RTF scan: most citations not recognized

qrypt
Recently I successfully imported a large Endnote library into Zotero; I also imported a smaller number of citations from a colleague. I am attempting to use RTF-scan to create a bibliography using citations from both groups -- but RTF-scan seems to recognize only the colleague's citations, not mine.

I have a document with hundreds of citations as follows:
{Adams, 2010}
{Berman, 2006}
etc.

In the RTF-scan process, most of these are not appearing in the dialogue window that resolves ambiguous citations. It's as if Zotero doesn't even know that the citations imported from my Endnote library are there -- except that they don't show up as "unmapped", they're simply absent.

I'd be grateful for any suggestions.

Q
  • adamsmith
    edited March 3, 2013
    with which word processor are you creating the RTF?
    While we're working on improving RTF scan to some degree, it hasn't been worked at for a long time and even after improvements will likely only work so-so. If you have a bigger problem, I'd recommend replacing RTF citations with proper Zotero citation - that will take you a couple of hourse but make your life much easier later on.
  • qrypt
    Thanks for the reply. I used Mac-Word. I've now gotten quite a bit further by saving the file with TextEdit. Odd, though, that the original problem was only with the ones imported from my own Endnote library...

    One remaining problem has to do with citations having surnames with accented characters -- e.g. Agustín. I've done some reading in previous threads, and I think my knowledge of encoding and character sets isn't sufficient to make much progress with that one.

    But I now have a much bigger bibliography than I had a few hours ago. For the remaining entries I can just do them manually.

    Thanks again...
  • adamsmith
    edited March 3, 2013
    I'm working on accents and umlauts, that'll likely make it into Zotero 3.1
    edit: and yes, the Word processor would matter quite a bit. Zotero doesn't read RTF in a very sophisticated way, so the RTF scan breaks if the processor poduces overly messy code.
    I'm still in the process of testing some programs. Libre Office works well, good to hear the same about TextEdit, google docs won't currently work.
  • r0adi3
    Glad I found this. I was trying to download a google doc as rtf, then do the scan and it wasn't working, even when I opened and resaved the .rtf in Word. But, when I used the scan on a Word .rtf file, it worked beautifully.

    I am trying to come up with a good process for collaboration utilizing shared google docs and zotero, so any suggestions would be welcome.
  • adamsmith
    we're working on a different solution that will be significantly more robust and work well with google docs. Unfortunately (for a general idea look at the ODF scan feature discussed here https://forums.zotero.org/discussion/18064/2/please-add-better-integration-with-scrivener/ ) Currently we're working both on adding a GUI version so no one needs to run python from the command line and adding group functionality, which won't work properly as things are.

    Until then, RTF scan is your best bet. Libre Office will convert the RTF formatting, so you should be able to author in google doc, then open & save as RTF with LibreOffice, then run the RTF scan.
  • r0adi3
    OK. Thank you. From reading, I think the following may be the option I will test for a collaborative Google doc paper. I need it to be very simple to get people started quickly.

    - Create a group library with a sub-collection.
    - When members add a citation* to the document, the reference is added to the sub-collection
    - When the document is finished,all the references in the sub-collection are selected and drug into the document for the bibliography.

    The one thing that would make this work better is knowing how to drag-and-drop or copy-paste the reference as (Name, date) instead of the entire reference. Can that be done? I'm new to Zotero and haven't found that yet. I've been using it with the Word plug-in.
  • adamsmith
    shift+drag will drop the citation (if you prefer keyboard shortcuts, check the shortcuts tab in the Zotero preferences - it varies by OS, but is something like ctrl+alt+a).

    Note that, as opposed to the plugin, this won't remember what you've cited, so you'll still have to manually create the bibliography from the subcollection in the end as you describe.
  • r0adi3
    Hmm.

    I use shift+drag and get this:
    Harding, Kaczynski, and Wood, “Evaluation of Blended Learning.”

    When what I want is this:
    (Harding, Kaczynski, & Wood, 2012)

    Is there a setting I need to change?

    (I'm sorry to take this thread off-topic and will gladly take questions elsewhere, but will be floating the use of Zotero for this workshop in little over an hour and I'd like to test the idea before I sell it.)
  • adamsmith
    oh sorry - you set the export format drag&drop - both citations and bibliography - in the Export tab of the Zotero preferences.
  • r0adi3
    Works perfectly. THANK YOU.

    I will now be able to put together a demo of academic collaboration using G+, GDocs, and Zotero that is going to illustrate some key pedagogical approaches in action.

    This is going to be great. Thank you again for the speedy responses.
  • madpentiste
    I have the same problem as qrypt. I am working with EndNote & Word-2007 and try to share documents with collaborators who use Zotero. I am also trying to transition from EndNote to Zotero, and that limitations of RTF-scan is a deterrent, I have too many documents with too many EndNote citations to deal with, although the import of the references from EndNote to Zotero worked fine so far.

    In the RTF file to be scanned, my references are in the {Author, Year, Pages} format, with semicolon as separator for multiple citations. What is puzzling is that, unlike qrypt, most (but not all) citations are recognized. Furthermore, there is no typical pattern of "non-recognition" (neither number nor position of missed citation would be consistent from one set of citations to the next). More puzzling is that, if I modify the rtf file in a way that keeps the citations as they were, the recognition changes.

    The only consistencies I was able to find were the following:
    - short sets of citations were less problematic than long ones;
    - citations artificially positioned at the beginning of a paragraph tended to be adequately recognized (meaning that putting a paragraph mark before the { helped the recognition.

    If Scan-RTF has a hard time with messy RTF files, I am wondering if using more sophisticated marks for the citations would make things easier (i.e. {{ and }} or anything that would make these citation marks unique, no matter what.

    Any improvement on RTF-Scan will be more than welcome.
    Thanks.
  • adamsmith
    we have since released this: http://zotero-odf-scan.github.io/zotero-odf-scan/ plugin as an alternative to RTF-scan.
    Unfortunately, this won't help you with old Endnote documents, though.
    I don't believe there is currently anyone interested in improving RTF-scan itself.
Sign In or Register to comment.