Why do some accents export better than others?

atulchin

[background: I started using Zotero the other day, installed ML Zotero, had some difficulties, which I resolved thanks to Aurimas, but this evening I decided to revert to regular Zotero following the instructions posted here: https://forums.zotero.org/discussion/33217/i-dont-want-multi-lingual-zotero-anymore-/ which worked fine, once I remembered to reveal hidden folders]

I then discovered that when I use ctrl-shift-a to copy a citation into WordPerfect, diacritics work just fine for some references but sometimes I get little "?"s everywhere. MS Word is OK. My field is French history so proper export of diacritics is significant for me.

adamsmith

with the same format/citation style and the same items? There's absolutely no reason for that and it's quite odd that would be the case

atulchin

Thanks!

I re-checked the problem and discovered you were correct -- different items. I have just edited my original comment, which now I think describes the problem correctly.

adamsmith

My guess is you imported from Worldcat? There are currently some problems with importing special characters - umlauts, diacrits et al - from Worldcat that only manifest themselves once you create bibliographies. As far as I know, import from all other libraries and databases is fine.

atulchin

Thanks. Some were from WorldCat, some not. Examples:

Amazon (my own book!):
Allan A Tulchin, That Men Would Praise the Lord: The Triumph of Protestantism in Ni?mes, 1530-1570 (New York: Oxford University Press, 2010).

Yale library catalog:
Mario Turchetti, Concordia O Tolleranza?: Franc?ois Bauduin (1520-1573) E I “Moyenneurs,” Travaux D’humanisme et Renaissance no 200 (Gene?ve: Droz, 1984).

adamsmith

amazon scrapes from Worldcat via ISBN since they're not allowing Zotero to use their API anymore (for not entirely invalid reasons - it's not designed for what Zotero does), so that's the same.
But the Turchetti book from Yale works for me. Where are you importing from?
I'm looking at:
http://hdl.handle.net/10079/bibid/798106

atulchin

Thanks for checking. I entered a 2nd copy following your URL but got the same result... I think the first time I copied the same entry you did. Again, in MS Word there's no problem, just in WordPerfect.

adamsmith

I see - so while this is also a Zotero issue, what you're seeing is mainly a WordPerfect problem - it appearas that WordPerfect (or the font you're using) isn't handling decomposed characters:
https://en.wikipedia.org/wiki/Precomposed_character#Comparing_precomposed_and_decomposed_characters
and it should. Zotero should - and will - ideally normalize these, as there are a number of problems with mixing composition methods (e.g. searches and bibtex escaping), but WordPerfect should be able to render them.