[SPARK-45060][SQL] Fix an internal error from to_char()on NULL fo…

…rmat ### What changes were proposed in this pull request? In the PR, I propose to check the `format` argument is not a NULL in `ToCharacterBuilder`. And if it is, throw an `AnalysisException` with new error class `INVALID_PARAMETER_VALUE.NULL`. ### Why are the changes needed? To fix the issue demonstrated by the example: ```sql $ spark-sql (default)> SELECT to_char(x'537061726b2053514c', CAST(NULL AS STRING)); [INTERNAL_ERROR] The Spark SQL phase analysis failed with an internal error. You hit a bug in Spark or the Spark plugins you use. Please, report this bug to the corresponding communities or vendors, and provide the full stack trace. ``` ### Does this PR introduce _any_ user-facing change? No. ### How was this patch tested? By running new test: ``` $ build/sbt "test:testOnly *.StringFunctionsSuite" ``` ### Was this patch authored or co-authored using generative AI tooling? No. Closes apache#42781 from MaxGekk/fix-null-fmt-to_char. Authored-by: Max Gekk <max.gekk@gmail.com> Signed-off-by: Max Gekk <max.gekk@gmail.com>
iemejia · Sep 4, 2023 · d03ebce · d03ebce
1 parent f2a6c97
commit d03ebce
Show file tree

Hide file tree

Showing 5 changed files with 37 additions and 2 deletions.
diff --git a/common/utils/src/main/resources/error/error-classes.json b/common/utils/src/main/resources/error/error-classes.json
@@ -1808,6 +1808,11 @@
           "expects one of the units without quotes YEAR, QUARTER, MONTH, WEEK, DAY, DAYOFYEAR, HOUR, MINUTE, SECOND, MILLISECOND, MICROSECOND, but got the string literal <invalidValue>."
         ]
       },
+      "NULL" : {
+        "message" : [
+          "expects a non-NULL value."
+        ]
+      },
       "PATTERN" : {
         "message" : [
           "<value>."

diff --git a/docs/sql-error-conditions-invalid-parameter-value-error-class.md b/docs/sql-error-conditions-invalid-parameter-value-error-class.md
@@ -45,6 +45,10 @@ expects one of binary formats 'base64', 'hex', 'utf-8', but got `<invalidFormat>
 
 expects one of the units without quotes YEAR, QUARTER, MONTH, WEEK, DAY, DAYOFYEAR, HOUR, MINUTE, SECOND, MILLISECOND, MICROSECOND, but got the string literal `<invalidValue>`.
 
+## NULL
+
+expects a non-NULL value.
+
 ## PATTERN
 
 `<value>`.

diff --git a/...st/src/main/scala/org/apache/spark/sql/catalyst/expressions/numberFormatExpressions.scala b/...st/src/main/scala/org/apache/spark/sql/catalyst/expressions/numberFormatExpressions.scala
@@ -241,15 +241,19 @@ case class TryToNumber(left: Expression, right: Expression)
 object ToCharacterBuilder extends ExpressionBuilder {
   override def build(funcName: String, expressions: Seq[Expression]): Expression = {
     val numArgs = expressions.length
-    if (expressions.length == 2) {
+    if (numArgs == 2) {
       val (inputExpr, format) = (expressions(0), expressions(1))
       inputExpr.dataType match {
         case _: DatetimeType => DateFormatClass(inputExpr, format)
         case _: BinaryType =>
           if (!(format.dataType == StringType && format.foldable)) {
             throw QueryCompilationErrors.nonFoldableArgumentError(funcName, "format", StringType)
           }
-          format.eval().asInstanceOf[UTF8String].toString.toLowerCase(Locale.ROOT).trim match {
+          val fmt = format.eval()
+          if (fmt == null) {
+            throw QueryCompilationErrors.nullArgumentError(funcName, "format")
+          }
+          fmt.asInstanceOf[UTF8String].toString.toLowerCase(Locale.ROOT).trim match {
             case "base64" => Base64(inputExpr)
             case "hex" => Hex(inputExpr)
             case "utf-8" => new Decode(Seq(inputExpr, format))

diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/errors/QueryCompilationErrors.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/errors/QueryCompilationErrors.scala
@@ -166,6 +166,14 @@ private[sql] object QueryCompilationErrors extends QueryErrorsBase with Compilat
         "invalidFormat" -> toSQLValue(invalidFormat, StringType)))
   }
 
+  def nullArgumentError(funcName: String, parameter: String): Throwable = {
+    new AnalysisException(
+      errorClass = "INVALID_PARAMETER_VALUE.NULL",
+      messageParameters = Map(
+        "parameter" -> toSQLId(parameter),
+        "functionName" -> toSQLId(funcName)))
+  }
+
   def unorderablePivotColError(pivotCol: Expression): Throwable = {
     new AnalysisException(
       errorClass = "INCOMPARABLE_PIVOT_COLUMN",

diff --git a/sql/core/src/test/scala/org/apache/spark/sql/StringFunctionsSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/StringFunctionsSuite.scala
@@ -900,6 +900,20 @@ class StringFunctionsSuite extends QueryTest with SharedSparkSession {
           "actualNum" -> "3",
           "docroot" -> SPARK_DOC_ROOT),
         context = ExpectedContext("", "", 7, 21 + funcName.length, s"$funcName('a', 'b', 'c')"))
+      checkError(
+        exception = intercept[AnalysisException] {
+          sql(s"select $funcName(x'537061726b2053514c', CAST(NULL AS STRING))")
+        },
+        errorClass = "INVALID_PARAMETER_VALUE.NULL",
+        parameters = Map(
+          "functionName" -> s"`$funcName`",
+          "parameter" -> "`format`"),
+        context = ExpectedContext(
+          "",
+          "",
+          7,
+          51 + funcName.length,
+          s"$funcName(x'537061726b2053514c', CAST(NULL AS STRING))"))
     }
   }