Skip to content

Commit 14fffb8

Browse files
committed
fix comments.
1 parent e4bd579 commit 14fffb8

File tree

1 file changed

+10
-11
lines changed
  • datafusion/physical-plan/src/aggregates/group_values

1 file changed

+10
-11
lines changed

datafusion/physical-plan/src/aggregates/group_values/column.rs

Lines changed: 10 additions & 11 deletions
Original file line numberDiff line numberDiff line change
@@ -207,9 +207,8 @@ impl<const STREAMING: bool> GroupValuesColumn<STREAMING> {
207207

208208
/// Scalarized intern
209209
///
210-
/// This is used only for `streaming aggregation`,
211-
/// because it depends on the order between `input rows` and their corresponding
212-
/// `group indices`.
210+
/// This is used only for `streaming aggregation`, because `streaming aggregation`
211+
/// depends on the order between `input rows` and their corresponding `group indices`.
213212
///
214213
/// For example, assuming `input rows` in `cols` with 4 new rows
215214
/// (not equal to `exist rows` in `group_values`, and need to create
@@ -222,7 +221,7 @@ impl<const STREAMING: bool> GroupValuesColumn<STREAMING> {
222221
/// row4
223222
/// ```
224223
///
225-
/// # In [`GroupValuesColumn`], their `group indices` will be
224+
/// # In `scalarized_intern`, their `group indices` will be
226225
///
227226
/// ```text
228227
/// row1 --> 0
@@ -234,7 +233,7 @@ impl<const STREAMING: bool> GroupValuesColumn<STREAMING> {
234233
/// `Group indices` order agrees with their input order, and the `streaming aggregation`
235234
/// depends on this.
236235
///
237-
/// # However In [`VectorizedGroupValuesColumn`], their `group indices` will be
236+
/// # However In `vectorized_intern`, their `group indices` will be
238237
///
239238
/// ```text
240239
/// row1 --> 2
@@ -376,12 +375,12 @@ impl<const STREAMING: bool> GroupValuesColumn<STREAMING> {
376375
//
377376
// 3. Perform `vectorized_equal_to` for `vectorized_equal_to_row_indices`
378377
// and `vectorized_equal_to_group_indices`. If found some rows in input `cols`
379-
// not equal to `exist rows` in `group_values`, place them in `scalarized_indices`
380-
// and perform `scalarized_intern` for them similar as what in [`GroupValuesColumn`]
378+
// not equal to `exist rows` in `group_values`, place them in `remaining_row_indices`
379+
// and perform `scalarized_intern_remaining` for them similar as `scalarized_intern`
381380
// after.
382381
//
383-
// 4. Perform `scalarized_intern` for rows mentioned above, when we process like this
384-
// can see the comments of `scalarized_intern`.
382+
// 4. Perform `scalarized_intern_remaining` for rows mentioned above, about in what situation
383+
// we will process this can see the comments of `scalarized_intern_remaining`.
385384
//
386385

387386
// 1. Collect vectorized context by checking hash values of `cols` in `map`
@@ -394,7 +393,7 @@ impl<const STREAMING: bool> GroupValuesColumn<STREAMING> {
394393
self.vectorized_equal_to(cols, groups);
395394

396395
// 4. Perform scalarized inter for remaining rows
397-
// (about remaining rows, can see comments for `remaining_rows`)
396+
// (about remaining rows, can see comments for `remaining_row_indices`)
398397
self.scalarized_intern_remaining(cols, &batch_hashes, groups);
399398

400399
self.hashes_buffer = batch_hashes;
@@ -1020,7 +1019,7 @@ impl<const STREAMING: bool> GroupValues for GroupValuesColumn<STREAMING> {
10201019
self.hashes_buffer.clear();
10211020
self.hashes_buffer.shrink_to(count);
10221021

1023-
// Such structure is only used in `non-streaming` case
1022+
// Such structures are only used in `non-streaming` case
10241023
if !STREAMING {
10251024
self.group_index_lists.clear();
10261025
self.emit_group_index_list_buffer.clear();

0 commit comments

Comments
 (0)