Sync error: text highlights on PDF preventing library sync: report ID 816185811

Hi there,
I've just imported my library from Mendeley and struck an error message when attempting to sync to zotero.org. I sifted through the error messages and it seems that it's highlights on PDFs. Is there a way to keep the highlights or will I have to go through and delete them to be able to complete the sync?
  • ZoteroObjectUploadError: Annotation text '[…]' is too long for attachment V7ZW66ES"]
    At least based on the errors in your report, there's a problem with the highlights on a specific item. If you paste V7ZW66ES into the Zotero search bar in All Fields & Tags mode, you can see the PDF in question. If you open that in the PDF reader, you should see a number of annotations covering a large amount of text.

    I'd be curious to know if it seems like a mistake — e.g., a single highlight covering an entire page of text — or something that should be valid. If you don't mind, can you take a screenshot that shows one of the annotations from the error message and either upload it somewhere (e.g., Dropbox or Google Drive) and provide a link or email it to support@zotero.org with a link to this thread? (We'll respond here.)
  • Sure! Here's the first highlight linked to that first attachment reference code. It's a few lines of text but not a whole lot! There are other highlights through the same document but there's really no individual ones that are larger than this.
    https://www.dropbox.com/s/ug141gjift3knlr/Screenshot 2022-07-06 094043.png?dl=0

  • Hmm, that's weird. Can you click the parent item in the items list, run this in Tools → Run JavaScript, and email us the output with a link to this thread?

    return (await ZoteroPane.getSelectedItems()[0].getBestAttachment()).getAnnotations()
  • Sent to support@zotero email :)
  • OK, that's certainly not right! We'll investigate — thanks.
  • edited July 22, 2022
    Update: I decided to go through and delete the highlight annotations on the PDF which were causing the issue (originally done in Mendeley and imported when I imported my library) & recreate them using the Zotero highlight. Sync now completed. Seems only to have affected this one file though. Decided to go the manual fix for now :)

    If it's highlighted a bug that might affect others and can be resolved then I'm happy with that. Thanks for the help and I hope this helps others!
  • Yeah, it's specific to this PDF. We're going to put in a server-side fix that will avoid this problem when annotations get extracted this way, but we haven't gotten to it yet.
Sign In or Register to comment.