Import from EndNote X using RIS format

gjdv

I tried to import the RIS file I exported from Endnote X, however got the following error:

Fout: [Exception... "Component returned failure code: 0x80520012 (NS_ERROR_FILE_NOT_FOUND) [nsIFile.copyTo]" nsresult: "0x80520012 (NS_ERROR_FILE_NOT_FOUND)" location: "JS frame :: chrome://zotero/content/xpcom/attachments.js :: importFromFile :: line 78" data: no] Bronbestand: file:///C:/Documents%20and%20Settings/nlv20272/Application%20Data/Mozilla/Firefox/Profiles/3lvjdgko.default/zotero/translators/RIS.js Regel: 0

Upon closer inspection, the RIS file turned out to start with the following three characters  (or in hex: EF BB BF), which were nog generally shown by most editors. Once removed, Zotero gets one bit further, but now gives the message that it could not save the object, that Zotero is busy and I should try again once it finishes. From that moment, there is a progress bar in Zotero showing the process of importing objects, which stays there forever, the only solution is closing firefox and opening again, then it becomes visible that Zotero did create a folder to put the imported objects in, however it did not actually import anything.

I checked the permission to write in the output folder, no issue, there are also some other documents already placed there by Zotero. The L1 entry nicely contains "file://" and the absolute path to the pdf.

Does anyone know how to solve this?

noksagt

Those characters are a BOM & Zotero should really import the file whether or not they are present:
http://forums.zotero.org/discussion/7836/failed-import-on-ris-file-thats-been-created-with-save-as-#Item_5

How big is the file and how long have you waited? Have you tried to import smaller subsets of the file to start off with and/or just giving it longer to perform the import?

gjdv

Thank you for the reply, noksagt. I did not yet come across the exact forum-discussion you refered to, but it didn't fully solve my problem.
I've selected out of the ris file 1 single article description and tried to import that (through copying it into a new ris-file). The pdf that is refered to from the item is 186 kB in size and Zotero has been crunching for at least 1.5 hours now, still showing the exact same progress bar etc.

dstillman

The pdf that is refered to from the item is 186 kB in size and Zotero has been crunching for at least 1.5 hours now, still showing the exact same progress bar etc.

Then it failed, likely with an (uncaught) error in Report Errors. Is it the same error as above, or another? If the same as above, what's the path that triggered it?

gjdv

I'm sorry for not being fully clear.
Select import from the gear menu, select ris file (named EndNoteLibraryExportPart.ris) containing just one item refering to a 186 kB pdf, content:

---

TY - JOUR
AU - Alexander, D.M.
AU - Trengove, C.
AU - Johnston, P.
AU - Cooper, T.
AU - August, J.P.
AU - Gordon, E.
PY - 2005
TI - Separating individual skin conductance responses in a short interstimulus-interval paradigm
SP - 116-123
JF - Journal of Neuroscience Methods
VL - 146
N1 - Separating individual skin conductance responses in a short interstimulus-interval paradigm
L1 - file://H:/data/Data/Articles/2005_Alexander&al_Separating individual skin conductance responses in a short interstimulus-interval paradigm.pdf
ID - 114
ER -

---

On the right bottom of the screen a message of Zotero becomes visible (black letters, grey background, red outline) telling me it could not save the object, that Zotero is busy and I should try again once it finishes (I'm sorry for having the Dutch version of Zotero installed, so I can't give you the exact English formulation). A progress bar becomes available (and stays there through the night...) and Zotero is made inactive, that means, I can not check whether error reporting is available.
There is NO error/warning/info message in my firefox error log.
The only way to be able to check Zotero again is to restart firefox, but that resets Zotero's error own log.

[update]
With javascript.options.showInConsole on there are two warnings while starting firefox (and Zotero):
Waarschuwing: Fout tijdens het parsen van waarde voor ‘min-height’. Declaratie genegeerd.
Bronbestand: chrome://zotero/skin/bindings/itembox.css
Regel: 73
Waarschuwing: Selector verwacht. Regelset genegeerd vanwege foute selector.
Bronbestand: chrome://zotero-platform/content/itembox.css
Regel: 12
[first one says: ignoring declaration of min-height, second one says ignoring rule set due to wrong selector]

And during the execution of the import another (though not chrome://zotero***) appears:
Waarschuwing: Toetsgebeurtenis niet beschikbaar op sommige toetsenbordindelingen: key=“z” modifiers=“accel alt”
Bronbestand: chrome://browser/content/browser.xul
Regel: 0
[key-event not available on some keyboard settings]

dstillman

I'm guessing it's the ampersand in the filename that's causing the error. Can you rename the file to remove the ampersand and edit the RIS entry and try again?]

Also make sure that entering the file:// URL in Firefox brings up the file.

gjdv

Thank you, Dan, for your help.
I tried to rename the file to test.pdf, even moved it to C:\ (a local drive; and changing the L1 description to "L1 - file://C:/test.pdf").
The link file://C:/test.pdf nicely loads when put in the firefox location bar (just as well as the original filename)

Any more suggestions on getting the cause of the error?

dstillman

Does it import without the L1 line?

gjdv

That's a good one. I didn't even think of some other problem than storing a file to which the ris refers (also because other imports (using e.g. bibtex) still worked as well as manual inputs). Thanks to your comment I tried some more.

I found out the following: the original ris file (as saved by EndNote) is apparently badly formatted. When I copy all text, create a new blank file, insert the text and save it explicitely as a UTF8 file, Zotero processes the entries as it should, that is, it processed a bit, then stopped processing and became active again (that is, I could click buttons in Zotero). By binary searching :) I found the entries (only 4) that cause an issue and will try and find out the problem with those, but the good news is that all other entries were parsed well and files inserted by Zotero.

So apparently, Zotero is very sensitive to the exact formatting of the file (and with that I mean the UTF-8 with/without BOM etc).