Mark1626 commented on code in PR #20460:
URL: https://github.com/apache/datafusion/pull/20460#discussion_r2868877197
##########
datafusion/functions-aggregate/src/count.rs:
##########
@@ -841,6 +865,156 @@ impl Accumulator for DistinctCountAccumulator {
}
}
+#[derive(Debug)]
+struct MultiColumnDistinctCountAccumulator {
+ values: HashSet<Vec<ScalarValue>, RandomState>,
+ state_data_types: Vec<DataType>,
+}
+
+impl MultiColumnDistinctCountAccumulator {
+ fn new(state_data_types: Vec<DataType>) -> Self {
+ Self {
+ values: HashSet::default(),
+ state_data_types,
+ }
+ }
+
+ fn update(&mut self, values: &[ScalarValue]) -> Result<()> {
+ if !values.iter().any(|v| v.is_null()) {
+ self.values.insert(values.to_vec());
+ }
+ Ok(())
+ }
+
+ fn fixed_size(&self) -> usize {
+ std::mem::size_of_val(self)
Review Comment:
Done
##########
datafusion/functions-aggregate/src/count.rs:
##########
@@ -1038,4 +1212,82 @@ mod tests {
assert_eq!(merged.evaluate()?, ScalarValue::Int64(Some(3)));
Ok(())
}
+
+ #[test]
+ fn multi_column_accumulator_basic() -> Result<()> {
+ let mut acc = MultiColumnDistinctCountAccumulator::new(vec![
+ DataType::Int32,
+ DataType::Utf8,
+ ]);
+
+ // (1, a), (1, b), (1, a), (2, b), (3, b)
+ let col1 = Arc::new(Int32Array::from(vec![
+ Some(1),
+ Some(1),
+ Some(1),
+ Some(2),
+ Some(3),
+ ]));
+ let col2 = Arc::new(StringArray::from(vec![
+ Some("a"),
+ Some("b"),
+ Some("a"),
+ Some("b"),
+ Some("b"),
+ ])) as ArrayRef;
+
+ acc.update_batch(&[col1, col2])?;
+ // Expected (1, a), (1, b), (2, b), (3, b)
+ assert_eq!(acc.evaluate()?, ScalarValue::Int64(Some(4)));
+ Ok(())
+ }
+
+ #[test]
+ fn multi_column_accumulator_merge() -> Result<()> {
+ let mut acc1 = MultiColumnDistinctCountAccumulator::new(vec![
+ DataType::Int32,
+ DataType::Utf8,
+ ]);
+
+ // (1, a), (1, b)
+ let col1 = Arc::new(Int32Array::from(vec![Some(1), Some(1)]));
+ let col2 = Arc::new(StringArray::from(vec![Some("a"), Some("b")])) as
ArrayRef;
+
+ acc1.update_batch(&[col1, col2])?;
+
+ let mut acc2 = MultiColumnDistinctCountAccumulator::new(vec![
+ DataType::Int32,
+ DataType::Utf8,
+ ]);
+
+ // (1, a), (2, b), (3, b)
+ let col1 = Arc::new(Int32Array::from(vec![Some(1), Some(2), Some(3)]));
+ let col2 = Arc::new(StringArray::from(vec![Some("a"), Some("b"),
Some("b")]))
+ as ArrayRef;
+
+ acc2.update_batch(&[col1, col2])?;
+
+ let state_sv1 = acc1.state()?;
+ let state_sv2 = acc2.state()?;
+ let state_arr1: Vec<ArrayRef> = state_sv1
+ .into_iter()
+ .map(|sv| sv.to_array())
+ .collect::<Result<_>>()?;
+ let state_arr2: Vec<ArrayRef> = state_sv2
+ .into_iter()
+ .map(|sv| sv.to_array())
+ .collect::<Result<_>>()?;
+
+ let mut merged = MultiColumnDistinctCountAccumulator::new(vec![
+ DataType::Int32,
+ DataType::Utf8,
+ ]);
+ merged.merge_batch(&state_arr1)?;
+ merged.merge_batch(&state_arr2)?;
+
+ // Expected (1, a), (1, b), (1, a), (2, b), (3, b)
Review Comment:
Updating this
--
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]
