,6 +191,7 @@ print(first_crop.path, len(first_crop.image_bytes))
 
 When ``returnBytes=True`` the helper returns ``SignatureCrop`` objects containing the saved path,
 PNG bytes, and the originating signature metadata.
+Pass ``saveToDisk=False`` if you only want in-memory PNG bytes (no files on disk or ``crop_path`` updates).
 
 
 ## Result schema

{sigdetect-0.3.0 → sigdetect-0.3.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "sigdetect"
-version = "0.3.0"
+version = "0.3.1"
 description = "Signature detection and role attribution for PDFs"
 readme = "README.md"
 authors = [{ name = "BT Asmamaw", email = "basmamaw@angeiongroup.com" }]

{sigdetect-0.3.0 → sigdetect-0.3.1}/src/sigdetect/api.py

@@ -200,6 +200,7 @@ def CropSignatureImages(
     outputDirectory: str | Path,
     dpi: int = 200,
     returnBytes: Literal[False] = False,
+    saveToDisk: bool = True,
 ) -> list[Path]: ...
 
 
@@ -211,6 +212,7 @@ def CropSignatureImages(
     outputDirectory: str | Path,
     dpi: int,
     returnBytes: Literal[True],
+    saveToDisk: bool,
 ) -> list[SignatureCrop]: ...
 
 
@@ -221,12 +223,14 @@ def CropSignatureImages(
     outputDirectory: str | Path,
     dpi: int = 200,
     returnBytes: bool = False,
+    saveToDisk: bool = True,
 ) -> list[Path] | list[SignatureCrop]:
     """Crop detected signature regions to PNG files.
 
     Accepts either a :class:`FileResult` instance or the ``dict`` returned by
     :func:`DetectPdf`. Requires the optional ``pymupdf`` dependency.
-    Set ``returnBytes=True`` to also receive in-memory PNG bytes for each crop.
+    Set ``returnBytes=True`` to also receive in-memory PNG bytes for each crop. Set
+    ``saveToDisk=False`` to skip writing PNG files while still returning in-memory data.
     """
 
     from sigdetect.cropping import crop_signatures
@@ -238,6 +242,7 @@ def CropSignatureImages(
         output_dir=Path(outputDirectory),
         dpi=dpi,
         return_bytes=returnBytes,
+        save_files=saveToDisk,
     )
     if original_dict is not None:
         original_dict.clear()

{sigdetect-0.3.0 → sigdetect-0.3.1}/src/sigdetect/cropping.py

@@ -28,6 +28,7 @@ class SignatureCrop:
     path: Path
     image_bytes: bytes
     signature: Signature
+    saved_to_disk: bool = True
 
 
 @overload
@@ -62,27 +63,32 @@ def crop_signatures(
     dpi: int = 200,
     logger: logging.Logger | None = None,
     return_bytes: bool = False,
+    save_files: bool = True,
 ) -> list[Path] | list[SignatureCrop]:
     """Render each signature bounding box to a PNG image using PyMuPDF.
 
     Set ``return_bytes=True`` to collect in-memory PNG bytes for each crop while also writing
-    the files to ``output_dir``.
+    the files to ``output_dir``. Set ``save_files=False`` to skip writing PNGs to disk.
     """
 
     if fitz is None:  # pragma: no cover - exercised when dependency absent
         raise SignatureCroppingUnavailable(
             "PyMuPDF is required for PNG crops. Install 'pymupdf' or 'sigdetect[pymupdf]'."
         )
+    if not save_files and not return_bytes:
+        raise ValueError("At least one of save_files or return_bytes must be True")
 
     pdf_path = Path(pdf_path)
     output_dir = Path(output_dir)
-    output_dir.mkdir(parents=True, exist_ok=True)
+    if save_files:
+        output_dir.mkdir(parents=True, exist_ok=True)
     generated_paths: list[Path] = []
     generated_crops: list[SignatureCrop] = []
 
     with fitz.open(pdf_path) as document:  # type: ignore[attr-defined]
         per_document_dir = output_dir / pdf_path.stem
-        per_document_dir.mkdir(parents=True, exist_ok=True)
+        if save_files:
+            per_document_dir.mkdir(parents=True, exist_ok=True)
         scale = dpi / 72.0
         matrix = fitz.Matrix(scale, scale)
 
@@ -113,7 +119,8 @@ def crop_signatures(
             try:
                 image_bytes: bytes | None = None
                 pixmap = page.get_pixmap(matrix=matrix, clip=clip, alpha=False)
-                pixmap.save(destination)
+                if save_files:
+                    pixmap.save(destination)
                 if return_bytes:
                     image_bytes = pixmap.tobytes("png")
             except Exception as exc:  # pragma: no cover - defensive
@@ -129,8 +136,9 @@ def crop_signatures(
                     )
                 continue
 
-            signature.CropPath = str(destination)
-            generated_paths.append(destination)
+            if save_files:
+                signature.CropPath = str(destination)
+                generated_paths.append(destination)
             if return_bytes:
                 if image_bytes is None:  # pragma: no cover - defensive
                     continue
@@ -139,6 +147,7 @@ def crop_signatures(
                         path=destination,
                         image_bytes=image_bytes,
                         signature=signature,
+                        saved_to_disk=save_files,
                     )
                 )
 

{sigdetect-0.3.0 → sigdetect-0.3.1}/src/sigdetect.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sigdetect
-Version: 0.3.0
+Version: 0.3.1
 Summary: Signature detection and role attribution for PDFs
 Author-email: BT Asmamaw <basmamaw@angeiongroup.com>
 License: MIT
@@ -198,6 +198,7 @@ crops = CropSignatureImages(
     outputDirectory="./signature_crops",
     dpi=200,
     returnBytes=True,  # also returns in-memory PNG bytes for each crop
+    # saveToDisk=False,  # optional: skip writing PNGs to disk
 )
 
 first_crop = crops[0]
@@ -206,6 +207,7 @@ print(first_crop.path, len(first_crop.image_bytes))
 
 When ``returnBytes=True`` the helper returns ``SignatureCrop`` objects containing the saved path,
 PNG bytes, and the originating signature metadata.
+Pass ``saveToDisk=False`` if you only want in-memory PNG bytes (no files on disk or ``crop_path`` updates).
 
 
 ## Result schema

{sigdetect-0.3.0 → sigdetect-0.3.1}/tests/test_cropping.py

@@ -91,6 +91,29 @@ def test_crop_signature_images_returns_bytes(tmp_path: Path) -> None:
     assert result_dict["signatures"][0]["crop_path"] is not None
 
 
+def test_crop_signature_images_can_skip_disk(tmp_path: Path) -> None:
+    pdf_path = tmp_path / "doc.pdf"
+    _pdf_with_signature(pdf_path)
+
+    result_dict = DetectPdf(pdf_path, engineName="pymupdf")
+    out_dir = tmp_path / "dict_byte_crops_no_disk"
+    crops = CropSignatureImages(
+        pdf_path,
+        result_dict,
+        outputDirectory=out_dir,
+        returnBytes=True,
+        saveToDisk=False,
+    )
+
+    assert crops
+    first_crop = crops[0]
+    assert isinstance(first_crop, SignatureCrop)
+    assert first_crop.image_bytes
+    assert first_crop.saved_to_disk is False
+    assert not first_crop.path.exists()
+    assert result_dict["signatures"][0]["crop_path"] is None
+
+
 def test_crop_signatures_returns_bytes(tmp_path: Path) -> None:
     pdf_path = tmp_path / "doc.pdf"
     _pdf_with_signature(pdf_path)
@@ -111,3 +134,21 @@ def test_crop_signatures_returns_bytes(tmp_path: Path) -> None:
     assert isinstance(crops[0], SignatureCrop)
     assert crops[0].path.exists()
     assert crops[0].image_bytes
+
+
+def test_crop_signatures_requires_save_or_bytes(tmp_path: Path) -> None:
+    pdf_path = tmp_path / "doc.pdf"
+    _pdf_with_signature(pdf_path)
+
+    cfg = DetectConfiguration(pdf_root=tmp_path, out_dir=tmp_path, engine="pypdf2")
+    result = PyPDF2Detector(cfg).Detect(pdf_path)
+
+    with pytest.raises(ValueError):
+        crop_signatures(
+            pdf_path,
+            result,
+            output_dir=tmp_path / "unused",
+            dpi=120,
+            save_files=False,
+            return_bytes=False,
+        )