Apache
/
lucene
mirror of https://github.com/apache/lucene.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

SOLR-9935: UnifiedHighlighter, when hl.fragsize=0 don't do fragmenting

This commit is contained in:
David Smiley 2017-01-17 08:06:21 -05:00
parent 43874fc5b5
commit ed513fdee7
2 changed files with 12 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
solr/core/src/java/org/apache/solr/highlight/UnifiedSolrHighlighter.java
View File

@ -295,6 +295,13 @@ public class UnifiedSolrHighlighter extends SolrHighlighter implements PluginInf
@Override @Override
protected BreakIterator getBreakIterator(String field) { protected BreakIterator getBreakIterator(String field) {
// Use a default fragsize the same as the regex Fragmenter (original Highlighter) since we're
// both likely shooting for sentence-like patterns.
int fragsize = params.getFieldInt(field, HighlightParams.FRAGSIZE, LuceneRegexFragmenter.DEFAULT_FRAGMENT_SIZE);
if (fragsize == 0) { // special value; no fragmenting
return new WholeBreakIterator();
}
String language = params.getFieldParam(field, HighlightParams.BS_LANGUAGE); String language = params.getFieldParam(field, HighlightParams.BS_LANGUAGE);
String country = params.getFieldParam(field, HighlightParams.BS_COUNTRY); String country = params.getFieldParam(field, HighlightParams.BS_COUNTRY);
String variant = params.getFieldParam(field, HighlightParams.BS_VARIANT); String variant = params.getFieldParam(field, HighlightParams.BS_VARIANT);
@ -302,9 +309,6 @@ public class UnifiedSolrHighlighter extends SolrHighlighter implements PluginInf
String type = params.getFieldParam(field, HighlightParams.BS_TYPE); String type = params.getFieldParam(field, HighlightParams.BS_TYPE);
BreakIterator baseBI = parseBreakIterator(type, locale); BreakIterator baseBI = parseBreakIterator(type, locale);
// Use a default fragsize the same as the regex Fragmenter (original Highlighter) since we're
// both likely shooting for sentence-like patterns.
int fragsize = params.getFieldInt(field, HighlightParams.FRAGSIZE, LuceneRegexFragmenter.DEFAULT_FRAGMENT_SIZE);
if (fragsize <= 1 || baseBI instanceof WholeBreakIterator) { // no real minimum size if (fragsize <= 1 || baseBI instanceof WholeBreakIterator) { // no real minimum size
return baseBI; return baseBI;
} }

7
solr/core/src/test/org/apache/solr/highlight/TestUnifiedSolrHighlighter.java
View File

@ -79,7 +79,7 @@ public class TestUnifiedSolrHighlighter extends SolrTestCaseJ4 {
assertU(commit()); assertU(commit());
assertQ("multiple snippets test", assertQ("multiple snippets test",
req("q", "text:document", "sort", "id asc", "hl", "true", "hl.snippets", "2", "hl.bs.type", "SENTENCE", req("q", "text:document", "sort", "id asc", "hl", "true", "hl.snippets", "2", "hl.bs.type", "SENTENCE",
"hl.fragsize", "0"), "hl.fragsize", "-1"),
"count(//lst[@name='highlighting']/lst[@name='101']/arr[@name='text']/*)=2", "count(//lst[@name='highlighting']/lst[@name='101']/arr[@name='text']/*)=2",
"//lst[@name='highlighting']/lst[@name='101']/arr/str[1]='<em>Document</em> snippet one. '", "//lst[@name='highlighting']/lst[@name='101']/arr/str[1]='<em>Document</em> snippet one. '",
"//lst[@name='highlighting']/lst[@name='101']/arr/str[2]='<em>Document</em> snippet two.'"); "//lst[@name='highlighting']/lst[@name='101']/arr/str[2]='<em>Document</em> snippet two.'");
@ -214,9 +214,12 @@ public class TestUnifiedSolrHighlighter extends SolrTestCaseJ4 {
public void testBreakIteratorWhole() { public void testBreakIteratorWhole() {
assertU(adoc("text", "Document one has a first sentence. Document two has a second sentence.", "id", "103")); assertU(adoc("text", "Document one has a first sentence. Document two has a second sentence.", "id", "103"));
assertU(commit()); assertU(commit());
assertQ("different breakiterator", assertQ("WHOLE breakiterator",
req("q", "text:document", "sort", "id asc", "hl", "true", "hl.bs.type", "WHOLE", "hl.fragsize", "-1"), req("q", "text:document", "sort", "id asc", "hl", "true", "hl.bs.type", "WHOLE", "hl.fragsize", "-1"),
"//lst[@name='highlighting']/lst[@name='103']/arr[@name='text']/str='<em>Document</em> one has a first sentence. <em>Document</em> two has a second sentence.'"); "//lst[@name='highlighting']/lst[@name='103']/arr[@name='text']/str='<em>Document</em> one has a first sentence. <em>Document</em> two has a second sentence.'");
assertQ("hl.fragsize 0 is equivalent to WHOLE",
req("q", "text:document", "sort", "id asc", "hl", "true", "hl.fragsize", "0"),
"//lst[@name='highlighting']/lst[@name='103']/arr[@name='text']/str='<em>Document</em> one has a first sentence. <em>Document</em> two has a second sentence.'");
} }
public void testFragsize() { public void testFragsize() {