[SPARK-47444][SQL] Validate numeric table stats in ALTER TABLE SET TBLPROPERTIES

common/utils/src/main/resources/error/error-conditions.json

@@ -4657,6 +4657,12 @@
     },
     "sqlState" : "42823"
   },
+  "INVALID_TABLE_STATS_VALUE" : {
+    "message" : [
+      "The value <value> for table statistics property <key> is not a valid numeric value."
+    ],
+    "sqlState" : "22023"
+  },
   "INVALID_TEMP_OBJ_QUALIFIER" : {
     "message" : [
       "Temporary <objectType> <objectName> cannot be qualified with <qualifier>. Temporary objects can only be qualified with SESSION or SYSTEM.SESSION."

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/CheckAnalysis.scala

@@ -1203,6 +1203,16 @@ trait CheckAnalysis extends LookupCatalog with QueryErrorsBase with PlanToString
             }
           case _ =>
         }
+      // SPARK-47444: Validate that stats-related properties have numeric values.
+      case SetTableProperties(_, properties) =>
+        val numericStatsKeys = Set("numRows", "totalSize", "rawDataSize")
+        properties.foreach { case (key, value) =>
+          if (numericStatsKeys.contains(key) && scala.util.Try(BigInt(value)).isFailure) {
+            throw new AnalysisException(
+              errorClass = "INVALID_TABLE_STATS_VALUE",
+              messageParameters = Map("key" -> toSQLId(key), "value" -> s"'$value'"))
+          }
+        }
       case _ =>
     }
   }

sql/core/src/main/scala/org/apache/spark/sql/execution/command/ddl.scala

@@ -30,7 +30,7 @@ import org.apache.hadoop.mapred.{FileInputFormat, JobConf}
 
 import org.apache.spark.internal.{Logging, LogKeys}
 import org.apache.spark.internal.config.RDD_PARALLEL_LISTING_THRESHOLD
-import org.apache.spark.sql.{Row, SaveMode, SparkSession}
+import org.apache.spark.sql.{AnalysisException, Row, SaveMode, SparkSession}
 import org.apache.spark.sql.catalyst.TableIdentifier
 import org.apache.spark.sql.catalyst.analysis.{EliminateSubqueryAliases, Resolver}
 import org.apache.spark.sql.catalyst.catalog._
@@ -321,6 +321,15 @@ case class AlterTableSetPropertiesCommand(
   override def run(sparkSession: SparkSession): Seq[Row] = {
     val catalog = sparkSession.sessionState.catalog
     val table = catalog.getTableRawMetadata(tableName)
+    // SPARK-47444: Validate that stats-related properties have numeric values.
+    val numericStatsKeys = Set("numRows", "totalSize", "rawDataSize")
+    properties.foreach { case (key, value) =>
+      if (numericStatsKeys.contains(key) && scala.util.Try(BigInt(value)).isFailure) {
+        throw new AnalysisException(
+          errorClass = "INVALID_TABLE_STATS_VALUE",
+          messageParameters = Map("key" -> toSQLId(key), "value" -> s"'$value'"))
+      }
+    }
     // This overrides old properties and update the comment parameter of CatalogTable
     // with the newly added/modified comment since CatalogTable also holds comment as its
     // direct property.

sql/core/src/test/scala/org/apache/spark/sql/execution/command/AlterTableSetTblPropertiesSuiteBase.scala

@@ -126,4 +126,29 @@ trait AlterTableSetTblPropertiesSuiteBase extends QueryTest with DDLCommandTestU
       }
     }
   }
+
+  test("SPARK-47444: reject non-numeric values for table stats properties") {
+    withNamespaceAndTable("ns", "tbl") { t =>
+      sql(s"CREATE TABLE $t (col1 int) $defaultUsing")
+      Seq("numRows", "totalSize", "rawDataSize").foreach { key =>
+        checkError(
+          exception = intercept[AnalysisException] {
+            sql(s"ALTER TABLE $t SET TBLPROPERTIES ('$key'='')")
+          },
+          condition = "INVALID_TABLE_STATS_VALUE",
+          parameters = Map("key" -> s"`$key`", "value" -> "''")
+        )
+        checkError(
+          exception = intercept[AnalysisException] {
+            sql(s"ALTER TABLE $t SET TBLPROPERTIES ('$key'='abc')")
+          },
+          condition = "INVALID_TABLE_STATS_VALUE",
+          parameters = Map("key" -> s"`$key`", "value" -> "'abc'")
+        )
+      }
+      // Valid numeric values should succeed
+      sql(s"ALTER TABLE $t SET TBLPROPERTIES ('numRows'='100', 'totalSize'='5000')")
+    }
+  }
+
 }