That function inserts into a bibtex file, and if you don't save it then the new entry is not "findable", so it is somewhat critical to do that in my mind. It could make sense to check if there is a file attached to the buffer though, and only save in that case. That would preserve the current behavior, and solve your problem I think. WDYT?
Daniele Nicolodi <dani...@grinta.net> writes: > On 09-04-2020 13:43, John Kitchin wrote: >> This is basically what doi-add-bibtex-entry does. There isn't one >> function that does arxiv too, but there is arxiv-add-bibtex-entry. Those >> are both commands in org-ref. These don't pop up a buffer for approval, >> that is something you would have to build yourself. >> >> There are many workflows people want, the one above works for me when I >> am writing, and just want to add entries with minimal disruption. I >> usually polish the entries later if needed. I can see a slower, approach >> like you describe too. > > Hello John, > > I have been able to hack something combining org-capture and the utility > functions you mention that implements the workflow I described. > > There is only one issue: doi-utils-insert-bibtex-entry-from-doi contains > a (save-buffer) call that does not play nicely when I call it on a > temporary buffer. Why is the (save-buffer) there? > doi-utils-insert-bibtex-entry-from-doi looks like an utility function, > thus I think it should not try to save the current buffer. What do you > think? > > Thank you. > > Cheers, > Dan -- Professor John Kitchin Doherty Hall A207F Department of Chemical Engineering Carnegie Mellon University Pittsburgh, PA 15213 412-268-7803 @johnkitchin http://kitchingroup.cheme.cmu.edu
