NVIDIA / NVIDIA/cudf

[FEA] Add `sliced_child_begin()` and `sliced_child_end()` iterators for `column_view`

Open
#9,650 2 comments 0 reactions 1 assignee Claimed by @ttnghia View on GitHub
feature request improvement libcudf non-breaking proposal
Dominant language
C++
Stars
9.8k
Forks
1.1k
Avg merge
3d 6m
Merged PRs (30d)
278

Description

Currently, the `column_view` class offers the accessor `child_begin()` and `child_end()` iterators that allow iterating over the children of a column. However, when we want to access the sliced children column (in many places), we have to do that manually. It would be great if we can cache the sliced children and provide iterators `sliced_child_begin()` and `sliced_child_end()` that allow quickly iterating over them.

Since different classes (`structs_column_view` and `lists_column_view`) have different interfaces for generating `column_view` of sliced child, we may have to implement the proposed iterators separately in these classes instead of implementing in the `column_view` base class.

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.