sigdetect 0.1.1__tar.gz → 0.2.0__tar.gz

{sigdetect-0.1.1 → sigdetect-0.2.0}/PKG-INFO

@@ -1,13 +1,12 @@
 Metadata-Version: 2.4
 Name: sigdetect
-Version: 0.1.1
+Version: 0.2.0
 Summary: Signature detection and role attribution for PDFs
 Author-email: BT Asmamaw <basmamaw@angeiongroup.com>
 License: MIT
 Requires-Python: >=3.9
 Description-Content-Type: text/markdown
 Requires-Dist: pypdf>=4.0.0
-Requires-Dist: pandas>=2.0
 Requires-Dist: rich>=13.0
 Requires-Dist: typer>=0.12
 Requires-Dist: pydantic>=2.5
@@ -102,6 +101,8 @@ sigdetect detect \
 - `--profile` selects tuned role logic:
   - `hipaa` → patient / representative / attorney
   - `retainer` → client / firm (prefers detecting two signatures)
+- `--recursive/--no-recursive` toggles whether `sigdetect detect` descends into subdirectories when hunting for PDFs (recursive by default).
+- `--crop-signatures` enables PNG crops for each detected widget (requires installing the optional `pymupdf` dependency). Use `--crop-dir` to override the destination and `--crop-dpi` to choose rendering quality.
 - If the executable is not on `PATH`, you can always fall back to `python -m sigdetect.cli ...`.
 
 ### EDA (quick aggregate stats)
@@ -135,7 +136,7 @@ result = detector.Detect(Path("/path/to/pdfs/example.pdf"))
 print(result.to_dict())
 ~~~
 
-`Detect(Path)` returns a **FileResult** dataclass; call `.to_dict()` for the JSON-friendly representation (see [Result schema](#result-schema)).
+`Detect(Path)` returns a **FileResult** dataclass; call `.to_dict()` for the JSON-friendly representation (see [Result schema](#result-schema)). Each signature entry now exposes `bounding_box` coordinates (PDF points, origin bottom-left). When PNG cropping is enabled, `crop_path` points at the generated image.
 
 ---
 
@@ -146,7 +147,17 @@ Import from `sigdetect.api` and get plain dicts out (JSON-ready),
 with no I/O side effects by default:
 
 ~~~python
-from sigdetect.api import DetectPdf, DetectMany, ScanDirectory, ToCsvRow, Version
+from pathlib import Path
+
+from sigdetect.api import (
+    CropSignatureImages,
+    DetectMany,
+    DetectPdf,
+    ScanDirectory,
+    ToCsvRow,
+    Version,
+    get_detector,
+)
 
 print("sigdetect", Version())
 
@@ -178,6 +189,15 @@ for res in ScanDirectory(
     # store in DB, print, etc.
     pass
 
+# 3) Crop PNG snippets for FileResult objects (requires PyMuPDF)
+detector = get_detector(pdfRoot="/path/to/pdfs", profileName="hipaa")
+file_result = detector.Detect(Path("/path/to/pdfs/example.pdf"))
+CropSignatureImages(
+    "/path/to/pdfs/example.pdf",
+    file_result,
+    outputDirectory="./signature_crops",
+    dpi=200,
+)
 ~~~
 
 
@@ -205,7 +225,10 @@ High-level summary (per file):
       "score": 5,
       "scores": { "field": 3, "page_label": 2 },
       "evidence": ["field:patient", "page_label:patient"],
-      "hint": "AcroSig:sig_patient"
+      "hint": "AcroSig:sig_patient",
+      "render_type": "typed",
+      "bounding_box": [10.0, 10.0, 150.0, 40.0],
+      "crop_path": "signature_crops/example/sig_01_patient.png"
     },
     {
       "page": null,
@@ -214,7 +237,10 @@ High-level summary (per file):
       "score": 6,
       "scores": { "page_label": 4, "general": 2 },
       "evidence": ["page_label:representative(parent/guardian)", "pseudo:true"],
-      "hint": "VendorOrAcroOnly"
+      "hint": "VendorOrAcroOnly",
+      "render_type": "unknown",
+      "bounding_box": null,
+      "crop_path": null
     }
   ]
 }
@@ -227,6 +253,8 @@ High-level summary (per file):
 - **`mixed`** means both `esign_found` and `scanned_pdf` are `true`.  
 - **`roles`** summarizes unique non-`unknown` roles across signatures.
 - In retainer profile, emitter prefers two signatures (client + firm), often on the same page.
+- **`signatures[].bounding_box`** reports the widget rectangle in PDF points (origin bottom-left).  
+- **`signatures[].crop_path`** is populated when PNG crops are generated (via CLI `--crop-signatures` or `CropSignatureImages`).
 
 ---
 
@@ -252,6 +280,9 @@ engine: pypdf2
 pseudo_signatures: true
 recurse_xobjects: true
 profile: retainer    # or: hipaa
+crop_signatures: false   # enable to write PNG crops (requires pymupdf)
+# crop_output_dir: ./signature_crops
+crop_image_dpi: 200
 ~~~
 
 YAML files can be customized or load at runtime (see CLI `--config`, if available, or import and pass patterns into engine).

{sigdetect-0.1.1 → sigdetect-0.2.0}/README.md

@@ -85,6 +85,8 @@ sigdetect detect \
 - `--profile` selects tuned role logic:
   - `hipaa` → patient / representative / attorney
   - `retainer` → client / firm (prefers detecting two signatures)
+- `--recursive/--no-recursive` toggles whether `sigdetect detect` descends into subdirectories when hunting for PDFs (recursive by default).
+- `--crop-signatures` enables PNG crops for each detected widget (requires installing the optional `pymupdf` dependency). Use `--crop-dir` to override the destination and `--crop-dpi` to choose rendering quality.
 - If the executable is not on `PATH`, you can always fall back to `python -m sigdetect.cli ...`.
 
 ### EDA (quick aggregate stats)
@@ -118,7 +120,7 @@ result = detector.Detect(Path("/path/to/pdfs/example.pdf"))
 print(result.to_dict())
 ~~~
 
-`Detect(Path)` returns a **FileResult** dataclass; call `.to_dict()` for the JSON-friendly representation (see [Result schema](#result-schema)).
+`Detect(Path)` returns a **FileResult** dataclass; call `.to_dict()` for the JSON-friendly representation (see [Result schema](#result-schema)). Each signature entry now exposes `bounding_box` coordinates (PDF points, origin bottom-left). When PNG cropping is enabled, `crop_path` points at the generated image.
 
 ---
 
@@ -129,7 +131,17 @@ Import from `sigdetect.api` and get plain dicts out (JSON-ready),
 with no I/O side effects by default:
 
 ~~~python
-from sigdetect.api import DetectPdf, DetectMany, ScanDirectory, ToCsvRow, Version
+from pathlib import Path
+
+from sigdetect.api import (
+    CropSignatureImages,
+    DetectMany,
+    DetectPdf,
+    ScanDirectory,
+    ToCsvRow,
+    Version,
+    get_detector,
+)
 
 print("sigdetect", Version())
 
@@ -161,6 +173,15 @@ for res in ScanDirectory(
     # store in DB, print, etc.
     pass
 
+# 3) Crop PNG snippets for FileResult objects (requires PyMuPDF)
+detector = get_detector(pdfRoot="/path/to/pdfs", profileName="hipaa")
+file_result = detector.Detect(Path("/path/to/pdfs/example.pdf"))
+CropSignatureImages(
+    "/path/to/pdfs/example.pdf",
+    file_result,
+    outputDirectory="./signature_crops",
+    dpi=200,
+)
 ~~~
 
 
@@ -188,7 +209,10 @@ High-level summary (per file):
       "score": 5,
       "scores": { "field": 3, "page_label": 2 },
       "evidence": ["field:patient", "page_label:patient"],
-      "hint": "AcroSig:sig_patient"
+      "hint": "AcroSig:sig_patient",
+      "render_type": "typed",
+      "bounding_box": [10.0, 10.0, 150.0, 40.0],
+      "crop_path": "signature_crops/example/sig_01_patient.png"
     },
     {
       "page": null,
@@ -197,7 +221,10 @@ High-level summary (per file):
       "score": 6,
       "scores": { "page_label": 4, "general": 2 },
       "evidence": ["page_label:representative(parent/guardian)", "pseudo:true"],
-      "hint": "VendorOrAcroOnly"
+      "hint": "VendorOrAcroOnly",
+      "render_type": "unknown",
+      "bounding_box": null,
+      "crop_path": null
     }
   ]
 }
@@ -210,6 +237,8 @@ High-level summary (per file):
 - **`mixed`** means both `esign_found` and `scanned_pdf` are `true`.  
 - **`roles`** summarizes unique non-`unknown` roles across signatures.
 - In retainer profile, emitter prefers two signatures (client + firm), often on the same page.
+- **`signatures[].bounding_box`** reports the widget rectangle in PDF points (origin bottom-left).  
+- **`signatures[].crop_path`** is populated when PNG crops are generated (via CLI `--crop-signatures` or `CropSignatureImages`).
 
 ---
 
@@ -235,6 +264,9 @@ engine: pypdf2
 pseudo_signatures: true
 recurse_xobjects: true
 profile: retainer    # or: hipaa
+crop_signatures: false   # enable to write PNG crops (requires pymupdf)
+# crop_output_dir: ./signature_crops
+crop_image_dpi: 200
 ~~~
 
 YAML files can be customized or load at runtime (see CLI `--config`, if available, or import and pass patterns into engine).

{sigdetect-0.1.1 → sigdetect-0.2.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "sigdetect"
-version = "0.1.1"
+version = "0.2.0"
 description = "Signature detection and role attribution for PDFs"
 readme = "README.md"
 authors = [{ name = "BT Asmamaw", email = "basmamaw@angeiongroup.com" }]
@@ -12,7 +12,6 @@ license = { text = "MIT" }
 requires-python = ">=3.9"
 dependencies = [
   "pypdf>=4.0.0",
-  "pandas>=2.0",
   "rich>=13.0",
   "typer>=0.12",
   "pydantic>=2.5",

sigdetect-0.2.0/src/sigdetect/api.py

@@ -0,0 +1,261 @@
+"""Public helpers for programmatic use of the signature detection engine."""
+
+from __future__ import annotations
+
+from contextlib import contextmanager
+from pathlib import Path
+from typing import Any, Generator, Iterable, Iterator, Literal
+
+from sigdetect.config import DetectConfiguration
+from sigdetect.detector import BuildDetector, Detector, FileResult, Signature
+
+EngineName = Literal["pypdf2", "pypdf", "pymupdf"]
+ProfileName = Literal["hipaa", "retainer"]
+
+
+def DetectPdf(
+    pdfPath: str | Path,
+    *,
+    profileName: ProfileName = "hipaa",
+    engineName: EngineName = "pypdf2",
+    includePseudoSignatures: bool = True,
+    recurseXObjects: bool = True,
+    detector: Detector | None = None,
+) -> dict[str, Any]:
+    """Detect signature evidence and assign roles for a single PDF."""
+
+    resolvedPath = Path(pdfPath)
+    activeDetector = detector or get_detector(
+        pdfRoot=resolvedPath.parent,
+        profileName=profileName,
+        engineName=engineName,
+        includePseudoSignatures=includePseudoSignatures,
+        recurseXObjects=recurseXObjects,
+        outputDirectory=None,
+    )
+
+    result = activeDetector.Detect(resolvedPath)
+    return _ToPlainDictionary(result)
+
+
+def get_detector(
+    *,
+    pdfRoot: str | Path | None = None,
+    profileName: ProfileName = "hipaa",
+    engineName: EngineName = "pypdf2",
+    includePseudoSignatures: bool = True,
+    recurseXObjects: bool = True,
+    outputDirectory: str | Path | None = None,
+) -> Detector:
+    """Return a reusable detector instance configured with the supplied options."""
+
+    configuration = DetectConfiguration(
+        PdfRoot=Path(pdfRoot) if pdfRoot is not None else Path.cwd(),
+        OutputDirectory=Path(outputDirectory) if outputDirectory is not None else None,
+        Engine=engineName,
+        PseudoSignatures=includePseudoSignatures,
+        RecurseXObjects=recurseXObjects,
+        Profile=profileName,
+    )
+    return BuildDetector(configuration)
+
+
+def _ToPlainDictionary(candidate: Any) -> dict[str, Any]:
+    """Convert pydantic/dataclass instances to plain dictionaries."""
+
+    if hasattr(candidate, "to_dict"):
+        return candidate.to_dict()
+    if hasattr(candidate, "model_dump"):
+        return candidate.model_dump()  # type: ignore[attr-defined]
+    if hasattr(candidate, "dict"):
+        return candidate.dict()  # type: ignore[attr-defined]
+    try:
+        from dataclasses import asdict, is_dataclass
+
+        if is_dataclass(candidate):
+            return asdict(candidate)
+    except Exception:
+        pass
+    if isinstance(candidate, dict):
+        return {key: _ToPlainValue(candidate[key]) for key in candidate}
+    raise TypeError(f"Unsupported result type: {type(candidate)!r}")
+
+
+def _ToPlainValue(value: Any) -> Any:
+    """Best effort conversion for nested structures."""
+
+    if hasattr(value, "to_dict"):
+        return value.to_dict()
+    if hasattr(value, "model_dump") or hasattr(value, "dict"):
+        return _ToPlainDictionary(value)
+    try:
+        from dataclasses import asdict, is_dataclass
+
+        if is_dataclass(value):
+            return asdict(value)
+    except Exception:
+        pass
+    if isinstance(value, list):
+        return [_ToPlainValue(item) for item in value]
+    if isinstance(value, tuple):
+        return tuple(_ToPlainValue(item) for item in value)
+    if isinstance(value, dict):
+        return {key: _ToPlainValue(result) for key, result in value.items()}
+    return value
+
+
+def DetectMany(
+    pdfPaths: Iterable[str | Path],
+    *,
+    detector: Detector | None = None,
+    **kwargs: Any,
+) -> Iterator[dict[str, Any]]:
+    """Yield :func:`DetectPdf` results for each path in ``pdfPaths``."""
+
+    if detector is not None:
+        for pdfPath in pdfPaths:
+            yield _DetectWithDetector(detector, pdfPath)
+        return
+
+    for pdfPath in pdfPaths:
+        yield DetectPdf(pdfPath, **kwargs)
+
+
+def ScanDirectory(
+    pdfRoot: str | Path,
+    *,
+    globPattern: str = "**/*.pdf",
+    detector: Detector | None = None,
+    **kwargs: Any,
+) -> Iterator[dict[str, Any]]:
+    """Walk ``pdfRoot`` and yield detection output for every matching PDF."""
+
+    rootDirectory = Path(pdfRoot)
+    if globPattern == "**/*.pdf":
+        iterator = (path for path in rootDirectory.rglob("*") if path.is_file())
+    else:
+        iterator = (
+            rootDirectory.rglob(globPattern.replace("**/", "", 1))
+            if globPattern.startswith("**/")
+            else rootDirectory.glob(globPattern)
+        )
+
+    for pdfPath in iterator:
+        if pdfPath.is_file() and pdfPath.suffix.lower() == ".pdf":
+            yield DetectPdf(pdfPath, detector=detector, **kwargs)
+
+
+def ToCsvRow(result: dict[str, Any]) -> dict[str, Any]:
+    """Return a curated subset of keys suitable for CSV export."""
+
+    return {
+        "file": result.get("file"),
+        "size_kb": result.get("size_kb"),
+        "pages": result.get("pages"),
+        "esign_found": result.get("esign_found"),
+        "scanned_pdf": result.get("scanned_pdf"),
+        "mixed": result.get("mixed"),
+        "sig_count": result.get("sig_count"),
+        "sig_pages": result.get("sig_pages"),
+        "roles": result.get("roles"),
+        "hints": result.get("hints"),
+    }
+
+
+def Version() -> str:
+    """Expose the installed package version without importing the CLI stack."""
+
+    try:
+        from importlib.metadata import version as resolveVersion
+
+        return resolveVersion("sigdetect")
+    except Exception:
+        return "0.0.0-dev"
+
+
+def _DetectWithDetector(detector: Detector, pdfPath: str | Path) -> dict[str, Any]:
+    """Helper that runs ``detector`` and returns the plain dictionary result."""
+
+    resolvedPath = Path(pdfPath)
+    return _ToPlainDictionary(detector.Detect(resolvedPath))
+
+
+@contextmanager
+def detector_context(**kwargs: Any) -> Generator[Detector, None, None]:
+    """Context manager wrapper around :func:`get_detector`."""
+
+    detector = get_detector(**kwargs)
+    try:
+        yield detector
+    finally:
+        pass
+
+
+def CropSignatureImages(
+    pdfPath: str | Path,
+    fileResult: FileResult | dict[str, Any],
+    *,
+    outputDirectory: str | Path,
+    dpi: int = 200,
+) -> list[Path]:
+    """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.
+    """
+
+    from sigdetect.cropping import crop_signatures
+
+    file_result_obj, original_dict = _CoerceFileResult(fileResult)
+    paths = crop_signatures(
+        pdf_path=Path(pdfPath),
+        file_result=file_result_obj,
+        output_dir=Path(outputDirectory),
+        dpi=dpi,
+    )
+    if original_dict is not None:
+        original_dict.clear()
+        original_dict.update(file_result_obj.to_dict())
+    return paths
+
+
+def _CoerceFileResult(
+    candidate: FileResult | dict[str, Any]
+) -> tuple[FileResult, dict[str, Any] | None]:
+    if isinstance(candidate, FileResult):
+        return candidate, None
+    if not isinstance(candidate, dict):
+        raise TypeError("fileResult must be FileResult or dict")
+
+    signatures: list[Signature] = []
+    for entry in candidate.get("signatures") or []:
+        bbox = entry.get("bounding_box")
+        signatures.append(
+            Signature(
+                Page=entry.get("page"),
+                FieldName=str(entry.get("field_name") or ""),
+                Role=str(entry.get("role") or "unknown"),
+                Score=int(entry.get("score") or 0),
+                Scores=dict(entry.get("scores") or {}),
+                Evidence=list(entry.get("evidence") or []),
+                Hint=str(entry.get("hint") or ""),
+                RenderType=str(entry.get("render_type") or "unknown"),
+                BoundingBox=tuple(bbox) if bbox else None,
+                CropPath=entry.get("crop_path"),
+            )
+        )
+
+    file_result = FileResult(
+        File=str(candidate.get("file") or ""),
+        SizeKilobytes=candidate.get("size_kb"),
+        PageCount=int(candidate.get("pages") or 0),
+        ElectronicSignatureFound=bool(candidate.get("esign_found")),
+        ScannedPdf=candidate.get("scanned_pdf"),
+        MixedContent=candidate.get("mixed"),
+        SignatureCount=int(candidate.get("sig_count") or len(signatures)),
+        SignaturePages=str(candidate.get("sig_pages") or ""),
+        Roles=str(candidate.get("roles") or "unknown"),
+        Hints=str(candidate.get("hints") or ""),
+        Signatures=signatures,
+    )
+    return file_result, candidate