Have a way of keeping aliases for common papers #feature-request
slosar opened this issue · 4 comments
There should be a ~/.adstexrc file or something that would allow aliases for commonly used papers. E.g Planck18 should resolve to whatever the arxiv number is (but the search would be performed to ensure published papers become published once published :), i.e pure key-value resolution).
This is a good suggestion but I wonder if it can already be accomplished by current features.
One can already supply additional bibtex files to adstex with -r, so that adstex would know those citation keys exist. This is useful for citation keys like Planck18, numpy etc.
I understand that you would also want to these entries (adstex does not update the files specified by -r). But I am implementing #3, which let you use
adstex my_own_bib.bib
to update the entries. So you can run this on those custom bibtex files once a while to update them.
This involves more steps, but seems more consistent with the idea of using bibtex files as the libraries (so that we don't need to create a new library format).
Yes, but the entire point of adstex is that I can give up having to deal with my central bibtex file.
The issue there is that in a typical workflow, someone invites me to work on their overleaf paper that already comes seeded with some bib file. Then I copy my cosmo.bib over, add to repo and immediately there are duplications between the two and when I update I often forget to update cosmo.bib in my repo and instead update paper's cosmo.bib, i.e things go awry very soon.
But perhaps the solution is to use \defcitealias instead -- that has the advantage of taking care of duplicates... Let me think a little...
This is a relevant link:
https://tex.stackexchange.com/questions/44877/bibtex-with-multiple-aliases-for-the-same-reference
You probably want ID field with aliases in your bib file (second answer), but perhaps for another day!
the entire point of adstex is that I can give up having to deal with my central bibtex file
Indeed!
But my point is that maintaining a .adstexrc file is not that different from maintaining a bibtex file.
Citation aliases are indeed really difficult to deal with. The IDS field is a good solution but it only works with biblatex, and I think most astro people still use natbib.
For now, adstex does detect if there are citation keys that point to the same paper, and prints out warning messages at the end. It doesn't fix the issue for you though.