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

use singletons in FST outputs; add 2 commented out test cases showing non-minimality 

git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/trunk@1233696 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Michael McCandless 2012-01-19 23:54:55 +00:00
parent 4fdebf90e5
commit 5f33d8d2ae
3 changed files with 48 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
lucene/src/java/org/apache/lucene/util/fst/ByteSequenceOutputs.java
View File

@ -32,12 +32,13 @@ import org.apache.lucene.util.BytesRef;
public final class ByteSequenceOutputs extends Outputs<BytesRef> {
private final static BytesRef NO_OUTPUT = new BytesRef();
private final static ByteSequenceOutputs singleton = new ByteSequenceOutputs();
private ByteSequenceOutputs() {
}
public static ByteSequenceOutputs getSingleton() {
return new ByteSequenceOutputs();
return singleton;
}
@Override

3
lucene/src/java/org/apache/lucene/util/fst/IntSequenceOutputs.java
View File

@ -32,12 +32,13 @@ import org.apache.lucene.util.IntsRef;
public final class IntSequenceOutputs extends Outputs<IntsRef> {
private final static IntsRef NO_OUTPUT = new IntsRef();
private final static IntSequenceOutputs singleton = new IntSequenceOutputs();
private IntSequenceOutputs() {
}
public static IntSequenceOutputs getSingleton() {
return new IntSequenceOutputs();
return singleton;
}
@Override

44
lucene/src/test/org/apache/lucene/util/fst/TestFSTs.java
View File

@ -1055,6 +1055,50 @@ public class TestFSTs extends LuceneTestCase {
}
}
// NOTE: this test shows a case where our current builder
// fails to produce minimal FST:
/*
public void test3() throws Exception {
final PositiveIntOutputs outputs = PositiveIntOutputs.getSingleton(true);
Builder<Long> builder = new Builder<Long>(FST.INPUT_TYPE.BYTE1, outputs);
IntsRef scratchIntsRef = new IntsRef();
builder.add(Util.toIntsRef(new BytesRef("aa$"), scratchIntsRef), outputs.get(0));
builder.add(Util.toIntsRef(new BytesRef("aab$"), scratchIntsRef), 1L);
builder.add(Util.toIntsRef(new BytesRef("bbb$"), scratchIntsRef), 2L);
final FST<Long> fst = builder.finish();
//System.out.println("NODES " + fst.getNodeCount() + " ARCS " + fst.getArcCount());
// NOTE: we produce 7 nodes today
assertEquals(6, fst.getNodeCount());
// NOTE: we produce 8 arcs today
assertEquals(7, fst.getNodeCount());
//Writer w = new OutputStreamWriter(new FileOutputStream("out.dot"), "UTF-8");
//Util.toDot(fst, w, false, false);
//w.close();
}
*/
// NOTE: this test shows a case where our current builder
// fails to produce minimal FST:
/*
public void test4() throws Exception {
final ByteSequenceOutputs outputs = ByteSequenceOutputs.getSingleton();
Builder<BytesRef> builder = new Builder<BytesRef>(FST.INPUT_TYPE.BYTE1, outputs);
IntsRef scratchIntsRef = new IntsRef();
builder.add(Util.toIntsRef(new BytesRef("aa$"), scratchIntsRef), outputs.getNoOutput());
builder.add(Util.toIntsRef(new BytesRef("aab$"), scratchIntsRef), new BytesRef("1"));
builder.add(Util.toIntsRef(new BytesRef("bbb$"), scratchIntsRef), new BytesRef("11"));
final FST<BytesRef> fst = builder.finish();
//System.out.println("NODES " + fst.getNodeCount() + " ARCS " + fst.getArcCount());
// NOTE: we produce 7 nodes today
assertEquals(6, fst.getNodeCount());
// NOTE: we produce 8 arcs today
assertEquals(7, fst.getNodeCount());
//Writer w = new OutputStreamWriter(new FileOutputStream("out.dot"), "UTF-8");
//Util.toDot(fst, w, false, false);
//w.close();
}
*/
// Build FST for all unique terms in the test line docs
// file, up until a time limit
public void testRealTerms() throws Exception {