Upgrade StorageImporting special characters using RIS from Papers
Hi! I am trying the beta release (2.0b4). I have serious problems working woth the references in Word. The author's names are not displayed right when they have special characters (ñ, á, ö, you can get the idea). This is huge deal, because of the vast amount of papers published by people with last names in special languages.
Any comments or ideas about this? please.
Pablo
Any comments or ideas about this? please.
Pablo
a) With version 1, I was creating my libraries exporting a RIS file from Papers, and importing to Zotero. But this approach seems to be useless, because in every article in which the name of one or more authors have special characters, appears a lot of "%&/#ºA..." instead of "ñ" or "á", r "ö", for example.
b) I tried another approach. I exported the RIS file from Endnote. However, it takes too long importing the file to Zotero. And this time, the special characters appeared, but the whole idea of Zotero is to get rid of Endnote, right? Maybe I should try Mendeley, because it's easier to export a library from Papers, Mendeley or similar programs (Papers is very cheap compared to Sente, Endnote, etc). So, it will be great of we can still use Papers or Mendeley to export a RIS file, and use Zotero.
c) If I try to import a xml file, or a text file, a pop-up windows appears, showing a error message. I tried to send the report, but Firefox crashes. Data: I use Mac OS 10.5, Firefox 3.0.10. I hope that helps.
1. This is not a general Zotero problem: Zotero deals just fine with special characters (and Chines, arabic, etc. letters for that matter).
2. The problem is in the import/export. I don't quite understand why you do all the importing/exporting - while it's possible, it's very much not an ideal workflow. Especially since papers, Zotero, and Mendeley are pretty similar, why would you want to use more than 1 of them?
3. You can open the RIS file with a text editor. Look at the endnote created one and at the papers created one and what´s different in their treatment of special characters. You can probably 'fix' the RIS file with search-replace.
RIS is an old specification without up-to-date character set guidelines, but Zotero attempts to auto-detect the character set on import and should generally import UTF-8 RIS just fine. You'd have to check the character set Papers is using on export, however, either by asking someone or opening the file in a character-set-aware text editor such as jEdit (any platform) or TextWrangler (OS X) that will show you the character set of the document.
Once you find out, you can force a particular character set on import in the Advanced pane of the Zotero prefs, though we'd be interested to know if the character set of a file wasn't being detected properly.
I posted this because, using the same "workflow", I did not have any problems with the special characters using the previous version of Zotero (1.0.10). However, when I tried the beta version, the problem showed up.
Of course, the problem can be related to Firefox, but in any case it is related to the RIS file from Papers. Why should be working the RIS file with the 1.0.10, but not with the beta, version? I checked out the RIS file and the characters were OK, so the problem was after the creation of the RIS file.
Nonetheless. I started from scratch, deleting all the data from Zotero and starting again. It worked fine with the RIS file exported from Endnote (but that's not useful, because many Zotero users are running away from having to purchase Endnote, right?), and my new attempt with Papers worked fine. Thank you again for the help!
Here's what worked very well for me:
Suppose you want to fix your "WorldCat_2893657.ris" RIS file. Issue this command on Linux/Unix, which will make a corrected RIS file, named "WorldCat_2893657_2.ris" in this case: Essentially, you need to convert from UTF-8 to ISO-8859-1.