PDF saving problem

I have an issue with the PDF downloading when I save a reference on a website (Sciencedirecte...).
When I click on the icon to save the reference, a box appears in the bottom right what shows the saving state (snapshot and PDF article) but the PDF doesn't save (red cross).
In my library, the reference is there, the snapshot too but not the PDF article. Why doesn't this work?
I use Google Chrome and Zotero Standalone v 4.0.11
ID report 2014769436

Thank you!
  • What's the URL as you see it?
    Note that if you access sciencedirect through a proxy and use Standalone with a Connector, Zotero won't be able to get PDFs for some sites (including sciencedirect and proquest)
  • For exemple, this one:
    http://www.sciencedirect.com.bases-doc.univ-lorraine.fr/science/article/pii/S0269749110004720

    On 20 references, none has saved PDF article, and this on several websites (Sciencedirect, Wiley Online Library, SpringerLink...).
    Effectively, I use Standalone and a proxy. But I think a friend use the same configuration and it works...
  • Wiley should definitely work. Do you have a URL for that?
    You've made sure that you can open the PDF?
    If so provide two debug IDs - one from Chrome, one from Standalone - for importing that item:
    http://www.zotero.org/support/debug_output
  • For SpringerLink:
    http://link.springer.com.bases-doc.univ-lorraine.fr/article/10.1007/s11104-011-0948-y

    For Wiley, I just tried but there is only snapshot on the box in the bottom right and no PDF available.
    http://onlinelibrary.wiley.com.bases-doc.univ-lorraine.fr/doi/10.1111/ejss.12094/abstract

    I don't know if I'm clear: I can get the PDF if I click on the website but in my library, when I click on the little triangle in front of the reference, there is only snapshot. I think I should have the PDF as well.
    Debud ID D1624737290.
  • (2)(+0000003): Downloaded PDF did not have MIME type 'application/pdf' in Attachments.importFromURL()

    (3)(+0000000):

    <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">

    (3)(+0000000): Deleting item 49
    This is likely due to the proxy.

    When you first start a web browser and authenticate with the proxy, do you log in via a web page or via a pop-up username/password prompt from the browser itself?
  • Also, if you check your system proxy settings, are you using a PAC (proxy auto-configuration) URL or something else?
  • I'm not sure but to get some article on sciencedirect (for example), I need to log on through my university web page (workspace)with username and password. So, I don't know if this meet your expectations. I don't think that I check the configuration of my proxy (or VPN).
    Sorry, I'm not an expert in computer science!!
  • We understand you need to log in. But how does that look?
    do you log in via a web page or via a pop-up username/password prompt from the browser itself?
  • Via a web page
  • And can you say whether there's a system proxy configured? That's something you likely would have had to set up yourself, and you could check that in the network preferences of your OS (though it differs by OS, so we can't give you specific instructions).

    While we're working on this, your best bet is to use Zotero for Firefox, where this shouldn't be an issue. It will share you existing Zotero database, so switching between it and Zotero Standalone is easy. You just have to close Zotero Standalone before using Zotero for Firefox.
  • I don't think so. I have access to nothing. I log in the university web site, I click on "online resource", I choose the web site (sciencedirect...) and I am automatically on the homepage.
  • OK, thanks. I would give Zotero for Firefox a try.

    For the technical details here (not necessarily relevant to you, Pierz), Zotero Standalone 4.0.13 will fix compatibility with PAC-based setups using standard proxy authentication, but it won't help with proxies that require a webpage login. I suspect we have a couple different situations remaining: setups where there's a PAC file but login is still done via a webpage and setups where there's no system proxy but you have to log in via the library webpage before accessing resources. For the former situation, at least, we could show the login page in Standalone, but since they're both essentially just cookie-based proxies, we might be able to fix them the same way, with the connectors passing cookies to Standalone. I'm not sure how much (if any) of that we're doing now (or even if it's possible). Simon would know the details there.
  • Regarding connectors passing cookies to ZSA, there's some discussion here: https://github.com/zotero/zotero-connectors/issues/3
  • FWIW, Springer also fails for me with the Northwestern Proxy i.e. a perfectly standard non-system proxy (EZproxy I believe) when using Chrome and Standalone.

This is an old discussion that has not been active in a long time. Before commenting here, you should strongly consider starting a new discussion instead. If you think the content of this discussion is still relevant, you can link to it from your new discussion.

Sign In or Register to comment.