Zotero connector 5.0.111 failed to save snapshot of any webpage in incognito mode of Chrome.
After updating Zotero connector to the latest version this morning, it simply failed to save a snapshot of any webpage. I've tried almost all in https://www.zotero.org/support/troubleshooting_translator_issues.
================
Your debug output has been submitted.
The Debug ID is D1394924421.
================
And where can I download the previous version of Zotero connector? I couldn't find it anywhere, and Google chrome extension store doesn't provide previous versions for download.
================
Your debug output has been submitted.
The Debug ID is D1394924421.
================
And where can I download the previous version of Zotero connector? I couldn't find it anywhere, and Google chrome extension store doesn't provide previous versions for download.
Update: if I go to an arXiv.org page, and save it using the arXiv.org translator, it can save both the PDF and the webpage snapshot. However, if I save the same page using "Web Page with Snapshot" option, the snapshot won't be saved at all. So there seems to be something wrong with that translator or so.
Note: the previous version (before upgrading this morning) of Zotero connector worked properly when saving snapshot of a webpage. So there must be something wrong with the latest version.
Saving pages via multiple-item dialog and in some other instances trigger a different method of saving, which is why it works. There's a technical reason for it that I won't get into here. We still need to figure out why the basic use case is broken for you.
Meanwhile, I have been browsing all webpages in incognito mode exclusively, with cookies enabled but all 3rd-party cookies disabled. Could this be a reason?
Because a bug of a new version of Zotero Connector might be discovered only after it's released, it will be very important for a user to be able to download a last working version of the .crx file from zotero.org, especially for time-sensitive/critical situation like mine. Since a user can download the .xpi file of a previous version for Firefox easily from https://www.zotero.org/download/connector/dl?browser=firefox&version=5.0.110, is there any technical restriction that prohibits the hosting of the .crx file on zotero.org? I mean, there is no need to host all previous versions of the .crx files. It will suffice to host 1 or 2 previous versions, in case that the latest/updated version has a bug like now, and the user needs to revert back to the last working version without the needs to switch browser or build from source. I hope this will make sense.
If you want snapshot saving in a private window, you'll have to use Firefox, where this worked properly before and continues to work.
We've submitted 5.0.112 with this fix to the Chrome Web Store and Edge Add-ons site. The latter has already been approved, and the former will hopefully be available later today.
When running 5.0.112 with Zotero 6.0.26, the save popup will show a red X for the snapshot, but the snapshot will be saved nevertheless. With the next Zotero 7 beta, the save popup will display proper feedback for the snapshot.
Or you can just use the Zotero Connector in a new Firefox profile now and get correct behavior in all cases. Again, the Zotero Connector, like all public Chrome extensions, is distributed via the Chrome Web Store. A .crx can only be installed in developer mode and isn't equivalent to an XPI. It's not something we distribute, so you can stop asking. If you want to use a browser that allows file-based installs of public extensions, use Firefox.
My temporary solution yesterday (before the updated v5.0.112) was to use SingleFile (https://github.com/gildas-lormeau/SingleFile) to download a snapshot when browsing a webpage in incognito mode of Chrome, drag it into Zotero to the item (for which saving snapshot using Zotero connector in incognito mode failed), and then delete the snapshot downloaded by SingleFile. It looks like SingleFile can faithfully scrap the webpage as exactly shown in the incognito window, and when downloading, it will trigger a regular download from blob:chrome-extension://mpiodijhokgodhhofbcjdecpffjipkle/, for which I can see the download progress and history in chrome://downloads/. Since Zotero Connector is partly based on SingleFile (or SingleFile-Lite https://github.com/gildas-lormeau/SingleFile-Lite ?), could this saving mechanism of SingleFile be integrated/implemented in Zotero Connector?
Many thanks again.
Therefore, saving webpage's snapshots within the Zotero app is NOT a proper solution to this issue. So could you please integrate or implement the same or similar snapshot saving mechanism of SingleFile (which works properly in incognito mode of Chromium based browsers) in Zotero Connector?
We're done talking about this. This is a limitation of Chrome. Use Firefox if this is important to you. If you can't use Firefox for some reason, use SingleFile and attach the snapshot manually.