honeymoose
/
OpenSearch
mirror of https://github.com/honeymoose/OpenSearch.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Pass resolved extended bounds to unmapped histogram aggregator 

Previous to this change the unresolved extended bounds was passed into the histogram aggregator which meant extendedbounds.min and extendedbounds.max was passed through as null. This had two effects on the histogram aggregator:

1. If the histogram aggregator was unmapped across all shards, the reduce phase would not add buckets for the extended bounds and the response would contain zero buckets
2. If the histogram aggregator was not unmapped in some shards, the reduce phase might sometimes chose to reduce based on the unmapped shard response and therefore the extended bounds would be ignored.

This change resolves the extended bounds in the unmapped case and solves the above two issues.

Closes #19009
This commit is contained in:
Colin Goodheart-Smithe 2016-06-27 10:46:33 +01:00
parent cb0824e957
commit 108ba23073
2 changed files with 84 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
core/src/main/java/org/elasticsearch/search/aggregations/bucket/histogram/AbstractHistogramAggregatorFactory.java
View File

@ -69,9 +69,7 @@ public abstract class AbstractHistogramAggregatorFactory<AF extends AbstractHist
@Override
protected Aggregator createUnmapped(Aggregator parent, List<PipelineAggregator> pipelineAggregators, Map<String, Object> metaData)
throws IOException {
Rounding rounding = createRounding();
return new HistogramAggregator(name, factories, rounding, order, keyed, minDocCount, extendedBounds, null, config.format(),
histogramFactory, context, parent, pipelineAggregators, metaData);
return createAggregator(null, parent, pipelineAggregators, metaData);
}
protected Rounding createRounding() {
@ -92,6 +90,11 @@ public abstract class AbstractHistogramAggregatorFactory<AF extends AbstractHist
if (collectsFromSingleBucket == false) {
return asMultiBucketAggregator(this, context, parent);
}
return createAggregator(valuesSource, parent, pipelineAggregators, metaData);
}
private Aggregator createAggregator(ValuesSource.Numeric valuesSource, Aggregator parent, List<PipelineAggregator> pipelineAggregators,
Map<String, Object> metaData) throws IOException {
Rounding rounding = createRounding();
// we need to round the bounds given by the user and we have to do it
// for every aggregator we create

78
modules/lang-groovy/src/test/java/org/elasticsearch/messy/tests/HistogramTests.java
View File

@ -21,6 +21,7 @@ package org.elasticsearch.messy.tests;
import com.carrotsearch.hppc.LongHashSet;
import org.elasticsearch.action.index.IndexRequestBuilder;
import org.elasticsearch.action.search.SearchResponse;
import org.elasticsearch.index.query.QueryBuilders;
import org.elasticsearch.plugins.Plugin;
import org.elasticsearch.script.Script;
import org.elasticsearch.script.groovy.GroovyPlugin;
@ -825,6 +826,83 @@ public class HistogramTests extends ESIntegTestCase {
}
}
public void testEmptyWithExtendedBounds() throws Exception {
int lastDataBucketKey = (numValueBuckets - 1) * interval;
// randomizing the number of buckets on the min bound
// (can sometimes fall within the data range, but more frequently will fall before the data range)
int addedBucketsLeft = randomIntBetween(0, numValueBuckets);
long boundsMinKey = addedBucketsLeft * interval;
if (frequently()) {
boundsMinKey = -boundsMinKey;
} else {
addedBucketsLeft = 0;
}
long boundsMin = boundsMinKey + randomIntBetween(0, interval - 1);
// randomizing the number of buckets on the max bound
// (can sometimes fall within the data range, but more frequently will fall after the data range)
int addedBucketsRight = randomIntBetween(0, numValueBuckets);
long boundsMaxKeyDelta = addedBucketsRight * interval;
if (rarely()) {
addedBucketsRight = 0;
boundsMaxKeyDelta = -boundsMaxKeyDelta;
}
long boundsMaxKey = lastDataBucketKey + boundsMaxKeyDelta;
long boundsMax = boundsMaxKey + randomIntBetween(0, interval - 1);
// it could be that the random bounds.min we chose ended up greater than bounds.max - this should cause an
// error
boolean invalidBoundsError = boundsMin > boundsMax;
// constructing the newly expected bucket list
int bucketsCount = numValueBuckets + addedBucketsLeft + addedBucketsRight;
long[] extendedValueCounts = new long[bucketsCount];
System.arraycopy(valueCounts, 0, extendedValueCounts, addedBucketsLeft, valueCounts.length);
SearchResponse response = null;
try {
response = client().prepareSearch("idx")
.setQuery(QueryBuilders.termQuery("foo", "bar"))
.addAggregation(histogram("histo")
.field(SINGLE_VALUED_FIELD_NAME)
.interval(interval)
.minDocCount(0)
.extendedBounds(new ExtendedBounds(boundsMin, boundsMax)))
.execute().actionGet();
if (invalidBoundsError) {
fail("Expected an exception to be thrown when bounds.min is greater than bounds.max");
return;
}
} catch (Exception e) {
if (invalidBoundsError) {
// expected
return;
} else {
throw e;
}
}
assertSearchResponse(response);
Histogram histo = response.getAggregations().get("histo");
assertThat(histo, notNullValue());
assertThat(histo.getName(), equalTo("histo"));
List<? extends Bucket> buckets = histo.getBuckets();
assertThat(buckets.size(), equalTo(bucketsCount));
long key = Math.min(boundsMinKey, 0);
for (int i = 0; i < bucketsCount; i++) {
Histogram.Bucket bucket = buckets.get(i);
assertThat(bucket, notNullValue());
assertThat(((Number) bucket.getKey()).longValue(), equalTo(key));
assertThat(bucket.getDocCount(), equalTo(0L));
key += interval;
}
}
/**
* see issue #9634, negative interval in histogram should raise exception
*/