dongjoon-hyun commented on a change in pull request #30069:
URL: https://github.com/apache/spark/pull/30069#discussion_r506688087
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetRowConverter.scala
##########
@@ -300,15 +300,7 @@ private[parquet] class ParquetRowConverter(
new ParquetPrimitiveConverter(updater) {
// Converts nanosecond timestamps stored as INT96
override def addBinary(value: Binary): Unit = {
- assert(
- value.length() == 12,
- "Timestamps (with nanoseconds) are expected to be stored in
12-byte long binaries, " +
- s"but got a ${value.length()}-byte binary.")
Review comment:
~Shall we keep this assertion? Or, is it checked inside
`binaryToSQLTimestamp` already?~
Got it. I checked.
```
def binaryToSQLTimestamp(binary: Binary): Long = {
assert(binary.length() == 12, s"Timestamps (with nanoseconds) are
expected to be stored in" +
s" 12-byte long binaries. Found a ${binary.length()}-byte binary
instead.")
val buffer = binary.toByteBuffer.order(ByteOrder.LITTLE_ENDIAN)
val timeOfDayNanos = buffer.getLong
val julianDay = buffer.getInt
DateTimeUtils.fromJulianDay(julianDay, timeOfDayNanos)
}
```
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
