zeruibao commented on code in PR #41052:
URL: https://github.com/apache/spark/pull/41052#discussion_r1201271382
##########
connector/avro/src/main/scala/org/apache/spark/sql/avro/AvroDeserializer.scala:
##########
@@ -117,6 +119,19 @@ private[sql] class AvroDeserializer(
val incompatibleMsg = errorPrefix +
s"schema is incompatible (avroType = $avroType, sqlType =
${catalystType.sql})"
+ val realDataType = SchemaConverters.toSqlType(avroType).dataType
+ val confKey = SQLConf.LEGACY_AVRO_ALLOW_READING_WITH_INCOMPATIBLE_SCHEMA
+ val preventReadingIncorrectType = !SQLConf.get.getConf(confKey)
+ def isNotExactType: Boolean = {
+ preventReadingIncorrectType && !DataTypeUtils.sameType(realDataType,
catalystType)
Review Comment:
Yeah, I use
```
(avroType.getType, catalystType) match {
// check whether allow conversion of DecimalType
case (FIXED, dt: DecimalType) =>
val d = avroType.getLogicalType.asInstanceOf[LogicalTypes.Decimal]
if (preventReadingIncorrectType &&
d.getPrecision - d.getScale > dt.precision - dt.scale) {
throw lowerPrecisionException(dt)
}
case (BYTES, dt: DecimalType) =>
val d = avroType.getLogicalType.asInstanceOf[LogicalTypes.Decimal]
if (preventReadingIncorrectType &&
d.getPrecision - d.getScale > dt.precision - dt.scale) {
throw lowerPrecisionException(dt)
}
// show allow following conversion
case (RECORD, _: StructType) =>
case (LONG, _: TimestampType)
if !preventReadingIncorrectType ||
!realDataType.isInstanceOf[DayTimeIntervalType] =>
case (LONG, _: TimestampNTZType)
if !preventReadingIncorrectType ||
!realDataType.isInstanceOf[DayTimeIntervalType] =>
// check whether allow different type
case _ if isNotExactType =>
throw incorrectTypeException(catalystType)
case _ =>
}
```
to fail earier. I am not sure whether it makes the code simpler. What do you
think?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
