Mvexpand feature

core/src/main/java/org/opensearch/sql/analysis/Analyzer.java

@@ -78,6 +78,7 @@
 import org.opensearch.sql.ast.tree.Lookup;
 import org.opensearch.sql.ast.tree.ML;
 import org.opensearch.sql.ast.tree.Multisearch;
+import org.opensearch.sql.ast.tree.MvExpand;
 import org.opensearch.sql.ast.tree.Paginate;
 import org.opensearch.sql.ast.tree.Parse;
 import org.opensearch.sql.ast.tree.Patterns;
@@ -701,6 +702,11 @@ public LogicalPlan visitExpand(Expand expand, AnalysisContext context) {
     throw getOnlyForCalciteException("Expand");
   }
 
+  @Override
+  public LogicalPlan visitMvExpand(MvExpand node, AnalysisContext context) {
+    throw getOnlyForCalciteException("MvExpand");
+  }
+
   /** Build {@link LogicalTrendline} for Trendline command. */
   @Override
   public LogicalPlan visitTrendline(Trendline node, AnalysisContext context) {

core/src/main/java/org/opensearch/sql/ast/AbstractNodeVisitor.java

@@ -66,6 +66,7 @@
 import org.opensearch.sql.ast.tree.Lookup;
 import org.opensearch.sql.ast.tree.ML;
 import org.opensearch.sql.ast.tree.Multisearch;
+import org.opensearch.sql.ast.tree.MvExpand;
 import org.opensearch.sql.ast.tree.Paginate;
 import org.opensearch.sql.ast.tree.Parse;
 import org.opensearch.sql.ast.tree.Patterns;
@@ -451,4 +452,8 @@ public T visitAppend(Append node, C context) {
   public T visitMultisearch(Multisearch node, C context) {
     return visitChildren(node, context);
   }
+
+  public T visitMvExpand(MvExpand node, C context) {
+    return visitChildren(node, context);
+  }
 }

core/src/main/java/org/opensearch/sql/ast/dsl/AstDSL.java

@@ -62,6 +62,7 @@
 import org.opensearch.sql.ast.tree.Head;
 import org.opensearch.sql.ast.tree.Limit;
 import org.opensearch.sql.ast.tree.MinSpanBin;
+import org.opensearch.sql.ast.tree.MvExpand;
 import org.opensearch.sql.ast.tree.Parse;
 import org.opensearch.sql.ast.tree.Patterns;
 import org.opensearch.sql.ast.tree.Project;
@@ -136,6 +137,10 @@ public Expand expand(UnresolvedPlan input, Field field, String alias) {
     return new Expand(field, alias).attach(input);
   }
 
+  public static UnresolvedPlan mvexpand(UnresolvedPlan input, Field field, Integer limit) {
+    return new MvExpand(field, limit).attach(input);
+  }
+
   public static UnresolvedPlan projectWithArg(
       UnresolvedPlan input, List<Argument> argList, UnresolvedExpression... projectList) {
     return new Project(Arrays.asList(projectList), argList).attach(input);

core/src/main/java/org/opensearch/sql/ast/tree/MvExpand.java

@@ -0,0 +1,46 @@
+/*
+ * Copyright OpenSearch Contributors
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+package org.opensearch.sql.ast.tree;
+
+import com.google.common.collect.ImmutableList;
+import java.util.List;
+import javax.annotation.Nullable;
+import lombok.EqualsAndHashCode;
+import lombok.Getter;
+import lombok.ToString;
+import org.opensearch.sql.ast.AbstractNodeVisitor;
+import org.opensearch.sql.ast.expression.Field;
+
+/** AST node representing an {@code mvexpand <field> [limit N]} operation. */
+@ToString
+@EqualsAndHashCode(callSuper = false)
+public class MvExpand extends UnresolvedPlan {
+
+  private UnresolvedPlan child;
+  @Getter private final Field field;
+  @Getter @Nullable private final Integer limit;
+
+  public MvExpand(Field field, @Nullable Integer limit) {
+    this.field = field;
+    this.limit = limit;
+  }
+
+  @Override
+  public MvExpand attach(UnresolvedPlan child) {
+    this.child = child;
+    return this;
+  }
+
+  @Override
+  public List<UnresolvedPlan> getChild() {
+    return this.child == null ? ImmutableList.of() : ImmutableList.of(this.child);
+  }
+
+  @Override
+  public <T, C> T accept(AbstractNodeVisitor<T, C> nodeVisitor, C context) {
+    return nodeVisitor.visitMvExpand(this, context);
+  }
+}

core/src/main/java/org/opensearch/sql/calcite/CalciteRelNodeVisitor.java

@@ -91,6 +91,7 @@
 import org.opensearch.sql.ast.expression.ParseMethod;
 import org.opensearch.sql.ast.expression.PatternMethod;
 import org.opensearch.sql.ast.expression.PatternMode;
+import org.opensearch.sql.ast.expression.QualifiedName;
 import org.opensearch.sql.ast.expression.Span;
 import org.opensearch.sql.ast.expression.SpanUnit;
 import org.opensearch.sql.ast.expression.UnresolvedExpression;
@@ -120,6 +121,7 @@
 import org.opensearch.sql.ast.tree.Lookup.OutputStrategy;
 import org.opensearch.sql.ast.tree.ML;
 import org.opensearch.sql.ast.tree.Multisearch;
+import org.opensearch.sql.ast.tree.MvExpand;
 import org.opensearch.sql.ast.tree.Paginate;
 import org.opensearch.sql.ast.tree.Parse;
 import org.opensearch.sql.ast.tree.Patterns;
@@ -843,7 +845,11 @@ public RelNode visitPatterns(Patterns node, CalcitePlanContext context) {
                 .toList();
         context.relBuilder.aggregate(context.relBuilder.groupKey(groupByList), aggCall);
         buildExpandRelNode(
-            context.relBuilder.field(node.getAlias()), node.getAlias(), node.getAlias(), context);
+            context.relBuilder.field(node.getAlias()),
+            node.getAlias(),
+            node.getAlias(),
+            null,
+            context);
         flattenParsedPattern(
             node.getAlias(),
             context.relBuilder.field(node.getAlias()),
@@ -2810,11 +2816,102 @@ public RelNode visitExpand(Expand expand, CalcitePlanContext context) {
     RexInputRef arrayFieldRex = (RexInputRef) rexVisitor.analyze(arrayField, context);
     String alias = expand.getAlias();
 
-    buildExpandRelNode(arrayFieldRex, arrayField.getField().toString(), alias, context);
+    buildExpandRelNode(arrayFieldRex, arrayField.getField().toString(), alias, null, context);
 
     return context.relBuilder.peek();
   }
 
+  /**
+   * MVExpand command visitor.
+   *
+   * <p>For Calcite remote planning, mvexpand shares the same expansion mechanics as {@link Expand}:
+   * it unnests the target multivalue field and joins back to the original relation. The additional
+   * mvexpand semantics (such as an optional per-document limit) are surfaced via the MVExpand AST
+   * node but reuse the same underlying RelBuilder pipeline as expand at this layer.
+   *
+   * @param mvExpand MVExpand command to be visited
+   * @param context CalcitePlanContext containing the RelBuilder and other context
+   * @return RelNode representing records with the expanded multi-value field
+   */
+  @Override
+  public RelNode visitMvExpand(MvExpand mvExpand, CalcitePlanContext context) {
+    // 1. Visit children
+    visitChildren(mvExpand, context);
+
+    RelBuilder relBuilder = context.relBuilder;
+    RelDataType rowType = relBuilder.peek().getRowType();
+
+    Field field = mvExpand.getField();
+
+    String fieldName = extractFieldName(field);
+
+    // 2. Lookup field
+    RelDataTypeField matched = rowType.getField(fieldName, false, false);
+
+    // 2A. Missing field → true EMPTY relation (no schema, no rows)
+    if (matched == null) {
+      // Schema must include the missing field, even if no rows returned.
+      List<RelDataTypeField> fields = rowType.getFieldList();
+      List<RexNode> projects = new ArrayList<>();
+      List<String> names = new ArrayList<>();
+
+      // Keep existing fields
+      for (RelDataTypeField f : fields) {
+        projects.add(relBuilder.field(f.getIndex()));
+        names.add(f.getName());
+      }
+
+      // Add NULL for missing field
+      projects.add(relBuilder.literal(null));
+      names.add(fieldName);
+
+      relBuilder.project(projects, names);
+
+      // Now return 0 rows
+      relBuilder.filter(relBuilder.literal(false));
+
+      return relBuilder.peek();
+    }
+
+    // 2B. Non-array → SemanticCheckException (return immediately)
+    RelDataType type = matched.getType();
+    SqlTypeName sqlType = type.getSqlTypeName();
+
+    if (sqlType != SqlTypeName.ARRAY) {
+      throw new SemanticCheckException(
+          String.format(
+              "Cannot expand field '%s': expected ARRAY type but found %s",
+              fieldName, sqlType.getName()));
+    }
+
+    // 2C. Valid array → expand (with optional per-document limit)
+    int index = matched.getIndex();
+    RexInputRef fieldRef = context.rexBuilder.makeInputRef(type, index);
+
+    Integer limit = mvExpand.getLimit();
+    if (limit != null && limit <= 0) {
+      throw new SemanticCheckException(
+          String.format("mvexpand limit must be positive, but got %d", limit));
+    }
+    buildExpandRelNode(fieldRef, fieldName, fieldName, limit, context);
+
+    return relBuilder.peek();
+  }
+
+  private String extractFieldName(Field f) {
+    UnresolvedExpression inner = f.getField();
+
+    if (inner instanceof QualifiedName) {
+      List<String> parts = ((QualifiedName) inner).getParts();
+      if (!parts.isEmpty()) {
+        return String.join(".", parts);
+      }
+    }
+
+    // Fallback - return clean string
+    return inner.toString().replace("`", "");
+  }
+
   @Override
   public RelNode visitValues(Values values, CalcitePlanContext context) {
     if (values.getValues() == null || values.getValues().isEmpty()) {
@@ -3059,7 +3156,11 @@ private void flattenParsedPattern(
   }
 
   private void buildExpandRelNode(
-      RexInputRef arrayFieldRex, String arrayFieldName, String alias, CalcitePlanContext context) {
+      RexInputRef arrayFieldRex,
+      String arrayFieldName,
+      String alias,
+      @Nullable Integer perDocLimit,
+      CalcitePlanContext context) {
     // 3. Capture the outer row in a CorrelationId
     Holder<RexCorrelVariable> correlVariable = Holder.empty();
     context.relBuilder.variable(correlVariable::set);
@@ -3074,14 +3175,17 @@ private void buildExpandRelNode(
     RelNode leftNode = context.relBuilder.build();
 
     // 5. Build join right node and expand the array field using uncollect
-    RelNode rightNode =
-        context
-            .relBuilder
-            // fake input, see convertUnnest and convertExpression in Calcite SqlToRelConverter
-            .push(LogicalValues.createOneRow(context.relBuilder.getCluster()))
-            .project(List.of(correlArrayFieldAccess), List.of(arrayFieldName))
-            .uncollect(List.of(), false)
-            .build();
+    context
+        .relBuilder
+        // fake input, see convertUnnest and convertExpression in Calcite SqlToRelConverter
+        .push(LogicalValues.createOneRow(context.relBuilder.getCluster()))
+        .project(List.of(correlArrayFieldAccess), List.of(arrayFieldName))
+        .uncollect(List.of(), false);
+
+    if (perDocLimit != null) {
+      context.relBuilder.limit(0, perDocLimit);
+    }
+    RelNode rightNode = context.relBuilder.build();
 
     // 6. Perform a nested-loop join (correlate) between the original table and the expanded
     // array field.

core/src/main/java/org/opensearch/sql/expression/function/PPLFuncImpTable.java

@@ -1098,6 +1098,14 @@ void populate() {
                   OperandTypes.family(SqlTypeFamily.ARRAY, SqlTypeFamily.INTEGER)
                       .or(OperandTypes.family(SqlTypeFamily.MAP, SqlTypeFamily.ANY)),
               false));
+      // Allow using INTERNAL_ITEM when the element type is unknown/undefined at planning time.
+      // Some datasets (or Calcite's type inference) may give the element an UNDEFINED type.
+      // Accept a "ignore" first-argument family so INTERNAL_ITEM(elem, 'key') can still be planned
+      // and resolved at runtime (fallback semantics handled at execution side). - Used in MVEXPAND
+      registerOperator(
+          INTERNAL_ITEM,
+          SqlStdOperatorTable.ITEM,
+          PPLTypeChecker.family(SqlTypeFamily.IGNORE, SqlTypeFamily.CHARACTER));
       registerOperator(
           XOR,
           SqlStdOperatorTable.NOT_EQUALS,

docs/category.json

@@ -74,6 +74,59 @@
     "user/dql/complex.rst",
     "user/dql/metadata.rst"
   ],
+<<<<<<< HEAD
+=======
+  "ppl_cli_calcite": [
+    "user/ppl/cmd/ad.rst",
+    "user/ppl/cmd/append.rst",
+    "user/ppl/cmd/bin.rst",
+    "user/ppl/cmd/dedup.rst",
+    "user/ppl/cmd/describe.rst",
+    "user/ppl/cmd/eventstats.rst",
+    "user/ppl/cmd/eval.rst",
+    "user/ppl/cmd/fields.rst",
+    "user/ppl/cmd/fillnull.rst",
+    "user/ppl/cmd/grok.rst",
+    "user/ppl/cmd/head.rst",
+    "user/ppl/cmd/join.rst",
+    "user/ppl/cmd/lookup.rst",
+    "user/ppl/cmd/parse.rst",
+    "user/ppl/cmd/patterns.rst",
+    "user/ppl/cmd/rare.rst",
+    "user/ppl/cmd/regex.rst",
+    "user/ppl/cmd/rename.rst",
+    "user/ppl/cmd/multisearch.rst",
+    "user/ppl/cmd/replace.rst",
+    "user/ppl/cmd/rex.rst",
+    "user/ppl/cmd/search.rst",
+    "user/ppl/cmd/showdatasources.rst",
+    "user/ppl/cmd/sort.rst",
+    "user/ppl/cmd/spath.rst",
+    "user/ppl/cmd/stats.rst",
+    "user/ppl/cmd/streamstats.rst",
+    "user/ppl/cmd/subquery.rst",
+    "user/ppl/cmd/syntax.rst",
+    "user/ppl/cmd/chart.rst",
+    "user/ppl/cmd/timechart.rst",
+    "user/ppl/functions/statistical.rst",
+    "user/ppl/cmd/top.rst",
+    "user/ppl/cmd/trendline.rst",
+    "user/ppl/cmd/where.rst",
+    "user/ppl/functions/collection.rst",
+    "user/ppl/functions/condition.rst",
+    "user/ppl/functions/datetime.rst",
+    "user/ppl/functions/expressions.rst",
+    "user/ppl/functions/ip.rst",
+    "user/ppl/functions/json.rst",
+    "user/ppl/functions/math.rst",
+    "user/ppl/functions/relevance.rst",
+    "user/ppl/functions/string.rst",
+    "user/ppl/functions/conversion.rst",
+    "user/ppl/general/datatypes.rst",
+    "user/ppl/general/identifiers.rst",
+    "user/ppl/cmd/mvexpand.rst"
+  ],
+>>>>>>> mywork-backup
   "bash_settings": [
     "user/ppl/admin/settings.md"
   ]

docs/user/dql/metadata.rst

@@ -35,7 +35,7 @@ Example 1: Show All Indices Information
 SQL query::
 
     os> SHOW TABLES LIKE '%'
-    fetched rows / total rows = 23/23
+    fetched rows / total rows = 24/24
     +----------------+-------------+-------------------+------------+---------+----------+------------+-----------+---------------------------+----------------+
     | TABLE_CAT      | TABLE_SCHEM | TABLE_NAME        | TABLE_TYPE | REMARKS | TYPE_CAT | TYPE_SCHEM | TYPE_NAME | SELF_REFERENCING_COL_NAME | REF_GENERATION |
     |----------------+-------------+-------------------+------------+---------+----------+------------+-----------+---------------------------+----------------|
@@ -48,6 +48,7 @@ SQL query::
     | docTestCluster | null        | events_many_hosts | BASE TABLE | null    | null     | null       | null      | null                      | null           |
     | docTestCluster | null        | events_null       | BASE TABLE | null    | null     | null       | null      | null                      | null           |
     | docTestCluster | null        | json_test         | BASE TABLE | null    | null     | null       | null      | null                      | null           |
+    | docTestCluster | null        | mvexpand_logs     | BASE TABLE | null    | null     | null       | null      | null                      | null           |
     | docTestCluster | null        | nested            | BASE TABLE | null    | null     | null       | null      | null                      | null           |
     | docTestCluster | null        | nyc_taxi          | BASE TABLE | null    | null     | null       | null      | null                      | null           |
     | docTestCluster | null        | occupation        | BASE TABLE | null    | null     | null       | null      | null                      | null           |
@@ -120,3 +121,4 @@ SQL query::
     | docTestCluster | null        | accounts   | firstname   | null      | text      | null        | null          | null           | 10             | 2        | null    | null       | null          | null             | null              | 1                |             | null          | null         | null        | null             | NO               |                    |
     | docTestCluster | null        | accounts   | lastname    | null      | text      | null        | null          | null           | 10             | 2        | null    | null       | null          | null             | null              | 10               |             | null          | null         | null        | null             | NO               |                    |
     +----------------+-------------+------------+-------------+-----------+-----------+-------------+---------------+----------------+----------------+----------+---------+------------+---------------+------------------+-------------------+------------------+-------------+---------------+--------------+-------------+------------------+------------------+--------------------+
+