ES|QL: add local optimizations for constant_keyword

docs/changelog/127549.yaml

@@ -0,0 +1,5 @@
+pr: 127549
+summary: Add local optimizations for `constant_keyword`
+area: ES|QL
+type: enhancement
+issues: []

x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/optimizer/LocalLogicalPlanOptimizer.java

@@ -12,6 +12,7 @@
 import org.elasticsearch.xpack.esql.optimizer.rules.logical.local.InferIsNotNull;
 import org.elasticsearch.xpack.esql.optimizer.rules.logical.local.InferNonNullAggConstraint;
 import org.elasticsearch.xpack.esql.optimizer.rules.logical.local.LocalPropagateEmptyRelation;
+import org.elasticsearch.xpack.esql.optimizer.rules.logical.local.ReplaceConstantKeywords;
 import org.elasticsearch.xpack.esql.optimizer.rules.logical.local.ReplaceMissingFieldWithNull;
 import org.elasticsearch.xpack.esql.optimizer.rules.logical.local.ReplaceTopNWithLimitAndSort;
 import org.elasticsearch.xpack.esql.plan.logical.LogicalPlan;
@@ -38,6 +39,7 @@ public class LocalLogicalPlanOptimizer extends ParameterizedRuleExecutor<Logical
             new Batch<>(
                 "Local rewrite",
                 Limiter.ONCE,
+                new ReplaceConstantKeywords(),
                 new ReplaceTopNWithLimitAndSort(),
                 new ReplaceMissingFieldWithNull(),
                 new InferIsNotNull(),

x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/optimizer/rules/logical/local/ReplaceConstantKeywords.java

@@ -0,0 +1,67 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the Elastic License
+ * 2.0; you may not use this file except in compliance with the Elastic License
+ * 2.0.
+ */
+
+package org.elasticsearch.xpack.esql.optimizer.rules.logical.local;
+
+import org.elasticsearch.index.IndexMode;
+import org.elasticsearch.xpack.esql.core.expression.Attribute;
+import org.elasticsearch.xpack.esql.core.expression.Expression;
+import org.elasticsearch.xpack.esql.core.expression.FieldAttribute;
+import org.elasticsearch.xpack.esql.core.expression.Literal;
+import org.elasticsearch.xpack.esql.optimizer.LocalLogicalOptimizerContext;
+import org.elasticsearch.xpack.esql.plan.logical.EsRelation;
+import org.elasticsearch.xpack.esql.plan.logical.Eval;
+import org.elasticsearch.xpack.esql.plan.logical.Filter;
+import org.elasticsearch.xpack.esql.plan.logical.LogicalPlan;
+import org.elasticsearch.xpack.esql.plan.logical.OrderBy;
+import org.elasticsearch.xpack.esql.plan.logical.RegexExtract;
+import org.elasticsearch.xpack.esql.plan.logical.TopN;
+import org.elasticsearch.xpack.esql.rule.ParameterizedRule;
+
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * Look for any constant_keyword fields used in the plan and replaces them with their actual value.
+ */
+public class ReplaceConstantKeywords extends ParameterizedRule<LogicalPlan, LogicalPlan, LocalLogicalOptimizerContext> {
+
+    @Override
+    public LogicalPlan apply(LogicalPlan plan, LocalLogicalOptimizerContext localLogicalOptimizerContext) {
+        Map<Attribute, Expression> attrToValue = new HashMap<>();
+        plan.forEachUp(EsRelation.class, esRelation -> {
+            if (esRelation.indexMode() == IndexMode.STANDARD) {
+                for (Attribute attribute : esRelation.output()) {
+                    var val = localLogicalOptimizerContext.searchStats().constantValue(attribute.name());
+                    if (val != null) {
+                        attrToValue.put(attribute, Literal.of(attribute, val));
+                    }
+                }
+            }
+        });
+        if (attrToValue.isEmpty()) {
+            return plan;
+        }
+        return plan.transformUp(p -> replaceAttributes(p, attrToValue));
+    }
+
+    private LogicalPlan replaceAttributes(LogicalPlan plan, Map<Attribute, Expression> attrToValue) {
+        // This is slightly different from ReplaceMissingFieldWithNull.
+        // It's on purpose: reusing NameIDs is dangerous, and we have no evidence that adding an EVAL will actually lead to
+        // practical performance benefits
+        if (plan instanceof Eval
+            || plan instanceof Filter
+            || plan instanceof OrderBy
+            || plan instanceof RegexExtract
+            || plan instanceof TopN) {
+            return plan.transformExpressionsOnlyUp(FieldAttribute.class, f -> attrToValue.getOrDefault(f, f));
+        }
+
+        return plan;
+    }
+
+}

x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/stats/SearchContextStats.java

@@ -310,6 +310,41 @@ public boolean canUseEqualityOnSyntheticSourceDelegate(String name, String value
         return true;
     }
 
+    public String constantValue(String name) {
+        String val = null;
+        for (SearchExecutionContext ctx : contexts) {
+            MappedFieldType f = ctx.getFieldType(name);
+            if (f == null) {
+                return null;
+            }
+            if (f instanceof ConstantFieldType cf) {
+                var fetcher = cf.valueFetcher(ctx, null);
+                String thisVal = null;
+                try {
+                    // since the value is a constant, the doc _should_ be irrelevant
+                    List<Object> vals = fetcher.fetchValues(null, -1, null);
+                    Object objVal = vals.size() == 1 ? vals.get(0) : null;
+                    // we are considering only string values for now, since this can return "strange" things,
+                    // see IndexModeFieldType
+                    thisVal = objVal instanceof String ? (String) objVal : null;
+                } catch (IOException iox) {}
+
+                if (thisVal == null) {
+                    // Value not yet set
+                    return null;
+                }
+                if (val == null) {
+                    val = thisVal;
+                } else if (thisVal.equals(val) == false) {
+                    return null;
+                }
+            } else {
+                return null;
+            }
+        }
+        return val;
+    }
+
     private interface DocCountTester {
         Boolean test(LeafReader leafReader) throws IOException;
     }

x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/stats/SearchStats.java

@@ -39,6 +39,14 @@ public interface SearchStats {
 
     boolean canUseEqualityOnSyntheticSourceDelegate(String name, String value);
 
+    /**
+     * Returns the value for a field if it's a constant (eg. a constant_keyword with only one value for the involved indices).
+     * NULL if the field is not a constant.
+     */
+    default String constantValue(String name) {
+        return null;
+    }
+
     /**
      * When there are no search stats available, for example when there are no search contexts, we have static results.
      */

x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/optimizer/LocalPhysicalPlanOptimizerTests.java

@@ -59,6 +59,7 @@
 import org.elasticsearch.xpack.esql.plan.logical.Enrich;
 import org.elasticsearch.xpack.esql.plan.logical.LogicalPlan;
 import org.elasticsearch.xpack.esql.plan.physical.AggregateExec;
+import org.elasticsearch.xpack.esql.plan.physical.DissectExec;
 import org.elasticsearch.xpack.esql.plan.physical.EsQueryExec;
 import org.elasticsearch.xpack.esql.plan.physical.EsStatsQueryExec;
 import org.elasticsearch.xpack.esql.plan.physical.EsStatsQueryExec.Stat;
@@ -69,6 +70,7 @@
 import org.elasticsearch.xpack.esql.plan.physical.FilterExec;
 import org.elasticsearch.xpack.esql.plan.physical.LimitExec;
 import org.elasticsearch.xpack.esql.plan.physical.LocalSourceExec;
+import org.elasticsearch.xpack.esql.plan.physical.MvExpandExec;
 import org.elasticsearch.xpack.esql.plan.physical.PhysicalPlan;
 import org.elasticsearch.xpack.esql.plan.physical.ProjectExec;
 import org.elasticsearch.xpack.esql.plan.physical.TimeSeriesAggregateExec;
@@ -150,6 +152,18 @@ public boolean isSingleValue(String field) {
         }
     };
 
+    private final SearchStats CONSTANT_K_STATS = new TestSearchStats() {
+        @Override
+        public boolean isSingleValue(String field) {
+            return true;
+        }
+
+        @Override
+        public String constantValue(String name) {
+            return name.startsWith("constant_keyword") ? "foo" : null;
+        }
+    };
+
     @ParametersFactory(argumentFormatting = PARAM_FORMATTING)
     public static List<Object[]> readScriptSpec() {
         return settings().stream().map(t -> {
@@ -1856,6 +1870,113 @@ public void testPushDownFieldExtractToTimeSeriesSource() {
         assertTrue(timeSeriesSource.attrs().stream().noneMatch(EsQueryExec::isSourceAttribute));
     }
 
+    /**
+     * LimitExec[1000[INTEGER]]
+     * \_ExchangeExec[[!alias_integer, boolean{f}#415, byte{f}#416, constant_keyword-foo{f}#417, date{f}#418, date_nanos{f}#419,
+     *   double{f}#420, float{f}#421, half_float{f}#422, integer{f}#424, ip{f}#425, keyword{f}#426, long{f}#427, scaled_float{f}#423,
+     *   !semantic_text, short{f}#429, text{f}#430, unsigned_long{f}#428, version{f}#431, wildcard{f}#432], false]
+     *   \_ProjectExec[[!alias_integer, boolean{f}#415, byte{f}#416, constant_keyword-foo{f}#417, date{f}#418, date_nanos{f}#419,
+     *     double{f}#420, float{f}#421, half_float{f}#422, integer{f}#424, ip{f}#425, keyword{f}#426, long{f}#427, scaled_float{f}#423,
+     *     !semantic_text, short{f}#429, text{f}#430, unsigned_long{f}#428, version{f}#431, wildcard{f}#432]]
+     *     \_FieldExtractExec[!alias_integer, boolean{f}#415, byte{f}#416, consta..]
+     *       \_EsQueryExec[test], indexMode[standard], query[][_doc{f}#434], limit[1000], sort[] estimatedRowSize[412]
+     */
+    public void testConstantKeywordWithMatchingFilter() {
+        String queryText = """
+            from test
+            | where `constant_keyword-foo` == "foo"
+            """;
+        var analyzer = makeAnalyzer("mapping-all-types.json");
+        var plan = plannerOptimizer.plan(queryText, CONSTANT_K_STATS, analyzer);
+
+        var limit = as(plan, LimitExec.class);
+        var exchange = as(limit.child(), ExchangeExec.class);
+        var project = as(exchange.child(), ProjectExec.class);
+        var field = as(project.child(), FieldExtractExec.class);
+        var query = as(field.child(), EsQueryExec.class);
+        assertThat(as(query.limit(), Literal.class).value(), is(1000));
+        assertNull(query.query());
+    }
+
+    /**
+     * LimitExec[1000[INTEGER]]
+     * \_ExchangeExec[[!alias_integer, boolean{f}#4, byte{f}#5, constant_keyword-foo{f}#6, date{f}#7, date_nanos{f}#8, double{f}#9,
+     *    float{f}#10, half_float{f}#11, integer{f}#13, ip{f}#14, keyword{f}#15, long{f}#16, scaled_float{f}#12, !semantic_text,
+     *    short{f}#18, text{f}#19, unsigned_long{f}#17, version{f}#20, wildcard{f}#21], false]
+     *   \_LocalSourceExec[[!alias_integer, boolean{f}#4, byte{f}#5, constant_keyword-foo{f}#6, date{f}#7, date_nanos{f}#8, double{f}#9,
+     *     float{f}#10, half_float{f}#11, integer{f}#13, ip{f}#14, keyword{f}#15, long{f}#16, scaled_float{f}#12, !semantic_text,
+     *     short{f}#18, text{f}#19, unsigned_long{f}#17, version{f}#20, wildcard{f}#21], EMPTY]
+     */
+    public void testConstantKeywordWithNonMatchingFilter() {
+        String queryText = """
+            from test
+            | where `constant_keyword-foo` == "non-matching"
+            """;
+        var analyzer = makeAnalyzer("mapping-all-types.json");
+        var plan = plannerOptimizer.plan(queryText, CONSTANT_K_STATS, analyzer);
+
+        var limit = as(plan, LimitExec.class);
+        var exchange = as(limit.child(), ExchangeExec.class);
+        var source = as(exchange.child(), LocalSourceExec.class);
+    }
+
+    /**
+     * LimitExec[1000[INTEGER]]
+     * \_ExchangeExec[[!alias_integer, boolean{f}#6, byte{f}#7, constant_keyword-foo{r}#25, date{f}#9, date_nanos{f}#10, double{f}#1...
+     *   \_ProjectExec[[!alias_integer, boolean{f}#6, byte{f}#7, constant_keyword-foo{r}#25, date{f}#9, date_nanos{f}#10, double{f}#1...
+     *     \_FieldExtractExec[!alias_integer, boolean{f}#6, byte{f}#7, date{f}#9,
+     *       \_LimitExec[1000[INTEGER]]
+     *         \_FilterExec[constant_keyword-foo{r}#25 == [66 6f 6f][KEYWORD]]
+     *           \_MvExpandExec[constant_keyword-foo{f}#8,constant_keyword-foo{r}#25]
+     *             \_FieldExtractExec[constant_keyword-foo{f}#8]
+     *               \_EsQueryExec[test], indexMode[standard], query[][_doc{f}#26], limit[], sort[] estimatedRowSize[412]
+     */
+    public void testConstantKeywordExpandFilter() {
+        String queryText = """
+            from test
+            | mv_expand `constant_keyword-foo`
+            | where `constant_keyword-foo` == "foo"
+            """;
+        var analyzer = makeAnalyzer("mapping-all-types.json");
+        var plan = plannerOptimizer.plan(queryText, CONSTANT_K_STATS, analyzer);
+
+        var limit = as(plan, LimitExec.class);
+        var exchange = as(limit.child(), ExchangeExec.class);
+        var project = as(exchange.child(), ProjectExec.class);
+        var fieldExtract = as(project.child(), FieldExtractExec.class);
+        var limit2 = as(fieldExtract.child(), LimitExec.class);
+        var filter = as(limit2.child(), FilterExec.class);
+        var expand = as(filter.child(), MvExpandExec.class);
+        var field = as(expand.child(), FieldExtractExec.class); // MV_EXPAND is not optimized yet (it doesn't accept literals)
+        as(field.child(), EsQueryExec.class);
+    }
+
+    /**
+     * DissectExec[constant_keyword-foo{f}#8,Parser[pattern=%{bar}, appendSeparator=, ...
+     * \_LimitExec[1000[INTEGER]]
+     *   \_ExchangeExec[[!alias_integer, boolean{f}#6, byte{f}#7, constant_keyword-foo{f}#8, date{f}#9, date_nanos{f}#10, double{f}#11...
+     *     \_ProjectExec[[!alias_integer, boolean{f}#6, byte{f}#7, constant_keyword-foo{f}#8, date{f}#9, date_nanos{f}#10, double{f}#11...
+     *       \_FieldExtractExec[!alias_integer, boolean{f}#6, byte{f}#7, constant_k..]
+     *         \_EsQueryExec[test], indexMode[standard], query[][_doc{f}#25], limit[1000], sort[] estimatedRowSize[462]
+     */
+    public void testConstantKeywordDissectFilter() {
+        String queryText = """
+            from test
+            | dissect `constant_keyword-foo` "%{bar}"
+            | where `constant_keyword-foo` == "foo"
+            """;
+        var analyzer = makeAnalyzer("mapping-all-types.json");
+        var plan = plannerOptimizer.plan(queryText, CONSTANT_K_STATS, analyzer);
+
+        var dissect = as(plan, DissectExec.class);
+        var limit = as(dissect.child(), LimitExec.class);
+        var exchange = as(limit.child(), ExchangeExec.class);
+        var project = as(exchange.child(), ProjectExec.class);
+        var field = as(project.child(), FieldExtractExec.class);
+        var query = as(field.child(), EsQueryExec.class);
+        assertNull(query.query());
+    }
+
     private QueryBuilder wrapWithSingleQuery(String query, QueryBuilder inner, String fieldName, Source source) {
         return FilterTests.singleValueQuery(query, inner, fieldName, source);
     }