PDF Extractor

A small FastAPI service that extracts Markdown-ready text from PDF files using PyMuPDF and PyMuPDF4LLM.

Features

• Accepts PDF uploads via POST /extract
• Converts PDF content into Markdown text
• Returns page count and extracted character count
• Includes health check endpoint
• Supports Docker and local development

Requirements

• Python 3.12+
• uv

Environment

Create a .env file at the project root to customize runtime settings.

Supported variables:

• PORT — service port (default: 5000)
• MAX_FILE_SIZE_MB — maximum upload size in megabytes (default: 50)
• CORS_ORIGINS — allowed CORS origins, comma-separated (default: *)

Local development

uv venv
source .venv/bin/activate
uv sync

Run the API locally:

uv run uvicorn main:app --host 0.0.0.0 --port ${PORT:-5000} --reload

API Endpoints

Health check

GET /health

Response:

{ "status": "ok" }

Extract PDF

POST /extract
Content-Type: multipart/form-data

file: <PDF file>

Successful response:

{
  "text": "...",
  "pages": 10,
  "characters": 12345
}

Errors:

• 422 if the uploaded file is missing, empty, or not a PDF
• 413 if the file is larger than MAX_FILE_SIZE_MB
• 500 for extraction or server errors

Docker

Build and run with Docker Compose:

docker compose up --build

The app will be available at http://localhost:5000 by default.

Notes

• Only application/pdf uploads are supported.
• Extraction uses pymupdf4llm.to_markdown() to generate Markdown-friendly output.
• The service is intentionally small and focused on PDF text extraction.