How can I set it to automatically import RIS instead of saving file and importing later?

Firefox ALWAYS asks me what to do. I have to dl the RIS file and manually import it. Is there a way to make zotero automatically catch it?
  • First, make sure you have "use Zotero to open RIS/Refer files" checked in the general preferences of Zotero.
    Even if that is checked, often RIS files will not be recognized as such, because Zotero is not looking at the file extension, but at the Mime type - a number of sites incorrectly label RIS files as text files and not as RIS/Refer. Not much to be done there.
    If you provide a sample URL I can tell you which of the two you are dealing with.
  • edited November 11, 2009
    http://infoweb.newsbank.com/iw-search/we/InfoWeb?p_product=AWNB&p_theme=aggregated5&p_action=export&f_bibType=RIS&f_exportarticles=1201851115B9C098&submitRIS=.RIS

    Basically I push a button which then generates causes the download dialog to pop up asking what to do with the file.

    It would be really cool if zotero had an external script that any filetype could be run against to pipe the file back into zotero. That way, regardless of the mime-type, one could simply tell firefox to always use that script with that filetype and the script would pump it into zotero.

    Anyway, if there is any way to get this working it would be great. Jstor does the same thing. This is a Newbank link.\

    Edit: I just ran file on the RIS file and it is showing up as ASCII text. Would the appropriately labeled RIS files show up as something different?
  • How about this link?
    http://www.worldcat.org/oclc/41273112?page=endnote&client=worldcat.org-detailed_record
    this works for me as a direct import to Zotero? If the link doesn't work just try worldcat -> cite/export -> endnote

    RIS from JSTOR doesn't work either for me - but for JSTOR that's no problem of course as the regular Zotero translator works.
    Generally I agree - if there could be a way to do this in a more reliable fashion that'd be great.

    Btw. are you aware of the "import from Clipboard" option - it's sometimes a bit clumsy, but can be quite helpful.
  • I'm getting an authentication prompt on that file, so I can't test it, but you could check the content type it's being served as with LiveHTTPHeaders.

    Not really sure what you mean re: the external script, but it's certainly not necessary to use an external script for anything related to this.

    The output of 'file' doesn't make a difference. The content type is determined by the server sending the file. With the pref enabled, Zotero will automatically import the following MIME types:
    • application/x-endnote-refer
    • application/x-research-info-systems
    • text/x-research-info-systems
    • text/application/x-research-info-systems
    • text/ris
    • ris
    The first two are the only valid ones. The other ones are non-standard and were added to Zotero recently, possibly since 2.0b7.4, to support some badly behaving servers. But if a server is sending a RIS file as text/plain, there's really not much we can do about it short of inspecting every text file that passes over the wire, which doesn't seem like a very good solution (and wouldn't necessarily even be desired behavior).
  • Well I appreciate you taking the time to look into it for me. I emailed newsbank about the mimetype issue. Hopefully they take it into consideration. Wikipedia indicates that text/plain is the correct MIME type for RIS.
  • Yeah, that's wrong.
Sign In or Register to comment.