swarmauri_parser_fitzpdf 0.8.4.dev3

Fitz PDF Parser for Swarmauri.

Swarmauri Parser Fitz PDF

PDF-to-text parser for Swarmauri built on PyMuPDF (pymupdf). Extracts text from every page of a PDF and returns a Document object with the aggregated content and source metadata.

Features

• Opens PDFs via PyMuPDF and collects text per page.
• Emits a single Document with content containing the combined text and metadata['source'] holding the file path.
• Raises a clear error if the input is not a file path string; returns an empty list if PyMuPDF encounters parsing failures.

Prerequisites

• Python 3.10 or newer.
• PyMuPDF (pymupdf) along with system dependencies (X11 libraries on Linux, poppler on some distros). Install OS packages listed in PyMuPDF docs before pip installing if needed.
• Read access to the PDF files you plan to parse.

Installation

# pip
pip install swarmauri_parser_fitzpdf

# poetry
poetry add swarmauri_parser_fitzpdf

# uv (pyproject-based projects)
uv add swarmauri_parser_fitzpdf

Quickstart

from swarmauri_parser_fitzpdf import FitzPdfParser

parser = FitzPdfParser()
documents = parser.parse("reports/quarterly.pdf")

for doc in documents:
    print(doc.metadata["source"])
    print(doc.content[:500])

Handling Errors

from swarmauri_parser_fitzpdf import FitzPdfParser

parser = FitzPdfParser()
try:
    docs = parser.parse("missing.pdf")
    if not docs:
        print("Parsing failed or returned no content.")
except ValueError as exc:
    print(f"Bad input: {exc}")

Tips

• Pre-process PDFs (deskew, OCR) before parsing if they contain scanned pages without embedded text; PyMuPDF only extracts existing text objects.
• For multi-document pipelines, pair this parser with Swarmauri token-count measurements or summarizers to chunk large PDFs.
• Cache parsed output if the same PDF is accessed frequently—parsing large documents repeatedly is expensive.

Want to help?

If you want to contribute to swarmauri-sdk, read up on our guidelines for contributing that will help you get started.