suremarc commented on code in PR #15503:
URL: https://github.com/apache/datafusion/pull/15503#discussion_r2029801707
##########
datafusion/physical-plan/src/coalesce_batches.rs:
##########
@@ -195,6 +195,10 @@ impl ExecutionPlan for CoalesceBatchesExec {
Statistics::with_fetch(self.input.statistics()?, self.schema(),
self.fetch, 0, 1)
}
+ fn statistics_by_partition(&self) -> Result<Vec<Statistics>> {
+ Ok(vec![self.statistics()?])
+ }
Review Comment:
Seeing as `CoalesceBatchesExec` doesn't change partitioning of its input I
think this is the incorrect number of partitions? It should be repeated N
times, once for each partition.
##########
datafusion/physical-plan/src/execution_plan.rs:
##########
@@ -427,6 +427,16 @@ pub trait ExecutionPlan: Debug + DisplayAs + Send + Sync {
Ok(Statistics::new_unknown(&self.schema()))
}
+ /// Returns statistics for each partition of this `ExecutionPlan` node.
+ /// If statistics are not available, returns an array of
+ /// [`Statistics::new_unknown`] for each partition.
+ fn statistics_by_partition(&self) -> Result<Vec<Statistics>> {
+ Ok(vec![
+ Statistics::new_unknown(&self.schema());
+ self.properties().partitioning.partition_count()
+ ])
Review Comment:
Previously I had considered repeating the global statistics for each
partition, but I am not sure if this is correct. At a minimum we would need to
relax any exact statistics, and I guess we would also need to consider how the
node's partitioning could affect how data is distributed.
I guess it is safe no matter what to return unknown statistics and there
aren't any real benefits yet to trying to implement a better default. Curious
if you have any thoughts to add.
##########
datafusion/core/tests/physical_optimizer/partition_statistics.rs:
##########
@@ -0,0 +1,244 @@
+// Licensed to the Apache Software Foundation (ASF) under one
+// or more contributor license agreements. See the NOTICE file
+// distributed with this work for additional information
+// regarding copyright ownership. The ASF licenses this file
+// to you under the Apache License, Version 2.0 (the
+// "License"); you may not use this file except in compliance
+// with the License. You may obtain a copy of the License at
+//
+// http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing,
+// software distributed under the License is distributed on an
+// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+// KIND, either express or implied. See the License for the
+// specific language governing permissions and limitations
+// under the License.
+
+#[cfg(test)]
+mod test {
+ use arrow_schema::{DataType, Field, Schema, SortOptions};
+ use datafusion::datasource::listing::ListingTable;
+ use datafusion::prelude::SessionContext;
+ use datafusion_catalog::TableProvider;
+ use datafusion_common::stats::Precision;
+ use datafusion_common::{ColumnStatistics, ScalarValue, Statistics};
+ use datafusion_execution::config::SessionConfig;
+ use datafusion_expr_common::operator::Operator;
+ use datafusion_physical_expr::expressions::{binary, lit, Column};
+ use datafusion_physical_expr_common::physical_expr::PhysicalExpr;
+ use datafusion_physical_expr_common::sort_expr::{LexOrdering,
PhysicalSortExpr};
+ use datafusion_physical_plan::filter::FilterExec;
+ use datafusion_physical_plan::projection::ProjectionExec;
+ use datafusion_physical_plan::sorts::sort::SortExec;
+ use datafusion_physical_plan::union::UnionExec;
+ use datafusion_physical_plan::ExecutionPlan;
+ use std::sync::Arc;
+
+ /// Creates a test table with statistics from the test data directory.
+ ///
+ /// This function:
+ /// - Creates an external table from
'./tests/data/test_statistics_per_partition'
+ /// - If we set the `target_partition` to `2, the data contains 2
partitions, each with 2 rows
+ /// - Each partition has an "id" column (INT) with the following values:
+ /// - First partition: [3, 4]
+ /// - Second partition: [1, 2]
+ /// - Each row is 110 bytes in size
+ ///
+ /// @param target_partition Optional parameter to set the target partitions
+ /// @return ExecutionPlan representing the scan of the table with
statistics
+ async fn generate_listing_table_with_statistics(
+ target_partition: Option<usize>,
+ ) -> Arc<dyn ExecutionPlan> {
+ let mut session_config =
SessionConfig::new().with_collect_statistics(true);
+ if let Some(partition) = target_partition {
+ session_config = session_config.with_target_partitions(partition);
+ }
+ let ctx = SessionContext::new_with_config(session_config);
+ // Create table with partition
+ let create_table_sql = "CREATE EXTERNAL TABLE t1 (id INT not null,
date DATE) STORED AS PARQUET LOCATION
'./tests/data/test_statistics_per_partition' PARTITIONED BY (date) WITH ORDER
(id ASC);";
+ ctx.sql(create_table_sql)
+ .await
+ .unwrap()
+ .collect()
+ .await
+ .unwrap();
+ let table = ctx.table_provider("t1").await.unwrap();
+ let listing_table = table
+ .as_any()
+ .downcast_ref::<ListingTable>()
+ .unwrap()
+ .clone();
+ listing_table
+ .scan(&ctx.state(), None, &[], None)
+ .await
+ .unwrap()
+ }
+
+ /// Helper function to create expected statistics for a partition with
Int32 column
+ fn create_partition_statistics(
+ num_rows: usize,
+ total_byte_size: usize,
+ min_value: i32,
+ max_value: i32,
+ include_date_column: bool,
+ ) -> Statistics {
+ let mut column_stats = vec![ColumnStatistics {
+ null_count: Precision::Exact(0),
+ max_value: Precision::Exact(ScalarValue::Int32(Some(max_value))),
+ min_value: Precision::Exact(ScalarValue::Int32(Some(min_value))),
+ sum_value: Precision::Absent,
+ distinct_count: Precision::Absent,
+ }];
+
+ if include_date_column {
+ column_stats.push(ColumnStatistics {
+ null_count: Precision::Absent,
+ max_value: Precision::Absent,
+ min_value: Precision::Absent,
+ sum_value: Precision::Absent,
+ distinct_count: Precision::Absent,
+ });
+ }
+
+ Statistics {
+ num_rows: Precision::Exact(num_rows),
+ total_byte_size: Precision::Exact(total_byte_size),
+ column_statistics: column_stats,
+ }
+ }
+
+ #[tokio::test]
+ async fn test_statistics_by_partition_of_data_source() ->
datafusion_common::Result<()>
+ {
+ let scan = generate_listing_table_with_statistics(Some(2)).await;
+ let statistics = scan.statistics_by_partition()?;
+ let expected_statistic_partition_1 =
+ create_partition_statistics(2, 110, 3, 4, true);
+ let expected_statistic_partition_2 =
+ create_partition_statistics(2, 110, 1, 2, true);
+ // Check the statistics of each partition
+ assert_eq!(statistics.len(), 2);
+ assert_eq!(statistics[0], expected_statistic_partition_1);
+ assert_eq!(statistics[1], expected_statistic_partition_2);
+ Ok(())
+ }
Review Comment:
What do you think about adding a function that runs all partitions of an
`ExecutionPlan` (using
[execute_stream_partitioned](https://docs.rs/datafusion/latest/datafusion/physical_plan/fn.execute_stream_partitioned.html))
and checks if the min/max/etc. of each partition actually is consistent with
the `statistics_by_partition`? It would be a useful way to check that the
implementation matches what `statistics_by_partition` predicts, and I think we
should be able to write a single function to check this in all cases here.
--
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-unsubscr...@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For additional commands, e-mail: github-h...@datafusion.apache.org
