etseidl commented on code in PR #9855:
URL: https://github.com/apache/arrow-rs/pull/9855#discussion_r3173492988
##########
parquet/src/arrow/arrow_reader/mod.rs:
##########
@@ -3686,6 +3686,76 @@ pub(crate) mod tests {
}
}
+ // test that we can handle the UNKNOWN logical type annotation on any
physical type
+ #[test]
+ fn test_unknown_logical_type() {
+ let message_type = "message uk {
+ OPTIONAL INT32 uki32 (UNKNOWN);
+ OPTIONAL INT64 uki64 (UNKNOWN);
+ OPTIONAL INT96 uki96 (UNKNOWN);
+ OPTIONAL BOOLEAN ukbool (UNKNOWN);
+ OPTIONAL FLOAT ukfloat (UNKNOWN);
+ OPTIONAL DOUBLE ukdbl (UNKNOWN);
+ OPTIONAL BYTE_ARRAY ukbytes (UNKNOWN);
+ OPTIONAL FIXED_LEN_BYTE_ARRAY(10) ukflba (UNKNOWN);
+ }";
+
+ let schema = Arc::new(parse_message_type(message_type).unwrap());
+ let file = tempfile::tempfile().unwrap();
+
+ let mut writer =
+ SerializedFileWriter::new(file.try_clone().unwrap(), schema,
Default::default())
+ .unwrap();
+
+ let mut row_group_writer = writer.next_row_group().unwrap();
+
+ fn write_nulls<T: DataType>(row_group_writer: &mut
SerializedRowGroupWriter<'_, File>) {
+ let mut column_writer =
row_group_writer.next_column().unwrap().unwrap();
+ // write out a bunch of nulls
+ column_writer
+ .typed::<T>()
+ .write_batch(&[], Some(&[0, 0, 0, 0]), None)
+ .unwrap();
+ column_writer.close().unwrap();
+ }
+
+ // INT32
+ write_nulls::<Int32Type>(&mut row_group_writer);
+
+ // INT64
+ write_nulls::<Int64Type>(&mut row_group_writer);
+
+ // INT96
+ write_nulls::<Int96Type>(&mut row_group_writer);
+
+ // BOOLEAN
+ write_nulls::<BoolType>(&mut row_group_writer);
+
+ // FLOAT
+ write_nulls::<FloatType>(&mut row_group_writer);
+
+ // DOUBLE
+ write_nulls::<DoubleType>(&mut row_group_writer);
+
+ // BYTE_ARRAY
+ write_nulls::<ByteArrayType>(&mut row_group_writer);
+
+ // FIXED_LEN_BYTE_ARRAY
+ write_nulls::<FixedLenByteArrayType>(&mut row_group_writer);
+
+ row_group_writer.close().unwrap();
+
+ writer.close().unwrap();
+
+ let mut reader = ParquetRecordBatchReader::try_new(file, 4).unwrap();
+ let batch = reader.next().unwrap().unwrap();
+
+ for col in batch.columns() {
+ assert_eq!(col.len(), 4);
Review Comment:
Done in
https://github.com/apache/arrow-rs/pull/9855/commits/ecbeaf141d4b9b4865fdae734d55fa8a151efbf6
--
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]
