EzDevInfo.com

Solr/Siren - Insert/Index data using sparql

I am using siren in my application that interacts with solr. I know how to index data into Solr(/solr/update),but I can't figure out how to index using siren. Does it even support indexing? Also, can we use sparql queries to index data in Solr?

SIREn N-Tuples Results

Is it possible in SIREn to work with N-Tuples values(by names) in the SolrDocument? I want to parse query results without regexp of the response string.

Something like:

SolrDocument.getField("ntriple","name"); // get "Jim Maple" value for '<http:/example.org/schema/name> predicate

More details:

//N-Tuple from the IndexQueryNTriple.java file

 private final String[] doc1 = 

 { "http:/example.org/person/jim",

 "<http:/example.org/person/jim> <http:/example.org/schema/type> <http:/example.org
  /schema/Person> .\n" + 

  "<http:/example.org/person/jim> <http:/example.org/schema/name> \"Jim Maple\" .\n" 
 };

 final SolrInputDocument document1 = new SolrInputDocument();

document1.addField("url", doc1[0]);
document1.addField("ntriple", doc1[1]);
add(document1); 

final SolrQuery query2 = new SolrQuery();
query2.setQueryType("siren");
query2.set(SirenParams.NQ, "* <name> 'john AND gartner'"); // ntriple query
final QueryResponse response2 = server.query(query2);
for (SolrDocument d : response2.getResults()) {

///?????? want to get 'name' value from the long N-Triple string with ALL values
 d.getField("ntriple","name"); // get "Jim Maple" value

}

Solr siren, siren.xml

As siren documentation explains, the file located in tomcat/conf/Catalina/localhost must set the path of the war file. I have tried with the absolute path, but I wanted to used a relative path, because it is not only me who will use siren.

siren.xml

<Context docBase="/../../../webapps/siren" debug="0" crossContext="true" >
   <Environment name="siren/home" type="java.lang.String" value="/../../../webapps/siren" override="true" />
</Context>

But running my tomcat I get this error:

GRAVE: Error running static resources
java.lang.IllegalArgumentException: Document /webapps/siren does not exist or is not a readable directory

How could I set up this configuration??

Thanks in advance

Siren "more like this" query

i am using the newest Siren distribution for Solr to index my data and search it. (http://siren.solutions/siren/downloads/)

Is there a simple way to search similar documents in my indexed data. Something similar to the MoreLikeThis query of Solr (https://cwiki.apache.org/confluence/display/solr/MoreLikeThis).

My goal is to find documents that have a similar json structure that the one i am interested in.

best, Bernd

What is the way to execute SIREn query on an exact value search?

I am trying to use Siren as a plugin on ElasticSearch for a poc. I want to do searches which would give exact match on the string values. I set the analyzer in elasticsearch.yml file as follows :

siren: analysis: datatype: http://www.w3.org/2001/XMLSchema#string: index_analyzer: not_analyzed.

I tried to set index:not_analyzed in SIREn mapping file as follows :

{

    "workflow" : {
        "properties" : {
            "_siren_source" : {
                "index" : "not_analyzed",
                "analyzer" : "concise",
                "postings_format" : "Siren10AFor",
                "store" : "no",
                "type" : "string"

            }
        },
        "_siren" : {}
      }
}

But still I am not getting the exact search result. I am getting the partial search result only.

Please help me out.

Also I have data stored as file path values like 'user/cloudera/myFolder' which I need to search on. Right now when I do a search with a value like that the search fails with parse Errors. Why is it throwing the error?

Ruby Siren jsonquery

My json content stored in a file:

{
    "vms": [
        {
            "component": "pgdb",
            "count": "1",
            "endpoints": "80:8080,5432:5432"
        },
        {
            "component": "mq",
            "count": "1",
            "endpoints": "80:8080,5672:5672,15672:15672"
        },
        {
            "component": "ucms",
            "count": "1",
            "endpoints": "80:80,8080:8080"
        },
        {
            "component": "wsgw",
            "count": "1",
            "endpoints": "8080:8080,9093:9093"
        }
    ]
}

in irb :

require 'json'
require 'Siren'

json = File.read('c:\\stack1.json')
irb(main):022:0> Siren.query "$.vms..[? @.component == ucms ]", json
=> []

I am trying to draft a query that will search on component name and return the value of endpoints. Any help/pointers will be appreciated.

Thanx