jaylmiller commented on code in PR #5554:
URL: https://github.com/apache/arrow-datafusion/pull/5554#discussion_r1133280154
##########
datafusion/physical-expr/src/aggregate/count_distinct.rs:
##########
@@ -192,17 +230,150 @@ impl Accumulator for DistinctCountAccumulator {
}
}
}
+/// Special case accumulator for counting distinct values in a dict
+struct CountDistinctDictAccumulator<K>
+where
+ K: ArrowDictionaryKeyType + std::marker::Send + std::marker::Sync,
+{
+ /// `K` is required when casting to dict array
+ _dt: core::marker::PhantomData<K>,
+ /// laziliy initialized state that holds a boolean for each index.
+ /// the bool at each index indicates whether the value for that index has
been seen yet.
+ state: Option<Vec<bool>>,
+}
+
+impl<K> std::fmt::Debug for CountDistinctDictAccumulator<K>
+where
+ K: ArrowDictionaryKeyType + std::marker::Send + std::marker::Sync,
+{
+ fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+ f.debug_struct("CountDistinctDictAccumulator")
+ .field("state", &self.state)
+ .finish()
+ }
+}
+impl<K: ArrowDictionaryKeyType + std::marker::Send + std::marker::Sync>
+ CountDistinctDictAccumulator<K>
+{
+ fn new() -> Self {
+ Self {
+ _dt: core::marker::PhantomData,
+ state: None,
+ }
+ }
+}
+impl<K> Accumulator for CountDistinctDictAccumulator<K>
+where
+ K: ArrowDictionaryKeyType + std::marker::Send + std::marker::Sync,
+{
+ fn state(&self) -> Result<Vec<ScalarValue>> {
Review Comment:
This doesn't compile since `state` is part of `Accumulator` trait.
--
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]
