Citation 9 imports

I have used Citation for 15 years. The latest version (v. 9) was released in 2006 and there has been no support or updates available since. I would like to migrate my large Citation database to Zotero but have not yet been successful in getting a clean import/export. It would be great for Zotero to provide a direct import option from Citation for all of us who want to ditch Citation.
  • I would add my request to this. I started to research a large book about three years ago and have thousands of entries in Citation 9. My attempts to import them into Zotero -- by whatever means -- have all failed. I have several doctoral students with the same problem. Is there any hope of a solution?
  • What happens when you try to import a RIS file exported from Citation?
  • Citation exports a bad RIS file: No ER- tags and there's just too much junk created from notes, tags, etc.
  • And I don't suppose the Refer output is any better?
  • Refer output creates a file that stalls the Zotero import and generates an error messge. After I clear the error message I'm left with a slew of bad entries that I must delete before restoring my Zotero database.
  • Also see this thread:

    Can you make a custom export for Citation that gets RIS right? (Post-processing is probably possible (certainly easy enough to add ER where needed), but I don't know what some of the extensions to RIS that they've added do.)

    In that thread, they used the Endnote format with apparently reasonable results & you might try the same.
  • No luck with anything produced by Citation exports. The closest I've come to getting anything useful from Citation to Zotero is downloading an Endnote eval version, importing the Citation file and exporting it to RIS for import into Zotero. It's a messy import with lots of work needed with the individual records so until/if Zotero develops a Citation import filter, I guess I'm stuck.
  • Try the '' format.
  • Add me too. I have been using Citation for a decade and have v. 9x. I have thousands of items in my Citation data files. However, Citations dependence on ASCII is a major obstacle to my use (Middle Eastern Languages). I cannot enter a Czech author, a Hebrew title, etc. I have talked with tech support for Citation and they have no plans for moving to Unicode. So now I have just downloaded Zotero which sounds good. Now I see problems in converting my data base. BTW, could I use Zotero offline or must it always have internet connection?
  • Now I see problems in converting my data base.
    Have you used the export from citavi? If so, please expand as to what problems you are having.
    BTW, could I use Zotero offline or must it always have internet connection?
    You may use it with no internet connection.
  • RCB
    edited September 23, 2009
    I'm not following the instructions by noksagt. Are you suggesting that we convert the Citation 9 file to a citavi file?

    I'm generally confused by Citation 9. How do you export anything from Citation 9? I can see how to convert files into Citation, but not out of.
  • I think citavi is a typo - it's meant to read citation.
    He suggests exporting to a file from Citation.
    Can't help you on the export, sorry.
  • I saved the Citation file as a file. Then when I try to import, Zotero tries to download the file (dialog box for open with vs save, then download manager pops up). Then the progress bar hums along forever, never finishing.

    Any help?
  • what's forever? If this is a large database, it could well take zotero an hour or so to import.
  • dialog box for open with vs save, then download manager pops up
    If this happens, it will never import. Confirm your references are in the refer format. If they are not, paste the first few lines of the file here.

    If they are, you can try to convert the character encoding of the file or copy the contents to the clipboard & use Zotero 2's import from clipboard feature.
  • I got them into the format you suggest, noksagt. I opened the file in Notepad and that's just what it looks like.

    What exactly do you mean by "covert the character encoding"? I'm also not clear on how to get the "import from clipboard".

    Thanks for your help.

    @Adam Smith: How big is "big"? I have about 1300 entries (biblio & notes).

  • Ok, I got the file to convert--messy as it is. (It was saved as a .cit file before, so I just changed the extension to .txt. Is there a better extension that might help?)

    Does anyone have any suggestions on how to fix the messiness? For example
    1) all of my notes are separated from their original works

    2) all of my tags are gone

    Thanks for all the help so far. I'm getting there!
  • Are the notes and/or tags present in the exported Refer-formatted file?

    If so, how do they appear in that file?

    If not, the first thing you would have to do would be to modify the exporter for Citation (and would have to refer to their support for how to do that).
  • I'm sorry it's been a while.

    I tried what was suggested:
    1. I exported my .cit file to a file in format. (I couldn't figure out how to import into Zotero via clipboard.)
    2. I opened the file in Notepad and saved as a .txt.
    3. I imported the .txt file into Zotero.

    More or less, things worked. However I am trying to work out the major kinks.
    1. My notes are now disassociated from the citation of the work. The notes are in Zotero. They're just totally separate entries.
    2. My "series name" field comes up blank in Zotero.
    3. My tags are often in a list for a given entry, separated by semicolons. In Zotero, each series of tags is considered a single tag. So rather than "x; y; z" being three tags (x & y & z), it's considered a single tag, "x; y; z".

    Any suggestions? What questions am I not addressing?

  • I would encourage you to take a look at the .txt file to see how things are being mapped and then take a look at the source code of the Refer translator in Zotero (You can see it in your local copy of ReferBibIX.js or browse it online at

    The actual translator code is accessible enough to give you a good idea of what codes in the .txt file are being translated into what fields and item types in Zotero.

    If you find anything that should be revised in the Refer translator please report on it to the forums so that we can fix it globally.
  • So now I can see that my problem is with the Citation exporter. No matter what format I use (refer-en, refer, endnote, ris), I end up with big problems with info in the wrong categories and such.

    I looked at the custom format converter, and I don't understand the system, so I can't fix it. I'm not expecting much from Citation support.

    Any suggestions?
  • Citation has no intention to release her captured audience.
  • I've just changed the .cit extension to .txt, and the file is just a very simple text file. Surely, there are some intrepid programmers who who can write a small program to read this this into Zotero?
  • I'm sure it's possible. But writing import translators is relatively slow, tedious works, especially if formats are poorly documented as I would expect to be the case here, and .cit isn't exactly a common data format (I think this is the first time this has come up again after 5 years), so I wouldn't expect anyone to just come along and do it. For people able and interested in writing translators, there are higher-impact options (that's certainly the case for me. It'd be around number 200 on the list of small things I'd like to do in Zotero).

    I'd try Refer or RIS output and try to fix it with some light search&replace.
  • Thanks for your reply.

    Actually, the cit file format is pure text. The .cit extension is just a disguise. (I have included a sample from my library below, so you can get the idea).

    I think there are a lot of people stuck with Citation 9 because they have thousands of refs and can't get out. After reading the comments above, they pretty much give up. (I did, months ago, but now I'm getting desperate, that's why I commented here). There are instructions on the Citation page about how to export RIS but it either doesn't work or (intentionally?) mangles the records into something that can't easily be read.

    Here is my sample:


    >Article in a Journal
    >AU Masters, K
    >YR 2005
    >AT Flawed Evidence: a case study of misquoting and inaccurate referencing
    >PG 221-226
    >JR South African Journal of Libraries and Information Science
    >VO 71
    >IS 3
    >AB The need to quote and reference accurately is crucial to academic writing and debate. While small errors are excusable, obvious interpretations, alterations, additions and deletions are not, unless indicated. In addition, primary texts are preferred; if secondary texts are used, this should also be clearly indicated. This article details a case in which these rules have been ignored. It begins by identifying the scope of the particular case: the mis-quoting of Michaels Gibbons. It then traces the route of the inaccurate quotation as it moves from its source through various influential documents and journal articles, until it is firmly established, and easily cross-referenced, but as a vague resemblance of the original, not recognised by the original author. Although this article deals with this one case, it is obvious that the circumstances are not unique, and that there is little reason to suppose that these events cannot be replicated elsewhere. This case illustrates the need for writers to always, where possible, refer to the original documents when citing them.

    >KW evidence, plagiarism, quotation, Michael Gibbons
    >AP Masters 2005
    >Article in a Journal
    >AU Berners-Lee, T; Hendler, J; Lassila, O
    >YR 2001
    >AT The semantic Web
    >PG 29-37
    >JR Scientific American
    >VO 28
    >IS 4
    >AP Berners-Lee 2001
    >AU Horrigan, JB; Rainie, L
    >YR 2002
    >BT Counting on the Internet
    >PR Pew Internet and American Life project
    >AP Horrigan 2002
    >Article in a Journal
    >AU Coiera, E
    >YR 1995
    >AT Medical Informatics
    >PG 1381-1387
    >JR British Medical Journal
    >VO 310
    >AP Coiera 1995
    >Internet: WWW Book, report, paper, ms.
    >AU ITU
    >YR 2003
    >AT ITU Digital Access Index
    >PL Geneva
    >JR ITU
    >SR Press Release
    >LC 21/04/2006
    >AP ITU 2003
    >Technical Report
    >AU ITU
    >YR 2003
    >BT World Telecommunication Development Report (7th Edition)
    >PL Geneva
    >PR ITU
    >AP ITU 2003 b
    >Technical Report
    >AU ITU
    >YR 2007
    >BT World Information Society Report: Beyond WSIS
    >PL Geneva
    >PR ITU
    >AP ITU 2007


  • edited September 23, 2015
    right, but you say "pure text" as if that made it easier. If this were an XML format, _that_ would make it (somewhat) easier, but parsing pure text is a lot messier (where does a tag start? where does it end? how about an entry? etc.)
    It can be done -- we do it for RIS, which is only marginally better -- but it's no fun and it takes time to program.

    edit: and I'm sorry to say (and I do realize that sucks for you and other users) Citation really is a marginal product. It's so small it doesn't even make the super-extensive Wikipedia list of reference managers, which lists 32 different products...)
  • According to the manual you can export your library into different formats:
    If you find this is preferable, however, you can export your Citation datafiles in a number of
    different formats.
    Endnote Refer
    RIS (Reference Manager)
    Given your example above, I wouldn't expect that you loose much data (if any) by exporting into RIS or some other format.
  • right, though people have reported repeatedly above that the export formats like Refer and RIS aren't well done. But since those, too, are plain text formats, they're reasonably easy to fix using search&replace, which is what I'd recommend here.
  • edited September 23, 2015
    If I may nose-in and expand upon the suggestion immediately above. If you are not well-experienced with performing multiple search and replace actions on a file it is not too difficult if you are careful and keep working above a net (keep multiple copies of your file as you make progress). For those who are quite experienced with using a text editor, please excuse my intrusion.

    You will need to use a reasonably good text editor. I didn't immediately find your operating system but you really need something more than NotePad or TextEdit for this to not be too frustrating.

    Be sure to only work on a copy and save the original.

    Look closely at RIS standards:

    Plan ahead. You may find that the order of your actions can make it easier or more difficult.

    Except for eliminating the angle bracket, you shouldn't try to search/replace more than one tag at a time. This may require several iterations to get it right. (I recommend saving each step and then working on a new version of the file.) Each time you complete a search/replace cycle I recommend that you compare the product of your effort to the RIS standard.

    Be sure to pay attention to punctuation and spaces.

    Be aware that the exported files may not be consistently presented even within item type. If you find yourself wondering why your search/replace works for some records and not for others you may need to do some hand editing.
Sign In or Register to comment.