bibtex in/out and more advance exports

mcomp

Hi, just started using zotero and its best research management too I have seen, great work. I have one grumble and one request (both maybe just my inexperience).

1) I'm a Bibtex user so I imported a bibtex file to start a zotero collection. Two problems: entries that have { .... {BLAA}...} to preserve upper case have had the brackets stripped; and the pages have been changed from 34--45 to 34-45. I could go through and edit the entire zotero collection, but for big imports this will drive me crazy, is there a better way?

2) The JabRef tool seems to be able to do great exports - totally customizable. You can basically define your own export translation to anything you want (see some examples at http://www.markschenk.com/tools/jabref/). It would be great if zotero could have a programable export mechanism like this, so that one could automatically mark up document names, plus notes, plus web links, or whatever, to html pages with embedded javascript and the like.

bdarcus

1) what Zotero does on import is appropriate it seems to me. Zotero is a modern application with unicode support, so there's no reason to be, for example, using ASCII characters to represent en-dashes.

The question is, what do they do on export? Obviously, it should support the appropriate way to encode en-dashes depending on the output encoding (see other threads on the issues there).

The issue with acronyms and personal names is a general issue that needs a general solution; not simply preserving format-specific solutions. But indeed, it's an important issue.

2) I agree in principle and there's been some discussion of this sort of thing, though keep in mind JabRef is a dedicated BibTeX tool. Zotero is not. Again: we need general solutions, and ones that will work well in the context of contemporary technologies like XML and RDF.

mcomp

I agree with you completely on the difference between the internal and external representations, which is why I'm a little surprised by the arguments I have been reading in the other threads. I would have thought that the way it should work is exactly as you hint in your post.

Zotero should have exactly one output format: a nicely designed XML format. From this, all the other transformations are (essentially) XSLT (or something similar). So for example when I press 'generate Bibtex' zotero spits out its standard XML document and this is then passed through the Bibtex-XSLT transform and given to me. Same for all other outputs. This would mean that the zotero output wasn't an in-house solution, but (as you point out) an effective use of contemporary standards and technologies.

I know that there are concerns about having nice tools for allowing transformations to be defined graphically or through other non expert means. However, as a first cut, this would standardise the output generation and expose the full power of such manipulation to users who can take advantage of it.

(again, apologies for any blatant ignorance here - I haven't flipped the hood to see exactly how zotero works yet)

rock_fabric

How and where may I customize the BibTeX key Zotero assigns to each library entry. When I export to BibTeX format sometimes the keys are like
"mueller_computer_modelling"
and sometimes its like
"mueller_computer_2003"
Where can I specify a homogeneous syntax? One could use variables like they are used in MP3 tagging programs. Let's say
%first_author__%first_word_of_title__%year
or what ever one prefer.

dstillman

How and where may I customize the BibTeX key Zotero assigns to each library entry.

That sort of customizability (in addition to a field for manually overriding the key) is planned for Zotero 1.5, but in the meantime you'd have to edit the BibTeX translator yourself by copying it from scrapers.sql (in the extension directory or 1.0 SVN branch) to a separate text file, editing it, and importing it into Zotero using the SQLite command line client with Firefox closed.