Configure SIREn, Solr on Jetty

(Default settings: SIREn with Lucene 3.5, Solr 3.6 with default Jetty)

1. create a /lib folder under SOLR_HOME/example/solr, add the following from SIREn targets:

siren-core-0.2.3-SNAPSHOT.jar,
siren-qparser-0.2.3-SNAPSHOT.jar
siren-solr-0.2.3-SNAPSHOT.jar

2. modify solrconfig.xml, add following:

 <!-- Example of Registration of the siren query parser. -->  
  <queryParser name="siren" class="org.sindice.siren.solr.SirenQParserPlugin"/>  
  <requestHandler name="siren" class="solr.StandardRequestHandler">  
   <!-- default values for query parameters -->  
    <lst name="defaults">  
     <str name="defType">siren</str>  
     <str name="echoParams">explicit</str>  
                 <!-- Disable field query in keyword parser -->  
     <str name="disableField">true</str>  
     <str name="qf">  
      ntriple^1.0 url^1.2  
     </str>  
     <str name="nqf">  
      ntriple^1.0  
     </str>  
     <!-- the NTriple query multi-field operator:  
       - disjunction: the query should match in at least one of the fields  
       - scattered: each Ntriple patterns should match in at least on of the fields  
     -->   
     <str name="nqfo">scattered</str>  
     <str name="tqf">  
      tabular^1.0  
     </str>  
     <!-- the Tabular query multi-field operator:  
       - disjunction: the query should match in at least one of the fields  
       - scattered: each tabular patterns should match in at least on of the fields  
     -->  
     <str name="tqfo">scattered</str>  
     <str name="fl">  
      id  
     </str>  
    </lst>  
  </requestHandler>  

3. modify schema.xml, add following and rename fields url and id if they exist in the file already.

 <!-- The ID (URL) of the document   
         Use the 'string' field type (no tokenisation)  
      -->  
        <field name="id" type="string" indexed="true" stored="true" required="false"/>  
       <!-- The URL of the document   
         Use the 'text' field type in order to be tokenised  
      -->  
        <field name="url" type="uri" indexed="true" stored="true" required="true"/>  
 <!-- n-triple indexing scheme -->  
        <field name="ntriple" type="ntriple" indexed="true" stored="true" multiValued="false"/>  
     <!-- tabular indexing scheme -->  
     <field name="tabular" type="tabular" indexed="true" stored="false" multiValued="false"/>  

 <!-- A uri field that uses WhitespaceTokenizer and WordDelimiterFilter to   
      split URIs into multiple compoenents. Stopwords is customized by   
      external files.  
      omitNorms is true since it is a short field, and it does not make   
      really sense on URI.  
      Does not use the ASCIIFoldingExpansionFilter since URIs should not  
      contain accented characters.  
   -->  
   <fieldType name="uri" class="solr.TextField" omitNorms="true" positionIncrementGap="100">  
    <analyzer type="index">  
     <tokenizer class="solr.WhitespaceTokenizerFactory"/>  
     <!-- Splits words into subwords based on delimiters  
        - split subwords based on case change  
        - preserveOriginal="1" in order to preserve the original word.  
        Removed split based on numerics to fix SND-355 and SND-1283   
     -->  
     <filter class="solr.WordDelimiterFilterFactory"   
         generateWordParts="1"   
         generateNumberParts="1"   
         catenateWords="0"   
         catenateNumbers="0"   
         catenateAll="0"   
         splitOnCaseChange="1"  
         splitOnNumerics="0"  
         preserveOriginal="1"/>  
     <!-- Filters out those tokens *not* having length min through max   
        inclusive. -->  
     <filter class="solr.LengthFilterFactory" min="2" max="256"/>  
     <!-- Change to lowercase text -->  
     <filter class="solr.LowerCaseFilterFactory"/>  
     <!-- Case insensitive stop word removal.  
      add enablePositionIncrements=true in both the index and query  
      analyzers to leave a 'gap' for more accurate phrase queries.  
     -->  
     <filter class="solr.StopFilterFactory"  
         ignoreCase="true"  
         words="stopwords.txt"  
         enablePositionIncrements="true"  
         />  
    </analyzer>  
    <analyzer type="query">  
     <!-- whitespace tokenizer to not tokenize URI -->  
     <tokenizer class="solr.WhitespaceTokenizerFactory"/>  
     <!-- Filters out those tokens *not* having length min through max   
        inclusive. -->  
     <filter class="solr.LengthFilterFactory" min="2" max="256"/>  
     <filter class="solr.LowerCaseFilterFactory"/>  
     <filter class="solr.StopFilterFactory"  
         ignoreCase="true"  
         words="stopwords.txt"  
         enablePositionIncrements="true"  
         />  
     <!-- Replace Qnames by their name spaces in URIs. -->  
     <filter class="org.sindice.siren.solr.analysis.QNamesFilterFactory"   
         qnames="qnames.txt"/>  
    </analyzer>  
   </fieldType>  
 <!--  
            The SIREn field type:  
                The top-level analyzers must be defined in the top-level analyzer   
    configuration file (ntriple-analyzers.xml) and the datatype analyzers in   
    the datatype analyzer configuration file (ntriples-datatypes.xml).   
                Field norms are not useful for SIREn fields. Set omitNorms to true reduces  
                memory consumption, and improve ranking.  
    omitTermFreqAndPositions *must* be set to false.  
           -->  
   <fieldType name="ntriple" class="org.sindice.siren.solr.schema.SirenField"  
         omitNorms="true"   
         omitTermFreqAndPositions="false"  
         analyzerConfig="tuple-analyzers.xml"  
         datatypeConfig="tuple-datatypes.xml"/>  
   <fieldType name="tabular" class="org.sindice.siren.solr.schema.SirenField"  
         omitNorms="true"   
         omitTermFreqAndPositions="false"  
         analyzerConfig="tuple-analyzers.xml"  
         datatypeConfig="tuple-datatypes.xml"/>  

 <similarity class="org.sindice.siren.similarity.SirenSimilarity"/>  

4. copy the following files from SIREN_HOME/siren_solr/example/solr/config to SOLR_HOME/example/solr/config

tuple-analyzers.xml
tuple-datatypes.xml
qnames.txt

5. Restart default Jetty in Solr by java -jar start.jar

6. test with sample code in SIREN_HOME/siren_solr/example/

The examples are indexed successfully but the queries return no result.

P.S. SIREn doesn't support SPARQL.

sources:

• https://github.com/rdelbru/SIREn/blob/master/siren-solr/example/INSTALL.txt

Install SIREn on Ubuntu 12.04

1. check jdk and maven installation.

$sudo apt-get install maven (*maven 3 will be installed*)

2. run maven at the SIREn directory

$mvn package

3. check jars 

The jar files are located under /target in each folder.

P.S. If the following error occurs:

[ERROR] Failed to execute goal on project siren-core: Could not resolve dependencies for project org.sindice.siren:siren-core:jar:0.2.3-SNAPSHOT: Failure to find com.google.code.caliper:caliper:jar:1.0-SNAPSHOT in https://oss.sonatype.org/content/groups/public/ was cached in the local repository, resolution will not be reattempted until the update interval of oss-sonatype has elapsed or updates are forced -> [Help 1]

Modify pom.xml under siren-core

Change

 <dependency>  
    <groupId>com.google.code.caliper</groupId>  
    <artifactId>caliper</artifactId>  
    <version>1.0-SNAPSHOT</version>  
    <scope>test</scope>  
   </dependency>  

To

 <dependency>  
 <groupId>com.google.caliper</groupId>  
 <artifactId>caliper</artifactId>  
 <version>0.5-rc1</version>  
 <scope>test</scope>  
 </dependency>  

Sources:

• https://github.com/rdelbru/SIREn/wiki/Getting-Started