etseidl commented on code in PR #9700:
URL: https://github.com/apache/arrow-rs/pull/9700#discussion_r3164514879
##########
parquet/src/arrow/arrow_writer/mod.rs:
##########
@@ -4813,18 +4909,67 @@ mod tests {
let col0_meta = metadata.row_group(0).column(0);
let col1_meta = metadata.row_group(0).column(1);
- let get_dict_page_size = move |meta: &ColumnChunkMetaData| {
- let mut reader =
- SerializedPageReader::new(Arc::new(data.clone()), meta, 0,
None).unwrap();
- let page = reader.get_next_page().unwrap().unwrap();
- match page {
- Page::DictionaryPage { buf, .. } => buf.len(),
- _ => panic!("expected DictionaryPage"),
- }
- };
+ assert_eq!(get_dict_page_size(col0_meta, data.clone()), 1024 * 1024);
+ assert_eq!(get_dict_page_size(col1_meta, data.clone()), 1024 * 1024 *
4);
+ }
+
+ #[test]
+ fn test_dict_page_size_decided_by_compression_fallback() {
+ // Generate values that are well dispersed across a range
approximating (0..256 * 1024)
+ let array = Arc::new(Int32Array::from_iter(
+ (0i32..1024 * 1024).map(|x| x.wrapping_mul(163019) % 262139),
+ ));
Review Comment:
Since this is `i32`, the `wrapping_mul` ultimately does wrap around to
negative numbers, so the cardinality will actually approach 512k.
Overall I don't think this test is really necessary. The earlier test covers
the early fall back.
--
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]
