Add position() to V2 engine (#177)

Signed-off-by: Margarit Hakobyan <[email protected]>
opensearch-project · Nov 30, 2022 · 2964853 · 2964853
1 parent cfb2650
commit 2964853
Show file tree

Hide file tree

Showing 15 changed files with 319 additions and 23 deletions.
diff --git a/core/src/main/java/org/opensearch/sql/analysis/ExpressionAnalyzer.java b/core/src/main/java/org/opensearch/sql/analysis/ExpressionAnalyzer.java
@@ -35,6 +35,7 @@
 import org.opensearch.sql.ast.expression.Literal;
 import org.opensearch.sql.ast.expression.Not;
 import org.opensearch.sql.ast.expression.Or;
+import org.opensearch.sql.ast.expression.PositionFunction;
 import org.opensearch.sql.ast.expression.QualifiedName;
 import org.opensearch.sql.ast.expression.RelevanceFieldList;
 import org.opensearch.sql.ast.expression.Span;
@@ -213,6 +214,13 @@ public Expression visitHighlightFunction(HighlightFunction node, AnalysisContext
     return new HighlightExpression(expr);
   }
 
+  @Override
+  public Expression visitPositionFunction(PositionFunction node, AnalysisContext context) {
+    Expression stringPatternExpr = node.getStringPatternExpr().accept(this, context);
+    Expression searchStringExpr = node.getSearchStringExpr().accept(this, context);
+    return DSL.position(stringPatternExpr, searchStringExpr);
+  }
+
   @Override
   public Expression visitIn(In node, AnalysisContext context) {
     return visitIn(node.getField(), node.getValueList(), context);

diff --git a/core/src/main/java/org/opensearch/sql/ast/AbstractNodeVisitor.java b/core/src/main/java/org/opensearch/sql/ast/AbstractNodeVisitor.java
@@ -27,6 +27,7 @@
 import org.opensearch.sql.ast.expression.Map;
 import org.opensearch.sql.ast.expression.Not;
 import org.opensearch.sql.ast.expression.Or;
+import org.opensearch.sql.ast.expression.PositionFunction;
 import org.opensearch.sql.ast.expression.QualifiedName;
 import org.opensearch.sql.ast.expression.RelevanceFieldList;
 import org.opensearch.sql.ast.expression.Span;
@@ -278,6 +279,10 @@ public T visitHighlightFunction(HighlightFunction node, C context) {
     return visitChildren(node, context);
   }
 
+  public T visitPositionFunction(PositionFunction node, C context) {
+    return visitChildren(node, context);
+  }
+
   public T visitStatement(Statement node, C context) {
     return visit(node, context);
   }

diff --git a/core/src/main/java/org/opensearch/sql/ast/dsl/AstDSL.java b/core/src/main/java/org/opensearch/sql/ast/dsl/AstDSL.java
@@ -33,6 +33,7 @@
 import org.opensearch.sql.ast.expression.Not;
 import org.opensearch.sql.ast.expression.Or;
 import org.opensearch.sql.ast.expression.ParseMethod;
+import org.opensearch.sql.ast.expression.PositionFunction;
 import org.opensearch.sql.ast.expression.QualifiedName;
 import org.opensearch.sql.ast.expression.Span;
 import org.opensearch.sql.ast.expression.SpanUnit;
@@ -288,6 +289,11 @@ public UnresolvedExpression highlight(UnresolvedExpression fieldName,
     return new HighlightFunction(fieldName, arguments);
   }
 
+  public UnresolvedExpression position(UnresolvedExpression stringPatternExpr,
+                                       UnresolvedExpression searchStringExpr) {
+    return new PositionFunction(stringPatternExpr, searchStringExpr);
+  }
+
   public UnresolvedExpression window(UnresolvedExpression function,
                                      List<UnresolvedExpression> partitionByList,
                                      List<Pair<SortOption, UnresolvedExpression>> sortList) {

diff --git a/core/src/main/java/org/opensearch/sql/ast/expression/PositionFunction.java b/core/src/main/java/org/opensearch/sql/ast/expression/PositionFunction.java
@@ -0,0 +1,39 @@
+/*
+ * Copyright OpenSearch Contributors
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+package org.opensearch.sql.ast.expression;
+
+import java.util.Arrays;
+import java.util.List;
+import lombok.AllArgsConstructor;
+import lombok.EqualsAndHashCode;
+import lombok.Getter;
+import lombok.ToString;
+import org.opensearch.sql.ast.AbstractNodeVisitor;
+
+
+/**
+ * Expression node of Position function.
+ */
+@AllArgsConstructor
+@EqualsAndHashCode(callSuper = false)
+@Getter
+@ToString
+public class PositionFunction extends UnresolvedExpression {
+  @Getter
+  private UnresolvedExpression stringPatternExpr;
+  @Getter
+  private UnresolvedExpression searchStringExpr;
+
+  @Override
+  public List<UnresolvedExpression> getChild() {
+    return Arrays.asList(stringPatternExpr, searchStringExpr);
+  }
+
+  @Override
+  public <R, C> R accept(AbstractNodeVisitor<R, C> nodeVisitor, C context) {
+    return nodeVisitor.visitPositionFunction(this, context);
+  }
+}
diff --git a/core/src/main/java/org/opensearch/sql/expression/DSL.java b/core/src/main/java/org/opensearch/sql/expression/DSL.java
@@ -229,6 +229,10 @@ public static FunctionExpression cbrt(Expression... expressions) {
     return compile(BuiltinFunctionName.CBRT, expressions);
   }
 
+  public static FunctionExpression position(Expression... expressions) {
+    return compile(BuiltinFunctionName.POSITION, expressions);
+  }
+
   public static FunctionExpression truncate(Expression... expressions) {
     return compile(BuiltinFunctionName.TRUNCATE, expressions);
   }

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
@@ -153,23 +153,24 @@ public enum BuiltinFunctionName {
   /**
    * Text Functions.
    */
-  SUBSTR(FunctionName.of("substr")),
-  SUBSTRING(FunctionName.of("substring")),
-  RTRIM(FunctionName.of("rtrim")),
-  LTRIM(FunctionName.of("ltrim")),
-  TRIM(FunctionName.of("trim")),
-  UPPER(FunctionName.of("upper")),
-  LOWER(FunctionName.of("lower")),
-  REGEXP(FunctionName.of("regexp")),
+  ASCII(FunctionName.of("ascii")),
   CONCAT(FunctionName.of("concat")),
   CONCAT_WS(FunctionName.of("concat_ws")),
-  LENGTH(FunctionName.of("length")),
-  STRCMP(FunctionName.of("strcmp")),
-  RIGHT(FunctionName.of("right")),
   LEFT(FunctionName.of("left")),
-  ASCII(FunctionName.of("ascii")),
+  LENGTH(FunctionName.of("length")),
   LOCATE(FunctionName.of("locate")),
+  LOWER(FunctionName.of("lower")),
+  LTRIM(FunctionName.of("ltrim")),
+  POSITION(FunctionName.of("position")),
+  REGEXP(FunctionName.of("regexp")),
   REPLACE(FunctionName.of("replace")),
+  RIGHT(FunctionName.of("right")),
+  RTRIM(FunctionName.of("rtrim")),
+  STRCMP(FunctionName.of("strcmp")),
+  SUBSTR(FunctionName.of("substr")),
+  SUBSTRING(FunctionName.of("substring")),
+  TRIM(FunctionName.of("trim")),
+  UPPER(FunctionName.of("upper")),
 
   /**
    * NULL Test.

diff --git a/core/src/main/java/org/opensearch/sql/expression/text/TextFunction.java b/core/src/main/java/org/opensearch/sql/expression/text/TextFunction.java
@@ -39,22 +39,23 @@ public class TextFunction {
    * @param repository {@link BuiltinFunctionRepository}.
    */
   public void register(BuiltinFunctionRepository repository) {
-    repository.register(substr());
-    repository.register(substring());
-    repository.register(ltrim());
-    repository.register(rtrim());
-    repository.register(trim());
-    repository.register(lower());
-    repository.register(upper());
+    repository.register(ascii());
     repository.register(concat());
     repository.register(concat_ws());
-    repository.register(length());
-    repository.register(strcmp());
-    repository.register(right());
     repository.register(left());
-    repository.register(ascii());
+    repository.register(length());
     repository.register(locate());
+    repository.register(lower());
+    repository.register(ltrim());
+    repository.register(position());
     repository.register(replace());
+    repository.register(right());
+    repository.register(rtrim());
+    repository.register(strcmp());
+    repository.register(substr());
+    repository.register(substring());
+    repository.register(trim());
+    repository.register(upper());
   }
 
   /**
@@ -241,6 +242,18 @@ private DefaultFunctionResolver locate() {
                 TextFunction::exprLocate), INTEGER, STRING, STRING, INTEGER));
   }
 
+  /**
+   * Returns the position of the first occurrence of a substring in a string starting from 1.
+   * Returns 0 if substring is not in string.
+   * Returns NULL if any argument is NULL.
+   * Supports following signature:
+   * (STRING IN STRING) -> INTEGER
+   */
+  private DefaultFunctionResolver position() {
+    return define(BuiltinFunctionName.POSITION.getName(),
+            impl(nullMissingHandling(TextFunction::exprPosition), INTEGER, STRING, STRING));
+  }
+
   /**
    * REPLACE(str, from_str, to_str) returns the string str with all occurrences of
    * the string from_str replaced by the string to_str.
@@ -313,6 +326,10 @@ private static ExprValue exprLocate(ExprValue subStr, ExprValue str, ExprValue p
         str.stringValue().indexOf(subStr.stringValue(), pos.integerValue() - 1) + 1);
   }
 
+  private static ExprValue exprPosition(ExprValue subStr, ExprValue str) {
+    return exprLocate(subStr, str);
+  }
+
   private static ExprValue exprReplace(ExprValue str, ExprValue from, ExprValue to) {
     return new ExprStringValue(str.stringValue().replaceAll(from.stringValue(), to.stringValue()));
   }

diff --git a/core/src/test/java/org/opensearch/sql/analysis/NamedExpressionAnalyzerTest.java b/core/src/test/java/org/opensearch/sql/analysis/NamedExpressionAnalyzerTest.java
@@ -16,6 +16,7 @@
 import org.opensearch.sql.ast.expression.Alias;
 import org.opensearch.sql.ast.expression.HighlightFunction;
 import org.opensearch.sql.ast.expression.Literal;
+import org.opensearch.sql.ast.expression.PositionFunction;
 import org.opensearch.sql.expression.NamedExpression;
 import org.springframework.context.annotation.Configuration;
 import org.springframework.test.context.ContextConfiguration;
@@ -48,4 +49,14 @@ void visit_highlight() {
     NamedExpression analyze = analyzer.analyze(alias, analysisContext);
     assertEquals("highlight(fieldA)", analyze.getNameOrAlias());
   }
+
+  @Test
+  void visit_position() {
+    Alias alias = AstDSL.alias("position(fieldA IN fieldB)",
+        new PositionFunction(AstDSL.stringLiteral("fieldA"), AstDSL.stringLiteral("fieldB")));
+    NamedExpressionAnalyzer analyzer = new NamedExpressionAnalyzer(expressionAnalyzer);
+
+    NamedExpression analyze = analyzer.analyze(alias, analysisContext);
+    assertEquals("position(fieldA IN fieldB)", analyze.getNameOrAlias());
+  }
 }
diff --git a/core/src/test/java/org/opensearch/sql/expression/text/TextFunctionTest.java b/core/src/test/java/org/opensearch/sql/expression/text/TextFunctionTest.java
@@ -377,6 +377,26 @@ void locate() {
         DSL.locate(missingRef, DSL.literal("hello"), DSL.literal(1))));
   }
 
+  @Test
+  void position() {
+    FunctionExpression expression = DSL.position(
+        DSL.literal("world"),
+        DSL.literal("helloworldworld"));
+    assertEquals(INTEGER, expression.type());
+    assertEquals(6, eval(expression).integerValue());
+
+    expression = DSL.position(
+            DSL.literal("abc"),
+            DSL.literal("hello world"));
+    assertEquals(INTEGER, expression.type());
+    assertEquals(0, eval(expression).integerValue());
+
+    when(nullRef.type()).thenReturn(STRING);
+    assertEquals(nullValue(), eval(DSL.position(nullRef, DSL.literal("hello"))));
+    when(missingRef.type()).thenReturn(STRING);
+    assertEquals(missingValue(), eval(DSL.position(missingRef, DSL.literal("hello"))));
+  }
+
   @Test
   void replace() {
     FunctionExpression expression = DSL.replace(

diff --git a/docs/user/dql/functions.rst b/docs/user/dql/functions.rst
@@ -2331,6 +2331,31 @@ Example::
     +---------------------+---------------------+
 
 
+POSITION
+------
+
+Description
+>>>>>>>>>>>
+
+Usage: The syntax POSITION(substr IN str) returns the position of the first occurrence of substring substr in string str. Returns 0 if substr is not in str. Returns NULL if any argument is NULL.
+
+Argument type: STRING, STRING, INTEGER
+
+Return type integer:
+
+(STRING IN STRING) -> INTEGER
+
+Example::
+
+    os> SELECT POSITION('world' IN 'helloworld')
+    fetched rows / total rows = 1/1
+    +-------------------------------------+---------------------------------------+
+    | POSITION('world' IN 'helloworld')   | POSITION('invalid' IN 'helloworld')   |
+    |-------------------------------------+---------------------------------------|
+    | 6                                   | 0                                     |
+    +-------------------------------------+---------------------------------------+
+
+
 REPLACE
 -------