Add support for custom date format and openSearch date format for dat…

…e fields as part of Lucene query (#2762) (#2849)

Github Issue - #2700

(cherry picked from commit 0fad56d)

Signed-off-by: Manasvini B S <[email protected]>
(cherry picked from commit 02d57e0)
Signed-off-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

docs/user/general/datatypes.rst

@@ -429,6 +429,48 @@ Querying such index will provide a response with ``schema`` block as shown below
         "status": 200
     }
 
+If the sql query contains an `IndexDateField` and a literal value with an operator (such as a term query or a range query), then the literal value can be in the `IndexDateField` format.
+
+.. code-block:: json
+
+    {
+        "mappings" : {
+            "properties" : {
+                "release_date" : {
+                    "type" : "date",
+                    "format": "dd-MMM-yy"
+                }
+            }
+        }
+    }
+
+Querying such an `IndexDateField` (``release_date``) will provide a response with ``schema`` and ``datarows`` blocks as shown below.
+
+.. code-block:: json
+
+    {
+        "query" : "SELECT release_date FROM test_index WHERE release_date = \"03-Jan-21\""
+    }
+
+.. code-block:: json
+
+    {
+      "schema": [
+        {
+          "name": "release_date",
+          "type": "date"
+        }
+      ],
+      "datarows": [
+        [
+          "2021-01-03"
+        ]
+      ],
+      "total": 1,
+      "size": 1,
+      "status": 200
+    }
+
 String Data Types
 =================
 

opensearch/src/main/java/org/opensearch/sql/opensearch/data/type/OpenSearchDataType.java

@@ -62,19 +62,23 @@ public String toString() {
   @EqualsAndHashCode.Exclude @Getter protected MappingType mappingType;
 
   // resolved ExprCoreType
-  protected ExprCoreType exprCoreType;
+  @Getter protected ExprCoreType exprCoreType;
 
   /**
    * Get a simplified type {@link ExprCoreType} if possible. To avoid returning `UNKNOWN` for
-   * `OpenSearch*Type`s, e.g. for IP, returns itself.
+   * `OpenSearch*Type`s, e.g. for IP, returns itself. If the `exprCoreType` is {@link
+   * ExprCoreType#DATE}, {@link ExprCoreType#TIMESTAMP}, {@link ExprCoreType#TIME}, or {@link
+   * ExprCoreType#UNKNOWN}, it returns the current instance; otherwise, it returns `exprCoreType`.
    *
    * @return An {@link ExprType}.
    */
   public ExprType getExprType() {
-    if (exprCoreType != ExprCoreType.UNKNOWN) {
-      return exprCoreType;
-    }
-    return this;
+    return (exprCoreType == ExprCoreType.DATE
+            || exprCoreType == ExprCoreType.TIMESTAMP
+            || exprCoreType == ExprCoreType.TIME
+            || exprCoreType == ExprCoreType.UNKNOWN)
+        ? this
+        : exprCoreType;
   }
 
   /**
@@ -230,6 +234,9 @@ public String legacyTypeName() {
     if (mappingType == null) {
       return exprCoreType.typeName();
     }
+    if (mappingType.toString().equalsIgnoreCase("DATE")) {
+      return exprCoreType.typeName();
+    }
     return mappingType.toString().toUpperCase();
   }
 

opensearch/src/main/java/org/opensearch/sql/opensearch/data/type/OpenSearchDateType.java

@@ -11,11 +11,16 @@
 import static org.opensearch.sql.data.type.ExprCoreType.TIME;
 import static org.opensearch.sql.data.type.ExprCoreType.TIMESTAMP;
 
+import java.time.ZoneOffset;
+import java.time.ZonedDateTime;
+import java.time.temporal.TemporalAccessor;
 import java.util.List;
 import java.util.Objects;
 import java.util.stream.Collectors;
+import java.util.stream.Stream;
 import lombok.EqualsAndHashCode;
 import org.opensearch.common.time.DateFormatter;
+import org.opensearch.common.time.DateFormatters;
 import org.opensearch.common.time.FormatNames;
 import org.opensearch.sql.data.type.ExprCoreType;
 import org.opensearch.sql.data.type.ExprType;
@@ -137,6 +142,11 @@ public class OpenSearchDateType extends OpenSearchDataType {
 
   private static final String CUSTOM_FORMAT_DATE_SYMBOLS = "FecEWwYqQgdMLDyuG";
 
+  private static final List<DateFormatter> OPENSEARCH_DEFAULT_FORMATTERS =
+      Stream.of("strict_date_time_no_millis", "strict_date_optional_time", "epoch_millis")
+          .map(DateFormatter::forPattern)
+          .collect(Collectors.toList());
+
   @EqualsAndHashCode.Exclude private final List<String> formats;
 
   private OpenSearchDateType() {
@@ -235,6 +245,59 @@ public List<DateFormatter> getAllCustomFormatters() {
         .collect(Collectors.toList());
   }
 
+  /**
+   * Retrieves a list of custom formatters and OpenSearch named formatters defined by the user, and
+   * attempts to parse the given date/time string using these formatters.
+   *
+   * @param dateTime The date/time string to parse.
+   * @return A ZonedDateTime representing the parsed date/time in UTC, or null if parsing fails.
+   */
+  public ZonedDateTime getParsedDateTime(String dateTime) {
+    List<DateFormatter> dateFormatters =
+        Stream.concat(this.getAllNamedFormatters().stream(), this.getAllCustomFormatters().stream())
+            .collect(Collectors.toList());
+    ZonedDateTime zonedDateTime = null;
+
+    // check if dateFormatters are empty, then set default ones
+    if (dateFormatters.isEmpty()) {
+      dateFormatters = OPENSEARCH_DEFAULT_FORMATTERS;
+    }
+    // parse using OpenSearch DateFormatters
+    for (DateFormatter formatter : dateFormatters) {
+      try {
+        TemporalAccessor accessor = formatter.parse(dateTime);
+        zonedDateTime = DateFormatters.from(accessor).withZoneSameLocal(ZoneOffset.UTC);
+        break;
+      } catch (IllegalArgumentException ignored) {
+        // nothing to do, try another format
+      }
+    }
+    return zonedDateTime;
+  }
+
+  /**
+   * Returns a formatted date string using the internal formatter, if available.
+   *
+   * @param accessor The TemporalAccessor object containing the date/time information.
+   * @return A formatted date string if a formatter is available, otherwise null.
+   */
+  public String getFormattedDate(TemporalAccessor accessor) {
+    if (hasNoFormatter()) {
+      return OPENSEARCH_DEFAULT_FORMATTERS.get(0).format(accessor);
+    }
+    // Use the first available format string to create the formatter
+    return DateFormatter.forPattern(this.formats.get(0)).format(accessor);
+  }
+
+  /**
+   * Checks if the formatter is not initialized.
+   *
+   * @return True if the formatter is not set, otherwise false.
+   */
+  public boolean hasNoFormatter() {
+    return this.formats.isEmpty();
+  }
+
   /**
    * Retrieves a list of named formatters that format for dates.
    *

opensearch/src/main/java/org/opensearch/sql/opensearch/data/value/OpenSearchExprValueFactory.java

@@ -226,7 +226,7 @@ private Optional<ExprType> type(String field) {
   private static ExprValue parseDateTimeString(String value, OpenSearchDateType dataType) {
     List<DateFormatter> formatters = dataType.getAllNamedFormatters();
     formatters.addAll(dataType.getAllCustomFormatters());
-    ExprCoreType returnFormat = (ExprCoreType) dataType.getExprType();
+    ExprCoreType returnFormat = dataType.getExprCoreType();
 
     for (DateFormatter formatter : formatters) {
       try {
@@ -268,8 +268,7 @@ private static ExprValue parseDateTimeString(String value, OpenSearchDateType da
 
   private static ExprValue createOpenSearchDateType(Content value, ExprType type) {
     OpenSearchDateType dt = (OpenSearchDateType) type;
-    ExprType returnFormat = dt.getExprType();
-
+    ExprCoreType returnFormat = dt.getExprCoreType();
     if (value.isNumber()) { // isNumber
       var numFormatters = dt.getNumericNamedFormatters();
       if (numFormatters.size() > 0 || !dt.hasFormats()) {
@@ -282,7 +281,7 @@ private static ExprValue createOpenSearchDateType(Content value, ExprType type)
           epochMillis = value.longValue();
         }
         Instant instant = Instant.ofEpochMilli(epochMillis);
-        switch ((ExprCoreType) returnFormat) {
+        switch (returnFormat) {
           case TIME:
             return new ExprTimeValue(LocalTime.from(instant.atZone(UTC_ZONE_ID)));
           case DATE:

opensearch/src/main/java/org/opensearch/sql/opensearch/storage/script/aggregation/dsl/BucketAggregationBuilder.java

@@ -24,6 +24,7 @@
 import org.opensearch.sql.ast.expression.SpanUnit;
 import org.opensearch.sql.expression.NamedExpression;
 import org.opensearch.sql.expression.span.SpanExpression;
+import org.opensearch.sql.opensearch.data.type.OpenSearchDateType;
 import org.opensearch.sql.opensearch.storage.serialization.ExpressionSerializer;
 
 /** Bucket Aggregation Builder. */
@@ -66,7 +67,10 @@ private CompositeValuesSourceBuilder<?> buildCompositeValuesSourceBuilder(
               .missingOrder(missingOrder)
               .order(sortOrder);
       // Time types values are converted to LONG in ExpressionAggregationScript::execute
-      if (List.of(TIMESTAMP, TIME, DATE, DATETIME).contains(expr.getDelegated().type())) {
+      if ((expr.getDelegated().type() instanceof OpenSearchDateType
+              && List.of(TIMESTAMP, TIME, DATE, DATETIME)
+                  .contains(((OpenSearchDateType) expr.getDelegated().type()).getExprCoreType()))
+          || List.of(TIMESTAMP, TIME, DATE, DATETIME).contains(expr.getDelegated().type())) {
         sourceBuilder.userValuetypeHint(ValueType.LONG);
       }
       return helper.build(expr.getDelegated(), sourceBuilder::field, sourceBuilder::script);