This is a script that handles some of the more common tasks I need to do with my bibtex-files.
See
bibmanip -h
for a list of the available subcommands (this list is also in HELP.txt).
For some examples, see Examples.
I have some functionality for "tags" which I use pretty much as citeulike does. Any bibtex-entry may have a field
tags={tag1;tag2;tag3}
and the scripts can operate on them (e.g., split a master.bib file into separate tag1.bib, tag2.bib, ... files)
bibmanip does not have a "real" parser. It relies purely on regular expressions, to parse the bibtex files and I had to add a some regex magic to get it to work.
So far, it worked for all bibtex-files I used (and those came from citeulike, mendely, Papers (mac) and handwritten files).
I'm not sure that I didn't miss some specific example in which my regular expressions do not work though. Be warned!