Using bib entries exported from http://adsabs.harvard.edu sometimes the field archivePrefix is missing. One example is this paper: http://adsabs.harvard.edu/abs/2003A%26A...410..389R

As a result of the missing prefix, the hyperlink to the preprint is build incorrectly and does not work.

Of course I could add it manually each time, but as this effects roughly 20% of the ADS preprints (based on the limited and maybe biased survey of papers I cite in my work) I'd like to have some workaround.


The following workaround makes arxiv the default value of archivePrefix


In case you specify some value in your .bib file your value will be used, arxiv is only used if nothing is specified.

For me this workaround works flawlessly, but depending on the style you use, this might have undesired effects, so when applying this fix, please check carefully if you still get your desired output.

  • With biblatex archivePrefix is an alias for eprinttype, so you could also set eprinttype to arxiv if you wanted. – moewe Aug 12 '17 at 18:42
  • @moewe Thanks for the hint. Do you see any advantages/disadvantages for one or the other? I briefly considered using eprinttype, as this field is ultimately called when glueing together the link, but as I could not find any different behaviour I thought archivePrefix to be more intuitive as this is what ADS normally exports - but if there is a real difference I would gladly change this. – samcarter_is_at_topanswers.xyz Aug 12 '17 at 19:07
  • No, there is a source-mapping that maps archivePrefix to eprinttype so they are equivalent. I prefer eprinttype because archivePrefix is the alias, and eprinttype is what the bibmacros actually use. But it makes sense to use archivePrefix if that is what ADS normally uses and is missing here. – moewe Aug 13 '17 at 13:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.