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

LUCENE-9212: Intervals.multiterm() should take CompiledAutomaton

This commit is contained in:
Alan Woodward 2020-02-24 11:08:48 +00:00
parent 663611c99c
commit ffb7cafe93
3 changed files with 80 additions and 25 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
lucene/CHANGES.txt
View File

@ -114,6 +114,9 @@ API Changes
* LUCENE-9218: XY geometries API works in float space. (Ignacio Vera)
* LUCENE-9212: Intervals.multiterm() takes CompiledAutomaton rather than plain Automaton
(Alan Woodward)
New Features
---------------------

34
lucene/queries/src/java/org/apache/lucene/queries/intervals/Intervals.java
View File

@ -203,7 +203,10 @@ public final class Intervals {
* @param pattern string representation of the given automaton, mostly used in exception messages
*
* @throws IllegalStateException if the automaton accepts more than 128 terms
*
* @deprecated use {@link #multiterm(CompiledAutomaton, String)}
*/
@Deprecated
public static IntervalsSource multiterm(Automaton automaton, String pattern) {
return multiterm(automaton, 128, pattern);
}
@ -219,11 +222,42 @@ public final class Intervals {
* @param pattern string representation of the given automaton, mostly used in exception messages
*
* @throws IllegalStateException if the automaton accepts more than {@code maxExpansions} terms
*
* @deprecated use {@link #multiterm(CompiledAutomaton, int, String)}
*/
@Deprecated
public static IntervalsSource multiterm(Automaton automaton, int maxExpansions, String pattern) {
CompiledAutomaton ca = new CompiledAutomaton(automaton);
return new MultiTermIntervalsSource(ca, maxExpansions, pattern);
}
/**
* Expert: Return an {@link IntervalsSource} over the disjunction of all terms that's accepted by the given automaton
*
* @param ca an automaton accepting matching terms
* @param pattern string representation of the given automaton, mostly used in exception messages
*
* @throws IllegalStateException if the automaton accepts more than 128 terms
*/
public static IntervalsSource multiterm(CompiledAutomaton ca, String pattern) {
return multiterm(ca, 128, pattern);
}
/**
* Expert: Return an {@link IntervalsSource} over the disjunction of all terms that's accepted by the given automaton
*
* WARNING: Setting {@code maxExpansions} to higher than the default value of 128
* can be both slow and memory-intensive
*
* @param ca an automaton accepting matching terms
* @param maxExpansions the maximum number of terms to expand to
* @param pattern string representation of the given automaton, mostly used in exception messages
*
* @throws IllegalStateException if the automaton accepts more than {@code maxExpansions} terms
*/
public static IntervalsSource multiterm(CompiledAutomaton ca, int maxExpansions, String pattern) {
return new MultiTermIntervalsSource(ca, maxExpansions, pattern);
}
/**
* Create an {@link IntervalsSource} that filters a sub-source by the width of its intervals

68
lucene/queries/src/test/org/apache/lucene/queries/intervals/TestIntervals.java
View File

@ -20,7 +20,6 @@ package org.apache.lucene.queries.intervals;
import java.io.IOException;
import java.util.Arrays;
import java.util.HashSet;
import java.util.List;
import java.util.Set;
import java.util.stream.Collectors;
@ -44,7 +43,6 @@ import org.apache.lucene.index.Term;
import org.apache.lucene.search.BooleanClause;
import org.apache.lucene.search.IndexSearcher;
import org.apache.lucene.search.MatchesIterator;
import org.apache.lucene.search.PrefixQuery;
import org.apache.lucene.search.Query;
import org.apache.lucene.search.QueryVisitor;
import org.apache.lucene.search.TermQuery;
@ -52,6 +50,8 @@ import org.apache.lucene.store.Directory;
import org.apache.lucene.util.BytesRef;
import org.apache.lucene.util.IOUtils;
import org.apache.lucene.util.LuceneTestCase;
import org.apache.lucene.util.automaton.CompiledAutomaton;
import org.apache.lucene.util.automaton.RegExp;
import org.junit.AfterClass;
import org.junit.BeforeClass;
@ -886,36 +886,30 @@ public class TestIntervals extends LuceneTestCase {
}
public void testPrefix() throws IOException {
for (IntervalsSource source : List.of(Intervals.prefix(new BytesRef("p")),
Intervals.multiterm(PrefixQuery.toAutomaton(new BytesRef("p")), "p*" ) )) {
checkIntervals(source, "field1", 5, new int[][]{
{},
{ 0, 0, 1, 1, 3, 3, 4, 4, 6, 6, 7, 7, 10, 10, 27, 27 },
{ 0, 0, 1, 1, 3, 3, 4, 4, 6, 6, 7, 7, 10, 10 },
{ 7, 7 },
{ 0, 0, 1, 1, 3, 3, 4, 4, 6, 6, 7, 7, 10, 10 },
{ 0, 0 }
});
MatchesIterator mi = getMatches(source, 1, "field1");
assertNotNull(mi);
assertMatch(mi, 0, 0, 0, 5);
assertMatch(mi, 1, 1, 6, 14);
}
IntervalsSource source = Intervals.prefix(new BytesRef("p"));
checkIntervals(source, "field1", 5, new int[][]{
{},
{0, 0, 1, 1, 3, 3, 4, 4, 6, 6, 7, 7, 10, 10, 27, 27},
{0, 0, 1, 1, 3, 3, 4, 4, 6, 6, 7, 7, 10, 10},
{7, 7},
{0, 0, 1, 1, 3, 3, 4, 4, 6, 6, 7, 7, 10, 10},
{0, 0}
});
MatchesIterator mi = getMatches(source, 1, "field1");
assertNotNull(mi);
assertMatch(mi, 0, 0, 0, 5);
assertMatch(mi, 1, 1, 6, 14);
for (IntervalsSource noSuch : List.of(Intervals.prefix(new BytesRef("qqq")),
Intervals.multiterm(PrefixQuery.toAutomaton(new BytesRef("qqq")), "qqq*" ))) {
checkIntervals(noSuch, "field1", 0, new int[][]{});
}
IntervalsSource noSuch = Intervals.prefix(new BytesRef("qqq"));
checkIntervals(noSuch, "field1", 0, new int[][]{});
for (IntervalsSource source : List.of(Intervals.prefix(new BytesRef("p"), 1),
Intervals.multiterm(PrefixQuery.toAutomaton(new BytesRef("p")), 1, "p*")) ) {
IntervalsSource s = Intervals.prefix(new BytesRef("p"), 1);
IllegalStateException e = expectThrows(IllegalStateException.class, () -> {
for (LeafReaderContext ctx : searcher.getIndexReader().leaves()) {
source.intervals("field1", ctx);
s.intervals("field1", ctx);
}
});
assertEquals("Automaton [p*] expanded to too many terms (limit 1)", e.getMessage());
}
checkVisits(Intervals.prefix(new BytesRef("p")), 1);
}
@ -965,4 +959,28 @@ public class TestIntervals extends LuceneTestCase {
}
public void testMultiTerm() throws IOException {
RegExp re = new RegExp("p.*e");
IntervalsSource source = Intervals.multiterm(new CompiledAutomaton(re.toAutomaton()), re.toString());
checkIntervals(source, "field1", 5, new int[][]{
{},
{ 0, 0, 1, 1, 3, 3, 4, 4, 6, 6, 7, 7 },
{ 0, 0, 1, 1, 3, 3, 4, 4, 6, 6, 7, 7 },
{ 7, 7 },
{ 0, 0, 1, 1, 3, 3, 4, 4, 6, 6, 7, 7 },
{ 0, 0 }
});
IllegalStateException e = expectThrows(IllegalStateException.class, () -> {
IntervalsSource s = Intervals.multiterm(new CompiledAutomaton(re.toAutomaton()), 1, re.toString());
for (LeafReaderContext ctx : searcher.getIndexReader().leaves()) {
s.intervals("field1", ctx);
}
});
assertEquals("Automaton [\\p(.)*\\e] expanded to too many terms (limit 1)", e.getMessage());
checkVisits(source, 1);
}
}