Question

我正在尝试比较SPARQL中的RDF列表，我有以下的龟文件（下面只是其中的一部分）：

dyldo_o:ob388bc6eb70fae0379746174f3092f22 a qb:Observation ; # One observation 
per seed URI.
  :hasSnapshot dyldo_s:s2012-05-13 ;
  :hasSeedURI "http://whatev.er/3" ;
  :hasRequestChain (
    [ a http:Request ;
      http:requestURI "http://whatev.er/3" ;
      http:resp dyldo_r:rb388bc6eb70fae0379746174f3092f22 
        ]) .
dyldo_r:rb388bc6eb70fae0379746174f3092f22 a http:Response ;
        http:statusCodeValue 200 ;
   http:body (
          [ a rdf:Statement ; 
            rdf:subject <http://ex.org/s3> ;
            rdf:predicate <http://ex.org/p2> ;
            rdf:object <http://ex.org/o1> 
            # Context in the quads equals request URI
          ]
          [ a rdf:Statement ;
            rdf:subject <http://ex.org/s0> ;
            rdf:predicate <http://ex.org/p9> ;
            rdf:object <http://ex.org/o8> 
            # Context in the quads equals request URI
          ]
        ) .

整个文件包含许多快照，每个快照有许多不同的URI。我的SPARQL查询必须能够比较的是http：body，与RDF语句的顺序无关。它应该找出在具有不同http响应主体的给定快照之后拍摄的快照（包含或不包含不在初始快照中的三元组）。

不幸的是，这个和其他变体不起作用：

PREFIX : <http://dyldo.info/vocab#> 
PREFIX http: <http://www.w3.org/2011/http#>
PREFIX dyldo: <http://dyldo.info/#>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>

select distinct ?seedURI ?subject1
WHERE {
  ?observation1 :hasSnapshot ?snapshot1;
                :hasSeedURI ?seedURI ;
                :hasRequestChain/rdf:rest*/rdf:first ?item1 .
  ?snapshot1 <http://purl.org/dc/terms/created> ?date1 .
  ?item1 http:resp ?hash1 .
  ?hash1 http:body/rdf:rest*/rdf:first ?body1 .
  ?observation2 :hasSnapshot ?snapshot2;
                :hasSeedURI ?seedURI2;
                :hasRequestChain/rdf:rest*/rdf:first ?item2 .
  ?snapshot2 <http://purl.org/dc/terms/created> ?date2 .
  ?item2 http:resp ?hash2 .
  ?hash2 http:body/rdf:rest*/rdf:first ?body2 .
  filter(?date1 = "2012-05-06"^^xsd:date && str(?seedURI) = str(?seedURI2) && ?date2 > ?date1)
  filter (exists {
    ?body1  rdf:subject ?subject1;
            rdf:predicate ?predicate1;
            rdf:object ?object1 .
      filter not exists {
      ?body2 rdf:subject ?subject1;
             rdf:predicate ?predicate1;
             rdf:object ?object1 .
    }
    })
}

修改下面的查询似乎有效，至少在我的示例数据上，可能有更好的方法来实现相同的结果。仍然存在的问题是存在空白节点，如果它们只有空白节点，则2个相等的RDF内容将被标记为不相等

PREFIX : <http://dyldo.info/vocab#>
PREFIX http: <http://www.w3.org/2011/http#>
PREFIX dyldo: <http://dyldo.info/#>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>
PREFIX list: <http://jena.hpl.hp.com/ARQ/list#>

select distinct ?seedURI1
WHERE {
?observation1 :hasSnapshot ?snapshot1;
                         :hasSeedURI ?seedURI1 ;
                         :hasRequestChain/rdf:rest* ?rest1 .
?snapshot1 <http://purl.org/dc/terms/created> ?date1 .
?rest1 rdf:first ?requestChain1; rdf:rest rdf:nil .
?requestChain http:resp/http:body ?body1 .
?observation2 :hasSnapshot ?snapshot2;
                  :hasSeedURI ?seedURI2;
                         :hasRequestChain/rdf:rest* ?rest2 .
?snapshot2 <http://purl.org/dc/terms/created> ?date2 .
?rest2 rdf:first ?requestChain2; rdf:rest rdf:nil .
?requestChain http:resp/http:body ?body2 .
filter(?date2 > ?date1 && str(?seedURI1) = str(?seedURI2))
filter (exists {
    ?body1 list:member ?member1 .
    ?member1 rdf:subject ?subject1 .
    ?member1 rdf:object ?object1
    filter not exists {
     ?body2 list:member ?member1 .
     ?member2 rdf:subject ?subject1 .
     ?member2 rdf:object ?object1
            }
}
|| exists {
    ?body2 list:member ?member1 .
     ?member2 rdf:subject ?subject2 .
     ?member2 rdf:object ?object2
    filter not exists {
     ?body1 list:member ?member1 .
     ?member1 rdf:subject ?subject2 .
     ?member1 rdf:object ?object2
            }
    })
}

将N-Triples的RDF列表与SPARQL进行比较

0 个答案: