On Monday 05 December 2005 12:06 am, Bruce D'Arcus wrote: > Two quick things David: > > 1) I think it'd be helpful to have these sorts of documents as a wiki, > so it's easier and quicker to revise. Do we have that available to us? Yes, we have the 'forum' area. I have put my message there and could try to use it for the discussion.
http://bibliographic.openoffice.org/servlets/ForumMessageList?forumID=5 > 2) On this: > > 5. Refine and Improve the CITEPROC bibliographic formating engine > > I'm not so sure if we want to put it this way. It might be better to > think of citeproc as a solid proof-of-concept of what we need, so that > we simply want citeproc-like functionality. E.g. citeproc is > implemented in XSLT (2.0), but that's just an implementation detail. I > think people skilled in other languages could port it without too much > work. > > Discussions with Florian suggest that it might be a problem to use XSLT > for this sort of core formatting, mostly because of language issues > within OOo (apparently using XSLT requires a Java dependency). We want > to avoid, put simply, that core code like this would have additional > dependencies (like Java, or even Python). > There are at least 2 C++ XSLT processors Sablotron and Xalan, however both are (XSLT) Version 1.0 and the XML Path Language (XPath) Version 1.0. Would these be an option ? > I wonder, though, if another possibility if someone with the skills has > some time and interest, is to port citeproc to Python, and maybe allow > that to be a blueprint of sorts for a C++ coder to reimplement it > there? > It is worth considering as it would get around the complexities of using XSLT as you could implement just the transformation functions you require and not have to deal with XSLT workarounds. > Sort of an aside: when I was playing a bit with a little Ruby ris to > rdf conversion script awhile ago, it occurred to me that doing > citations and bibliographies are really just a specific kind of output > format. It's just that typical bib conversion tools assume fixed > formats, while something like citeproc has a dedicated XML languages to > configure the output. > > I know Ed was thinking of porting bibutils to Python. I wonder if such > a thing could be easily extended to have a kind of CSL module to > configure the output? Hmm ... > It is worth approaching him. We know he is a good programmer/ > Anyway, back to the webpage, I like the plan. We desperately need to > start getting movement here, and breaking things down into the most > essential components is a good step. My work over the last year has > basically been doing that with the OpenDocument XML stuff (well, and > citeproc), but now we need to get something done in OOo per se. > > My only worry is that we can't just rely on C++ people to make any > progress. > Do you mean by "C++ people" the Sun developers or the hard to find C++ programmers ? > Bruce > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] -- ------------------- David N. Wilson Co-Project Lead for the Bibliographic OpenOffice Project http://bibliographic.openoffice.org --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
