Connector snapshot fails for upload to Zotero.org cloud library

Firefox 70.0
Zotero Connector 5.0.60
(Zotero desktop not running)

# There is no problem saving into Zotero Desktop #

I'm able to save some items to Zotero.org from the FF connector, but trying to save pdfs (http://unec.edu.az/application/uploads/2014/12/pdf-sample.pdf) to my Zotero.org account fails with the following message in the log:
```
[JavaScript Error: "Method saveSnapshot failed
@moz-extension://92af9fd4-50bb-4907-b0dd-ff923836d6c6/connector.js:227:49
"]
```

for some pdfs the connector offers 'save to zotero (web page with snapshot)', and those fail (I get a red 'X');
for other pdfs, the connector offers 'save to Zotero (PDF)', and those also fail (silently);
for regular webpages, the connector offers 'save to Zotero (Embedded Metadata)', and those correctly appear in my Zotero.org library.

What's going on here?
  • CORRECTION:

    "Save to Zotero (PDF)" fails with red "X" and "Save to Zotero (Web Page with Snapshot)" fails silently.
  • UPDATE: disabling "Automatically take snapshots when saving items" has no effect
  • For various technical reasons, it's not currently possible to save PDFs directly to zotero.org. You'll need the Zotero program open for that, or you can save the PDF to disk and upload it to the online library. (Saving article pages that attach PDFs should work, though, which is what you should always do when you have the choice.)
  • is it possible to just add an entry with metadata?
  • I'm not sure what you mean. What kind of metadata?

    PDF metadata retrieval happens in Zotero after you've save the PDF itself, so if you're saving a PDF directly, there's no metadata to upload to zotero.org without Zotero being open.

    And the same issues that currently keep us from being able to upload PDFs directly keep us from adding a simple placeholder item with a link to the PDF URL.
  • yeah, that's exactly what I was asking and what I was afraid you'd say.
    I guess the scraper/interpreter code is too large to reside in the connector?
  • no, the translator code gets loaded into the connector, but that doesn't do anything for pages with just PDFs either -- in fact Firefox doesn't even allow it to run there.
  • I see, I see. Thank you both for your answers!
Sign In or Register to comment.