Support Percentile in PPL

* Support Percentile in PPL Signed-off-by: Lantao Jin <[email protected]> * Remove ANSI SQL percentile syntax Signed-off-by: Lantao Jin <[email protected]> * add more unit tests and increase test coverage Signed-off-by: Lantao Jin <[email protected]> * increase test coverage Signed-off-by: Lantao Jin <[email protected]> * address comments and add docs Signed-off-by: Lantao Jin <[email protected]> * add examples in doc Signed-off-by: Lantao Jin <[email protected]> * fix doctest failure and add more integ tests Signed-off-by: Lantao Jin <[email protected]> * remove useless code and antlr4 files Signed-off-by: Lantao Jin <[email protected]> --------- Signed-off-by: Lantao Jin <[email protected]>
mengweieric · Jun 6, 2024 · d767868 · d767868
1 parent 3f1e3bd
commit d767868
Show file tree

Hide file tree

Showing 27 changed files with 1,375 additions and 71 deletions.
diff --git a/core/build.gradle b/core/build.gradle
@@ -55,6 +55,7 @@ dependencies {
     api "com.fasterxml.jackson.core:jackson-databind:${versions.jackson_databind}"
     api "com.fasterxml.jackson.core:jackson-annotations:${versions.jackson}"
     api group: 'com.google.code.gson', name: 'gson', version: '2.8.9'
+    api group: 'com.tdunning', name: 't-digest', version: '3.3'
     api project(':common')
 
     testImplementation('org.junit.jupiter:junit-jupiter:5.9.3')

diff --git a/core/src/main/java/org/opensearch/sql/expression/DSL.java b/core/src/main/java/org/opensearch/sql/expression/DSL.java
@@ -735,6 +735,18 @@ public static Aggregator max(Expression... expressions) {
     return aggregate(BuiltinFunctionName.MAX, expressions);
   }
 
+  /**
+   * OpenSearch uses T-Digest to approximate percentile, so PERCENTILE and PERCENTILE_APPROX are the
+   * same function.
+   */
+  public static Aggregator percentile(Expression... expressions) {
+    return percentileApprox(expressions);
+  }
+
+  public static Aggregator percentileApprox(Expression... expressions) {
+    return aggregate(BuiltinFunctionName.PERCENTILE_APPROX, expressions);
+  }
+
   private static Aggregator aggregate(BuiltinFunctionName functionName, Expression... expressions) {
     return compile(FunctionProperties.None, functionName, expressions);
   }

diff --git a/core/src/main/java/org/opensearch/sql/expression/aggregation/AggregatorFunction.java b/core/src/main/java/org/opensearch/sql/expression/aggregation/AggregatorFunction.java
@@ -57,6 +57,7 @@ public static void register(BuiltinFunctionRepository repository) {
     repository.register(stddevSamp());
     repository.register(stddevPop());
     repository.register(take());
+    repository.register(percentileApprox());
   }
 
   private static DefaultFunctionResolver avg() {
@@ -235,4 +236,46 @@ private static DefaultFunctionResolver take() {
                 .build());
     return functionResolver;
   }
+
+  private static DefaultFunctionResolver percentileApprox() {
+    FunctionName functionName = BuiltinFunctionName.PERCENTILE_APPROX.getName();
+    DefaultFunctionResolver functionResolver =
+        new DefaultFunctionResolver(
+            functionName,
+            new ImmutableMap.Builder<FunctionSignature, FunctionBuilder>()
+                .put(
+                    new FunctionSignature(functionName, ImmutableList.of(INTEGER, DOUBLE)),
+                    (functionProperties, arguments) ->
+                        PercentileApproximateAggregator.percentileApprox(arguments, INTEGER))
+                .put(
+                    new FunctionSignature(functionName, ImmutableList.of(INTEGER, DOUBLE, DOUBLE)),
+                    (functionProperties, arguments) ->
+                        PercentileApproximateAggregator.percentileApprox(arguments, INTEGER))
+                .put(
+                    new FunctionSignature(functionName, ImmutableList.of(LONG, DOUBLE)),
+                    (functionProperties, arguments) ->
+                        PercentileApproximateAggregator.percentileApprox(arguments, LONG))
+                .put(
+                    new FunctionSignature(functionName, ImmutableList.of(LONG, DOUBLE, DOUBLE)),
+                    (functionProperties, arguments) ->
+                        PercentileApproximateAggregator.percentileApprox(arguments, LONG))
+                .put(
+                    new FunctionSignature(functionName, ImmutableList.of(FLOAT, DOUBLE)),
+                    (functionProperties, arguments) ->
+                        PercentileApproximateAggregator.percentileApprox(arguments, FLOAT))
+                .put(
+                    new FunctionSignature(functionName, ImmutableList.of(FLOAT, DOUBLE, DOUBLE)),
+                    (functionProperties, arguments) ->
+                        PercentileApproximateAggregator.percentileApprox(arguments, FLOAT))
+                .put(
+                    new FunctionSignature(functionName, ImmutableList.of(DOUBLE, DOUBLE)),
+                    (functionProperties, arguments) ->
+                        PercentileApproximateAggregator.percentileApprox(arguments, DOUBLE))
+                .put(
+                    new FunctionSignature(functionName, ImmutableList.of(DOUBLE, DOUBLE, DOUBLE)),
+                    (functionProperties, arguments) ->
+                        PercentileApproximateAggregator.percentileApprox(arguments, DOUBLE))
+                .build());
+    return functionResolver;
+  }
 }
diff --git a/.../main/java/org/opensearch/sql/expression/aggregation/PercentileApproximateAggregator.java b/.../main/java/org/opensearch/sql/expression/aggregation/PercentileApproximateAggregator.java
@@ -0,0 +1,98 @@
+/*
+ * Copyright OpenSearch Contributors
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+package org.opensearch.sql.expression.aggregation;
+
+import static org.opensearch.sql.data.model.ExprValueUtils.doubleValue;
+import static org.opensearch.sql.utils.ExpressionUtils.format;
+
+import com.tdunning.math.stats.AVLTreeDigest;
+import java.util.List;
+import org.opensearch.sql.common.utils.StringUtils;
+import org.opensearch.sql.data.model.ExprNullValue;
+import org.opensearch.sql.data.model.ExprValue;
+import org.opensearch.sql.data.type.ExprCoreType;
+import org.opensearch.sql.expression.Expression;
+import org.opensearch.sql.expression.function.BuiltinFunctionName;
+
+/** Aggregator to calculate approximate percentile. */
+public class PercentileApproximateAggregator
+    extends Aggregator<PercentileApproximateAggregator.PercentileApproximateState> {
+
+  public static Aggregator percentileApprox(List<Expression> arguments, ExprCoreType returnType) {
+    return new PercentileApproximateAggregator(arguments, returnType);
+  }
+
+  public PercentileApproximateAggregator(List<Expression> arguments, ExprCoreType returnType) {
+    super(BuiltinFunctionName.PERCENTILE_APPROX.getName(), arguments, returnType);
+    if (!ExprCoreType.numberTypes().contains(returnType)) {
+      throw new IllegalArgumentException(
+          String.format("percentile aggregation over %s type is not supported", returnType));
+    }
+  }
+
+  @Override
+  public PercentileApproximateState create() {
+    if (getArguments().size() == 2) {
+      return new PercentileApproximateState(getArguments().get(1).valueOf().doubleValue());
+    } else {
+      return new PercentileApproximateState(
+          getArguments().get(1).valueOf().doubleValue(),
+          getArguments().get(2).valueOf().doubleValue());
+    }
+  }
+
+  @Override
+  protected PercentileApproximateState iterate(ExprValue value, PercentileApproximateState state) {
+    state.evaluate(value);
+    return state;
+  }
+
+  @Override
+  public String toString() {
+    return StringUtils.format("%s(%s)", "percentile", format(getArguments()));
+  }
+
+  /**
+   * PercentileApproximateState is used to store the AVLTreeDigest state for percentile estimation.
+   */
+  protected static class PercentileApproximateState extends AVLTreeDigest
+      implements AggregationState {
+    // The compression level for the AVLTreeDigest, keep the same default value as OpenSearch core.
+    public static final double DEFAULT_COMPRESSION = 100.0;
+    private final double percent;
+
+    PercentileApproximateState(double percent) {
+      super(DEFAULT_COMPRESSION);
+      if (percent < 0.0 || percent > 100.0) {
+        throw new IllegalArgumentException("out of bounds percent value, must be in [0, 100]");
+      }
+      this.percent = percent / 100.0;
+    }
+
+    /**
+     * Constructor for specifying both percent and compression level.
+     *
+     * @param percent the percent to compute, must be in [0, 100]
+     * @param compression the compression factor of the t-digest sketches used
+     */
+    PercentileApproximateState(double percent, double compression) {
+      super(compression);
+      if (percent < 0.0 || percent > 100.0) {
+        throw new IllegalArgumentException("out of bounds percent value, must be in [0, 100]");
+      }
+      this.percent = percent / 100.0;
+    }
+
+    public void evaluate(ExprValue value) {
+      this.add(value.doubleValue());
+    }
+
+    @Override
+    public ExprValue result() {
+      return this.size() == 0 ? ExprNullValue.of() : doubleValue(this.quantile(percent));
+    }
+  }
+}
diff --git a/core/src/main/java/org/opensearch/sql/expression/function/BuiltinFunctionName.java b/core/src/main/java/org/opensearch/sql/expression/function/BuiltinFunctionName.java
@@ -175,6 +175,8 @@ public enum BuiltinFunctionName {
   STDDEV_POP(FunctionName.of("stddev_pop")),
   // take top documents from aggregation bucket.
   TAKE(FunctionName.of("take")),
+  // t-digest percentile which is used in OpenSearch core by default.
+  PERCENTILE_APPROX(FunctionName.of("percentile_approx")),
   // Not always an aggregation query
   NESTED(FunctionName.of("nested")),
 
@@ -279,6 +281,8 @@ public enum BuiltinFunctionName {
           .put("stddev_pop", BuiltinFunctionName.STDDEV_POP)
           .put("stddev_samp", BuiltinFunctionName.STDDEV_SAMP)
           .put("take", BuiltinFunctionName.TAKE)
+          .put("percentile", BuiltinFunctionName.PERCENTILE_APPROX)
+          .put("percentile_approx", BuiltinFunctionName.PERCENTILE_APPROX)
           .build();
 
   public static Optional<BuiltinFunctionName> of(String str) {