Allow doc-values only search on boolean fields (#82925)

Allows searching on boolean fields when those fields are not indexed (index: false) but just doc values are enabled.

This enables searches on archive data, which has access to doc values but not index structures. When combined with
searchable snapshots, it allows downloading only data for a given (doc value) field to quickly filter down to a select set
of documents.

Relates #81210 and #52728

docs/reference/mapping/params/doc-values.asciidoc

@@ -17,7 +17,8 @@ makes this data access pattern possible. They store the same values as the
 sorting and aggregations. Doc values are supported on almost all field types,
 with the __notable exception of `text` and `annotated_text` fields__.
 
-<<number,Numeric types>>, <<date,date types>>, and the <<keyword, keyword type>>
+<<number,Numeric types>>, <<date,date types>>, the <<boolean,boolean type>>
+and the <<keyword,keyword type>>
 can also be queried using term or range-based queries
 when they are not <<mapping-index,indexed>> but only have doc values enabled.
 Query performance on doc values is much slower than on index structures, but

docs/reference/mapping/types/boolean.asciidoc

@@ -174,7 +174,10 @@ The following parameters are accepted by `boolean` fields:
 
 <<mapping-index,`index`>>::
 
-    Should the field be searchable? Accepts `true` (default) and `false`.
+    Should the field be quickly searchable? Accepts `true` (default) and
+    `false`. Fields that only have <<doc-values,`doc_values`>>
+    enabled can still be queried using term or range-based queries,
+    albeit slower.
 
 <<null-value,`null_value`>>::
 

docs/reference/query-dsl.asciidoc

@@ -33,7 +33,7 @@ the stability of the cluster. Those queries can be categorised as follows:
 
 * Queries that need to do linear scans to identify matches:
 ** <<query-dsl-script-query,`script` queries>>
-** queries on <<number,numeric>>, <<date,date>>, or <<keyword,keyword>> fields that are not indexed
+** queries on <<number,numeric>>, <<date,date>>, <<boolean,boolean>>, or <<keyword,keyword>> fields that are not indexed
    but have <<doc-values,doc values>> enabled
 
 * Queries that have a high up-front cost:

rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/field_caps/10_basic.yml

@@ -89,6 +89,9 @@ setup:
                   non_indexed_keyword:
                     type:     keyword
                     index:    false
+                  non_indexed_boolean:
+                    type:     boolean
+                    index:    false
                   geo:
                     type:     keyword
                   object:
@@ -240,6 +243,18 @@ setup:
 
   - match: {fields.non_indexed_keyword.keyword.searchable:                 true}
 
+---
+"Field caps for boolean field with only doc values":
+  - skip:
+      version: " - 8.0.99"
+      reason: "doc values search was added in 8.1.0"
+  - do:
+      field_caps:
+        index: 'test1,test2,test3'
+        fields: non_indexed_boolean
+
+  - match: {fields.non_indexed_boolean.boolean.searchable:                 true}
+
 ---
 "Get object and nested field caps":
 

rest-api-spec/src/yamlRestTest/resources/rest-api-spec/test/search/390_doc_values_search.yml

@@ -39,6 +39,9 @@ setup:
               keyword:
                 type: keyword
                 index: false
+              boolean:
+                type: boolean
+                index: false
 
   - do:
       index:
@@ -54,6 +57,7 @@ setup:
           short: 1
           date: "2017/01/01"
           keyword: "key1"
+          boolean: "false"
 
   - do:
       index:
@@ -69,6 +73,7 @@ setup:
           short: 2
           date: "2017/01/02"
           keyword: "key2"
+          boolean: "true"
 
   - do:
       indices.refresh: {}
@@ -252,3 +257,30 @@ setup:
         index: test
         body: { query: { range: { keyword: { gte: "key1" } } } }
   - length:   { hits.hits: 2  }
+
+---
+"Test match query on boolean field where only doc values are enabled":
+
+  - do:
+      search:
+        index: test
+        body: { query: { match: { boolean: { query: "false" } } } }
+  - length:   { hits.hits: 1  }
+
+---
+"Test terms query on boolean field where only doc values are enabled":
+
+  - do:
+      search:
+        index: test
+        body: { query: { terms: { boolean: [ "false", "true" ] } } }
+  - length:   { hits.hits: 2  }
+
+---
+"Test range query on boolean field where only doc values are enabled":
+
+  - do:
+      search:
+        index: test
+        body: { query: { range: { boolean: { gte: "false" } } } }
+  - length:   { hits.hits: 2  }

server/src/main/java/org/elasticsearch/index/mapper/BooleanFieldMapper.java

@@ -14,6 +14,10 @@
 import org.apache.lucene.document.StoredField;
 import org.apache.lucene.index.IndexOptions;
 import org.apache.lucene.index.LeafReaderContext;
+import org.apache.lucene.search.BooleanClause;
+import org.apache.lucene.search.BooleanQuery;
+import org.apache.lucene.search.ConstantScoreQuery;
+import org.apache.lucene.search.MatchNoDocsQuery;
 import org.apache.lucene.search.Query;
 import org.apache.lucene.search.TermRangeQuery;
 import org.apache.lucene.util.BytesRef;
@@ -37,6 +41,7 @@
 
 import java.io.IOException;
 import java.time.ZoneId;
+import java.util.Collection;
 import java.util.Collections;
 import java.util.List;
 import java.util.Map;
@@ -155,18 +160,27 @@ public BooleanFieldType(
         }
 
         public BooleanFieldType(String name) {
-            this(name, true, false, true, false, null, Collections.emptyMap());
+            this(name, true);
         }
 
-        public BooleanFieldType(String name, boolean searchable) {
-            this(name, searchable, false, true, false, null, Collections.emptyMap());
+        public BooleanFieldType(String name, boolean isIndexed) {
+            this(name, isIndexed, true);
+        }
+
+        public BooleanFieldType(String name, boolean isIndexed, boolean hasDocValues) {
+            this(name, isIndexed, isIndexed, hasDocValues, false, null, Collections.emptyMap());
         }
 
         @Override
         public String typeName() {
             return CONTENT_TYPE;
         }
 
+        @Override
+        public boolean isSearchable() {
+            return isIndexed() || hasDocValues();
+        }
+
         @Override
         public ValueFetcher valueFetcher(SearchExecutionContext context, String format) {
             if (format != null) {
@@ -209,6 +223,15 @@ public BytesRef indexedValueForSearch(Object value) {
             };
         }
 
+        private long docValueForSearch(Object value) {
+            BytesRef ref = indexedValueForSearch(value);
+            if (Values.TRUE.equals(ref)) {
+                return 1;
+            } else {
+                return 0;
+            }
+        }
+
         @Override
         public Boolean valueForDisplay(Object value) {
             if (value == null) {
@@ -234,6 +257,30 @@ public DocValueFormat docValueFormat(@Nullable String format, ZoneId timeZone) {
             return DocValueFormat.BOOLEAN;
         }
 
+        @Override
+        public Query termQuery(Object value, SearchExecutionContext context) {
+            failIfNotIndexedNorDocValuesFallback(context);
+            if (isIndexed()) {
+                return super.termQuery(value, context);
+            } else {
+                return SortedNumericDocValuesField.newSlowExactQuery(name(), docValueForSearch(value));
+            }
+        }
+
+        @Override
+        public Query termsQuery(Collection<?> values, SearchExecutionContext context) {
+            failIfNotIndexedNorDocValuesFallback(context);
+            if (isIndexed()) {
+                return super.termsQuery(values, context);
+            } else {
+                BooleanQuery.Builder builder = new BooleanQuery.Builder();
+                for (Object value : values) {
+                    builder.add(termQuery(value, context), BooleanClause.Occur.SHOULD);
+                }
+                return new ConstantScoreQuery(builder.build());
+            }
+        }
+
         @Override
         public Query rangeQuery(
             Object lowerTerm,
@@ -242,14 +289,35 @@ public Query rangeQuery(
             boolean includeUpper,
             SearchExecutionContext context
         ) {
-            failIfNotIndexed();
-            return new TermRangeQuery(
-                name(),
-                lowerTerm == null ? null : indexedValueForSearch(lowerTerm),
-                upperTerm == null ? null : indexedValueForSearch(upperTerm),
-                includeLower,
-                includeUpper
-            );
+            failIfNotIndexedNorDocValuesFallback(context);
+            if (isIndexed()) {
+                return new TermRangeQuery(
+                    name(),
+                    lowerTerm == null ? null : indexedValueForSearch(lowerTerm),
+                    upperTerm == null ? null : indexedValueForSearch(upperTerm),
+                    includeLower,
+                    includeUpper
+                );
+            } else {
+                long l = 0;
+                long u = 1;
+                if (lowerTerm != null) {
+                    l = docValueForSearch(lowerTerm);
+                    if (includeLower == false) {
+                        l = Math.max(1, l + 1);
+                    }
+                }
+                if (upperTerm != null) {
+                    u = docValueForSearch(upperTerm);
+                    if (includeUpper == false) {
+                        l = Math.min(0, l - 1);
+                    }
+                }
+                if (l > u) {
+                    return new MatchNoDocsQuery();
+                }
+                return SortedNumericDocValuesField.newSlowRangeQuery(name(), l, u);
+            }
         }
     }
 

server/src/main/java/org/elasticsearch/index/mapper/KeywordFieldMapper.java

@@ -20,6 +20,7 @@
 import org.apache.lucene.index.MultiTerms;
 import org.apache.lucene.index.Terms;
 import org.apache.lucene.index.TermsEnum;
+import org.apache.lucene.sandbox.search.DocValuesTermsQuery;
 import org.apache.lucene.search.MultiTermQuery;
 import org.apache.lucene.search.Query;
 import org.apache.lucene.util.BytesRef;
@@ -50,6 +51,7 @@
 
 import java.io.IOException;
 import java.io.UncheckedIOException;
+import java.util.Collection;
 import java.util.Collections;
 import java.util.List;
 import java.util.Locale;
@@ -339,13 +341,51 @@ public KeywordFieldType(String name, NamedAnalyzer analyzer) {
         }
 
         @Override
-        protected boolean allowDocValueBasedQueries() {
-            return true;
+        public boolean isSearchable() {
+            return isIndexed() || hasDocValues();
         }
 
         @Override
-        public boolean isSearchable() {
-            return isIndexed() || hasDocValues();
+        public Query termQuery(Object value, SearchExecutionContext context) {
+            failIfNotIndexedNorDocValuesFallback(context);
+            if (isIndexed()) {
+                return super.termQuery(value, context);
+            } else {
+                return SortedSetDocValuesField.newSlowExactQuery(name(), indexedValueForSearch(value));
+            }
+        }
+
+        @Override
+        public Query termsQuery(Collection<?> values, SearchExecutionContext context) {
+            failIfNotIndexedNorDocValuesFallback(context);
+            if (isIndexed()) {
+                return super.termsQuery(values, context);
+            } else {
+                BytesRef[] bytesRefs = values.stream().map(this::indexedValueForSearch).toArray(BytesRef[]::new);
+                return new DocValuesTermsQuery(name(), bytesRefs);
+            }
+        }
+
+        @Override
+        public Query rangeQuery(
+            Object lowerTerm,
+            Object upperTerm,
+            boolean includeLower,
+            boolean includeUpper,
+            SearchExecutionContext context
+        ) {
+            failIfNotIndexedNorDocValuesFallback(context);
+            if (isIndexed()) {
+                return super.rangeQuery(lowerTerm, upperTerm, includeLower, includeUpper, context);
+            } else {
+                return SortedSetDocValuesField.newSlowRangeQuery(
+                    name(),
+                    lowerTerm == null ? null : indexedValueForSearch(lowerTerm),
+                    upperTerm == null ? null : indexedValueForSearch(upperTerm),
+                    includeLower,
+                    includeUpper
+                );
+            }
         }
 
         @Override

server/src/main/java/org/elasticsearch/index/mapper/StringFieldType.java

@@ -9,7 +9,6 @@
 package org.elasticsearch.index.mapper;
 
 import org.apache.lucene.analysis.Analyzer;
-import org.apache.lucene.document.SortedSetDocValuesField;
 import org.apache.lucene.index.Term;
 import org.apache.lucene.search.AutomatonQuery;
 import org.apache.lucene.search.FuzzyQuery;
@@ -211,27 +210,13 @@ public Query rangeQuery(
                     + "' is set to false."
             );
         }
-        if (allowDocValueBasedQueries()) {
-            failIfNotIndexedNorDocValuesFallback(context);
-        } else {
-            failIfNotIndexed();
-        }
-        if (isIndexed()) {
-            return new TermRangeQuery(
-                name(),
-                lowerTerm == null ? null : indexedValueForSearch(lowerTerm),
-                upperTerm == null ? null : indexedValueForSearch(upperTerm),
-                includeLower,
-                includeUpper
-            );
-        } else {
-            return SortedSetDocValuesField.newSlowRangeQuery(
-                name(),
-                lowerTerm == null ? null : indexedValueForSearch(lowerTerm),
-                upperTerm == null ? null : indexedValueForSearch(upperTerm),
-                includeLower,
-                includeUpper
-            );
-        }
+        failIfNotIndexed();
+        return new TermRangeQuery(
+            name(),
+            lowerTerm == null ? null : indexedValueForSearch(lowerTerm),
+            upperTerm == null ? null : indexedValueForSearch(upperTerm),
+            includeLower,
+            includeUpper
+        );
     }
 }