TLDR: This script retrieves information from DBLP to update your BibTex files.

python rebib.py --bibfile xxx.bib

1. It first parses the bib entries in xxx.bib.
2. For each entry, it queries DBLP using the title and the first author to retrieve the accurate bibliographical information.
   1. If there is only one match, that's it!
   2. If there is no match, skip it.
   3. If there are two matches, one is arXiv and the other is non-arXiv, it chooses the non-arXiv one.
   4. Otherwise, it lists two most relevant results and ask you to choose one with your keyboard.
      You can pass --interactive=False to just skip this.
3. The updated entries are stored in xxx_updated.bib and the skipped ones are stored in xxx_untouched.bib.

Use it like a pro:

• Change desired to decide what fields you want to keep in the updated entries.
• Change query to decide what you want to send to DBLP.
• Change num_workers to decide parallelization (don't be too greedy, you will be banned by DBLP!).