You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@datafusion.apache.org by "kumarUjjawal (via GitHub)" <gi...@apache.org> on 2026/04/21 06:09:36 UTC
[PR] perf(substr_index): speed up scalar and Utf8View [datafusion]
kumarUjjawal opened a new pull request, #21754:
URL: https://github.com/apache/datafusion/pull/21754
## Which issue does this PR close?
<!--
We generally require a GitHub issue to be filed for all bug fixes and enhancements and this helps us generate change logs for our releases. You can link an issue to this PR using the GitHub syntax. For example `Closes #123` indicates that this PR will close issue #123.
-->
- Part of #20585.
## Rationale for this change
`substr_index` is part of the Utf8View epic. This function was still missing the Utf8View optimization path, and it also did extra work when delimiter and count were constant scalars.
This PR adds the same kind of optimization used in similar string work and includes benchmark coverage.
<!--
Why are you proposing this change? If this is already explained clearly in the issue then this section is not needed.
Explaining clearly why changes are proposed helps reviewers understand your changes and offer better suggestions for fixes.
-->
## What changes are included in this PR?
- Return the input string type directly, including `Utf8View`.
- Add a zero-copy `Utf8View` path for `substr_index`.
- Add a scalar fast path when delimiter and count are constant.
- Keep the hot path specialized after checking a cleaner shared-helper version and benchmarking it.
- Add SQL coverage for `Utf8View` type and value results.
- Add unit tests for the scalar fast path, sliced `Utf8View` arrays, and the unchanged original-view case.
- Extend the benchmark to cover `Utf8`, `Utf8View`, array arguments, scalar delimiter/count, and positive and negative counts.
<!--
There is no need to duplicate the description in the issue here but it is sometimes worth providing a summary of the individual changes in this PR.
-->
## Are these changes tested?
Yes
<!--
We typically require tests for all PRs in order to:
1. Prevent the code from being accidentally broken by subsequent changes
2. Serve as another way to document the expected behavior of the code
If tests are not included in your PR, please explain why (for example, are they covered by existing tests)?
-->
## Are there any user-facing changes?
No
<!--
If there are user-facing changes then we may require documentation to be updated before approving the PR.
-->
<!--
If there are any breaking changes to public APIs, please add the `api change` label.
-->
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] perf(substr_index): speed up scalar and Utf8View [datafusion]
Posted by "kumarUjjawal (via GitHub)" <gi...@apache.org>.
kumarUjjawal commented on PR #21754:
URL: https://github.com/apache/datafusion/pull/21754#issuecomment-4294803989
> @kumarUjjawal Thanks for working on this! Can you post some benchmark data, please?
| Case | Before | After | Result |
|---|---:|---:|---:|
| `Utf8` scalar, single delimiter, `count = 1` | `20.18 us` | `11.29 us` | `1.79x faster` |
| `Utf8` scalar, long delimiter, `count = 1` | `39.82 us` | `13.69 us` | `2.91x faster` |
| `Utf8View` array, single delimiter | `26.64 us` | `11.77 us` | `2.26x faster` |
| `Utf8View` scalar, long delimiter, `count = 1` | `40.59 us` | `12.88 us` | `3.15x faster` |
| `Utf8View` array, long delimiter | `52.47 us` | `46.28 us` | `1.13x faster` |
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] perf(substr_index): speed up scalar and Utf8View [datafusion]
Posted by "kumarUjjawal (via GitHub)" <gi...@apache.org>.
kumarUjjawal commented on PR #21754:
URL: https://github.com/apache/datafusion/pull/21754#issuecomment-4298563865
Thanks @neilconway
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] perf(substr_index): speed up scalar and Utf8View [datafusion]
Posted by "alamb (via GitHub)" <gi...@apache.org>.
alamb merged PR #21754:
URL: https://github.com/apache/datafusion/pull/21754
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] perf(substr_index): speed up scalar and Utf8View [datafusion]
Posted by "neilconway (via GitHub)" <gi...@apache.org>.
neilconway commented on code in PR #21754:
URL: https://github.com/apache/datafusion/pull/21754#discussion_r3118375117
##########
datafusion/functions/src/unicode/substrindex.rs:
##########
@@ -130,126 +146,469 @@ fn substr_index(args: &[ArrayRef]) -> Result<ArrayRef> {
let string_array = str.as_string::<i32>();
let delimiter_array = delim.as_string::<i32>();
let count_array: &PrimitiveArray<Int64Type> = count.as_primitive();
- substr_index_general::<Int32Type, _, _>(
+ substr_index_general(
string_array,
delimiter_array,
count_array,
+ GenericStringBuilder::<i32>::with_capacity(
+ string_array.len(),
+ string_array.value_data().len(),
Review Comment:
Here and below, this over-allocates the buffer capacity for sliced arrays.
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] perf(substr_index): speed up scalar and Utf8View [datafusion]
Posted by "kumarUjjawal (via GitHub)" <gi...@apache.org>.
kumarUjjawal commented on PR #21754:
URL: https://github.com/apache/datafusion/pull/21754#issuecomment-4286307570
cc @neilconway since you are working on this area.
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] perf(substr_index): speed up scalar and Utf8View [datafusion]
Posted by "neilconway (via GitHub)" <gi...@apache.org>.
neilconway commented on PR #21754:
URL: https://github.com/apache/datafusion/pull/21754#issuecomment-4289219898
@kumarUjjawal Thanks for working on this! Can you post some benchmark data, please?
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org