You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@arrow.apache.org by GitBox <gi...@apache.org> on 2022/11/02 03:07:41 UTC
[GitHub] [arrow-datafusion] Ted-Jiang commented on a diff in pull request #3967: Support pushdown multi-columns in PageIndex pruning.
Ted-Jiang commented on code in PR #3967:
URL: https://github.com/apache/arrow-datafusion/pull/3967#discussion_r1011107824
##########
datafusion/core/src/physical_plan/file_format/parquet.rs:
##########
@@ -497,37 +500,67 @@ impl FileOpener for ParquetOpener {
&file_metrics,
);
- if enable_page_index && check_page_index_push_down_valid(&pruning_predicate) {
+ let page_index_predicates = extract_page_index_push_down_predicates(
+ &pruning_predicate,
+ builder.schema().clone(),
+ )?;
+
+ if enable_page_index && !page_index_predicates.is_empty() {
let file_offset_indexes = file_metadata.offset_indexes();
let file_page_indexes = file_metadata.page_indexes();
if let (Some(file_offset_indexes), Some(file_page_indexes)) =
(file_offset_indexes, file_page_indexes)
{
- let mut selectors = Vec::with_capacity(row_groups.len());
- for r in &row_groups {
- selectors.extend(
- prune_pages_in_one_row_group(
- &groups[*r],
- pruning_predicate.clone(),
- file_offset_indexes.get(*r),
- file_page_indexes.get(*r),
- &file_metrics,
- )
- .map_err(|e| {
- ArrowError::ParquetError(format!(
- "Fail in prune_pages_in_one_row_group: {}",
- e
- ))
- }),
+ let mut row_selections =
+ VecDeque::with_capacity(page_index_predicates.len());
+ for predicate in page_index_predicates {
+ // `extract_page_index_push_down_predicates` only return predicate with one col.
+ let col_id =
+ *predicate.need_input_columns_ids().iter().next().unwrap();
+ let mut selectors = Vec::with_capacity(row_groups.len());
+ for r in &row_groups {
+ let rg_offset_indexes = file_offset_indexes.get(*r);
+ let rg_page_indexes = file_page_indexes.get(*r);
+ if let (Some(rg_page_indexes), Some(rg_offset_indexes)) =
+ (rg_page_indexes, rg_offset_indexes)
+ {
+ selectors.extend(
+ prune_pages_in_one_row_group(
+ &groups[*r],
+ &predicate,
+ rg_offset_indexes.get(col_id),
+ rg_page_indexes.get(col_id),
+ &file_metrics,
+ )
+ .map_err(|e| {
+ ArrowError::ParquetError(format!(
+ "Fail in prune_pages_in_one_row_group: {}",
+ e
+ ))
+ }),
+ );
+ } else {
+ // fallback select all rows
+ let all_selected = vec![RowSelector::select(
+ groups[*r].num_rows() as usize,
+ )];
+ selectors.push(all_selected);
+ }
+ }
+ debug!(
+ "Use filter and page index create RowSelection {:?} from predicate:{:?}",
+ &selectors, predicate
+ );
+ row_selections.push_back(
+ selectors.into_iter().flatten().collect::<Vec<_>>(),
);
}
- debug!(
- "Use filter and page index create RowSelection {:?} ",
- &selectors
- );
- builder = builder.with_row_selection(RowSelection::from(
- selectors.into_iter().flatten().collect::<Vec<_>>(),
- ));
+ let acc = row_selections.pop_front().unwrap_or_default();
+ let final_selection = row_selections
+ .into_iter()
+ .fold(acc, |acc, x| intersect_row_selection(acc, x, true));
+ builder =
+ builder.with_row_selection(RowSelection::from(final_selection));
Review Comment:
Thanks! forget the `reduce` operator.😂
--
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: github-unsubscribe@arrow.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org