alamb commented on code in PR #4589:
URL: https://github.com/apache/arrow-rs/pull/4589#discussion_r1280448636
##########
parquet/src/file/writer.rs:
##########
@@ -1648,4 +1650,61 @@ mod tests {
let reader = SerializedFileReader::new_with_options(file,
options).unwrap();
test_read(reader);
}
+
+ #[test]
+ fn test_disabled_statistics() {
+ let message_type = "
+ message test_schema {
+ REQUIRED INT32 a;
+ REQUIRED INT32 b;
+ }
+ ";
+ let schema = Arc::new(parse_message_type(message_type).unwrap());
+ let props = WriterProperties::builder()
+ .set_statistics_enabled(EnabledStatistics::None)
+ .set_column_statistics_enabled("a".into(), EnabledStatistics::Page)
+ .build();
+ let mut file = Vec::with_capacity(1024);
+ let mut file_writer =
+ SerializedFileWriter::new(&mut file, schema,
Arc::new(props)).unwrap();
+
+ let mut row_group_writer = file_writer.next_row_group().unwrap();
+ let mut a_writer = row_group_writer.next_column().unwrap().unwrap();
+ let col_writer = a_writer.typed::<Int32Type>();
+ col_writer.write_batch(&[1, 2, 3], None, None).unwrap();
+ a_writer.close().unwrap();
+
+ let mut b_writer = row_group_writer.next_column().unwrap().unwrap();
+ let col_writer = b_writer.typed::<Int32Type>();
+ col_writer.write_batch(&[4, 5, 6], None, None).unwrap();
+ b_writer.close().unwrap();
+ row_group_writer.close().unwrap();
+
+ let metadata = file_writer.close().unwrap();
+ assert_eq!(metadata.row_groups.len(), 1);
+ let row_group = &metadata.row_groups[0];
+ assert_eq!(row_group.columns.len(), 2);
+ assert!(row_group.columns[0].offset_index_offset.is_some());
+ assert!(row_group.columns[0].column_index_offset.is_some());
+ // Should only have offset index
Review Comment:
```suggestion
// Column "b" should only have offset index
```
##########
parquet/src/file/writer.rs:
##########
@@ -1648,4 +1650,61 @@ mod tests {
let reader = SerializedFileReader::new_with_options(file,
options).unwrap();
test_read(reader);
}
+
+ #[test]
+ fn test_disabled_statistics() {
+ let message_type = "
+ message test_schema {
+ REQUIRED INT32 a;
+ REQUIRED INT32 b;
+ }
+ ";
+ let schema = Arc::new(parse_message_type(message_type).unwrap());
+ let props = WriterProperties::builder()
+ .set_statistics_enabled(EnabledStatistics::None)
+ .set_column_statistics_enabled("a".into(), EnabledStatistics::Page)
+ .build();
+ let mut file = Vec::with_capacity(1024);
+ let mut file_writer =
+ SerializedFileWriter::new(&mut file, schema,
Arc::new(props)).unwrap();
+
+ let mut row_group_writer = file_writer.next_row_group().unwrap();
+ let mut a_writer = row_group_writer.next_column().unwrap().unwrap();
+ let col_writer = a_writer.typed::<Int32Type>();
+ col_writer.write_batch(&[1, 2, 3], None, None).unwrap();
+ a_writer.close().unwrap();
+
+ let mut b_writer = row_group_writer.next_column().unwrap().unwrap();
+ let col_writer = b_writer.typed::<Int32Type>();
+ col_writer.write_batch(&[4, 5, 6], None, None).unwrap();
+ b_writer.close().unwrap();
+ row_group_writer.close().unwrap();
+
+ let metadata = file_writer.close().unwrap();
+ assert_eq!(metadata.row_groups.len(), 1);
+ let row_group = &metadata.row_groups[0];
+ assert_eq!(row_group.columns.len(), 2);
+ assert!(row_group.columns[0].offset_index_offset.is_some());
Review Comment:
```suggestion
// Column "a" has both offset and column index, as requested
assert!(row_group.columns[0].offset_index_offset.is_some());
```
--
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]
