Re: Using RDF queries with beagle

["D Bera" <dbera web gmail com>]

>  Further,
>   mono RdfQueryTool.exe "uid:jQFjVL1tIE+OAnJArQhca" "" "" ""
>  should return everything about the resource 'uid:jQFjVL1tIE+OAnJArQhca',
>   mono RdfQueryTool.exe "" "beagle:MimeType" "Keyword" "text/plain"
>  should return all resources that are of mime-type 'text/plain', and
>   mono RdfQueryTool.exe "" "beagle:HitType" "Keyword" "File"
>  returns only files. However, those three queries do not work for me
>  right now :-(

Fixed, fixed and fixed (1 was a //TODO, 1 was a doubt I had and was
marked //FIXME and the third was a bug).

>  This is what dBera meant with "rough" ;-)

Well... structured query is not for human beings anyway, they are for
softwares :) People like me will rather use the advanced query syntax
;-)

>  @dBera: I would suggest to modify the RDFQuery so that it only takes
>  subject, predicate and object, where the predicate is the one dumped by
>  beagle-dump-index --fields. The property type can then be derived from
>  the property name. I know that this exposes the internal property name
>  like
>   prop:k:beagle:MimeType
>  but since users of the RDFQuery have to specify Text, Keyword or
>  Internal anyways, they can take the long property name in the first
>  place. This makes things easier.

Go for it. This might need some change in the LuceneQueryingDriver
code; if you get lost there, let me know.

- dBera

-- 
-----------------------------------------------------
Debajyoti Bera @ http://dtecht.blogspot.com
beagle / KDE fan
Mandriva / Inspiron-1100 user