Maniphest T432076

VE's citation tool incorrectly pulling author information
Open, Needs TriagePublicBUG REPORT
Actions

Assigned To

None

Authored By

	Nil_NZ
	Jul 13 2026, 11:29 PM

Project Tags

Referenced Files

None

Subscribers

Aklapper

Mvolz

Nil_NZ

Description

Per this discussion at AfC, VE is pulling the names of the webpages social media links to populate its "author" field, rather than the author listed in the page's html info. In this case, it should be "Web Desk", which while not an actual author, is a lot more correct than "Fb; Tw; Linkdin; Whatsapp; Us, GNFollow."

Steps to replicate the issue (include links if applicable):

Paste https://www.asianetnews.com/entertainment/box-office/vaazha-malayalam-movie-three-week-box-office-vipin-das-anand-menen-sjdzn6 into VE's automatic citation generator.

Event Timeline

Nil_NZ created this task.Jul 13 2026, 11:29 PM

Restricted Application added a subscriber: Aklapper. · View Herald TranscriptJul 13 2026, 11:29 PM

• matmarex added a project: Citoid.Jul 14 2026, 2:41 AM

Mvolz moved this task from Backlog to Zotero on the Citoid board.Jul 14 2026, 8:14 AM

This is coming from the Embedded metadata translator in Zotero, which has a generic web translator which doesn't always work all that well.

(3)(+0000000): Translate: 2 bylines found:

(3)(+0000001): Translate: 2 Min read Author : Web Desk Published : Sep 06 2024, 03:55 PM ISTShare this ArticleFBTWLinkdinWhatsappGNFollow Us
Author : Web Desk Published : Sep 06 2024, 03:55 PM IST

(3)(+0000000): Translate: Locating the one closest to title.

(3)(+0000000): Translate: Looking for title using: .//*[normalize-space(translate(text(),"ABCDEFGHJIKLMNOPQRSTUVWXYZ ","abcdefghjiklmnopqrstuvwxyz "))="നാലാം വാരവും തിയറ്റര്‍ വിടാതെ 'വാഴ'; മൂന്നാഴ്ച കൊണ്ട് ബോക്സ് ഓഫീസില്‍ നിന്ന് നേടിയത്"]

(3)(+0000000): Translate: Parent level 1

(3)(+0000001): Translate: Extracting author(s) from byline: 2 Min read Author : Web Desk Published : Sep 06 2024, 03:55 PM ISTShare this ArticleFBTWLinkdinWhatsappGNFollow Us

You can try reporting the issue upstream here: https://github.com/zotero/translators/issues to better pull things from the byline, but realistically making a generic translator that works on every website is slightly more tricky than one that works on only one, so a dedicated translator might be more appropriate.

I also notice the page has JSON-LD data which actually looks pretty good and much less messy than pulling from the byline. Unfortunately a patch to add json-ld to the embedded metadata translator looks stalled upstream? https://github.com/zotero/translators/pull/1897

There is also https://meta.wikimedia.org/wiki/Web2Cit which is a user script that lets wiki users write their own translators on-wiki for specific sites and enable them, which doesn't require merging things upstream with Zotero.

VE's citation tool incorrectly pulling author informationOpen, Needs TriagePublicBUG REPORTActions

Description

Event Timeline

VE's citation tool incorrectly pulling author information
Open, Needs TriagePublicBUG REPORT
Actions