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

LUCENE-4732: Fixed TermsEnum.seekCeil/seekExact on term vectors. 

git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/trunk@1440308 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Adrien Grand 2013-01-30 09:08:06 +00:00
parent 4ec7239ce8
commit a81ac90986
4 changed files with 55 additions and 30 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
lucene/CHANGES.txt
View File

@ -115,6 +115,9 @@ Bug Fixes
you have a taxonomy index created with such strings, you should rebuild it.
(Michael McCandless, Shai Erera)
* LUCENE-4732: Fixed TermsEnum.seekCeil/seekExact on term vectors.
(Adrien Grand, Robert Muir)
======================= Lucene 4.1.0 =======================
Changes in backwards compatibility policy

16
lucene/core/src/java/org/apache/lucene/codecs/compressing/CompressingTermVectorsReader.java
View File

@ -832,10 +832,18 @@ public final class CompressingTermVectorsReader extends TermVectorsReader implem
}
}
// linear scan
do {
next();
} while (ord < numTerms - 1 && term().compareTo(text) < 0);
return term().equals(text) ? SeekStatus.FOUND : SeekStatus.END;
while (true) {
final BytesRef term = next();
if (term == null) {
return SeekStatus.END;
}
final int cmp = term.compareTo(text);
if (cmp > 0) {
return SeekStatus.NOT_FOUND;
} else if (cmp == 0) {
return SeekStatus.FOUND;
}
}
}
@Override

2
lucene/core/src/java/org/apache/lucene/codecs/lucene40/Lucene40TermVectorsReader.java
View File

@ -422,7 +422,7 @@ public class Lucene40TermVectorsReader extends TermVectorsReader implements Clos
this.storePayloads = storePayloads;
nextTerm = 0;
tvf.seek(tvfFPStart);
tvfFP = 1+tvfFPStart;
tvfFP = tvfFPStart;
positions = null;
startOffsets = null;
endOffsets = null;

64
lucene/core/src/test/org/apache/lucene/index/TestDuelingCodecs.java
View File

@ -30,7 +30,6 @@ import java.util.TreeSet;
import org.apache.lucene.analysis.Analyzer;
import org.apache.lucene.analysis.MockAnalyzer;
import org.apache.lucene.codecs.Codec;
import org.apache.lucene.document.Document;
import org.apache.lucene.index.TermsEnum.SeekStatus;
import org.apache.lucene.search.DocIdSetIterator;
import org.apache.lucene.store.Directory;
@ -39,6 +38,7 @@ import org.apache.lucene.util.BytesRef;
import org.apache.lucene.util.FixedBitSet;
import org.apache.lucene.util.LineFileDocs;
import org.apache.lucene.util.LuceneTestCase;
import org.apache.lucene.util._TestUtil;
import org.apache.lucene.util.automaton.AutomatonTestUtil;
import org.apache.lucene.util.automaton.CompiledAutomaton;
import org.apache.lucene.util.automaton.RegExp;
@ -231,11 +231,11 @@ public class TestDuelingCodecs extends LuceneTestCase {
private void assertTermsSeeking(Terms leftTerms, Terms rightTerms) throws Exception {
TermsEnum leftEnum = null;
TermsEnum rightEnum = null;
// just an upper bound
int numTests = atLeast(20);
Random random = random();
// collect this number of terms from the left side
HashSet<BytesRef> tests = new HashSet<BytesRef>();
int numPasses = 0;
@ -259,36 +259,50 @@ public class TestDuelingCodecs extends LuceneTestCase {
byte newbytes[] = new byte[term.length+5];
System.arraycopy(term.bytes, term.offset, newbytes, 5, term.length);
tests.add(new BytesRef(newbytes, 5, term.length));
} else if (code == 3) {
switch (random().nextInt(3)) {
case 0:
tests.add(new BytesRef()); // before the first term
break;
case 1:
tests.add(new BytesRef(new byte[] {(byte) 0xFF, (byte) 0xFF})); // past the last term
break;
case 2:
tests.add(new BytesRef(_TestUtil.randomSimpleString(random()))); // random term
break;
default:
throw new AssertionError();
}
}
}
numPasses++;
}
rightEnum = rightTerms.iterator(rightEnum);
ArrayList<BytesRef> shuffledTests = new ArrayList<BytesRef>(tests);
Collections.shuffle(shuffledTests, random);
for (BytesRef b : shuffledTests) {
leftEnum = leftTerms.iterator(leftEnum);
rightEnum = rightTerms.iterator(rightEnum);
assertEquals(info, leftEnum.seekExact(b, false), rightEnum.seekExact(b, false));
assertEquals(info, leftEnum.seekExact(b, true), rightEnum.seekExact(b, true));
SeekStatus leftStatus;
SeekStatus rightStatus;
leftStatus = leftEnum.seekCeil(b, false);
rightStatus = rightEnum.seekCeil(b, false);
assertEquals(info, leftStatus, rightStatus);
if (leftStatus != SeekStatus.END) {
assertEquals(info, leftEnum.term(), rightEnum.term());
if (rarely()) {
// reuse the enums
leftEnum = leftTerms.iterator(leftEnum);
rightEnum = rightTerms.iterator(rightEnum);
}
leftStatus = leftEnum.seekCeil(b, true);
rightStatus = rightEnum.seekCeil(b, true);
assertEquals(info, leftStatus, rightStatus);
if (leftStatus != SeekStatus.END) {
assertEquals(info, leftEnum.term(), rightEnum.term());
final boolean useCache = random().nextBoolean();
final boolean seekExact = random().nextBoolean();
if (seekExact) {
assertEquals(info, leftEnum.seekExact(b, useCache), rightEnum.seekExact(b, useCache));
} else {
SeekStatus leftStatus = leftEnum.seekCeil(b, useCache);
SeekStatus rightStatus = rightEnum.seekCeil(b, useCache);
assertEquals(info, leftStatus, rightStatus);
if (leftStatus != SeekStatus.END) {
assertEquals(info, leftEnum.term(), rightEnum.term());
assertTermStats(leftEnum, rightEnum);
}
}
}
}