importing keywords/tags

TessaC

I was glad to see that when I imported an RIS file generated by EndNote X, that keywords were put in the tags. I wasn't sure if that mapping existed. However, zotero created 1 tag with all of the keywords as that tag.

So, what is the delimiter on tags when importing? Or is there one? EndNote uses a line break to delimit its keywords. I'd like to be able to import keywords into tags such that each keyword created 1 tag. Is there a way to do this?

Thanks,
Tessa

bdarcus

FYI, this is an incredibly old bug in Endnote's RIS export (which I'm surprised they STILL haven't fixed!). Each keyword should get its own tag. So:

KY - one
KY - two
KY - three

It probably wouldn't be too hard for the Zotero guys to work around the bug, though I don't want to speak for them. In the meantime, maybe submit another bug report to ISI? It really is their problem.

TessaC

Oh, ok. I'll try to do some sort of regex to stick that in.

bdarcus

You probably already picked up my mistake; am pretty sure what I have as "KY" ought to be "KW."

gm

I encountered this also. Thanks to the comments here, I edited the
file exported from EndNote before reading it into Zotero. But it raises a general point - eventually Zotero will have to import files generated by EndNote just about perfectly. I am having other problems too, such as my abstract in EndNote disappearing. I am having to check manually every reference I imported from EndNote.

migugg

I have the problem that in my endnote library I separate keywords with a comma and when I import my endnote library into Zotero this leads to tags that comprise several keywords at once. Thus if in have in endnote the keywords "power, buildings, France" as three keywords, they end up in zotero as one tag "power, buildings, France".

It would be nice, if you could adjust either the import filter to make this work or to implement a "search and replace" function to get rid of the commas.
(This is the only reason why I haven't migrated my extensive endnote-database yet to your wonderful tool!).

migugg

Sorry, just found out that I can overcome the problem by using another endnote output style (bibtex export), but then there is still the problem that several fields (such as "url" or "link to pdf" and my own custom fields) are not imported and that several reference types (such as "electronic source") are not imported as these types, but as the reference type "book".
I think the latter problem could be overcome by generating new reference types in the endnote "output style", but to do so, I would have to know if and how zotero reads these reference types in its import filter.
is there a way to tinker with zoteros own import filter or at least to have a look what kind of reference types it knows? I could not find any description of its location in the forums.
I am using Endnote 9 on Mac OS X 10.4 and the latest Zotero release.

PeterSmith

I'm having problems too importing from EndNote. I've about 1,300 and I'm using BibText format. I've tried separting my keywords with comma, e.g. "strategy as practice, middle managers, becoming". This results in three tags "strategy as practice, "middle managers" and "becoming". That is exactly what I want to happen.

Alas, if there is only one 'keyword', say "strategy as practice", the result is three tags; "stategy", "as", and "practice". This isn't ideal.

Any ideas?

noksagt

Just thinking out loud, but a potential work around (dirty hack) may be to use a trailing comma and/or a trailing comma plus a dummy tag that would be applied to every reference.

This dummy tag could be useful (date and/or method of import and/or name of original endnote file, for example) or could be something that you remove after import.

PeterSmith

Thanks noksagt, I had thought of something similar -- the challenge is finding a way to do that automatically in EndNote (the thing I'm moving from). EndNote's search and replace isn't quite up to that.

noksagt

You can either customize the Endnote export style or could peform a regular expression search/replace on the generated file (in a text editor, for instance).

philosopherdog

Hi,
I'm having the same problem. Has anyone found a solution to this problem that is more elegant than adding all these additional tags to the output filter?

philosopherdog

I think the solution is to use endnote to replace commas and put a carriage return in its place. Now it should import correctly!

dstillman

We've put in a workaround for this in the RIS translator, but the new version requires additional features added since 1.0.7 and hasn't been pushed to existing clients. 1.0.8 will include the updated version.

1.0 dev branch users can also use "Reset Translators and Styles" in the Advanced pane of the Zotero prefs to get the latest version.

philosopherdog

Actually, in the mean time I have a work around for this issue in another post in the import/export section.