swarmauri_parser_pypdf2 0.11.0.dev1

Page-by-page PDF text parser for Swarmauri using PyPDF2 with path and bytes input support.

Swarmauri Parser PyPDF2

swarmauri_parser_pypdf2 is the Swarmauri PDF text parser for page-by-page extraction using PyPDF2. It converts a PDF file path or PDF bytes into a list of Swarmauri Document objects, preserving page numbers and source metadata for downstream chunking, indexing, and agent workflows.

Why Use Swarmauri Parser PyPDF2

• Extract embedded PDF text without introducing OCR when the document already contains readable text objects.
• Preserve page boundaries by returning one Document per page.
• Accept either on-disk PDFs or PDF bytes already loaded in memory.
• Keep document ingestion aligned with the Swarmauri parser interface used across other loaders and converters.

FAQ

What does this parser return?
A list of Swarmauri Document objects, usually one for each page that contains extractable text.

Can it parse PDF bytes instead of a file path?
Yes. parse() accepts either a string path or bytes.

Does it perform OCR on scanned PDFs?
No. PyPDF2 extracts embedded text. Scanned PDFs without text objects should be handled with OCR first.

What metadata is attached to each page?
Each returned document includes page_number and source.

Features

• Page-by-page PDF text extraction through PyPDF2.
• Supports parsing from local file paths and raw PDF bytes.
• Returns Swarmauri Document objects with content, page_number, and source.
• Clean fit for ingestion pipelines, retrieval systems, and document analysis workflows.
• Supports Python 3.10, 3.11, 3.12, 3.13, and 3.14.

Installation

uv add swarmauri_parser_pypdf2

pip install swarmauri_parser_pypdf2

Usage

from swarmauri_parser_pypdf2 import PyPDF2Parser

parser = PyPDF2Parser()
documents = parser.parse("manuals/device.pdf")

for document in documents:
    print(document.metadata["page_number"])
    print(document.content[:160])

Examples

Parse a PDF from bytes

from pathlib import Path
from swarmauri_parser_pypdf2 import PyPDF2Parser

pdf_bytes = Path("reports/statement.pdf").read_bytes()
parser = PyPDF2Parser()
pages = parser.parse(pdf_bytes)

for page in pages:
    print(page.metadata)

Send parsed pages to downstream chunking

from swarmauri_parser_pypdf2 import PyPDF2Parser

parser = PyPDF2Parser()
pages = parser.parse("contracts/master-service-agreement.pdf")

for page in pages:
    if page.content.strip():
        print(page.metadata["page_number"], len(page.content))

Related Packages

• swarmauri_parser_fitzpdf
• swarmauri_parser_pypdftk
• swarmauri_ocr_pytesseract
• swarmauri_parser_pypdf2

Swarmauri Foundations

• swarmauri
• swarmauri_core
• swarmauri_base
• swarmauri_standard

More Documentation

• PyPDF2 text extraction guide
• PyPDF2 documentation
• PyPDF2 on PyPI

Best Practices

• Use this parser when the PDF already contains embedded text.
• Use OCR first for scan-only PDFs, then parse or post-process the OCR output.
• Validate encrypted, malformed, or image-only PDFs earlier in the ingestion pipeline so downstream processing can route them correctly.
• Keep in mind that page text extraction quality depends on the PDF's internal structure and reading order.

License

This project is licensed under the Apache-2.0 License.