[SPARK-60118][SQL] Fix Int overflow in the fast-fail check of casting string to decimal

Compute the number of integral digits in `Decimal.fromString`/`fromStringANSI` as a Long,
so that strings with an exponent near Int.MaxValue (e.g. '1e2147483647') are rejected by
the fast-fail check instead of failing with a raw java.lang.ArithmeticException.

Co-authored-by: Isaac <no-reply@databricks.com>

Author: viirya

sql/api/src/main/scala/org/apache/spark/sql/types/Decimal.scala

@@ -635,8 +635,9 @@ object Decimal {
     }
   }
 
-  private def numDigitsInIntegralPart(bigDecimal: JavaBigDecimal): Int =
-    bigDecimal.precision - bigDecimal.scale
+  // Computed in Long since `precision - scale` can overflow Int, e.g. for "1e2147483647".
+  private def numDigitsInIntegralPart(bigDecimal: JavaBigDecimal): Long =
+    bigDecimal.precision.toLong - bigDecimal.scale
 
   private def stringToJavaBigDecimal(str: UTF8String): JavaBigDecimal = {
     // According the benchmark test,  `s.toString.trim` is much faster than `s.trim.toString`.

sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/CastWithAnsiOffSuite.scala

@@ -559,6 +559,9 @@ class CastWithAnsiOffSuite extends CastSuiteBase {
       Decimal("60000000000000000000000000000000000000"))
     checkEvaluation(cast("6E+38", DecimalType(38, 0)), null)
     checkEvaluation(cast("6E+37", DecimalType(38, 1)), null)
+    Seq("1e2147483647", "0e2147483647", "1.5e2147483647", "1e2147483648").foreach { s =>
+      checkEvaluation(cast(s, DecimalType(10, 2)), null)
+    }
 
     checkEvaluation(cast("abcd", DecimalType(38, 1)), null)
   }

sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/CastWithAnsiOnSuite.scala

@@ -475,6 +475,11 @@ class CastWithAnsiOnSuite extends CastSuiteBase with QueryErrorsBase {
     checkExceptionInExpression[ArithmeticException](
       cast("6E+37", DecimalType(38, 1)),
       "cannot be represented as Decimal(38, 1)")
+    Seq("1e2147483647", "0e2147483647", "1.5e2147483647", "1e2147483648").foreach { s =>
+      checkExceptionInExpression[ArithmeticException](
+        cast(s, DecimalType(10, 2)),
+        "NUMERIC_OUT_OF_SUPPORTED_RANGE")
+    }
 
     checkExceptionInExpression[NumberFormatException](
       cast("abcd", DecimalType(38, 1)),

sql/catalyst/src/test/scala/org/apache/spark/sql/types/DecimalSuite.scala

@@ -446,6 +446,19 @@ class DecimalSuite extends SparkFunSuite with PrivateMethodTester with SQLHelper
     }
   }
 
+  test("UTF8String to Decimal with exponent near Int.MaxValue") {
+    // The number of integral digits (precision - scale) of these values overflows Int.
+    Seq("1e2147483647", "-1E+2147483647", "0e2147483647", "12e2147483647",
+      "1.5e2147483647", "1e2147483648").foreach { string =>
+      assert(Decimal.fromString(UTF8String.fromString(string)) === null)
+      checkError(
+        exception = intercept[SparkArithmeticException](
+          Decimal.fromStringANSI(UTF8String.fromString(string))),
+        condition = "NUMERIC_OUT_OF_SUPPORTED_RANGE",
+        parameters = Map("value" -> string))
+    }
+  }
+
   // 18 is a max number of digits in Decimal's compact long
   test("SPARK-41554: decrease/increase scale by 18 and more on compact decimal") {
     val unscaledNums = Seq(