doi2bib3 1.1.1

Fetch BibTeX from DOI, arXiv ID, journal URL and article title

doi2bib3

doi2bib3 is a small Python utility to fetch BibTeX metadata for a DOI or to resolve arXiv identifiers to DOIs and fetch their BibTeX entries. It accepts DOI inputs, DOI URLs, arXiv IDs/URLs (modern and legacy), publisher landing pages, and uses a sequence of resolution strategies to return a BibTeX string. This tool combines the features of doi2bib and doi2bib2.

Key behaviors

• Accepts DOI, DOI URL, arXiv ID/URL, publisher URL, or article-title text.
• Resolves inputs to a DOI using URL metadata, arXiv metadata, Crossref lookup, and DOI content negotiation with Crossref fallback.
• Normalizes BibTeX output, including journal abbreviation mappings and selected publisher-specific cleanup.
• Full pipeline documentation (input -> output): docs/ALGORITHM.md
• Diagram version of the pipeline: docs/ALGORITHM_VISUALS.md

A cross-platform GUI frontend is available: Check out QuickBib and its webapp.

Installation

Install from pypi

pip install --user doi2bib3

Arch Linux

In Arch Linux you can install it from the AUR with the command yay -S python-doi2bib3.

Ubuntu

You can use our official PPA

sudo add-apt-repository ppa:apandada1/quickbib
sudo apt update
sudo apt install python3-doi2bib3

Installing from source

Create a virtual environment and install runtime dependencies:

git clone https://github.com/archisman-panigrahi/doi2bib3.git
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Install the package for local development:

pip install -e .

CLI usage

The CLI accepts a single positional identifier and an optional -o/--out path to save the BibTeX output. When installed, the package installs a console script named doi2bib3 (configured in pyproject.toml). From the repository root you can run the local script wrapper at scripts/doi2bib3.

# using the local wrapper script from repo root
python scripts/doi2bib3 <identifier> [-o OUT]

# or when installed as console script
doi2bib3 <identifier> -o references.bib

Examples

Fetch by DOI (bare DOI or DOI URL):

doi2bib3 10.1038/nphys1170
doi2bib3 https://doi.org/10.1038/nphys1170

ArXiv inputs (detected automatically):

doi2bib3 https://arxiv.org/abs/2411.08091
doi2bib3 arxiv.org/abs/2411.08091
doi2bib3 www.arxiv.org/abs/2411.08091
doi2bib3 arXiv:2411.08091
doi2bib3 2411.08091
doi2bib3 hep-th/9901001

Name of the paper (includes fuzzy search):

doi2bib3 "Projected Topological Branes"

Publisher/article pages (Supports APS, AMS, ACS, IOP Science, Nature, PNAS, SciPost, and ScienceDirect journals):

doi2bib3 https://www.pnas.org/doi/10.1073/pnas.2305943120
doi2bib3 https://iopscience.iop.org/article/10.1088/1402-4896/ad995f/pdf
doi2bib3 https://www.scipost.org/SciPostPhys.20.3.082/
doi2bib3 https://www.scipost.org/SciPostPhys.20.3.082/pdf
doi2bib3 https://www.sciencedirect.com/science/article/pii/S0003491605000096?via%3Dihub

Save to a file:

doi2bib3 https://doi.org/10.1038/nphys1170 -o paper.bib

Note: If the tool is not installed, you can run python scripts/doi2bib3 https://doi.org/10.1038/nphys1170.

Supported journal groups

doi2bib3 directly supports many APS, AMS, ACS, Nature, PNAS, SciPost, ScienceDirect and IOP groups of journals. For other journals, the DOI link works, but the paper's URL would not work..

Programmatic usage

Public API

The Python API exposes one primary function:

• doi2bib3.fetch_bibtex(identifier: str, timeout: int = 15) -> str

Example:

from doi2bib3 import fetch_bibtex

bib = fetch_bibtex('https://www.pnas.org/doi/10.1073/pnas.2305943120')
print(bib)

Programmatic CLI entry

Use subprocess with scripts/doi2bib3 (or installed doi2bib3 command) for automated CLI tests.

Internal module layout

• doi2bib3/backend.py: input resolution and network fetch logic
• doi2bib3/normalize.py: BibTeX normalization/transforms
• doi2bib3/io.py: file output helpers
• scripts/doi2bib3: command-line argument parsing and output handling

License

This project is distributed under the GNU General Public License v3 (GPL-3.0-only).

Acknowledgements

Parts of the code and documentation were assisted by copilot and codex.