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

LUCENE-4951: DrillSideways now uses Scorer.cost() to decide which scorer impl to use 

git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/trunk@1471705 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Michael McCandless 2013-04-24 21:43:57 +00:00
parent 160438c43a
commit dacc7f0166
3 changed files with 17 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
lucene/CHANGES.txt
View File

@ -55,7 +55,9 @@ Optimizations
a common divisor. In particular, this improves the compression ratio of dates
without time when they are encoded as milliseconds since Epoch. Also support
TABLE compressed numerics in the Disk codec. (Robert Muir, Adrien Grand)
* LUCENE-4951: DrillSideways uses the new Scorer.cost() method to make
better decisions about which scorer to use internally. (Mike McCandless)
New Features

2
lucene/facet/src/java/org/apache/lucene/facet/search/DrillSidewaysQuery.java
View File

@ -129,7 +129,7 @@ class DrillSidewaysQuery extends Query {
dims[dim].docsEnums = new DocsEnum[drillDownTerms[dim].length];
for(int i=0;i<drillDownTerms[dim].length;i++) {
if (termsEnum.seekExact(drillDownTerms[dim][i].bytes(), false)) {
dims[dim].freq = Math.max(dims[dim].freq, termsEnum.docFreq());
dims[dim].maxFreq = Math.max(dims[dim].maxFreq, termsEnum.docFreq());
dims[dim].docsEnums[i] = termsEnum.docs(null, null);
}
}

23
lucene/facet/src/java/org/apache/lucene/facet/search/DrillSidewaysScorer.java
View File

@ -76,7 +76,7 @@ class DrillSidewaysScorer extends Scorer {
assert baseScorer != null;
// Position all scorers to their first matching doc:
int baseDocID = baseScorer.nextDoc();
baseScorer.nextDoc();
for(DocsEnumsAndFreq dim : dims) {
for(DocsEnum docsEnum : dim.docsEnums) {
@ -90,18 +90,21 @@ class DrillSidewaysScorer extends Scorer {
DocsEnum[][] docsEnums = new DocsEnum[numDims][];
Collector[] sidewaysCollectors = new Collector[numDims];
int maxFreq = 0;
long drillDownCost = 0;
for(int dim=0;dim<numDims;dim++) {
docsEnums[dim] = dims[dim].docsEnums;
sidewaysCollectors[dim] = dims[dim].sidewaysCollector;
maxFreq = Math.max(maxFreq, dims[dim].freq);
for(DocsEnum de : dims[dim].docsEnums) {
if (de != null) {
drillDownCost += de.cost();
}
}
}
// TODO: if we add cost API to Scorer, switch to that!
int estBaseHitCount = context.reader().maxDoc() / (1+baseDocID);
long baseQueryCost = baseScorer.cost();
/*
System.out.println("\nbaseDocID=" + baseDocID + " est=" + estBaseHitCount);
System.out.println("\nbaseDocID=" + baseScorer.docID() + " est=" + estBaseHitCount);
System.out.println(" maxDoc=" + context.reader().maxDoc());
System.out.println(" maxFreq=" + maxFreq);
System.out.println(" dims[0].freq=" + dims[0].freq);
@ -110,10 +113,10 @@ class DrillSidewaysScorer extends Scorer {
}
*/
if (estBaseHitCount < maxFreq/10) {
if (baseQueryCost < drillDownCost/10) {
//System.out.println("baseAdvance");
doBaseAdvanceScoring(collector, docsEnums, sidewaysCollectors);
} else if (numDims > 1 && (dims[1].freq < estBaseHitCount/10)) {
} else if (numDims > 1 && (dims[1].maxFreq < baseQueryCost/10)) {
//System.out.println("drillDownAdvance");
doDrillDownAdvanceScoring(collector, docsEnums, sidewaysCollectors);
} else {
@ -630,13 +633,13 @@ class DrillSidewaysScorer extends Scorer {
static class DocsEnumsAndFreq implements Comparable<DocsEnumsAndFreq> {
DocsEnum[] docsEnums;
// Max docFreq for all docsEnums for this dim:
int freq;
int maxFreq;
Collector sidewaysCollector;
String dim;
@Override
public int compareTo(DocsEnumsAndFreq other) {
return freq - other.freq;
return maxFreq - other.maxFreq;
}
}
}