Web Login University Proxy

MikeDacre

I am using the standalone Zotero both on my Linux machine at work and on my Apple laptop at home.

I work at Stanford and use the university proxy to acess papers. The problem is that the university proxy is uses a web-based authentication, and then stores a cookie for one day. My Firefox or Chrome sessions are obviously not passed to my Zotero session, and so when I use the Chrome or Firefox extensions, Zotero fails to download the pdf, and also fails to make a copy of the web page - the web page it saves is just a plain HTML copy of my university's login page.

Is there any way to just log in with Zotero? I have tried in vain to find a way, all I need is a simple HTML browser and cookie storage, then I can just log in, save the session cookie, and use Zotero as I want.

Is that at all possible? Or do I have to switch back to zotero in firefox and abandon Chrome and the standalone app?

dstillman

Yeah, right now Zotero Standalone won't work (for saving files, at least) with proxies that require web-based authentication. It's something we hope to be able to address, but for now Zotero for Firefox is your best option for gated resources.

aurimas

Your analysis is correct. We have a fix for this under review. If you're interested: https://github.com/zotero/zotero/pull/409

MikeDacre

Thanks for the info guys. I switched to the Firefox plugin for now, although aurlmas' fix looks like a good solution to me. I will wait for it to be (hopefully) included in a major release though.

kdonovan11

Hello,

I'm having the same problem and hoping for the solution mentioned on Github. Any chance there's an update on its status?

Thanks,
Kevin

kamigoto45

Hello,

I just submitted an error report - 467076505
I logged in through my university to JSTOR; was prompted by MLZ-Firefox to store a proxy, so I said yes (first time ever).
I then clicked to download the PDF (from the url bar) and it failed. So I chose "Download PDF" from the JSTOR menu, Accepted its terms, and was taken to a PDF page.
I downloaded there, but Zotfile did not kick in to rename the file, and no record was created (no metadata available ?!?); no metadata was recognizable later, when I found the "bare PDF" in my library and right-clicked to find metadata from within MLZ.

kamigoto45

I just tried another thru my university to JSTOR; first "you can't make changes to the selected collection" (I had the trash folder open), so I selected "my library" -- 1345792532 is the error number, it asked for the prozy (I refused) and then said could not save item.

So I chose "View PDF" and then chose "Save to Zotero" - this time it worked.

What's different this time, I wonder?

kamigoto45

Wow, that was confusing. Let me try again:

1. Find file in university library
2. Click link, JSTOR opens on reference with preview of article below
3. Click the article icon in the address bar -> error about "collection"
4. Move from "Trash" to "My Library"
5. Return to JSTOR and click the article icon "you can't save" error, get report number shown above
6. Return to JSTOR and choose "View PDF"; it launches in a new window and a menu launches "save file? save to zotero?" I choose save to zotero. Success! (But I had to go into zotero and choose "rename" (Zotfile) manually - that hadn't launched at download.)

OK, hope that's clearer.

Andalman

I also find this issue frustrating.

Looking through the github fix mentioned above, the comments suggest this fix might not work.

Is there a fix in the pipeline for this problem?

Thanks for the update,
Aaron

aurimas

The fix mentioned above should fix most of the cookie/proxy issues with connectors. Not sure which comments you're referring to or what issue you're facing exactly though.