[[query-dsl-queries]]
== Query DSL - Queries
elasticsearch provides a full Java query dsl in a similar manner to the
REST {ref}/query-dsl.html[Query DSL]. The factory for query
builders is `QueryBuilders`. Once your query is ready, you can use the
<<search,Search API>>.
See also how to build <<query-dsl-filters,Filters>>
To use `QueryBuilders` or `FilterBuilders` just import them in your class:
[source,java]
--------------------------------------------------
import static org.elasticsearch.index.query.QueryBuilders.*;
import static org.elasticsearch.index.query.FilterBuilders.*;
--------------------------------------------------
Note that you can easily print (aka debug) JSON generated queries using
`toString()` method on `QueryBuilder` object.
The `QueryBuilder` can then be used with any API that accepts a query,
such as `count` and `search`.
[[match]]
=== Match Query
See {ref}/query-dsl-match-query.html[Match Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = matchQuery(
"name", <1>
"kimchy elasticsearch" <2>
);
--------------------------------------------------
<1> field
<2> text
[[multimatch]]
=== MultiMatch Query
See {ref}/query-dsl-multi-match-query.html[MultiMatch
Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = multiMatchQuery(
"kimchy elasticsearch", <1>
"user", "message" <2>
);
--------------------------------------------------
<1> text
<2> fields
[[bool]]
=== Boolean Query
See {ref}/query-dsl-bool-query.html[Boolean Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = boolQuery()
.must(termQuery("content", "test1")) <1>
.must(termQuery("content", "test4")) <1>
.mustNot(termQuery("content", "test2")) <2>
.should(termQuery("content", "test3")); <3>
--------------------------------------------------
<1> must query
<2> must not query
<3> should query
[[boosting]]
=== Boosting Query
See {ref}/query-dsl-boosting-query.html[Boosting Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = boostingQuery()
.positive(termQuery("name","kimchy")) <1>
.negative(termQuery("name","dadoonet")) <2>
.negativeBoost(0.2f); <3>
--------------------------------------------------
<1> query that will promote documents
<2> query that will demote documents
<3> negative boost
[[ids]]
=== IDs Query
See {ref}/query-dsl-ids-query.html[IDs Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = idsQuery().ids("1", "2"); <1>
--------------------------------------------------
<1> document ids
[[constant-score]]
=== Constant Score Query
See {ref}/query-dsl-constant-score-query.html[Constant
Score Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = constantScoreQuery(
termFilter("name","kimchy") <1>
)
.boost(2.0f); <2>
--------------------------------------------------
<1> you can use a filter
<2> filter score
[source,java]
--------------------------------------------------
QueryBuilder qb = constantScoreQuery(
termQuery("name","kimchy") <1>
)
.boost(2.0f); <2>
--------------------------------------------------
<1> you can use a query
<2> query score
[[dismax]]
=== Disjunction Max Query
See {ref}/query-dsl-dis-max-query.html[Disjunction Max
Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = disMaxQuery()
.add(termQuery("name","kimchy")) <1>
.add(termQuery("name","elasticsearch")) <2>
.boost(1.2f) <3>
.tieBreaker(0.7f); <4>
--------------------------------------------------
<1> add your queries
<2> add your queries
<3> boost factor
<4> tie breaker
[[flt]]
=== Fuzzy Like This (Field) Query (flt and flt_field)
See:
* {ref}/query-dsl-flt-query.html[Fuzzy Like This Query]
* {ref}/query-dsl-flt-field-query.html[Fuzzy Like This Field Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = fuzzyLikeThisQuery("name.first", "name.last") <1>
.likeText("text like this one") <2>
.maxQueryTerms(12); <3>
--------------------------------------------------
<1> fields
<2> text
<3> max num of Terms in generated queries
[source,java]
--------------------------------------------------
QueryBuilder qb = fuzzyLikeThisFieldQuery("name.first") <1>
.likeText("text like this one") <2>
.maxQueryTerms(12); <3>
--------------------------------------------------
<1> field
<2> text
<3> max num of Terms in generated queries
[[fuzzy]]
=== FuzzyQuery
See {ref}/query-dsl-fuzzy-query.html[Fuzzy Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = fuzzyQuery(
"name", <1>
"kimzhy" <2>
);
--------------------------------------------------
<1> field
<2> text
[[has-child-parent]]
=== Has Child / Has Parent
See:
* {ref}/query-dsl-has-child-query.html[Has Child Query]
* {ref}/query-dsl-has-parent-query.html[Has Parent]
[source,java]
--------------------------------------------------
// Has Child
QueryBuilder qb = hasChildQuery(
"blog_tag", <1>
termQuery("tag","something") <2>
);
--------------------------------------------------
<1> child type to query against
<2> query (could be also a filter)
// Has Parent
[source,java]
--------------------------------------------------
QueryBuilder qb = hasParentQuery(
"blog", <1>
termQuery("tag","something") <2>
);
--------------------------------------------------
<1> parent type to query against
<2> query (could be also a filter)
[[match-all]]
=== MatchAll Query
See {ref}/query-dsl-match-all-query.html[Match All
Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = matchAllQuery();
--------------------------------------------------
[[mlt]]
=== More Like This (Field) Query (mlt and mlt_field)
See:
* {ref}/query-dsl-mlt-query.html[More Like This Query]
* {ref}/query-dsl-mlt-field-query.html[More Like This Field Query]
[source,java]
--------------------------------------------------
// mlt Query
QueryBuilder qb = moreLikeThisQuery("name.first", "name.last") <1>
.likeText("text like this one") <2>
.minTermFreq(1) <3>
.maxQueryTerms(12); <4>
--------------------------------------------------
<1> fields
<2> text
<3> ignore threshold
<4> max num of Terms in generated queries
[source,java]
--------------------------------------------------
// mlt_field Query
QueryBuilder qb = moreLikeThisFieldQuery("name.first") <1>
.likeText("text like this one") <2>
.minTermFreq(1) <3>
.maxQueryTerms(12); <4>
--------------------------------------------------
<1> field
<2> text
<3> ignore threshold
<4> max num of Terms in generated queries
[[prefix]]
=== Prefix Query
See {ref}/query-dsl-prefix-query.html[Prefix Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = prefixQuery(
"brand", <1>
"heine" <2>
);
--------------------------------------------------
<1> field
<2> prefix
[[query-string]]
=== QueryString Query
See {ref}/query-dsl-query-string-query.html[QueryString Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = queryString("+kimchy -elasticsearch"); <1>
--------------------------------------------------
<1> text
[[java-range]]
=== Range Query
See {ref}/query-dsl-range-query.html[Range Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = rangeQuery("price") <1>
.from(5) <2>
.to(10) <3>
.includeLower(true) <4>
.includeUpper(false); <5>
--------------------------------------------------
<1> field
<2> from
<3> to
<4> include lower value means that `from` is `gt` when `false` or `gte` when `true`
<5> include upper value means that `to` is `lt` when `false` or `lte` when `true`
=== Span Queries (first, near, not, or, term)
See:
* {ref}/query-dsl-span-term-query.html[Span Term Query]
* {ref}/query-dsl-span-first-query.html[Span First Query]
* {ref}/query-dsl-span-near-query.html[Span Near Query]
* {ref}/query-dsl-span-not-query.html[Span Not Query]
* {ref}/query-dsl-span-or-query.html[Span Or Query]
[source,java]
--------------------------------------------------
// Span Term
QueryBuilder qb = spanTermQuery(
"user", <1>
"kimchy" <2>
);
--------------------------------------------------
<1> field
<2> value
[source,java]
--------------------------------------------------
// Span First
QueryBuilder qb = spanFirstQuery(
spanTermQuery("user", "kimchy"), <1>
3 <2>
);
--------------------------------------------------
<1> query
<2> max end position
[source,java]
--------------------------------------------------
// Span Near
QueryBuilder qb = spanNearQuery()
.clause(spanTermQuery("field","value1")) <1>
.clause(spanTermQuery("field","value2")) <1>
.clause(spanTermQuery("field","value3")) <1>
.slop(12) <2>
.inOrder(false) <3>
.collectPayloads(false); <4>
--------------------------------------------------
<1> span term queries
<2> slop factor: the maximum number of intervening unmatched positions
<3> whether matches are required to be in-order
<4> collect payloads or not
[source,java]
--------------------------------------------------
// Span Not
QueryBuilder qb = spanNotQuery()
.include(spanTermQuery("field","value1")) <1>
.exclude(spanTermQuery("field","value2")); <2>
--------------------------------------------------
<1> span query whose matches are filtered
<2> span query whose matches must not overlap those returned
[source,java]
--------------------------------------------------
// Span Or
QueryBuilder qb = spanOrQuery()
.clause(spanTermQuery("field","value1")) <1>
.clause(spanTermQuery("field","value2")) <1>
.clause(spanTermQuery("field","value3")); <1>
--------------------------------------------------
<1> span term queries
[[term]]
=== Term Query
See {ref}/query-dsl-term-query.html[Term Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = termQuery(
"name", <1>
"kimchy"); <2>
--------------------------------------------------
<1> field
<2> value
[[java-terms]]
=== Terms Query
See {ref}/query-dsl-terms-query.html[Terms Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = termsQuery("tags", <1>
"blue", "pill") <2>
.minimumMatch(1); <3>
--------------------------------------------------
<1> field
<2> values
<3> how many terms must match at least
[[top-children]]
=== Top Children Query
See {ref}/query-dsl-top-children-query.html[Top Children Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = topChildrenQuery(
"blog_tag", <1>
termQuery("tag", "something") <2>
)
.score("max") <3>
.factor(5) <4>
.incrementalFactor(2); <5>
--------------------------------------------------
<1> field
<2> query
<3> `max`, `sum` or `avg`
<4> how many hits are asked for in the first child query run (defaults to 5)
<5> if not enough parents are found, and there are still more child docs to query, then the child search hits are
expanded by multiplying by the incremental_factor (defaults to 2).
[[wildcard]]
=== Wildcard Query
See {ref}/query-dsl-wildcard-query.html[Wildcard Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = wildcardQuery("user", "k?mc*");
--------------------------------------------------
[[nested]]
=== Nested Query
See {ref}/query-dsl-nested-query.html[Nested Query]
[source,java]
--------------------------------------------------
QueryBuilder qb = nestedQuery(
"obj1", <1>
boolQuery() <2>
.must(matchQuery("obj1.name", "blue"))
.must(rangeQuery("obj1.count").gt(5))
)
.scoreMode("avg"); <3>
--------------------------------------------------
<1> path to nested document
<2> your query. Any fields referenced inside the query must use the complete path (fully qualified).
<3> score mode could be `max`, `total`, `avg` (default) or `none`
[[indices]]
=== Indices Query
See {ref}/query-dsl-indices-query.html[Indices Query]
[source,java]
--------------------------------------------------
// Using another query when no match for the main one
QueryBuilder qb = indicesQuery(
termQuery("tag", "wow"), <1>
"index1", "index2" <2>
)
.noMatchQuery(termQuery("tag", "kow")); <3>
--------------------------------------------------
<1> query to be executed on selected indices
<2> selected indices
<3> query to be executed on non matching indices
[source,java]
--------------------------------------------------
// Using all (match all) or none (match no documents)
QueryBuilder qb = indicesQuery(
termQuery("tag", "wow"), <1>
"index1", "index2" <2>
)
.noMatchQuery("all"); <3>
--------------------------------------------------
<1> query to be executed on selected indices
<2> selected indices
<3> `none` (to match no documents), and `all` (to match all documents). Defaults to `all`.
[[geo-shape]]
=== GeoShape Query
See {ref}/query-dsl-geo-shape-query.html[GeoShape Query]
Note: the `geo_shape` type uses `Spatial4J` and `JTS`, both of which are
optional dependencies. Consequently you must add `Spatial4J` and `JTS`
to your classpath in order to use this type:
[source,xml]
--------------------------------------------------
<dependency>
<groupId>com.spatial4j</groupId>
<artifactId>spatial4j</artifactId>
<version>0.4.1</version> <1>
</dependency>
<dependency>
<groupId>com.vividsolutions</groupId>
<artifactId>jts</artifactId>
<version>1.13</version> <2>
<exclusions>
<exclusion>
<groupId>xerces</groupId>
<artifactId>xercesImpl</artifactId>
</exclusion>
</exclusions>
</dependency>
--------------------------------------------------
<1> check for updates in http://search.maven.org/#search%7Cga%7C1%7Cg%3A%22com.spatial4j%22%20AND%20a%3A%22spatial4j%22[Maven Central]
<2> check for updates in http://search.maven.org/#search%7Cga%7C1%7Cg%3A%22com.vividsolutions%22%20AND%20a%3A%22jts%22[Maven Central]
[source,java]
--------------------------------------------------
// Import Spatial4J shapes
import com.spatial4j.core.context.SpatialContext;
import com.spatial4j.core.shape.Shape;
import com.spatial4j.core.shape.impl.RectangleImpl;
// Also import ShapeRelation
import org.elasticsearch.common.geo.ShapeRelation;
--------------------------------------------------
[source,java]
--------------------------------------------------
// Shape within another
QueryBuilder qb = geoShapeQuery(
"location", <1>
new RectangleImpl(0,10,0,10,SpatialContext.GEO) <2>
)
.relation(ShapeRelation.WITHIN); <3>
--------------------------------------------------
<1> field
<2> shape
<3> relation
[source,java]
--------------------------------------------------
// Intersect shapes
QueryBuilder qb = geoShapeQuery(
"location", <1>
new PointImpl(0, 0, SpatialContext.GEO) <2>
)
.relation(ShapeRelation.INTERSECTS); <3>
--------------------------------------------------
<1> field
<2> shape
<3> relation
[source,java]
--------------------------------------------------
// Using pre-indexed shapes
QueryBuilder qb = geoShapeQuery(
"location", <1>
"New Zealand", <2>
"countries") <3>
.relation(ShapeRelation.DISJOINT); <4>
--------------------------------------------------
<1> field
<2> indexed shape id
<3> index type of the indexed shapes
<4> relation