Apache/lucene
1
0
Fork 0
mirror of https://github.com/apache/lucene.git synced 2025-02-08 19:15:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

SOLR-6281: Made PostingsSolrHighlighter more configurable

Browse Source 
git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/trunk@1614702 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
David Wayne Smiley 2014-07-30 16:27:13 +00:00
parent 3f4b2b472f
commit 0ffce18e13
2 changed files with 103 additions and 87 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
solr/CHANGES.txt
View File

@ -314,6 +314,8 @@ Other Changes
* SOLR-6290: Harden and speed up CollectionsAPIAsyncDistributedZkTest. (Mark Miller, shalin) * SOLR-6290: Harden and speed up CollectionsAPIAsyncDistributedZkTest. (Mark Miller, shalin)
* SOLR-6281: Made PostingsSolrHighlighter more configurable via subclass extension. (David Smiley)
================== 4.9.0 ================== ================== 4.9.0 ==================
Versions of Major Components Versions of Major Components

184
solr/core/src/java/org/apache/solr/highlight/PostingsSolrHighlighter.java
View File

@ -17,13 +17,6 @@ package org.apache.solr.highlight;
* limitations under the License. * limitations under the License.
*/ */
import java.io.IOException;
import java.text.BreakIterator;
import java.util.Collections;
import java.util.Locale;
import java.util.Map;
import java.util.Set;
import org.apache.lucene.analysis.Analyzer; import org.apache.lucene.analysis.Analyzer;
import org.apache.lucene.index.StoredDocument; import org.apache.lucene.index.StoredDocument;
import org.apache.lucene.search.Query; import org.apache.lucene.search.Query;
@ -47,6 +40,13 @@ import org.apache.solr.search.DocList;
import org.apache.solr.search.SolrIndexSearcher; import org.apache.solr.search.SolrIndexSearcher;
import org.apache.solr.util.plugin.PluginInfoInitialized; import org.apache.solr.util.plugin.PluginInfoInitialized;
import java.io.IOException;
import java.text.BreakIterator;
import java.util.Collections;
import java.util.Locale;
import java.util.Map;
import java.util.Set;
/** /**
* Highlighter impl that uses {@link PostingsHighlighter} * Highlighter impl that uses {@link PostingsHighlighter}
* <p> * <p>
@ -119,86 +119,32 @@ public class PostingsSolrHighlighter extends SolrHighlighter implements PluginIn
final SolrParams params = req.getParams(); final SolrParams params = req.getParams();
// if highlighting isnt enabled, then why call doHighlighting? // if highlighting isnt enabled, then why call doHighlighting?
if (isHighlightingEnabled(params)) { if (!isHighlightingEnabled(params))
SolrIndexSearcher searcher = req.getSearcher();
int[] docIDs = toDocIDs(docs);
// fetch the unique keys
String[] keys = getUniqueKeys(searcher, docIDs);
// query-time parameters
int maxLength = params.getInt(HighlightParams.MAX_CHARS, PostingsHighlighter.DEFAULT_MAX_LENGTH);
String[] fieldNames = getHighlightFields(query, req, defaultFields);
int maxPassages[] = new int[fieldNames.length];
for (int i = 0; i < fieldNames.length; i++) {
maxPassages[i] = params.getFieldInt(fieldNames[i], HighlightParams.SNIPPETS, 1);
}
final IndexSchema schema = req.getSchema();
PostingsHighlighter highlighter = new PostingsHighlighter(maxLength) {
@Override
protected Passage[] getEmptyHighlight(String fieldName, BreakIterator bi, int maxPassages) {
boolean defaultSummary = params.getFieldBool(fieldName, HighlightParams.DEFAULT_SUMMARY, true);
if (defaultSummary) {
return super.getEmptyHighlight(fieldName, bi, maxPassages);
} else {
return new Passage[0];
}
}
@Override
protected PassageFormatter getFormatter(String fieldName) {
String preTag = params.getFieldParam(fieldName, HighlightParams.TAG_PRE, "<em>");
String postTag = params.getFieldParam(fieldName, HighlightParams.TAG_POST, "</em>");
String ellipsis = params.getFieldParam(fieldName, HighlightParams.TAG_ELLIPSIS, "... ");
String encoder = params.getFieldParam(fieldName, HighlightParams.ENCODER, "simple");
return new DefaultPassageFormatter(preTag, postTag, ellipsis, "html".equals(encoder));
}
@Override
protected PassageScorer getScorer(String fieldName) {
float k1 = params.getFieldFloat(fieldName, HighlightParams.SCORE_K1, 1.2f);
float b = params.getFieldFloat(fieldName, HighlightParams.SCORE_B, 0.75f);
float pivot = params.getFieldFloat(fieldName, HighlightParams.SCORE_PIVOT, 87f);
return new PassageScorer(k1, b, pivot);
}
@Override
protected BreakIterator getBreakIterator(String field) {
String language = params.getFieldParam(field, HighlightParams.BS_LANGUAGE);
String country = params.getFieldParam(field, HighlightParams.BS_COUNTRY);
String variant = params.getFieldParam(field, HighlightParams.BS_VARIANT);
Locale locale = parseLocale(language, country, variant);
String type = params.getFieldParam(field, HighlightParams.BS_TYPE);
return parseBreakIterator(type, locale);
}
@Override
protected char getMultiValuedSeparator(String field) {
String sep = params.getFieldParam(field, HighlightParams.MULTI_VALUED_SEPARATOR, " ");
if (sep.length() != 1) {
throw new IllegalArgumentException(HighlightParams.MULTI_VALUED_SEPARATOR + " must be exactly one character.");
}
return sep.charAt(0);
}
@Override
protected Analyzer getIndexAnalyzer(String field) {
if (params.getFieldBool(field, HighlightParams.HIGHLIGHT_MULTI_TERM, false)) {
return schema.getIndexAnalyzer();
} else {
return null;
}
}
};
Map<String,String[]> snippets = highlighter.highlightFields(fieldNames, query, searcher, docIDs, maxPassages);
return encodeSnippets(keys, fieldNames, snippets);
} else {
return null; return null;
SolrIndexSearcher searcher = req.getSearcher();
int[] docIDs = toDocIDs(docs);
// fetch the unique keys
String[] keys = getUniqueKeys(searcher, docIDs);
// query-time parameters
String[] fieldNames = getHighlightFields(query, req, defaultFields);
int maxPassages[] = new int[fieldNames.length];
for (int i = 0; i < fieldNames.length; i++) {
maxPassages[i] = params.getFieldInt(fieldNames[i], HighlightParams.SNIPPETS, 1);
} }
PostingsHighlighter highlighter = getHighlighter(req);
Map<String,String[]> snippets = highlighter.highlightFields(fieldNames, query, searcher, docIDs, maxPassages);
return encodeSnippets(keys, fieldNames, snippets);
}
/** Creates an instance of the Lucene PostingsHighlighter. Provided for subclass extension so that
* a subclass can return a subclass of {@link PostingsSolrHighlighter.SolrExtendedPostingsHighlighter}. */
protected PostingsHighlighter getHighlighter(SolrQueryRequest req) {
return new SolrExtendedPostingsHighlighter(req);
} }
/** /**
@ -262,6 +208,74 @@ public class PostingsSolrHighlighter extends SolrHighlighter implements PluginIn
} }
} }
/** From {@link #getHighlighter(org.apache.solr.request.SolrQueryRequest)}. */
public class SolrExtendedPostingsHighlighter extends PostingsHighlighter {
protected final SolrParams params;
protected final IndexSchema schema;
public SolrExtendedPostingsHighlighter(SolrQueryRequest req) {
super(req.getParams().getInt(HighlightParams.MAX_CHARS, PostingsHighlighter.DEFAULT_MAX_LENGTH));
this.params = req.getParams();
this.schema = req.getSchema();
}
@Override
protected Passage[] getEmptyHighlight(String fieldName, BreakIterator bi, int maxPassages) {
boolean defaultSummary = params.getFieldBool(fieldName, HighlightParams.DEFAULT_SUMMARY, true);
if (defaultSummary) {
return super.getEmptyHighlight(fieldName, bi, maxPassages);
} else {
//TODO reuse logic of DefaultSolrHighlighter.alternateField
return new Passage[0];
}
}
@Override
protected PassageFormatter getFormatter(String fieldName) {
String preTag = params.getFieldParam(fieldName, HighlightParams.TAG_PRE, "<em>");
String postTag = params.getFieldParam(fieldName, HighlightParams.TAG_POST, "</em>");
String ellipsis = params.getFieldParam(fieldName, HighlightParams.TAG_ELLIPSIS, "... ");
String encoder = params.getFieldParam(fieldName, HighlightParams.ENCODER, "simple");
return new DefaultPassageFormatter(preTag, postTag, ellipsis, "html".equals(encoder));
}
@Override
protected PassageScorer getScorer(String fieldName) {
float k1 = params.getFieldFloat(fieldName, HighlightParams.SCORE_K1, 1.2f);
float b = params.getFieldFloat(fieldName, HighlightParams.SCORE_B, 0.75f);
float pivot = params.getFieldFloat(fieldName, HighlightParams.SCORE_PIVOT, 87f);
return new PassageScorer(k1, b, pivot);
}
@Override
protected BreakIterator getBreakIterator(String field) {
String language = params.getFieldParam(field, HighlightParams.BS_LANGUAGE);
String country = params.getFieldParam(field, HighlightParams.BS_COUNTRY);
String variant = params.getFieldParam(field, HighlightParams.BS_VARIANT);
Locale locale = parseLocale(language, country, variant);
String type = params.getFieldParam(field, HighlightParams.BS_TYPE);
return parseBreakIterator(type, locale);
}
@Override
protected char getMultiValuedSeparator(String field) {
String sep = params.getFieldParam(field, HighlightParams.MULTI_VALUED_SEPARATOR, " ");
if (sep.length() != 1) {
throw new IllegalArgumentException(HighlightParams.MULTI_VALUED_SEPARATOR + " must be exactly one character.");
}
return sep.charAt(0);
}
@Override
protected Analyzer getIndexAnalyzer(String field) {
if (params.getFieldBool(field, HighlightParams.HIGHLIGHT_MULTI_TERM, false)) {
return schema.getIndexAnalyzer();
} else {
return null;
}
}
}
/** parse a break iterator type for the specified locale */ /** parse a break iterator type for the specified locale */
protected BreakIterator parseBreakIterator(String type, Locale locale) { protected BreakIterator parseBreakIterator(String type, Locale locale) {
if (type == null || "SENTENCE".equals(type)) { if (type == null || "SENTENCE".equals(type)) {