I've had the same problem, through the University of Wisconsin's ezproxy server. Zotero recognizes sites only when I'm not using the proxy, but I can only actually read the articles when I am using the proxy. Quite the catch-22.

Likewise via Columbia and Rutgers resources. This is obviously very important to many users.

Zotero does have some EZProxy support. While I have access to only two EZProxy-based systems, it works fine with both. It is quite possible that it doesn't work with others, but unfortunately, my ability to test your specific sites is limited.

*Still* failing for Google Scholar through Harvard's EZProxy. For example, on this search:

http://scholar.google.com/scholar?sourceid=Mozilla-search&q=%22Dispositional+theories+of+value%22

the references save fine when outside of EZProxy. But through Harvard's portal:

http://scholar.google.com.ezp1.harvard.edu/scholar?q=%22Dispositional+theories+of+value%22&hl=en&lr=&btnG=Search

I can see the address bar folder icon and select items to save, but nothing will save. Instead, on selecting items to save, Zotero gives me an empty rectangular frame ("Saving item..." style, but without any text or other interior) and fails silently.

Simon, Zotero is great but it *must* work with the institutional access portals for sites like this. How can we help you make it happen?

I followed the instructions, Sean, but no errors were reported in the console (information and warning messages were reported, but none from Zotero). Maybe I should clarify that Zotero isn't crashing, it's just failing to do anything to save the item... it doesn't even freeze up and the empty frame goes away if I click on it.

... I did manage to stumble upon an unrelated bug. This error pops up every time I open a New Tab, either blank or by right clicking on a link. I'm on XP and I use TabMixPlus. It doesn't occur when I open a New Window.

Error: [Exception... "Component returned failure code: 0x80004005 (NS_ERROR_FAILURE) [nsIURI.hostPort]" nsresult: "0x80004005 (NS_ERROR_FAILURE)" location: "JS frame :: chrome://zotero/content/xpcom/ingester.js :: proxyToProper :: line 147" data: no]
Source File: chrome://zotero/content/xpcom/ingester.js
Line: 147

Is there anything else we can do? I might be able to work through my library contacts to get you a Harvard login for testing of the EZProxy sites.

sean,

I am having the same issue with EZProxy and below is what I got (at the Mac OS X Terminal display) after following your instruction.

I was trying to save an IEEE Xplore article through MIT library's EZProxy. The citation was saved, together with a snapshot called "IEEE Explore Full Text PDF" which turned out not to be a pdf file but the html page asking me to log in for access.

This will be rather long and I am not sure if it will fit in a single post. If there is an email I could send to, please let me know.

lwc: You can send it to support@zo.....org

The best we can suggest is for anyone who's experiencing this to install the LiveHTTPHeaders extension and send us the output (clearing the headers right before clicking the URL bar icon and clicking Save All right after). That might give us clues as to why Zotero isn't detecting the EZProxy connection on certain library sites.

Anything new on this one?

Zotero isn't detecting the EZProxy connnection at my institution too, which in turn creates problems with ieeexplore, sciencedirect and others (gets the citation info but fails to retrieve the pdfs). Using 1.0 b4-r5 and Firefox 2.0.0.2 under linux

Let me know if you need more debugging information.

Could this be resolved with an additional configuration option? Basically have the user specify that EZProxy is used and what is appended to the regular url.

Could this be resolved with an additional configuration option? Basically have the user specify that EZProxy is used and what is appended to the regular url.

This may be one way to go. I'll talk to Simon (who wrote our EZProxy code) about it.

One question for people having problems with EZProxy: when you authenticate with your institution's server, is it on a different domain from the one on which you browse catalogs? The current code looks for a particular response from the EZProxy server when you log in, but if that happens on a different domain Zotero wouldn't detect the use of EZProxy on the second domain. It's possible we could change the code to support such setups (which use a cookie to track the EZProxy session), but it'd be good to know if there are other issues as well.

dalai: Thanks. If you could send the headers to support@zo.....org, that'd be great. Be sure to strip out any passwords before sending.

We'll see if we can get this working with the EZProxy cookie setup.

Using Harvard's EZProxy, I am indeed taken to a different domain for authentication. I have emailed you the LiveHTTPHeaders for a representative issue with Google Scholar. I just ran a couple of tests on JSTOR as well, but couldn't reproduce a problem so far today!

I'm having the same problem with all JSTOR documents--also at Harvard. Love zotero, but this does make it a LOT less useful. Please help!

Similar problem:
Our university uses a proxy-solution from juniper networks which changes the above link:
http://scholar.google.com/scholar?sourceid=Mozilla-search&q="Dispositional+theories+of+value"
into:
https://crypto.unil.ch/scholar,DanaInfo=scholar.google.com,SSO=U+?sourceid=Mozilla-search&q="Dispositional+theories+of+value"

So it basically transforms
http://domain/page?parameter
into
https://cryptoproxydomain/page,DanaInfo=domain,SSO=U+?parameter

Not sure, how easy it would be to catch that and pass it on, but it shouldn't be too difficult. Would write it myself in Scaffold, with "DanaInfo" as target, but I have no idea how to pass it on to other translators.

I just wanted to report that the problems I've previously reported with respect to JSTOR on the Harvard ezproxy appear to be fixed. A big Thankyou to the developers!!!