@impracticaldogg reports here that 26 entries are missing when opening the library in Jabref, after it was exported from Zotero.
Zotero entries: 3816 entries → export → Opening the resulting .bib file with Jabref: 3790
I decided to move this problem into a different thread, because in the original post more problems are being reported, which very likely are unrelated. I therefore would ask you to post here, if you think it is related to this specific issue and not in the other thread. Thanks
To be honest, i am not sure if this export issue is even related to Jabref.
The way forward would be to show some examples of these 26 entries that were not exported by Zotero, and to check if they are present in the bib file (you can use any text editor) to open the .bib file.
If they are NOT present in the bib file, Zotero is the culprit.
If they are present in the bib file, but not present after opening the file with Jabref, then Jabref is the culprit.
The test process sounds good in principle. My problem is that I don’t have an automated way of identifying which 26 entries didn’t go through. My preferred route would be to export both data sets to an intermediate format (maybe CSV), import them both to a DBMS, set up a compound index (eg: Year and Title) and then run a join between the tables.
I’ve no idea how to do an export to CSV or similar flat format from Zotero or JabRef. If I had the time I could probably hack a shell script together to extract the records from the Bib file.
EDIT: For my information and others with a similar question. This page says that a custom export filter can be set up in JabRef building on existing functionality. There is a list of existing custom filters here, with links to the layout code (including CSV).
I can recommend Winmerge to find differences in two files that are supposed to be largely similar. Winmerge is designed to be run under Windows. Of course there are alternatives for Linux.
