Ted-Jiang commented on code in PR #2335:
URL: https://github.com/apache/arrow-rs/pull/2335#discussion_r939543038
##########
parquet/src/arrow/async_reader.rs:
##########
@@ -271,25 +304,122 @@ impl<T: AsyncFileReader>
ParquetRecordBatchStreamBuilder<T> {
None => (0..self.metadata.row_groups().len()).collect(),
};
+ let reader = ReaderFactory {
+ input: self.input,
+ filter: self.filter,
+ metadata: self.metadata.clone(),
+ schema: self.schema.clone(),
+ };
+
Ok(ParquetRecordBatchStream {
+ metadata: self.metadata,
+ batch_size: self.batch_size,
row_groups,
projection: self.projection,
- batch_size: self.batch_size,
- metadata: self.metadata,
+ selection: self.selection,
schema: self.schema,
- input: Some(self.input),
+ reader: Some(reader),
state: StreamState::Init,
})
}
}
+type ReadResult<T> = Result<(ReaderFactory<T>,
Option<ParquetRecordBatchReader>)>;
+
+/// [`ReaderFactory`] is used by [`ParquetRecordBatchStream`] to create
+/// [`ParquetRecordBatchReader`]
+struct ReaderFactory<T> {
+ metadata: Arc<ParquetMetaData>,
+
+ schema: SchemaRef,
+
+ input: T,
+
+ filter: Option<RowFilter>,
+}
+
+impl<T> ReaderFactory<T>
+where
+ T: AsyncFileReader + Send,
+{
+ /// Reads the next row group with the provided `selection`, `projection`
and `batch_size`
+ ///
+ /// Note: this captures self so that the resulting future has a static
lifetime
+ async fn read_row_group(
+ mut self,
+ row_group_idx: usize,
+ mut selection: Option<RowSelection>,
Review Comment:
Is there a situation we need read row groups with previous row group
selection? 🤔
Edit: sorry it may come from `pageIndex`, forgive me
##########
parquet/src/arrow/async_reader.rs:
##########
@@ -271,25 +304,122 @@ impl<T: AsyncFileReader>
ParquetRecordBatchStreamBuilder<T> {
None => (0..self.metadata.row_groups().len()).collect(),
};
+ let reader = ReaderFactory {
+ input: self.input,
+ filter: self.filter,
+ metadata: self.metadata.clone(),
+ schema: self.schema.clone(),
+ };
+
Ok(ParquetRecordBatchStream {
+ metadata: self.metadata,
+ batch_size: self.batch_size,
row_groups,
projection: self.projection,
- batch_size: self.batch_size,
- metadata: self.metadata,
+ selection: self.selection,
schema: self.schema,
- input: Some(self.input),
+ reader: Some(reader),
state: StreamState::Init,
})
}
}
+type ReadResult<T> = Result<(ReaderFactory<T>,
Option<ParquetRecordBatchReader>)>;
+
+/// [`ReaderFactory`] is used by [`ParquetRecordBatchStream`] to create
+/// [`ParquetRecordBatchReader`]
+struct ReaderFactory<T> {
+ metadata: Arc<ParquetMetaData>,
+
+ schema: SchemaRef,
+
+ input: T,
+
+ filter: Option<RowFilter>,
+}
+
+impl<T> ReaderFactory<T>
+where
+ T: AsyncFileReader + Send,
+{
+ /// Reads the next row group with the provided `selection`, `projection`
and `batch_size`
+ ///
+ /// Note: this captures self so that the resulting future has a static
lifetime
+ async fn read_row_group(
+ mut self,
+ row_group_idx: usize,
+ mut selection: Option<RowSelection>,
Review Comment:
Maybe we need the previous each filter rate 😂(just a idea)
--
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]
