[SPARK-48441][SQL] Fix StringTrim behaviour for non-UTF8_BINARY collations #46762

uros-db · 2024-05-27T20:45:38Z

What changes were proposed in this pull request?

String searching in UTF8_LCASE now works on character-level, rather than on byte-level. For example: ltrim("İ", "i") now returns "İ", because there exist no characters in "İ", starting from the left, such that lowercased version of those characters are equal to "i". Note, however, that there is a byte subsequence of "İ" such that lowercased version of that UTF-8 byte sequence equals to "i" (so the new behaviour is different than the old behaviour).

Also, translation for ICU collations works by repeatedly trimming the longest possible substring that matches a character in the trim string, starting from the left side of the input string, until trimming is done.

Why are the changes needed?

Fix functions that give unusable results due to one-to-many case mapping when performing string search under UTF8_LCASE (see example above).

Does this PR introduce any user-facing change?

Yes, behaviour of trim* expressions is changed for collated strings for edge cases with one-to-many case mapping.

How was this patch tested?

New unit tests in CollationSupportSuite and new e2e sql tests in CollationStringExpressionsSuite.

Was this patch authored or co-authored using generative AI tooling?

No.

common/unsafe/src/main/java/org/apache/spark/sql/catalyst/util/CollationSupport.java

### What changes were proposed in this pull request? CollationFactory has been updated to no longer mark UNICODE as collation that supportsBinaryCollation. To reflect these changes, various tests have been updated. However, some tests have been (temporarily) removed because StringTrim no longer supports UNICODE collation given the new UNICODE definition in CollationFactory. At this time, StringTrim expression only supports UTF8_BINARY & UTF8_BINARY_LCASE, but not ICU collations. This work is in progress (#46762), so we'll ensure appropriate test coverage with those changes. ### Why are the changes needed? UNICODE collation should not support binary collation. Note: in the future, we may want to consider a collation such as UNICODE_BINARY, which will support binary equality, but also maintain UNICODE ordering. ### Does this PR introduce _any_ user-facing change? Yes, UNICODE is no longer treated as a binary collation. This affects how equality works for UNICODE, and also which codepath is taken for various collation-aware string expression given UNICODE collated string arguments. ### How was this patch tested? Updated existing unit and e2e sql test for UNICODE collation. ### Was this patch authored or co-authored using generative AI tooling? No. Closes #46772 from uros-db/fix-unicode. Authored-by: Uros Bojanic <157381213+uros-db@users.noreply.github.com> Signed-off-by: Wenchen Fan <wenchen@databricks.com>

### What changes were proposed in this pull request? Renaming `UTF8_BINARY_LCASE` collation to `UTF8_LCASE`. ### Why are the changes needed? As part of the collation effort in Spark, we've moved away from byte-by-byte logic towards character-by-character logic, so what we used to call `UTF8_BINARY_LCASE` is now more precisely `UTF8_LCASE`. For example, string searching in UTF8_LCASE now works on character-level (rather than on byte-level), which is reflected in this PRs: #46511, #46589, #46682, #46761, #46762. In addition, string comparison also works on character-level now, as per the changes introduced in this PR: #46700. ### Does this PR introduce _any_ user-facing change? Yes, what was previously named `UTF8_BINARY_LCASE` collation, will from now on be named `UTF8_LCASE`. ### How was this patch tested? Existing tests. ### Was this patch authored or co-authored using generative AI tooling? No. Closes #46924 from uros-db/rename-lcase. Authored-by: Uros Bojanic <157381213+uros-db@users.noreply.github.com> Signed-off-by: Wenchen Fan <wenchen@databricks.com>

common/unsafe/src/main/java/org/apache/spark/sql/catalyst/util/CollationAwareUTF8String.java

common/unsafe/src/main/java/org/apache/spark/unsafe/types/UTF8String.java

common/unsafe/src/test/java/org/apache/spark/unsafe/types/CollationSupportSuite.java

mkaravel · 2024-07-08T23:34:30Z

@uros-db Can we please update the PR description to refer to the UTF8_LCASE collation as opposed to UTF8_BINARY_LCASE?

mkaravel · 2024-07-12T18:27:15Z

In the PR description it is stated:

Does this PR introduce any user-facing change?

Yes, behaviour of trim* expressions is changed for edge cases with one-to-many case mapping.

I think it is worth adding that the changes refer to collated strings only. So something like:

Yes, behaviour of trim* expressions for non-default collated strings is changed for edge cases with one-to-many case mapping.

mkaravel

PR looks good, nothing blocking.
My comments are about:

Improving the code documentation.
Potentially improving the performance for ICU collations (by de-duplicating the trimming characters).
Coding style.

I am approving this PR (as none of these is a blocker), but please address them before merging.

common/unsafe/src/main/java/org/apache/spark/sql/catalyst/util/CollationAwareUTF8String.java

uros-db · 2024-07-15T06:58:09Z

@cloud-fan ready for review

cloud-fan · 2024-07-15T08:32:54Z

thanks, merging to master!

…tions ### What changes were proposed in this pull request? String searching in UTF8_LCASE now works on character-level, rather than on byte-level. For example: `ltrim("İ", "i")` now returns `"İ"`, because there exist **no characters** in `"İ"`, starting from the left, such that lowercased version of those characters are equal to `"i"`. Note, however, that there is a byte subsequence of `"İ"` such that lowercased version of that UTF-8 byte sequence equals to `"i"` (so the new behaviour is different than the old behaviour). Also, translation for ICU collations works by repeatedly trimming the longest possible substring that matches a character in the trim string, starting from the left side of the input string, until trimming is done. ### Why are the changes needed? Fix functions that give unusable results due to one-to-many case mapping when performing string search under UTF8_LCASE (see example above). ### Does this PR introduce _any_ user-facing change? Yes, behaviour of `trim*` expressions is changed for collated strings for edge cases with one-to-many case mapping. ### How was this patch tested? New unit tests in `CollationSupportSuite` and new e2e sql tests in `CollationStringExpressionsSuite`. ### Was this patch authored or co-authored using generative AI tooling? No. Closes apache#46762 from uros-db/alter-trim. Authored-by: Uros Bojanic <157381213+uros-db@users.noreply.github.com> Signed-off-by: Wenchen Fan <wenchen@databricks.com>

Initial commit

0209fc4

github-actions bot added the SQL label May 27, 2024

HyukjinKwon reviewed May 27, 2024

View reviewed changes

common/unsafe/src/main/java/org/apache/spark/sql/catalyst/util/CollationSupport.java Outdated Show resolved Hide resolved

uros-db changed the title ~~[WIP][SQL] Fix StringTrim behaviour for non-UTF8_BINARY collations~~ [SPARK-48441][SQL] Fix StringTrim behaviour for non-UTF8_BINARY collations May 28, 2024

uros-db changed the title ~~[SPARK-48441][SQL] Fix StringTrim behaviour for non-UTF8_BINARY collations~~ [WIP][SPARK-48441][SQL] Fix StringTrim behaviour for non-UTF8_BINARY collations May 28, 2024

uros-db added 4 commits May 28, 2024 20:10

Small fixes

1f6881d

Merge branch 'apache:master' into alter-trim

204d26e

Fix trim logic

5039ddf

Lint fixes

361c7b1

uros-db mentioned this pull request Jun 6, 2024

[SPARK-48435][SQL] UNICODE collation should not support binary equality #46772

Closed

uros-db changed the title ~~[WIP][SPARK-48441][SQL] Fix StringTrim behaviour for non-UTF8_BINARY collations~~ [SPARK-48441][SQL] Fix StringTrim behaviour for non-UTF8_BINARY collations Jun 10, 2024

uros-db mentioned this pull request Jun 10, 2024

[SPARK-48576][SQL] Rename UTF8_BINARY_LCASE to UTF8_LCASE #46924

Closed

uros-db added 2 commits June 10, 2024 19:55

Merge branch 'master' into alter-trim

aaa46ff

Update CollationSupportSuite.java

1af57bf

mkaravel reviewed Jun 20, 2024

View reviewed changes

uros-db added 8 commits July 4, 2024 15:41

Merge branch 'master' into alter-trim

544548c

Merge branch 'apache:master' into alter-trim

e559e83

Fix test

292acf8

Refactor trim

097a846

Add tests

176b148

Merge branch 'apache:master' into alter-trim

9192b55

Remove unused code

598f522

Fix java lint

12cc084

uros-db requested a review from mkaravel July 8, 2024 13:04

uros-db added 3 commits July 9, 2024 20:28

Fix

469c325

Merge branch 'apache:master' into alter-trim

c2133d7

Fix comments

06575fc

Merge branch 'master' into alter-trim

4cf90d7

mkaravel approved these changes Jul 12, 2024

View reviewed changes

uros-db added 2 commits July 14, 2024 13:42

Fixes

8c5787d

Fix tests

d15d92a

cloud-fan approved these changes Jul 15, 2024

View reviewed changes

cloud-fan closed this in 95de02e Jul 15, 2024

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[SPARK-48441][SQL] Fix StringTrim behaviour for non-UTF8_BINARY collations #46762

[SPARK-48441][SQL] Fix StringTrim behaviour for non-UTF8_BINARY collations #46762

uros-db commented May 27, 2024 •

edited

Loading

mkaravel commented Jul 8, 2024

mkaravel commented Jul 12, 2024

Does this PR introduce any user-facing change?

mkaravel left a comment

uros-db commented Jul 15, 2024

cloud-fan commented Jul 15, 2024

[SPARK-48441][SQL] Fix StringTrim behaviour for non-UTF8_BINARY collations #46762

[SPARK-48441][SQL] Fix StringTrim behaviour for non-UTF8_BINARY collations #46762

Conversation

uros-db commented May 27, 2024 • edited Loading

What changes were proposed in this pull request?

Why are the changes needed?

Does this PR introduce any user-facing change?

How was this patch tested?

Was this patch authored or co-authored using generative AI tooling?

mkaravel commented Jul 8, 2024

mkaravel commented Jul 12, 2024

Does this PR introduce any user-facing change?

mkaravel left a comment

Choose a reason for hiding this comment

uros-db commented Jul 15, 2024

cloud-fan commented Jul 15, 2024

uros-db commented May 27, 2024 •

edited

Loading