Allow parameter for ES|QL SAMPLE (elastic#129392)

* Allow parameter for ES|QL SAMPLE

* fix test to work around issue elastic#120272

* remove unused postAnalysisVerification

Author: jan-elastic

x-pack/plugin/esql/src/main/antlr/EsqlBaseParser.g4

@@ -389,5 +389,5 @@ completionCommand
     ;
 
 sampleCommand
-    : DEV_SAMPLE probability=decimalValue
+    : DEV_SAMPLE probability=constant
     ;

x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/action/EsqlCapabilities.java

@@ -928,7 +928,7 @@ public enum Cap {
         TO_LOWER_EMPTY_STRING,
 
         /**
-         * Support parameters for LiMIT command.
+         * Support parameters for LIMIT command.
          */
         PARAMETER_FOR_LIMIT,
 
@@ -965,7 +965,12 @@ public enum Cap {
         /**
          * Support for the SAMPLE command
          */
-        SAMPLE_V3(Build.current().isSnapshot());
+        SAMPLE_V3(Build.current().isSnapshot()),
+
+        /**
+         * Support parameters for SAMPLE command.
+         */
+        PARAMETER_FOR_SAMPLE(Build.current().isSnapshot());
 
         private final boolean enabled;
 

x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/parser/EsqlBaseParser.interp

@@ -707,7 +707,15 @@ public Literal inferenceId(EsqlBaseParser.IdentifierOrParameterContext ctx) {
     }
 
     public PlanFactory visitSampleCommand(EsqlBaseParser.SampleCommandContext ctx) {
-        var probability = visitDecimalValue(ctx.probability);
-        return plan -> new Sample(source(ctx), probability, plan);
+        Source source = source(ctx);
+        Object val = expression(ctx.probability).fold(FoldContext.small() /* TODO remove me */);
+        if (val instanceof Double probability && probability > 0.0 && probability < 1.0) {
+            return input -> new Sample(source, new Literal(source, probability, DataType.DOUBLE), input);
+        } else {
+            throw new ParsingException(
+                source(ctx),
+                "invalid value for SAMPLE probability [" + val + "], expecting a number between 0 and 1, exclusive"
+            );
+        }
     }
 }

x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/parser/EsqlBaseParser.java

@@ -9,23 +9,16 @@
 import org.elasticsearch.common.io.stream.NamedWriteableRegistry;
 import org.elasticsearch.common.io.stream.StreamInput;
 import org.elasticsearch.common.io.stream.StreamOutput;
-import org.elasticsearch.search.aggregations.bucket.sampler.random.RandomSamplingQuery;
-import org.elasticsearch.xpack.esql.capabilities.PostAnalysisVerificationAware;
 import org.elasticsearch.xpack.esql.capabilities.TelemetryAware;
-import org.elasticsearch.xpack.esql.common.Failures;
 import org.elasticsearch.xpack.esql.core.expression.Expression;
-import org.elasticsearch.xpack.esql.core.expression.FoldContext;
-import org.elasticsearch.xpack.esql.core.expression.Foldables;
 import org.elasticsearch.xpack.esql.core.tree.NodeInfo;
 import org.elasticsearch.xpack.esql.core.tree.Source;
 import org.elasticsearch.xpack.esql.io.stream.PlanStreamInput;
 
 import java.io.IOException;
 import java.util.Objects;
 
-import static org.elasticsearch.xpack.esql.common.Failure.fail;
-
-public class Sample extends UnaryPlan implements TelemetryAware, PostAnalysisVerificationAware {
+public class Sample extends UnaryPlan implements TelemetryAware {
     public static final NamedWriteableRegistry.Entry ENTRY = new NamedWriteableRegistry.Entry(LogicalPlan.class, "Sample", Sample::new);
 
     private final Expression probability;
@@ -92,13 +85,4 @@ public boolean equals(Object obj) {
 
         return Objects.equals(probability, other.probability) && Objects.equals(child(), other.child());
     }
-
-    @Override
-    public void postAnalysisVerification(Failures failures) {
-        try {
-            RandomSamplingQuery.checkProbabilityRange((double) Foldables.valueOf(FoldContext.small(), probability));
-        } catch (IllegalArgumentException e) {
-            failures.add(fail(probability, e.getMessage()));
-        }
-    }
 }

x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/parser/LogicalPlanBuilder.java

@@ -112,7 +112,6 @@
 import static org.elasticsearch.xpack.esql.analysis.AnalyzerTestUtils.defaultInferenceResolution;
 import static org.elasticsearch.xpack.esql.analysis.AnalyzerTestUtils.indexWithDateDateNanosUnionType;
 import static org.elasticsearch.xpack.esql.analysis.AnalyzerTestUtils.loadMapping;
-import static org.elasticsearch.xpack.esql.analysis.AnalyzerTestUtils.randomValueOtherThanTest;
 import static org.elasticsearch.xpack.esql.analysis.AnalyzerTestUtils.tsdbIndexResolution;
 import static org.elasticsearch.xpack.esql.core.tree.Source.EMPTY;
 import static org.elasticsearch.xpack.esql.core.type.DataType.DATETIME;
@@ -3166,20 +3165,6 @@ private boolean isMultiTypeEsField(Expression e) {
         return e instanceof FieldAttribute fa && fa.field() instanceof MultiTypeEsField;
     }
 
-    public void testRandomSampleProbability() {
-        assumeTrue("requires SAMPLE capability", EsqlCapabilities.Cap.SAMPLE_V3.isEnabled());
-
-        var e = expectThrows(VerificationException.class, () -> analyze("FROM test | SAMPLE 1."));
-        assertThat(e.getMessage(), containsString("RandomSampling probability must be strictly between 0.0 and 1.0, was [1.0]"));
-
-        e = expectThrows(VerificationException.class, () -> analyze("FROM test | SAMPLE .0"));
-        assertThat(e.getMessage(), containsString("RandomSampling probability must be strictly between 0.0 and 1.0, was [0.0]"));
-
-        double p = randomValueOtherThanTest(d -> 0 < d && d < 1, () -> randomDoubleBetween(0, Double.MAX_VALUE, false));
-        e = expectThrows(VerificationException.class, () -> analyze("FROM test | SAMPLE " + p));
-        assertThat(e.getMessage(), containsString("RandomSampling probability must be strictly between 0.0 and 1.0, was [" + p + "]"));
-    }
-
     private void verifyUnsupported(String query, String errorMessage) {
         verifyUnsupported(query, errorMessage, "mapping-multi-field-variation.json");
     }

x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/plan/logical/Sample.java

@@ -168,6 +168,25 @@ public void testInvalidLimit() {
         assertEquals("1:13: Invalid value for LIMIT [-1], expecting a non negative integer", error("row a = 1 | limit -1"));
     }
 
+    public void testInvalidSample() {
+        assertEquals(
+            "1:13: invalid value for SAMPLE probability [foo], expecting a number between 0 and 1, exclusive",
+            error("row a = 1 | sample \"foo\"")
+        );
+        assertEquals(
+            "1:13: invalid value for SAMPLE probability [-1.0], expecting a number between 0 and 1, exclusive",
+            error("row a = 1 | sample -1.0")
+        );
+        assertEquals(
+            "1:13: invalid value for SAMPLE probability [0], expecting a number between 0 and 1, exclusive",
+            error("row a = 1 | sample 0")
+        );
+        assertEquals(
+            "1:13: invalid value for SAMPLE probability [1], expecting a number between 0 and 1, exclusive",
+            error("row a = 1 | sample 1")
+        );
+    }
+
     private String error(String query) {
         ParsingException e = expectThrows(ParsingException.class, () -> defaultAnalyzer.analyze(parser.createStatement(query)));
         String message = e.getMessage();

x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/analysis/AnalyzerTests.java

@@ -103,6 +103,7 @@
 import static org.hamcrest.Matchers.hasSize;
 import static org.hamcrest.Matchers.instanceOf;
 import static org.hamcrest.Matchers.is;
+import static org.hamcrest.Matchers.startsWith;
 
 //@TestLogging(value = "org.elasticsearch.xpack.esql:TRACE", reason = "debug")
 public class StatementParserTests extends AbstractStatementParserTests {
@@ -3363,8 +3364,15 @@ public void testSample() {
         assumeTrue("SAMPLE requires corresponding capability", EsqlCapabilities.Cap.SAMPLE_V3.isEnabled());
         expectError("FROM test | SAMPLE .1 2", "line 1:23: extraneous input '2' expecting <EOF>");
         expectError("FROM test | SAMPLE .1 \"2\"", "line 1:23: extraneous input '\"2\"' expecting <EOF>");
-        expectError("FROM test | SAMPLE 1", "line 1:20: mismatched input '1' expecting {DECIMAL_LITERAL, '+', '-'}");
-        expectError("FROM test | SAMPLE", "line 1:19: mismatched input '<EOF>' expecting {DECIMAL_LITERAL, '+', '-'}");
+        expectError(
+            "FROM test | SAMPLE 1",
+            "line 1:13: invalid value for SAMPLE probability [1], expecting a number between 0 and 1, exclusive"
+        );
+        expectThrows(
+            ParsingException.class,
+            startsWith("line 1:19: mismatched input '<EOF>' expecting {"),
+            () -> statement("FROM test | SAMPLE")
+        );
     }
 
     static Alias alias(String name, Expression value) {

x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/analysis/ParsingTests.java

@@ -359,20 +359,22 @@ setup:
   - match: {values.2: ["1",2.0,null,true,123,1674835275193]}
 
 ---
-"Test Unnamed Input Params Also For Limit":
+"Test Unnamed Input Params Also For Limit And Sample":
   - requires:
         test_runner_features: [ capabilities ]
         capabilities:
           - method: POST
             path: /_query
             parameters: [ ]
-            capabilities: [ parameter_for_limit ]
+            capabilities: [ parameter_for_limit, parameter_for_sample ]
         reason: "named or positional parameters"
   - do:
       esql.query:
         body:
-          query: 'from test | eval x = ?, y = ?, z = ?, t = ?, u = ?, v = ? | keep x, y, z, t, u, v | limit ?'
-          params: ["1", 2.0, null, true, 123, 1674835275193, 3]
+          # The "| sort time" is to work around https://github.com/elastic/elasticsearch/issues/120272
+          # TODO: remove it when the issue is fixed
+          query: 'from test | sort time | sample ? | eval x = ?, y = ?, z = ?, t = ?, u = ?, v = ? | keep x, y, z, t, u, v | limit ?'
+          params: [0.999999999999, "1", 2.0, null, true, 123, 1674835275193, 3]
 
   - length: {columns: 6}
   - match: {columns.0.name: "x"}
@@ -401,14 +403,16 @@ setup:
         - method: POST
           path: /_query
           parameters: [ ]
-          capabilities: [ parameter_for_limit ]
+          capabilities: [ parameter_for_limit, parameter_for_sample ]
       reason: "named or positional parameters"
 
   - do:
       esql.query:
         body:
-          query: 'from test | eval x = ?, y = ?, z = ?, t = ?, u = ?, v = ? | keep x, y, z, t, u, v | limit ?'
-          params: [{"n1" : "1"}, {"n2" : 2.0}, {"n3" : null}, {"n4" : true}, {"n5" : 123}, {"n6": 1674835275193}, {"l": 3}]
+          # The "| sort time" is to work around https://github.com/elastic/elasticsearch/issues/120272
+          # TODO: remove it when the issue is fixed
+          query: 'from test | sort time | sample ? | eval x = ?, y = ?, z = ?, t = ?, u = ?, v = ? | keep x, y, z, t, u, v | limit ?'
+          params: [{"s": 0.99999999999}, {"n1" : "1"}, {"n2" : 2.0}, {"n3" : null}, {"n4" : true}, {"n5" : 123}, {"n6": 1674835275193}, {"l": 3}]
 
   - length: {columns: 6}
   - match: {columns.0.name: "x"}
@@ -465,14 +469,16 @@ setup:
         - method: POST
           path: /_query
           parameters: [ ]
-          capabilities: [ parameter_for_limit ]
+          capabilities: [ parameter_for_limit, parameter_for_sample ]
       reason: "named or positional parameters for field names"
 
   - do:
       esql.query:
         body:
-          query: 'from test | stats x = count(?f1), y = sum(?f2) by ?f3 | sort ?f3 | keep ?f3, x, y  | limit ?l'
-          params: [{"f1" : {"identifier" : "time"}}, {"f2" : { "identifier" : "count" }}, {"f3" : { "identifier" : "color"}}, {"l": 3}]
+          # The "| sort time" is to work around https://github.com/elastic/elasticsearch/issues/120272
+          # TODO: remove it when the issue is fixed
+          query: 'from test | sort time | sample ?s | stats x = count(?f1), y = sum(?f2) by ?f3 | sort ?f3 | keep ?f3, x, y  | limit ?l'
+          params: [{"s": 0.99999999999}, {"f1" : {"identifier" : "time"}}, {"f2" : { "identifier" : "count" }}, {"f3" : { "identifier" : "color"}}, {"l": 3}]
 
   - length: {columns: 3}
   - match: {columns.0.name: "color"}