gCubeSystem
/
social-data-search-client
16
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

added method to query a single field. Minor methods/classes'names fix 

git-svn-id: http://svn.research-infrastructures.eu/public/d4science/gcube/trunk/social-networking/social-data-search-client@141582 82a268e6-3cf1-43bd-a215-b396298e98cf
This commit is contained in:
Costantino Perciante 2017-01-15 18:50:20 +00:00
parent d9702ae4ee
commit 306defc345
5 changed files with 127 additions and 55 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
distro/changelog.xml
View File

@ -1,4 +1,8 @@
<ReleaseNotes>
<Changeset component="org.gcube.socialnetworking.social-data-search-client.1-2-0"
date="2017-02-01">
<Change>Added method to query a single field</Change>
</Changeset>
<Changeset component="org.gcube.socialnetworking.social-data-search-client.1-1-0"
date="2016-10-01">
<Change>Minor bug fixes</Change>

2
pom.xml
View File

@ -11,7 +11,7 @@
<groupId>org.gcube.socialnetworking</groupId>
<artifactId>social-data-search-client</artifactId>
<version>1.1.0-SNAPSHOT</version>
<version>1.2.0-SNAPSHOT</version>
<packaging>jar</packaging>
<name>Social Data Search Client</name>

34
src/main/java/org/gcube/socialnetworking/social_data_search_client/ElasticSearchClientInterface.java
View File

@ -1,34 +0,0 @@
package org.gcube.socialnetworking.social_data_search_client;
import java.util.List;
import java.util.Set;
import org.gcube.portal.databook.shared.EnhancedFeed;
/**
* The ElasticSearch client interface.
* @author Costantino Perciante at ISTI-CNR
* (costantino.perciante@isti.cnr.it)
*
*/
public interface ElasticSearchClientInterface {
/**
* Given a query, the method find matching enhanced feeds into the elasticsearch index and return
* at most <b>quantity</b> hits starting from <b>from</b>.
* @param query the query to match
* @param vreIDS specifies the vre(s) to which the returning feeds must belong
* @param from start hits index
* @param quantity max number of hits to return starting from <b>from</b>
* @return A list of matching enhanced feeds or nothing
*/
List<EnhancedFeed> searchInEnhancedFeeds(String query, Set<String> vreIDS, int from, int quantity);
/**
* Delete from the index a document with id docID.
* @param docID the id of the doc to delete
* @return true on success, false otherwise
*/
boolean deleteDocument(String docID);
}

45
src/main/java/org/gcube/socialnetworking/social_data_search_client/GCubeSearchClient.java Normal file
View File

@ -0,0 +1,45 @@
package org.gcube.socialnetworking.social_data_search_client;
import java.util.List;
import java.util.Set;
import org.gcube.socialnetworking.social_data_indexing_common.utils.SearchableFields;
import org.gcube.portal.databook.shared.EnhancedFeed;
/**
* The GCubeSearchClient client interface to search in social data.
* @author Costantino Perciante at ISTI-CNR (costantino.perciante@isti.cnr.it)
*/
public interface GCubeSearchClient {
/**
* Given a query, the method find matching enhanced feeds into the elasticsearch index and return
* at most <b>quantity</b> hits starting from <b>from</b>. A multimatch query is performed against all
* searchable fields.
* @param query the query to match
* @param vreIDS specifies the vre(s) to which the returning feeds must belong
* @param from start hits index
* @param quantity max number of hits to return starting from <b>from</b>
* @return A list of matching enhanced feeds or nothing
*/
List<EnhancedFeed> search(String query, Set<String> vreIDS, int from, int quantity);
/**
* Given a query, the method find matching enhanced feeds into the elasticsearch index and return
* at most <b>quantity</b> hits starting from <b>from</b>. The query is performed against one of the searchable fields.
* @param query the query to match
* @param vreIDS specifies the vre(s) to which the returning feeds must belong
* @param from start hits index
* @param quantity max number of hits to return starting from <b>from</b>
* @param field the field against which the query is performed
* @return A list of matching enhanced feeds or nothing
*/
List<EnhancedFeed> searchInField(String query, Set<String> vreIDS, int from, int quantity, SearchableFields field);
/**
* Delete from the index a document with id docID.
* @param docID the id of the doc to delete
* @return true on success, false otherwise
*/
boolean deleteDocument(String docID);
}

97
src/main/java/org/gcube/socialnetworking/social_data_search_client/ElasticSearchClientImpl.java → src/main/java/org/gcube/socialnetworking/social_data_search_client/GCubeSearchClientImpl.java
View File

@ -6,7 +6,7 @@ import java.net.UnknownHostException;
import java.util.ArrayList;
import java.util.List;
import java.util.Set;
import org.gcube.socialnetworking.social_data_indexing_common.utils.SearchableFields;
import org.codehaus.jackson.map.ObjectMapper;
import org.elasticsearch.action.delete.DeleteResponse;
import org.elasticsearch.action.search.SearchResponse;
@ -20,22 +20,18 @@ import org.elasticsearch.index.query.QueryBuilders;
import org.elasticsearch.index.query.TermsQueryBuilder;
import org.elasticsearch.search.SearchHit;
import org.gcube.portal.databook.shared.EnhancedFeed;
import org.gcube.socialnetworking.social_data_indexing_common.utils.ElasticSearchRunningCluster;
import org.gcube.socialnetworking.social_data_indexing_common.utils.IndexFields;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
/**
* The elasticsearch client for gcube portlets.
* @author Costantino Perciante at ISTI-CNR
* (costantino.perciante@isti.cnr.it)
*
* @author Costantino Perciante at ISTI-CNR (costantino.perciante@isti.cnr.it)
*/
public class ElasticSearchClientImpl implements ElasticSearchClientInterface{
public class GCubeSearchClientImpl implements GCubeSearchClient{
//logger
public static final Logger logger = LoggerFactory.getLogger(ElasticSearchClientImpl.class);
public static final Logger logger = LoggerFactory.getLogger(GCubeSearchClientImpl.class);
// the elasticsearch client
private TransportClient client;
@ -54,7 +50,7 @@ public class ElasticSearchClientImpl implements ElasticSearchClientInterface{
* @param scope the scope in the infrastructure
* @throws Exception
*/
public ElasticSearchClientImpl(String scope) throws Exception {
public GCubeSearchClientImpl(String scope) throws Exception {
// retrieve ElasticSearch Endpoint and set hosts/port number
ElasticSearchRunningCluster elasticCluster = new ElasticSearchRunningCluster(scope);
@ -96,34 +92,33 @@ public class ElasticSearchClientImpl implements ElasticSearchClientInterface{
}
}
logger.debug("Connection to ElasticSearch cluster done.");
logger.info("Connection to ElasticSearch cluster done.");
}
@Override
public List<EnhancedFeed> searchInEnhancedFeeds(String query, Set<String> vreIDS, int from, int quantity){
public List<EnhancedFeed> search(String query, Set<String> vreIDS, int from, int quantity){
List<EnhancedFeed> toReturn = new ArrayList<>();
if(from < 0 || quantity <= 0)
if(from < 0 || quantity <= 0 || vreIDS.isEmpty())
return toReturn;
// build the query
MultiMatchQueryBuilder mq = QueryBuilders.multiMatchQuery(
query,
IndexFields.EF_FEED_AUTHOR + "^3", // weight of 3 for feed's author
IndexFields.EF_FEED_TEXT + "^2", // weight of 2 for feed's description
IndexFields.EF_ATTACHMENT_NAME,
IndexFields.EF_PREVIEW_DESCRIPTION,
IndexFields.EF_COMMENT_TEXT + "^2",
IndexFields.EF_COMMENT_FULL_NAME)
SearchableFields.POST_AUTHOR + "^3", // weight of 3 for feed's author
SearchableFields.POST_TEXT + "^2", // weight of 2 for feed's description
SearchableFields.ATTACHMENT_NAME,
SearchableFields.PREVIEW_DESCRIPTION,
SearchableFields.COMMENT_TEXT + "^2",
SearchableFields.COMMENT_AUTHOR)
.type(Type.MOST_FIELDS);
// logger.debug(mq.toString());
// filter on vre
BoolQueryBuilder filter = QueryBuilders.boolQuery();
TermsQueryBuilder queryFilter = QueryBuilders.termsQuery(IndexFields.EF_FEED_VRE_ID, vreIDS);
TermsQueryBuilder queryFilter = QueryBuilders.termsQuery(SearchableFields.POST_VRE_ID, vreIDS);
filter.should(queryFilter);
//logger.debug(filter.toString());
@ -167,6 +162,68 @@ public class ElasticSearchClientImpl implements ElasticSearchClientInterface{
logger.debug("Returning " + toReturn.size() + " results");
return toReturn;
}
@Override
public List<EnhancedFeed> searchInField(String query, Set<String> vreIDS,
int from, int quantity, SearchableFields field) {
List<EnhancedFeed> toReturn = new ArrayList<>();
if(from < 0 || quantity <= 0 || field == null || vreIDS.isEmpty())
return toReturn;
// build the query
MultiMatchQueryBuilder mq = QueryBuilders.matchQuery(field, query);
// logger.debug(mq.toString());
// filter on vre
BoolQueryBuilder filter = QueryBuilders.boolQuery();
TermsQueryBuilder queryFilter = QueryBuilders.termsQuery(SearchableFields.POST_VRE_ID, vreIDS);
filter.should(queryFilter);
//logger.debug(filter.toString());
// final filtered query
BoolQueryBuilder filteredQuery = QueryBuilders.boolQuery();
filteredQuery.must(mq);
filteredQuery.filter(filter);
//logger.debug(filteredQuery.toString());
SearchResponse response = client.prepareSearch(IndexFields.INDEX_NAME)
.setQuery(filteredQuery)
.setFrom(from)
.setSize(quantity)
.setExplain(true)
.execute()
.actionGet();
logger.debug("The search took " + response.getTookInMillis() + " ms");
SearchHit[] results = response.getHits().getHits();
logger.debug("Number of hits is " + results.length);
ObjectMapper mapper = new ObjectMapper();
// rebuild objects
for (SearchHit hit : results) {
EnhancedFeed enhFeed;
try {
enhFeed = mapper.readValue(hit.getSourceAsString(), EnhancedFeed.class);
toReturn.add(enhFeed);
} catch (IOException e) {
logger.error(e.toString());
}
}
logger.debug("Returning " + toReturn.size() + " results");
return toReturn;
}
@Override
public boolean deleteDocument(String docID) {