Automated BibTeX Cleaner & Double-Checker

Try it online: https://bibtex-clean-tool.streamlit.app/

Demo Video: https://www.bilibili.com/video/BV1XcRvYhE27

Overview

Inspired by the cool repos bibtex-tidy and arxiv-latex-cleaner, this repository automates the process of cleaning, organizing, and verifying .bib bibliography files.

It consists of two primary components:

1. BibTeX Cleaner & Formatter (cleaner.py) - Reorders .bib entries based on citation order in the .tex file, removes duplicates, and appends unused references at the end. This was originally adapted from the repo SFRL/clean_bibtex.
2. Double-Checker (checker.py) - Searches IEEE Xplore for accurate BibTeX citations and updates .bib entries while preserving the original keys. This is necessary because sources like Google Scholar often have incorrect years, incomplete metadata, or lack authoritative information.

Features

🧹 Cleaner (cleaner.py)

• 🔄 Reorder .bib entries according to citation order in the .tex file.
• 🏷️ Add reference comments (% reference 01, % reference 02, etc.) to track ordering.
• ❌ Remove duplicate citations, ensuring a concise bibliography.
• 📌 Remove/Preserve unused entries, appending them at the end for later use.
• 🔤 Ensure proper acronym formatting by wrapping specified terms in \text{} within the title field.

🔍 Double-Checker (checker.py)

• 🌐 Automatically search IEEE Xplore, ensuring accurate metadata.
• 🔑 Maintain original BibTeX keys, replacing only outdated information.
• ⏳ Process a configurable number of entries (default: 60) with a progress bar.

Installation

Ensure you have Python installed, then install dependencies:

pip install -r requirements.txt

You also need to install the appropriate WebDriver (e.g., ChromeDriver for Google Chrome) if using automated web scraping.

Local Streamlit Usage

streamlit run app.py

Command-Line Usage

Running the Cleaner

python cleaner.py [bib_file] [tex_file] [--keep] [--wrap-text] [--remove-review-textcolor]

• bib_file (optional, default: ref.bib): Path to your .bib file.
• tex_file (optional, default: main.tex): Path to your .tex file.
• --keep: Preserve unused entries in the cleaned bibliography.
• --wrap-text: Wrap the first word in the title field with \text{} for proper LaTeX formatting.
• --remove-review-textcolor: Remove textcolor markup from the output files.

Running the Double-Checker

python checker.py [bib_file] [--num <number_of_entries>] [--remove_unselected]

• bib_file (optional, default: cleaned_ref.bib): Path to your .bib file.
• --num (optional, default: 60): Number of bibliography entries to check and update.
• --remove_unselected: Remove entries that were not selected during the checking process.

---

Contributing & Support

Developed and tested on Apple M1, macOS Sonoma 14.7.3.

Feel free to open issues or pull requests if you encounter any problems or have suggestions for improvement. If you find this project useful, consider giving it a ⭐ on GitHub!

Additional Notes

Actually, after the repo was built, I found that there are also many other repositories, like bib-cleaner and bib-world, that can also remove duplicates and unused entries. I particularly like the use case of bib-cleaner.

However, our tool provides more features, such as automatic ordering of .bib entries according to citation sequence, IEEE Xplore verification. Hope it helps you to some extent!!! 😊