Jason Ronallo wrote:
> Hi,
> I'm using Zotero 1.0.7 with Zotz 1.8.2. When uploading a collection I
> get the following error:
>
> edu.mit.simile.babel.bibtex.ParseException "}" at line 1071, column 1.
> Was expecting: "=" ...
>
> If I export my whole collection as Bibtex first and try to upload it I
> get this error:
>
> edu.mit.simile.babel.bibtex.ParseException: Encountered "}" at line
> 1168, column 1.
> Was expecting:
> "=" ...
>
> I think the relevant lines from the Bibtex file are:
>
> _at_misc{__????-2
> },
>
> Removing those lines it gets further until it hits a similar line:
>
> _at_misc{__????-5
> },
>
> Continuing on I deleted several similar lines.
>
> Looks like it might be an empty citation?
Jason,
we did encounter those empty citations before but we did allow Citeline
to parse them and ignore them. I'm not sure the problem is that the
citations are empty, but those "????" characters might not be part of
the 7bit ASCII charset (as BibTeX requires) so the parser won't be able
to continue there.
Try changing those "__????-2" with "blah" and see what happens.
> Do you know a way to find
> these within Zotero so I can delete them, or can Zotz be made to
> handle these gracefully?
Zotz is a very thin layer around zotero that just uses Zotero's own
internal BibTeX exporter and sends it to the entry Citeline URL. We
could do some post cleanup of the bibtex, but it's a lot faster to
update a server than a lot of installed zotz.
We already perform a considerable set of pre-processing cleanup on the
incoming BibTeX... we have issues with non-ASCII characters though,
because if we start to allow multiple encodings in there, it will become
a huge mess, since BibTeX is only expecting ASCII (the non-ASCII should
be escaped with LaTeX codes such as \'a ) and there would be no way for
us to know what encoding a particular string comes in with (since bibtex
won't be telling us).
We preferred to make the parser fail if the encoding is wrong, if only
to apply a little incentive on the users (and, by consequence, to the
bibtex exporting tool) to cleanup their bibtex nuances in respect to
encoding.
One way to work around this issue, though, is to create a new
'collection' and manually copy the publications you're interested in
from your zotero library to another collection and export only that. In
my experience, that seems to get rid of empty citations (which seem to
be index left-overs and happen only in the main zotero library).
HTH
--
Stefano Mazzocchi
Digital Libraries Research Group Research Scientist
Massachusetts Institute of Technology
E25-131, 77 Massachusetts Ave skype: stefanomazzocchi
Cambridge, MA 02139-4307, USA email: stefanom at mit . edu
-------------------------------------------------------------------
Received on Mon Aug 25 2008 - 12:35:46 EDT