It takes a bibtex file and for each entry will call the arXiv API and try and find its arXiv category from the title. Supposing it is able to find it, it will then write that entry into a file of that name. After doing this, there is another script that will go on to combine these files into "main" bibtex files which can then be browser conveniently in, say, JabRef.

Like me, you may have far too many bibtex entries in a single file, and so you'd like to split them in some fashion that makes sense. I happen to think that this way makes sense.

clone it from github:

git clone git://github.com/silky/bibcat.git

go into the directionry

cd bibcat

run bibcat on your bibtex file

python bibcat.py sample.bib

observe that many .bib files have been created in your current directory. Hence, you may now optionally run do_group.py, which will trivially group the bib files by concatinating them in the wat you specify

python do_group.py

observe that grouped bibtex files have now been created in your current directory. rejoice.

Python. and you'll want to pip install the relevant dependencies (these will be apparent by running it).

The exact dependencies are:

• Python 2.7+ (Works on this version at least)
• pip (then "pip install feedparser")
• cat (if you're running linux or mac you probably have this)
• A "nicely" formatted bibtex file. Specifically, each key/value should be on a seperate line and preferably be always surrounded by curly-braces: {}; this is the standard representation in JabRef.

By editing the do_groups.py file; it allows you to specify files and wildcards for the output (of bibcat) that it should include in this file. Notably, it works by just concatenating the files together.

Note that if it can't locate a file on the arXiv it will write it to the "nocat.bib" file.

None yet; email me if you'd like help using it though.