The new Ovid might be difficult/impossible to translate!

Hi everybody!

For some time now, there have been various threads from people seeking help saving references from OVID with this outstanding little tool, called Zotero!
I personally was amongst these people and was told that OVID was still supported. Never the less I am (like many others) unable to save ANY information or documents from OVID pages by using Zotero.
I finally reviewed the source of an OVID page and found - that there was no information to scrape from the HTML code! It seems the new Ovid displays all citation fields dynamically by calling Java scripts. So a new translator probably will have to call these java scripts and then scrape the information derived from there. Sadly, at present I know close to nothing about Java script, so I cannot act on this myself (maybe I should get a book). However by reviewing the js.file from the old tranlator I gather it looks for the citation information in HTML tags. Thus, I am quite sure that the old translator cannot deal with this problem at all. So one might as well remove it and declare OVID support for temporarily broken.

Best of regards to anyone. What has been acomplished so far here, is already outstanding.

Nino
  • by reviewing the js.file from the old tranlator I gather it looks for the citation information in HTML tags. Thus, I am quite sure that the old translator cannot deal with this problem at all.
    That's not a problem. Generally speaking, Zotero translators don't look at "HTML tags"—they look at the document object model, a.k.a. DOM, which is the internal representation of a web page and includes changes made by JavaScript.

    The Ovid translator works fine for me. Journals@Ovid Full Text -> "sociology" -> yellow folder icon -> first result, "Commentary: On Regulation and Medical Education[...]".

    If it's not working for you, provide steps to reproduce and include a Report ID from Report Errors (ideally on an another Ovid thread, not this one).
  • edited June 2, 2009
    That is strange as Zotero consistently fails to retrieve any information from OVID for me. Maybe I should try under Windows (I am using Linux).
    I am interested in Articles from Psyclit and Pubmed, not sociology. However, I find it peculiar that it should make any difference, which databases you search, as long the webinterface stays the same.

    I will provide steps for this error in another thread as advised.

    Thanks
  • edited June 2, 2009
    Accessed the article you mentioned, Dan:

    Journals@Ovid - Searched for
    "Regulation and Medical Education"
    in title

    Result:

    Commentary: On Regulation and Medical Education: Sociology, Learning, and Accountability.
    Durning, Steven J. MD; Artino, Anthony R. Jr PhD; Holmboe, Eric MD

    Academic Medicine. 84(5):545-547, May 2009.

    and other articles.

    The yellow folder symbol is visible. I click and check the appropriate box to scrape article info. Red box appears in lower right corner, but shows no content. No new bibliography item is added. Nothing as usual.
    Errors reported so far: ID 85886664
    [JavaScript Error: "itemGroup.ref.hasItem is not a function" {file: "chrome://zotero/content/xpcom/itemTreeView.js" line: 2028}]
    [JavaScript Error: "tipElement.ownerDocument is null" {file: "chrome://browser/content/browser.js" line: 2210}]

    Will try to access DOM info instead of the source and also try Windows instead of Linux.

    IMPORTANT: should i start a new OVID thread with this? None of the old ones seems to be active!

    Regards
  • Will try to access DOM info instead of the source and also try Windows instead of Linux.
    This isn't something for you to do—it's just in the translator, which works as is.

    Are you behind a proxy server?
    should i start a new OVID thread with this? None of the old ones seems to be active!
    No, actually. This thread is fine, since this doesn't appear to be a general issue with the Ovid translator.
  • Ok Dan!
    Just tried the same procedure in windows xp, FF 3.0.7. First in Zotero v.1.5, then again after update to 2.0b4. No luck. And initially no errors to report! So I don't think, that the java script errors I reported earlier actually are related to my problems.
    I access OVIDSP at home through Copenhagen Library Homeaccess (KB Fjernadgang). In my Proxy settings I found this one setting:
    http://%h.ep.fjernadgang.kb.dk/%p
    Here an example from an OVIDSP url:
    http://ovidsp.uk.ovid.com.ep.fjernadgang.kb.dk/spa/ovidweb.cgi
    ovidsp.uk.ovid.com is also associated to this proxy.
    As far as I can see there should be nothing wrong with this. The same proxy setting works fine with PubMed, Science Direct and others.
    How can I resolve this problem further?

    Thanks for your support btw.
Sign In or Register to comment.