Tpt commented on code in PR #18254:
URL: https://github.com/apache/datafusion/pull/18254#discussion_r2466077757
##########
datafusion/physical-plan/src/recursive_query.rs:
##########
@@ -434,5 +452,55 @@ impl RecordBatchStream for RecursiveQueryStream {
}
}
+/// Deduplicator based on a hash table.
+struct DistinctDeduplicator {
+ /// Grouped rows used for distinct
+ group_values: Box<dyn GroupValues>,
+ reservation: MemoryReservation,
+ intern_output_buffer: Vec<usize>,
+}
+
+impl DistinctDeduplicator {
+ fn new(schema: SchemaRef, task_context: &TaskContext) -> Result<Self> {
+ let group_values = new_group_values(schema, &GroupOrdering::None)?;
+ let reservation = MemoryConsumer::new("RecursiveQueryHashTable")
+ .register(task_context.memory_pool());
+ Ok(Self {
+ group_values,
+ reservation,
+ intern_output_buffer: Vec::new(),
+ })
+ }
+
+ fn deduplicate(&mut self, batch: &RecordBatch) -> Result<RecordBatch> {
+ // We use the hash table to allocate new group ids.
Review Comment:
Indeed. I have moved the comment as a doc comment and rephrased it to be
hopefully clearer.
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
