[Nepomuk] Resource clouds

[Marco Martin]

Hi all,

a thing i'm experimenting with for the resource browser of active is being 
able to have informations from nepomuk as "clouds" of attributes, so apart the 
tag cloud (easy, built in ;) i need a similar thing for other properties as 
well, so results of the form property/occurrences

i did some experiments and did manage only in pure Sparql so far (see below)..

would those be possible in the c++ api? also because i would like to 
arbitrarly mix them to have things like
"cardinalities of ratings per activity"
or
"cardinalities of tags of things in a certain date interval"
and in Sparql can become very messy quite quickly ;)


#how many resources of each type

SELECT ?o count(*) as ?count
WHERE {?s rdf:type ?o} 
group by ?o




#how many resources of each rating

SELECT ?rating count(*) as ?count
WHERE {?s 
<http://www.semanticdesktop.org/ontologies/2007/08/15/nao#numericRating> 
?rating} 
group by ?rating




#how many resources of each type, related to the activity 
nepomuk:/res/47d7281f-dd5c-4162-8026-5bc069f91945

select distinct ?t count(*) as ?count
where { <nepomuk:/res/47d7281f-dd5c-4162-8026-5bc069f91945> 
<http://www.semanticdesktop.org/ontologies/2007/08/15/nao#isRelated> ?r .
?r rdf:type ?t .
?r <http://www.semanticdesktop.org/ontologies/2007/08/15/nao#userVisible> ?v1 
.
FILTER(?v1>0) . }
group by ?t



#how many resources per activity

select distinct ?a count(*) as ?count
where { ?a 
<http://www.semanticdesktop.org/ontologies/2007/08/15/nao#isRelated> ?r 
. ?a rdf:type kext:Activity
} group by ?a

--
Marco Martin