RTF scan output to full ZOTERO functionality

Dear all,

I am new to Zotero, but very interested in it. I've just migrated from Citation and I have successfully managed to import all my references (via Biblioscape import-export).
I am now handling a paper I wrote with Citation. I have managed to change all the citations to a format which is recognized by the RTF scan.
I have scanned the doc and everything is fine. All the citation are now formatted. I have saved it into .doc format.
I would like to be able to change and edit the new document, using all the functionality of Zotero, such as the possibility to change the citation style of the whole document. It is not possible in my current .doc file. It says "you must insert a citation before performing this operation".
Any idea about solving this?

Thank you very much!
  • Sorry...I must correct the info!
    Actually the rtf scan is not really working...It has just changed correctly the citation without page numbers, all the others remained as in the original rtf file (example: {McQuade III, 2006, 16-17; Marler, 2002, 185; Gordon & Ford, 2006}).
    Also, Zotero does not seem to recognize the citation that were correctly changed (cannot change their style)...
    I also have difficulties in inserting new citations and references with Zotero in the document (trying to substitute the wrong citations). It pops up an error. The citation is inserted, but every attempt to modify it causes the same error message...
    I think it is getting complicated...
  • edited March 30, 2010
    The RTF scan does not work with the word processor plugins. The scan just converts correctly typed references into citations in your choice of style (and creates bibliographies). Once it is converted, Zotero is done with it. If you want to be able to change your document style, edit citations, and have Zotero interact with the document you have to enter the citations into it with the Zotero plugin for you word processor.

    Edit: I'm not sure about your other issues. For the references that did not get converted in the scan, did you get a dialogue asking you to confirm the references being used?
  • edited March 31, 2010
    Thank you very much for this. So, there is no hope to work it out. Better to know it since the beginning ;)
    Maybe I should work on the source file and format it only at the end...I will see.

    Has anybody got a clue about the problem with the citations with pages? Currently they are in the format copied above in this discussion. The rtf does not seem to catch them...but they are in one of the supported formats. Example from zotero pages: {Smith, 2009, 10-14}

    btw, it is a true pity that you cannot get full ZOTERO interaction after RTF scans...I thought that since the scan recognized the citation fields and matched them with your database, the full ZOTERO interaction would have been the logical consequence...

  • Still having problems with the RTF filters. I do not know why but some citations are scanned and formatted, some are just forgotten....
    Any solution?
  • ...no ideas? Everybody has a rtf filter working perfectly?
  • Perhaps you should start a new thread or look for a previous one about RTF scan errors. The title of this thread is not very descriptive of your current problem. You might also want to post the exact text entry that is not being converted properly and the associated bibliographical information from the Zotero item you were trying to cite. Finally, also answer whether you were given options for those citations in the RTF scan dialogue, or whether Zotero did not even notice them at all.
  • also, the RTF scan is not a feature that has so far been heavily developed - I think partly because as it works now there are just some inherent limitations - I think there is a case to be made for Zotero to be able to work with some type of citation key, and that has been discussed, but it entails a whole larger set of issues and I don't think is on the short-term agenda.
    That means that a) yes, RTF scan doesn't work perfectly and b) most people don't use it:
    The number of people in academia who don't work with either Word, Ooo, or LaTeX/Lyx is just very small, so my sense (based on talking to users and observing what is asked on the forum) is simply that RTF scan isn't used all that much.
    I realize that your case is a different usage scenario - i.e. dealing with old documents created by a different bib software, but in the medium run those cases are going to be relatively rare, too.

    And considering that there are a whole bunch of rather essential features that have yet to be implemented - I'd say duplicate detection, batch-editing, and the implementation of the new CSL processor are the top three, with better note management a runner up - I just don't think this has super high priority.

    On the other hand it's something that could easily be worked on by an independent interested developer, either as a plugin (see e.g. the Lyz plugin) or as a patch/addition to the current Zotero.
  • edited June 17, 2010
    I hope there is a developer who will improve the rtf scan to full functionality. Because I think the rtfscan feature is an easy solution for Citation Compatibility with endnote and other reference manger (like discussed here: http://forums.zotero.org/discussion/9498/endnote-citation-compatiblity/ or similar here: http://forums.zotero.org/discussion/8291/using-zotero-with-a-large-previously-fully-referenced-document/ and here: http://forums.zotero.org/discussion/8758/using-a-documant-with-zotero-that-has-endnote-citations/ ). Additional it would be helpful for interoperability between different office solutions and OSs.

    I think rtfscan or additional plaintext scan (http://forums.zotero.org/discussion/12280/plain-text-version-of-rtf-scan/ ) is a possible approach or at least a workaround to bridge the current or probably perpetual situation that there is no citation standard and also that many people who are working together using different citation manager.

    With only two clicks (1.covert to unformatted in endnote 2. rtfscan in zotero) you are able to convert a text written with endnote to a text with citations in zoterostyle and backwards provided that everybody has the citied literature.

    This more technical discussion (http://forums.zotero.org/discussion/6968/how-does-the-new-rtfscan-feature-work/ ) is maybe a helpful source for developers.
  • cumuluss makes an excellent point here, especially about collaboration.

    More generally, this aspect of zotero - relying on OO-or-word macros - has always made me queasy. I've even reached the conclusion I don't want to use this feature for my dissertation (although I do for shorter documents).

    Not much to add except I wonder about using a word processor macro to invoke rtfscan somehow to produce the finished document (hence: one click, not two or three). Seems like that macro could be simple and robust, and not prone to the various quirks we have now. (However I know very little about OO/word macros so I could be wrong here.)
  • edited June 20, 2010
    The number of people in academia who don't work with either Word, Ooo, or LaTeX/Lyx is just very small, so my sense (based on talking to users and observing what is asked on the forum) is simply that RTF scan isn't used all that much.
    Can't claim empirical data but anecdotal evidence from my vantage point leads me to disagree with this. There are a number of folks writing academically in Mellel, Nisus, Scrivener (for good reasons) that would all benefit from having better interop with their Zotero libraries through an improved rtf-scan-feature or one that even goes to the heights of cumuluss's suggestion
    I hope there is a developer who will improve the rtf scan to full functionality.
    I can't contribute code, understand limited developer resources (and generally agree with adamsmith's list of needed features above) but wish to voice user interest in continued development of Zotero's rtf-scan abilities. Thanks for listening.
  • On the use case at the top of the thread (saving an rtf file in doc format, and having the document behave as if the references had been originally entered with a Zotero wordprocessor plugin), is there a mechanism in RTF for embedding hidden metadata in a document, or is RTF purely a display/rendering language? I don't know the answer, but whether it is even possible to do what kalder wanted to do would turn on it.
  • Between fields, bookmarks, and user properties, RTF supports all of the features necessary to do this. (See p. 39 and p. 195 in the RTF 1.9.1 spec, although it looks like the latter references the OOXML spec.) Whether these features are implemented in any word processor but Microsoft Word is another question.
  • Are there any news about the rtf-scan feature.
  • I think that first we need to explore Simon's question-- are the advanced RTF features supported by any word processors apart from Microsoft Word? It would be useful if someone tested such features in the leading RTF-using word processors.
  • Like kithairon above, I would also like to voice user interest in a more reliable RTF scan feature from Zotero. I do most of my writing in google docs because I never find myself in one place throughout the day - exporting to RFT and using 'RFT scan' currently gives mixed results (see comments dated 1/16 at http://forums.zotero.org/discussion/14913/rtf-scan-problem/).

    These two would be a powerful combo if they were reliable together. That being said, I too can't contribute to code/development...(beggars can't be choosers?...)

  • The current version of RTF Scan is unusable. It is badly hindered by a bug that replaces the citations mistakenly. I can not imagine using it for a large document and then finding out that the some of the citations have been replaced with the wrong ones.
  • I agree with those who argued that RTF-scan should have a much higher priority. The fact that this feature does not attract many users is probably due to the fact that it is not well known and does not work very well right now. Despite this fact, there are good reasons to push its development:

    - It seems simpler (at least to me as a non-developer) to improve and maintain RTF-scan than to improve and maintain different plugins for (all versions of) all word-processors that are currently in use.

    - Those who use alternative word-processors (AbiWord etc.) would also get a chance to use automatic citation features. All those who can export in RTF would benefit immediately.

    - It would be advantageous to have a fallback solution in case the word-processor plugin fails. In the forums, one finds a lot of reports about plugin failures, many of them caused by broken dependencies, due to differences in platforms and software versions / integrations.

    - As already mentioned by cumluss, collaboration between authors who use different word processors would become easier.

    - Use of server-side text-processors (like Google docs) in combination with Zotero would be possible as well. Given the growing importance of online editing, this seems an important advantage.
  • The citation key syntax used by Pandoc, which has been implemented also in zot4rst, provides a working de facto standard for plain text representation of citation keys and associated content. Although it hasn't yet been implemented in the Zotero RTF-scan feature, there is a benefit in the hiatus in RTF-scan development, in that this syntax, which has been proven to work well elsewhere, can be picked up for its use as well.

    Someone just needs to implement the necessary parser code in Javascript. It's not a massive barrier, but someone will need to be seized by inspiration of given an incentive. When that happens, I'm sure this will move forward. It's not necessarily the responsibility of the core team to push this functionality forward; it's a self-contained task where anyone could lend a hand.
  • but someone will need to be seized by inspiration of given an incentive. When that happens, I'm sure this will move forward. It's not necessarily the responsibility of the core team to push this functionality forward; it's a self-contained task where anyone could lend a hand.
    Zotero has apparently been thinking about instituting some type of bounty system. This may be a place where that could work.
  • Like many others, I have moved to Zotero (from EndNote) only recently. I am using Zotero with Google Docs, so I rely on the RTF Scan. I wonder if it would be too hard for an RTF Scan to have an option to either replace curly brackets with formatted text or to replace the in-text citations with proper Zotero fields, which could then be editable by the Zotero plugin in Word or OpenOffice (so may be we may need two options for the RTF Scan). Such a facility would improve the functionality of Zotero by a mile!
  • not terribly hard - a version of this is working already - look at the ODF Scan tool here:

    though it's not very polished and regular Zotero won't currently produce the right cite keys for it I believe.
  • @adamsmith: Thanks for the mention! On the technical side, the font format and conversion method have been tested against real-world documents, and Multilingual Zoter (MLZ) has reasonably smooth options for controlling when and how "drag-and-drop" cites are generated.

    The converter is currently a standalone Python script that runs from the command line. It seems to work quite well, but I'll admit that a command line tool will be unfamiliar to many users. The conversion is done with regular expressions, and it should be possible to recast it in JavaScript at some point, to embed it in Zotero directly or run it as a Firefox plugin.
  • edited November 24, 2012
    @ironfrown: I should add one note about the MLZ ODF Scan method when used with Google Docs. The drag-and-drop cite placeholders contain URLs that are meant to be clickable, so that you can easily access the original Zotero item.

    Unfortunately, Google Docs doesn't recognize the addresses (i.e. zotero://select/items/0_ABCDEFGH), so the links don't work there. There is not much we can do about that; Google would need to recognise the zotero protocol in their servers to make it work. Conversion still works, though, as far as I know.
  • Dear Frank,
    A question. We are currently writing up a report with an international (distributed) team across the globe. All of our (multilingual) references are in MLZ (although, as you know, we have been experiencing some problems with syncing those across Zotero group users). Last time we did this, we collaboratively drafted the report in Google Docs, then exported it to Word, and then (manually) entered all footnotes through the Word Zotero add-in. Which was a nightmare. Is there a smarter way to do this now? Can we use you odf-scan tool for that? And if so, could you please give us some pointers on how to do that on a Windows machine? From what I gather from your github readme, I’d have to install the python module (I do have python installed on my machines and I have the python directory in my path - does that mean I can just download and double-click the setup.py file?); then I’d have to enter the citation from MLZ into Google Docs (after having checked the QuickCopy tick-box in Preferences -> Export); then I could export the file from Google Docs as an *.odt file; then I’d convert the QuickCopy linked citations to Zotero citations by running the “zodfscan mydoc.odt mynewdoc.odt” command from the windows command-line; and the I could open the file in Word, for instance, and it’d be all ok? Is that the way to do it? Or do I have it all wrong? :) As always, any help would be greatly appreciated! Thanks.
  • Stephan,
    That's how it's supposed to work. As far as I know, the tool will work with Google Docs saved in ODF format. The links won't be active in Google Docs itself, which is unfortunate, but the conversion to live references should work fine. It's a very simple procedure and a small script, so if problems arise we should be able to sort them out pretty quickly.
  • I'm not sure what the steps are to install Python modules on a Windows machine, but the zodfscan script is packaged in the standard way, and should be easy to install, following generic instructions for that platform.
  • Frank, I did figure out that first step. For those interested:
    1. Install Python and make sure the path is set to the right folder (but I think the python installer for Windows does that automatically - see http://www.python.org/getit/)
    2. Download Frank's zipfile and unzip it somehwere (remember where you put it)
    3. Open a Command-prompt and cd your way to the folder where you unzipped the zip-file (in my case "cd D:\Users\Stephan\Downloads\fbennett-zodfscan-39684b5")
    4. enter "python setup.py install". This is what you should see
    D:\Users\Stephan\Downloads\fbennett-zodfscan-39684b5>python setup.py install
    running install
    running bdist_egg
    running egg_info
    creating zodfscan.egg-info
    writing zodfscan.egg-info\PKG-INFO
    writing top-level names to zodfscan.egg-info\top_level.txt
    writing dependency_links to zodfscan.egg-info\dependency_links.txt
    writing manifest file 'zodfscan.egg-info\SOURCES.txt'
    reading manifest file 'zodfscan.egg-info\SOURCES.txt'
    writing manifest file 'zodfscan.egg-info\SOURCES.txt'
    installing library code to build\bdist.win32\egg
    running install_lib
    warning: install_lib: 'build\lib' does not exist -- no Python modules to install

    creating build
    creating build\bdist.win32
    creating build\bdist.win32\egg
    creating build\bdist.win32\egg\EGG-INFO
    installing scripts to build\bdist.win32\egg\EGG-INFO\scripts
    running install_scripts
    running build_scripts
    creating build\scripts-2.7
    copying and adjusting bin\zodfscan -> build\scripts-2.7
    creating build\bdist.win32\egg\EGG-INFO\scripts
    copying build\scripts-2.7\zodfscan -> build\bdist.win32\egg\EGG-INFO\scripts
    copying zodfscan.egg-info\PKG-INFO -> build\bdist.win32\egg\EGG-INFO
    copying zodfscan.egg-info\SOURCES.txt -> build\bdist.win32\egg\EGG-INFO
    copying zodfscan.egg-info\dependency_links.txt -> build\bdist.win32\egg\EGG-INFO

    copying zodfscan.egg-info\top_level.txt -> build\bdist.win32\egg\EGG-INFO
    zip_safe flag not set; analyzing archive contents...
    creating dist
    creating 'dist\zodfscan-0.1-py2.7.egg' and adding 'build\bdist.win32\egg' to it
    removing 'build\bdist.win32\egg' (and everything under it)
    Processing zodfscan-0.1-py2.7.egg
    creating c:\python27\lib\site-packages\zodfscan-0.1-py2.7.egg
    Extracting zodfscan-0.1-py2.7.egg to c:\python27\lib\site-packages
    Adding zodfscan 0.1 to easy-install.pth file
    Installing zodfscan script to C:\Python27\Scripts

    Installed c:\python27\lib\site-packages\zodfscan-0.1-py2.7.egg
    Processing dependencies for zodfscan==0.1
    Finished processing dependencies for zodfscan==0.1


    All done!
  • Frank, I invited you to a rizzoma Wave, where I am documenting my feeble attempts to make this work. I think I nailed the first step. But when I try the QuickCopy (after having made the change in the export preferences) from my MLZ (in Firefox, version 3.0m310), I just get the following (after doing Ctrl-Alt-C with a record selected):
    Городниченко, А. “Войны Современности Как Отражение Мировой Конфликтной Среды.” In Белорусская Политология: Многообразие в Единстве –v: Политическое Знание в Современном Социальном и Образовательном Пространстве, 146–914. Гродно: ГрГУ им. Я. Купалы, 2012. http://www.lib.grsu.by/library/data/resources/catalog/168096-372858.pdf.
    Now that's not a 'Zotero link wrapper', is it? Any idea what I could be doing wrong here? Because I guess that once I have that one straightened out, the other steps will probably work.
  • Try Ctrl-Alt-A.

    The label on the second toggle is a little misleading. Citation and bibliography entry output are only switched for the drag/shift-drag operations. For the hotkeys, the assignments stay the same, and wrapped cites are only triggered for citations, not for bibliography entries.
  • Frenk - sorry that was a typo: I DID do Ctrl-Alt-A. But I still get sthg like this: Белоусов, Краткий Отчет о Научно-исследовательской Работе По Теме: «Исследование Взаимосвязей Важнейших Параметров Социально-экономического, Научно-технологического и Инновационного Развития На Период До 2030 Года».
Sign In or Register to comment.