edgeparse 0.2.0__cp313-cp313-win_amd64.whl

edgeparse/__init__.py

@@ -0,0 +1,107 @@
+"""edgeparse — High-performance PDF extraction (Rust engine)."""
+
+from __future__ import annotations
+
+from pathlib import Path
+from typing import List, Optional, Tuple, Union
+
+from edgeparse._edgeparse import (
+    convert as _convert,
+    convert_file as _convert_file,
+    version as _version,
+)
+
+PathLike = Union[str, Path]
+
+
+def convert(
+    input_path: PathLike,
+    *,
+    format: str = "markdown",
+    pages: Optional[str] = None,
+    password: Optional[str] = None,
+    reading_order: str = "xycut",
+    table_method: str = "default",
+    image_output: str = "off",
+) -> str:
+    """Convert a PDF file and return the extracted content as a string.
+
+    Parameters
+    ----------
+    input_path:
+        Path to the PDF file.
+    format:
+        Output format. Valid values: ``markdown``, ``json``, ``html``, ``text``.
+        Default: ``markdown``.
+    pages:
+        Optional page range string, e.g. ``"1,3,5-7"``.
+    password:
+        Optional password for encrypted PDFs.
+    reading_order:
+        Reading order algorithm. ``"xycut"`` (default) or ``"off"``.
+    table_method:
+        Table detection method. ``"default"`` or ``"cluster"``.
+    image_output:
+        Image output mode. ``"off"`` (default), ``"embedded"``, or ``"external"``.
+
+    Returns
+    -------
+    str
+        The extracted content in the requested format.
+    """
+    return _convert(
+        str(input_path),
+        format=format,
+        pages=pages,
+        password=password,
+        reading_order=reading_order,
+        table_method=table_method,
+        image_output=image_output,
+    )
+
+
+def convert_file(
+    input_path: PathLike,
+    output_dir: PathLike = "output",
+    *,
+    format: str = "markdown",
+    pages: Optional[str] = None,
+    password: Optional[str] = None,
+) -> str:
+    """Convert a PDF file and write the output to a file.
+
+    Parameters
+    ----------
+    input_path:
+        Path to the PDF file.
+    output_dir:
+        Directory to write output files. Created if it does not exist.
+    format:
+        Output format. Valid values: ``markdown``, ``json``, ``html``, ``text``.
+        Default: ``markdown``.
+    pages:
+        Optional page range string, e.g. ``"1,3,5-7"``.
+    password:
+        Optional password for encrypted PDFs.
+
+    Returns
+    -------
+    str
+        Path to the created output file.
+    """
+    return _convert_file(
+        str(input_path),
+        str(output_dir),
+        format=format,
+        pages=pages,
+        password=password,
+    )
+
+
+def version() -> str:
+    """Return the edgeparse version string."""
+    return _version()
+
+
+__all__ = ["convert", "convert_file", "version"]
+__version__ = _version()

edgeparse/_types.py

@@ -0,0 +1,18 @@
+"""Type definitions for edgeparse."""
+
+from __future__ import annotations
+
+from typing import Optional
+
+
+# Valid output format strings
+FORMATS = ("markdown", "json", "html", "text")
+
+# Valid reading order algorithms
+READING_ORDERS = ("xycut", "off")
+
+# Valid table detection methods
+TABLE_METHODS = ("default", "cluster")
+
+# Valid image output modes
+IMAGE_OUTPUTS = ("off", "embedded", "external")

edgeparse/cli.py

@@ -0,0 +1,50 @@
+"""CLI entry point — mirrors the Rust CLI but invoked via ``edgeparse`` console script."""
+
+from __future__ import annotations
+
+import argparse
+import sys
+
+from . import convert_file
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(
+        prog="edgeparse",
+        description="Convert PDF files to Markdown, JSON, HTML, or plain text.",
+    )
+    parser.add_argument("input", nargs="+", help="PDF file(s) to convert")
+    parser.add_argument(
+        "-o", "--output-dir", default="output", help="Output directory (default: output)"
+    )
+    parser.add_argument(
+        "-f",
+        "--format",
+        default="markdown",
+        help="Output format: markdown, json, html, text (default: markdown)",
+    )
+    parser.add_argument("--pages", help="Page range, e.g. 1,3,5-7")
+    parser.add_argument("-p", "--password", help="Password for encrypted PDFs")
+    args = parser.parse_args()
+
+    has_errors = False
+    for input_path in args.input:
+        try:
+            out = convert_file(
+                input_path,
+                output_dir=args.output_dir,
+                format=args.format,
+                pages=args.pages,
+                password=args.password,
+            )
+            print(out)
+        except Exception as e:
+            print(f"Error processing {input_path}: {e}", file=sys.stderr)
+            has_errors = True
+
+    if has_errors:
+        sys.exit(1)
+
+
+if __name__ == "__main__":
+    main()

edgeparse-0.2.0.dist-info/METADATA

@@ -0,0 +1,72 @@
+Metadata-Version: 2.4
+Name: edgeparse
+Version: 0.2.0
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Rust
+Classifier: Topic :: Software Development :: Libraries
+Classifier: Topic :: Text Processing
+Summary: High-performance PDF-to-structured-data extraction — Rust engine, Python interface
+Keywords: pdf,extraction,markdown,rag,llm
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown; charset=UTF-8; variant=GFM
+Project-URL: Documentation, https://github.com/raphaelmansuy/edgeparse/tree/main/docs
+Project-URL: Homepage, https://github.com/raphaelmansuy/edgeparse
+Project-URL: Repository, https://github.com/raphaelmansuy/edgeparse
+
+# edgeparse
+
+High-performance PDF-to-structured-data extraction for Python — powered by a Rust engine via PyO3.
+
+## Install
+
+```bash
+pip install edgeparse
+```
+
+Pre-built wheels are available for **macOS**, **Linux** (x86_64, arm64), and **Windows** (x64).
+No system dependencies or compilation required.
+
+## Quick start
+
+```python
+import edgeparse
+
+# Convert a PDF to Markdown
+result = edgeparse.convert("document.pdf")
+print(result.markdown)
+
+# Convert with options
+result = edgeparse.convert(
+    "document.pdf",
+    format="markdown",      # "markdown" | "json" | "html"
+    extract_images=False,
+    page_range=None,        # None = all pages, or [0, 5] for pages 1–6
+)
+```
+
+## CLI
+
+```bash
+edgeparse document.pdf                     # → Markdown on stdout
+edgeparse document.pdf --format json       # → JSON
+edgeparse /path/to/dir/ --output-dir out/  # batch convert
+```
+
+## Performance
+
+`edgeparse` consistently leads open benchmarks for PDF-to-Markdown extraction quality across 200-document test suites.
+
+## Links
+
+- [GitHub](https://github.com/raphaelmansuy/edgeparse)
+- [crates.io (Rust)](https://crates.io/crates/edgeparse-core)
+- [npm (edgeparse)](https://www.npmjs.com/package/edgeparse)
+

edgeparse-0.2.0.dist-info/RECORD

@@ -0,0 +1,10 @@
+edgeparse/__init__.py,sha256=NlSYQTGNeIpGNMINXHWkCsynhbUNEPvZMyerLKz66vc,2766
+edgeparse/_edgeparse.cp313-win_amd64.pyd,sha256=maAo-lXTzjg2GsUBi_hTIm4a65viJhLs1XoerNKXJsA,4347392
+edgeparse/_types.py,sha256=aYnqZW0mYfep8vIItF2cqD2_zNaGveQeXmBM8mFHyB4,416
+edgeparse/cli.py,sha256=LrE7xftA0_4KbkXmMcAz0BxX9JztzGAp6D6yMP3SIsk,1491
+edgeparse/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+edgeparse-0.2.0.dist-info/METADATA,sha256=30OXpQcKMnCTw4QsfMQzQa-yQJVVqrH5qsA3X6rMOI4,2344
+edgeparse-0.2.0.dist-info/WHEEL,sha256=MeHAddfhFE43Hox4HMNv5ysU_Px_FPIOurt9DcMoh1g,97
+edgeparse-0.2.0.dist-info/entry_points.txt,sha256=XyI2HhMiCCI5ArQFB0nbp0Faewr1QKsDO7suWnXuU8A,47
+edgeparse-0.2.0.dist-info/sboms/edgeparse-python.cyclonedx.json,sha256=suJIStkgBxK1hwkH1DKvffn2QEmakIh-gChhe63nK5A,203706
+edgeparse-0.2.0.dist-info/RECORD,,

edgeparse-0.2.0.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: maturin (1.12.6)
+Root-Is-Purelib: false
+Tag: cp313-cp313-win_amd64

edgeparse-0.2.0.dist-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+edgeparse=edgeparse.cli:main