bib-extractor 0.1.0

Extract DOIs or titles from PDF papers and generate a BibTeX bibliography

Bib Extractor

A tiny, pip‑installable Python utility that scans a folder of PDF papers, extracts their DOI (or a title fallback), and produces a JSON file that can be turned into a BibTeX bibliography.

---

✨ Features

• Works on any folder of PDFs.
• Uses pdftotext (Poppler) to read the first two pages of each PDF.
• Detects DOI strings with a robust regular expression.
• Falls back to the first few non‑empty lines as a title when no DOI is found.
• Command‑line interface with --dir and --output options.
• Zero external Python dependencies (standard library only).

---

📦 Installation

From PyPI (recommended)

pip install bib-extractor

From source

1. Install Poppler – pdftotext is required.
   • Windows: download from https://github.com/oschwartz10612/poppler-windows/releases and add the bin folder to your PATH.
   • macOS: brew install poppler
   • Linux: sudo apt-get install poppler-utils
2. Clone the repository

git clone https://github.com/msrtarit/bib_extractor.git
cd bib_extractor

3. (Optional) Create a virtual environment and install the package in editable mode:

python -m venv .venv
.venv\\Scripts\\activate   # Windows
# or source .venv/bin/activate on Unix
pip install -e .

---

🚀 Usage

Extract DOIs / titles

# Using the installed command (if installed via pip)
bib-extractor --dir path/to/papers --output paper_info.json

# Or run the script directly from the source checkout
python extract_bib_info.py --dir path/to/papers --output paper_info.json

• --dir defaults to the current working directory.
• --output defaults to paper_info.json.

Fetch BibTeX entries

# From the repository root (or after installing via pip)
python fetch_bibtex.py --input paper_info.json --output papers.bib

The script reads the JSON produced by the extractor, fetches BibTeX records for entries that have a DOI, and writes them to papers.bib.

The extractor prints progress and writes a JSON array like:

[
  {"file": "1.pdf", "doi": "10.1109/XYZ.2023.123456"},
  {"file": "2.pdf", "title": "An Interesting Study on …"}
]

---

Next steps

• Convert the generated .bib file to the citation style you need.
• Extend the workflow with additional scripts or integrate into your bibliography manager.

---

🤝 Contributing

Please see the CONTRIBUTING.md for guidelines on how to fork the repo, set up a development environment, and submit pull requests.

---

📜 License

This project is licensed under the MIT License – see the LICENSE file for details.