[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#598160: RFP: biber -- BibTeX replacement for users of biblatex

Package: wnpp
Severity: wishlist

* Package name    : biber
  Version         : 0.5.5
  Upstream Author : Mailadress_was_not_Available@the_sourceforge_project.site
* URL             : http://biblatex-biber.sourceforge.net/
* License         : Artistic License, GNU General Public License (GPL)
  Programming Lang: Perl
  Description     : BibTeX replacement for users of biblatex

The biblatex package by Philipp Lehman is likely to become the definitive citation management tool for LaTeX users. Biblatex relies on the venerable BibTeX program only for sorting and generating a very generic bbl file without any formatting instruction. Everything else is taken care of by biblatex, which provides a powerful and flexible macro interface for authors of citation styles.

With Biber it is no longer necessary to rely on BibTeX. For maximal portability, the current version includes a Pure Perl BibTeX parser with a Parse::RecDescent grammar, but if available it will use the much faster Text::BibTeX module which relies on the btparse C library. The objective of the first development phase is to have a robust and reliable emulation of the BibTeX processor with the biblatex.bst style file. In other words, given the same data file as input, biber should output a functionally identical .bbl file as BibTeX.

In a future version, Biber will also support a new XML format tailored specifically for biblatex data, unimaginatively called BibLaTeXML. It owes its inspiration partly to the BibTeXML format, but it is far more complex. RelaXNG and W3C schemata will be provided, as well as a tool for converting .bib files to biblatexml. To parse BibLaTeXML databases the Perl module XML::LibXML is required. There is also experimental support for Berkeley DBXML databases (with the included Perl module Sleepycat::DbXml).

The advantages of no longer being dependent upon BibTeX are obvious: full Unicode support, no memory limitations, extensibility, etc. In the future, support for communication with embedded or relational databases (e.g. Berkeley DB, SQLite, MySQL, PostgreSQL) or remote resources via network protocols (such as SRU) might also be implemented.

Reply to: