Avoid NPE in `more_like_this` when field has zero tokens (#30365)

Fixes and edge case when using `more_like_this` where TermVectorsWriter
could throw an NPE when a field produced zero tokens after analysis. This
changes the implementation to use an empty list of tokens in this case.

Closes #30148
This commit is contained in:
aditya-agrawal 2018-05-08 18:43:07 +05:30 committed by Colin Goodheart-Smithe
parent 5d5acd5dc8
commit a28424ee5d
No known key found for this signature in database
GPG Key ID: F975E7BDD739B3C7
3 changed files with 38 additions and 0 deletions

View File

@ -104,6 +104,8 @@ ones that the user is authorized to access in case field level security is enabl
[float] [float]
=== Bug Fixes === Bug Fixes
Fix NPE in 'more_like_this' when field has zero tokens ({pull}30365[#30365])
Fixed prerelease version of elasticsearch in the `deb` package to sort before GA versions Fixed prerelease version of elasticsearch in the `deb` package to sort before GA versions
({pull}29000[#29000]) ({pull}29000[#29000])
@ -169,6 +171,8 @@ Added put index template API to the high level rest client ({pull}30400[#30400])
[float] [float]
=== Bug Fixes === Bug Fixes
Fix NPE in 'more_like_this' when field has zero tokens ({pull}30365[#30365])
Do not ignore request analysis/similarity settings on index resize operations when the source index already contains such settings ({pull}30216[#30216]) Do not ignore request analysis/similarity settings on index resize operations when the source index already contains such settings ({pull}30216[#30216])
Fix NPE when CumulativeSum agg encounters null value/empty bucket ({pull}29641[#29641]) Fix NPE when CumulativeSum agg encounters null value/empty bucket ({pull}29641[#29641])

View File

@ -70,6 +70,10 @@ final class TermVectorsWriter {
Terms topLevelTerms = topLevelFields.terms(field); Terms topLevelTerms = topLevelFields.terms(field);
// if no terms found, take the retrieved term vector fields for stats // if no terms found, take the retrieved term vector fields for stats
if (fieldTermVector == null) {
fieldTermVector = EMPTY_TERMS;
}
if (topLevelTerms == null) { if (topLevelTerms == null) {
topLevelTerms = EMPTY_TERMS; topLevelTerms = EMPTY_TERMS;
} }

View File

@ -91,6 +91,36 @@ public class MoreLikeThisIT extends ESIntegTestCase {
assertHitCount(response, 1L); assertHitCount(response, 1L);
} }
//Issue #30148
public void testMoreLikeThisForZeroTokensInOneOfTheAnalyzedFields() throws Exception {
CreateIndexRequestBuilder createIndexRequestBuilder = prepareCreate("test")
.addMapping("type", jsonBuilder()
.startObject().startObject("type")
.startObject("properties")
.startObject("myField").field("type", "text").endObject()
.startObject("empty").field("type", "text").endObject()
.endObject()
.endObject().endObject());
assertAcked(createIndexRequestBuilder);
ensureGreen();
client().index(indexRequest("test").type("type").id("1").source(jsonBuilder().startObject()
.field("myField", "and_foo").field("empty", "").endObject())).actionGet();
client().index(indexRequest("test").type("type").id("2").source(jsonBuilder().startObject()
.field("myField", "and_foo").field("empty", "").endObject())).actionGet();
client().admin().indices().refresh(refreshRequest()).actionGet();
SearchResponse searchResponse = client().prepareSearch().setQuery(
moreLikeThisQuery(new String[]{"myField", "empty"}, null, new Item[]{new Item("test", "type", "1")})
.minTermFreq(1).minDocFreq(1)
).get();
assertHitCount(searchResponse, 1L);
}
public void testSimpleMoreLikeOnLongField() throws Exception { public void testSimpleMoreLikeOnLongField() throws Exception {
logger.info("Creating index test"); logger.info("Creating index test");
assertAcked(prepareCreate("test") assertAcked(prepareCreate("test")