Zotero web extension not working even after following KB article

oaktreeadmirer
Hi,

I have tried without success to get the Zotero Connector extension working.

Computer: Windows 11 personal use computer
Browsers: Started in edge, then tried chrome
use case: university student
KB article: https://www.zotero.org/support/kb/connector_zotero_unavailable
Zotero desktop version: 6.0.27, updated in last day or 2.
Connector version: 5.0.112

I am just doing research right now, and the copy paste url to zotero web library works fine, but just want to make sure the connector is working for when I eventually write my project in google docs (about a week from now). The extension has never worked for me. Only started with Zotero ~2 weeks ago. The only obvious I have not tried is re-installing Zotero desktop, but am not sure why that would help. Or trying firefox or similar.

I followed the steps in the above KB article: Zotero desktop open before opening browser, http://127.0.0.1:23119/connector/ping gives me "Zotero is running". Uninstalled and re-installed connector in edge. Disabled Google docs offline, no other extensions installed. Tried Chrome and same steps. I have no security software running on the computer other than whatever windows has by default.

Thanks in advance. :)
  • dstillman
    edited September 14, 2023
    Could you say what's actually happening? Do you mean that you're getting the "Is Zotero Running?" dialog?

    Could you provide a Debug ID from the Connector for reloading a page and clicking the save button?
  • oaktreeadmirer
    What happens is that I right click on the zotero extension icon on MS edge, and click save to zotero ( I have tried both with or without snapshot). That does nothing, as in no feedback, so I hit the same menu item again, and a little window pops up in the browser's top corner, saying "saving to zotero.org [website name]". All this occurs while zotero desktop and zotero.org are open, and they are both logged in. I have seen the "Is zotero running dialog" once or twice, but I think that was occuring when I did not have zotero open, really early on in my troubleshooting.

    I will note that when I go to the connector preferences, that general>save to zotero.org>hitting authorize results in a message "Authorization could not be completed. an invalid response was received from the Zotero server.

    I tried submitting a debug ID for your above instructions, but got "http request to https://repo.zotero.org/repo/report?debug=1 rejected with status 0."

    The debug log is too long for the character limit of this post, it gave me an error.

    I guess the next step is resolving my inability to submit a debug file and get a debug ID? My internet connection is good, just a standard home network.

    I will note that the report errors section of the Zotero connector preferences lists:

    "[JavaScript Error: "Failed to retrieve translators from Zotero Repo Error: HTTP request to https://repo.zotero.org/repo/metadata?version=5.0.112&last=0 rejected with status 0" {file: "chrome-extension://nmhdhpibnnopknkmonacoephklnflpho/zotero.js" line: 306}]

    [JavaScript Error: "OAuth request failed with 0; response was undefined" {file: "chrome-extension://nmhdhpibnnopknkmonacoephklnflpho/zotero.js" line: 306}]

    [JavaScript Error: "OAuth request failed with 0; response was undefined" {file: "chrome-extension://nmhdhpibnnopknkmonacoephklnflpho/zotero.js" line: 306}]

    [JavaScript Error: "OAuth request failed with 0; response was undefined" {file: "chrome-extension://nmhdhpibnnopknkmonacoephklnflpho/zotero.js" line: 306}]"
  • dstillman
    What happens is that I right click on the zotero extension icon on MS edge
    (Shouldn't matter, but not sure why you're right-clicking. The standard action would just be just to click the button, and there's no need to add variables when troubleshooting.)

    The main problem here is that something in your browser or on your computer is preventing all network requests from the Zotero Connector to fail. You'll need to figure out what that is. Generally that would be another browser extension, but it could also be some sort of security software or a proxy configuration. If you can't figure it out, you can try in a new browser profile. But this isn't a problem in Zotero itself and isn't something we can help with.

    Beyond that, you should restart your browser, which should cause Zotero to re-prompt whether to save to the Zotero app or to your online library. It looks like you previously told it to save to the online library, which it will keep doing as long as it can't find the Zotero app, but you should save to the app if you have it. (And in this case both are failing for you, since all of Zotero's network requests are being blocked by something.)
  • oaktreeadmirer
    I changed "site access: Allow this extension to read and change all your data on websites you visit:" setting to on all sites from on click, and everything works now.

    I then went back to "on click" for site access, turned on the debbug logging, and tried to use the connector on a news site, by left clicking on the little Zotero connector icon near the url bar. I had zotero desktop open. I got a little "saving to zotero.org" window, but no actual success in generating a new item in my zotero library. I then turned site access back to "on all sites" and successfully submitted the debug id D1083429220 .

    I guess the connector only works with "on all sites" for site access? Perhaps that is by design, and it is not supposed to work with "on click"?

    As for whether I care about "on all sites" vs. "on click", I guess that is a personal preference issue of mine, and whether I care about any potential privacy implications (although I am sure that user privacy is a priority for the Zotero staff).

    Not sure if any of this is worth further investigation on your part. Thanks for the very quick responses.
  • dstillman
    To use "on click", you would have to click the button and then either click Reload if the browser prompts you (which it should at least the first time you do it) or reload the page manually.

    But to work normally, the Connector needs to run on every page: to update the icon to communicate what was detected on the page, to automatically import RIS/BibTeX/CSL files, and to perform automatic proxy redirection, and probably for other things in the future. It's possible we could make saving on click with automatic detection work without a reload, but given that other features wouldn't work, it's not something we've looked into.

    Re: privacy: https://www.zotero.org/support/privacy#zotero_connector
  • oaktreeadmirer
    Thanks. Great and fast answers. Questions resolved.
Sign In or Register to comment.