Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix range queries on _type field for singe type indices #31756

Merged
merged 5 commits into from
Jul 18, 2018
Merged
Show file tree
Hide file tree
Changes from 1 commit
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Original file line number Diff line number Diff line change
Expand Up @@ -34,7 +34,10 @@
import org.apache.lucene.search.Query;
import org.apache.lucene.search.TermInSetQuery;
import org.apache.lucene.search.TermQuery;
import org.apache.lucene.search.TermRangeQuery;
import org.apache.lucene.util.BytesRef;
import org.elasticsearch.common.logging.DeprecationLogger;
import org.elasticsearch.common.logging.ESLoggerFactory;
import org.elasticsearch.common.lucene.Lucene;
import org.elasticsearch.common.lucene.search.Queries;
import org.elasticsearch.common.xcontent.XContentBuilder;
Expand Down Expand Up @@ -90,6 +93,8 @@ public MetadataFieldMapper getDefault(MappedFieldType fieldType, ParserContext c

static final class TypeFieldType extends StringFieldType {

private static final DeprecationLogger DEPRECATION_LOGGER = new DeprecationLogger(ESLoggerFactory.getLogger(TypeFieldType.class));

TypeFieldType() {
}

Expand Down Expand Up @@ -154,6 +159,33 @@ public Query termsQuery(List<?> values, QueryShardContext context) {
}
}

@Override
public Query rangeQuery(Object lowerTerm, Object upperTerm, boolean includeLower, boolean includeUpper, QueryShardContext context) {
if (hasDocValues()) {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

we don't need this if block, do we? All 6.x and 7.x indices have a single type.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Don't we still support multi-type indices that have been created in 5.x on 6.x? I will remove this for the PR against master but I think we need it for 6.x in that case. Can you confirm?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

👍

return new TermRangeQuery(name(), lowerTerm == null ? null : indexedValueForSearch(lowerTerm),
upperTerm == null ? null : indexedValueForSearch(upperTerm), includeLower, includeUpper);
} else {
// this means the index has a single type and the type field is implicit
DEPRECATION_LOGGER.deprecatedAndMaybeLog("range_single_type",
"Running [range] query on [_type] field for an index with a single type. As types are deprecated, this functionality will be removed in future releases.");
Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Not sure if we really need the deprecation, but since it was mentioned at some point I added it for discussion purposes.

String type = context.getMapperService().documentMapper().type();
Query result = new MatchAllDocsQuery();
if (lowerTerm != null && lowerTerm instanceof BytesRef) {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

we shouldn't be lenient in case lowerTerm doesnt't implement BytesRef

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I was wondering the same, do we want to throw an IAE in those cases?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I would call indexedValueForSearch.

int comp = type.compareTo(((BytesRef) lowerTerm).utf8ToString());
Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Not sure if there is a better way of comparing BytesRef instances, maybe taking into account different lexicographical orderings?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

BytesRef implements Comparable, so you should be able to do something like:

int comp = lowerTerm.compareTo(new BytesRef(type));

if (comp < 0 || (comp == 0 && includeLower == false)) {
result = new MatchNoDocsQuery("[_type] was less then lower bound of range");
}
}
if (upperTerm != null && upperTerm instanceof BytesRef) {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

we shouldn't be lenient in case upperTerm doesnt't implement BytesRef

int comp = type.compareTo(((BytesRef) upperTerm).utf8ToString());
if (comp > 0 || (comp == 0 && includeUpper == false)) {
result = new MatchNoDocsQuery("[_type] was higher then upper bound of range");
}
}
return result;
}
}

}

/**
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -1818,4 +1818,40 @@ public void testRangeQueryRangeFields_24744() throws Exception {
SearchResponse searchResponse = client().prepareSearch("test").setQuery(range).get();
assertHitCount(searchResponse, 1);
}

public void testRangeQueryTypeField_31476() throws Exception {
assertAcked(prepareCreate("test").addMapping("foo", "field", "type=keyword"));

client().prepareIndex("test", "foo", "1").setSource("field", "value").get();
refresh();

RangeQueryBuilder range = new RangeQueryBuilder("_type").from("ape").to("zebra");
SearchResponse searchResponse = client().prepareSearch("test").setQuery(range).get();
assertHitCount(searchResponse, 1);

range = new RangeQueryBuilder("_type").from("monkey").to("zebra");
searchResponse = client().prepareSearch("test").setQuery(range).get();
assertHitCount(searchResponse, 0);

range = new RangeQueryBuilder("_type").from("ape").to("donkey");
searchResponse = client().prepareSearch("test").setQuery(range).get();
assertHitCount(searchResponse, 0);

range = new RangeQueryBuilder("_type").from("ape").to("foo").includeUpper(false);
searchResponse = client().prepareSearch("test").setQuery(range).get();
assertHitCount(searchResponse, 0);

range = new RangeQueryBuilder("_type").from("ape").to("foo").includeUpper(true);
searchResponse = client().prepareSearch("test").setQuery(range).get();
assertHitCount(searchResponse, 1);

range = new RangeQueryBuilder("_type").from("foo").to("zebra").includeLower(false);
searchResponse = client().prepareSearch("test").setQuery(range).get();
assertHitCount(searchResponse, 0);

range = new RangeQueryBuilder("_type").from("foo").to("zebra").includeLower(true);
searchResponse = client().prepareSearch("test").setQuery(range).get();
assertHitCount(searchResponse, 1);
}

}