Introduction

This wiki page provides information about how to implement a customized search site with functionalities similar to http://beta.vivosearch.org/.

8/2013: Note that the code linked from this page is obsolete and is being redeveloped as presented at the 2013 VIVO Conference

• see also the VIVO Multisite Search wiki space

Useful resources:

• mailing list: vivo-dev-all
• list archive: List Archive
• main wiki: Wiki Home
• Live IRC chat room: #VIVO

2011 Downloads

• Apache Solr from http://lucene.apache.org/solr/
• Linked Data Index Builder from https://sourceforge.net/p/vivo/tools/code/
• Drupal from http://drupal.org/
• VIVO search Drupal module from https://github.com/milesworthington/vivosolr
• MySQL from http://www.mysql.com/

Setting up the applications

Dates from late 2011

1. Apache Solr

• Unpack the downloaded Apache Solr file
• Copy the files schema.xml and solrconfig.xml from /LinkedDataIndexer/solr to $APACHE_SOLR_HOME/example/solr/conf
• Go to $APACHE_SOLR_HOME/example and run java -jar start.jar to start solr
• Open http://localhost:8983/solr and check if solr is running

2. LinkedDataIndexer

• Go to the directory of LinkedDataIndexer
• Copy the example file /src/main/scala/edu/cornell/indexbuilder/configurations/ExampleConfig.scala to your customized scala file (e.g. MyInstitutionTest.scala)

• Modify the copied file from:

  object ExampleConfig {
    val siteName = "Weill Cornell Medical College"
    val siteUrl = "http://vivo.med.cornell.edu"    
    val siteVivoVersion = VitroVersion.r1dot2
  
    val solrUrl = "http://exampleSolrServer.example.edu:8080/solr"
    val classUris = List( """http://vivoweb.org/ontology/core#Postdoc""" )
  }
  

to your customized institutional configurations:

object MyInstitutionTest {
  val siteName = "My Institution"
  val siteUrl = "http://vivo.myinstitution.edu"    
  val siteVivoVersion = VitroVersion.r1dot2

  val solrUrl = "http://localhost:8983/solr"
  val classUris = List( """http://vivoweb.org/ontology/core#Postdoc""" )
}

• The file /LinkedDataIndexer/src/main/resources/akka.conf needs to be modified by changing the statement: version = "1.1.3" to version = "1.2"
• The following are needed to compile or run the index builder
• : scala from http://www.scala-lang.org
• : sbt from https://github.com/harrah/xsbt/wiki/
• : slf4s_2.9.1-1.0.7.jar
• : slf4j-api-1.6.3.jar
• : akka-actor-1.2.jar
• : commons-httpclient.jar
• : slf4j-log4j12-1.6.3.jar

• Go to the directory of LinkedDataIndexer
• Type sbt to start up
• Type run
• : Note: You may encounter errors complaining about missing jar files. In this case, you will need to add more jar files to the lib directory.

• Select your customized configuration to run the builder
• : Note: Apache Solr needs to be running at the time of building the index.

3. Drupal

• Follow the instructions to install and configure a basic Drupal site
• Copy vivosolr to $DRUPAL_HOME/modules
• Open your local drupal site
• Log in as administrator
• Click on the Modules tab and enable "VIVO Solr Search Integration"
• Click on the Configuration tab and configure "VIVO search" to your solr installation for "Solr server URL".
• To allow all viewers to use the search form without logging in, do the following:
• : Click on the Modules tab
• : Click on the "Permissions" text link for "Block"
• : Check the boxes for "ANONYMOUS USER" and "AUTHENTICATED USER" for "Use search" and "Use advanced search"
• Log out
• Test the search capability