shared wasabi implementation

[Magnus Bergman]

On Thu, 15 Feb 2007 22:52:59 +0100
"Mikkel Kamstrup Erlandsen" <mikkel.kamstrup at gmail.com> wrote:

> 2007/2/15, Fabrice Colin <fabrice.colin at gmail.com>:
> >
> > On 2/15/07, Jos van den Oever <jvdoever at gmail.com> wrote:
> > > Hello wasabi people,
> > >
> > > Shouldnt we start sharing parts of the code required to implement
> > > wasabi? I'm mainly thinking about a struct in c representing a
> > > parsed wasabi query and functions to de- and serialize these from
> > > either the xml or the user language.
> > >
> > > This should be no more than a couple of files that we can put in
> > > an xsd svn. Requirements to make it palletable for most: few deps
> > > and written in c. For C++ fans we can add a small wrapper.
> > >
> > > What do you think? Can someone here open an svn project for this?
> > >
> > Jean-Francois and I have started a side project called Dijon, with
> > the aim of facilitating code sharing between Recoll and Pinot. Our
> > first target is filters, but this should eventually cover anything
> > that comes out of the Wasabi project, which is actually why we
> > chose the name Dijon ;-) The project's SVN repo is at
> > http://svn.berlios.de/wsvn/dijon There's a mailing list at
> > https://lists.berlios.de/mailman/listinfo/dijon-devel
> 
> 
> Great to hear! I like the name :-)
> 
> Do you have any plans for a standard way of installing new filters?
> Wasabi could use this (as has also been mentioned on Gnomes
> desktop-devel-list).

I haven't checked dijon yet and might have missed something. I assume
filtering refers to converting documents into formats that indexers can
understand (usually plain text and XML), right? Why not just use
gstreamer for this? I have done this myself, but not to gstreamers full
potential. The indexers could just tell gstreamer which mimetype(s) it
can handle and pass some sort of document identifier. Gstreamer can
then figure out how get the document and deliver it in a format the
indexer can understand.