test: add IN list slt coverage for temporal, Decimal128 and Interval types - #23875
Merged
Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #23875 +/- ##
=======================================
Coverage 80.86% 80.86%
=======================================
Files 1101 1101
Lines 374985 374985
Branches 374985 374985
=======================================
+ Hits 303214 303219 +5
+ Misses 53674 53668 -6
- Partials 18097 18098 +1 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
…types Part of apache#23307. Adds sqllogictest coverage for IN lists over Date32, Date64, Time32, Time64, Timestamp (with and without a timezone), Duration, Decimal128 and Interval(MonthDayNano), none of which had any SQL level coverage. Also adds a 17 element Int8/UInt8 list and a 9 element Float16 list so the bitmap specialization stays covered as shorter lists gain their own specializations. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
alamb
force-pushed
the
alamb/in-list-slt-temporal-decimal
branch
from
July 26, 2026 16:06
6d501e6 to
a3fd8f3
Compare
alamb
marked this pull request as ready for review
July 26, 2026 16:07
nuno-faria
approved these changes
Jul 29, 2026
nuno-faria
left a comment
nuno-faria
left a comment
Contributor
There was a problem hiding this comment.
Thanks @alamb, LGTM, I leave some minor suggestions below.
Also, according to the code coverage there are some uncovered IN cases with larger lists, like INT32. I don't know if it is important.
Comment on lines
+371
to
+380
| CREATE TABLE in_list_temporal AS | ||
| SELECT | ||
| label, | ||
| arrow_cast(arrow_cast(value, 'Int32'), 'Date32') AS d32, | ||
| arrow_cast(arrow_cast(value, 'Int64'), 'Date64') AS d64, | ||
| arrow_cast(arrow_cast(value, 'Int32'), 'Time32(Second)') AS t32s, | ||
| arrow_cast(arrow_cast(value, 'Int64'), 'Time64(Nanosecond)') AS t64ns, | ||
| arrow_cast(arrow_cast(value, 'Int64'), 'Timestamp(Nanosecond, None)') AS ts_ns, | ||
| arrow_cast(arrow_cast(value, 'Int64'), 'Timestamp(Second, Some("UTC"))') AS ts_s_utc, | ||
| arrow_cast(arrow_cast(value, 'Int64'), 'Duration(Second)') AS dur_s |
Contributor
There was a problem hiding this comment.
nit: The aliases are not aligned, if that was the idea 😄.
Comment on lines
+391
to
+397
| d32 IN (arrow_cast(arrow_cast(3, 'Int32'), 'Date32'), arrow_cast(arrow_cast(4, 'Int32'), 'Date32'), arrow_cast(arrow_cast(5, 'Int32'), 'Date32'), arrow_cast(arrow_cast(11, 'Int32'), 'Date32')), | ||
| d64 IN (arrow_cast(arrow_cast(3, 'Int64'), 'Date64'), arrow_cast(arrow_cast(4, 'Int64'), 'Date64'), arrow_cast(arrow_cast(5, 'Int64'), 'Date64'), arrow_cast(arrow_cast(11, 'Int64'), 'Date64')), | ||
| t32s IN (arrow_cast(arrow_cast(3, 'Int32'), 'Time32(Second)'), arrow_cast(arrow_cast(4, 'Int32'), 'Time32(Second)'), arrow_cast(arrow_cast(5, 'Int32'), 'Time32(Second)'), arrow_cast(arrow_cast(11, 'Int32'), 'Time32(Second)')), | ||
| t64ns IN (arrow_cast(arrow_cast(3, 'Int64'), 'Time64(Nanosecond)'), arrow_cast(arrow_cast(4, 'Int64'), 'Time64(Nanosecond)'), arrow_cast(arrow_cast(5, 'Int64'), 'Time64(Nanosecond)'), arrow_cast(arrow_cast(11, 'Int64'), 'Time64(Nanosecond)')), | ||
| ts_ns IN (arrow_cast(arrow_cast(3, 'Int64'), 'Timestamp(Nanosecond, None)'), arrow_cast(arrow_cast(4, 'Int64'), 'Timestamp(Nanosecond, None)'), arrow_cast(arrow_cast(5, 'Int64'), 'Timestamp(Nanosecond, None)'), arrow_cast(arrow_cast(11, 'Int64'), 'Timestamp(Nanosecond, None)')), | ||
| ts_s_utc IN (arrow_cast(arrow_cast(3, 'Int64'), 'Timestamp(Second, Some("UTC"))'), arrow_cast(arrow_cast(4, 'Int64'), 'Timestamp(Second, Some("UTC"))'), arrow_cast(arrow_cast(5, 'Int64'), 'Timestamp(Second, Some("UTC"))'), arrow_cast(arrow_cast(11, 'Int64'), 'Timestamp(Second, Some("UTC"))')), | ||
| dur_s IN (arrow_cast(arrow_cast(3, 'Int64'), 'Duration(Second)'), arrow_cast(arrow_cast(4, 'Int64'), 'Duration(Second)'), arrow_cast(arrow_cast(5, 'Int64'), 'Duration(Second)'), arrow_cast(arrow_cast(11, 'Int64'), 'Duration(Second)')) |
Contributor
There was a problem hiding this comment.
In this test and the others, I don't think the double arrow_cast is needed, only for Time32:
> select arrow_cast(3, 'Date32');
+-------------------------------------+
| arrow_cast(Int64(3),Utf8("Date32")) |
+-------------------------------------+
| 1970-01-04 |
+-------------------------------------+
> select arrow_cast(3, 'Time32');
Execution error: Unsupported type 'Time32'. Must be a supported arrow type name such as 'Int32' or 'Timestamp(ns)'. Error finding next token
Comment on lines
+463
to
+478
| #### | ||
| ## Decimal128 IN List Specializations | ||
| #### | ||
|
|
||
| statement ok | ||
| CREATE TABLE in_list_wide AS | ||
| SELECT * FROM (VALUES | ||
| ('match', arrow_cast(11, 'Decimal128(10, 2)'), INTERVAL '11 months'), | ||
| ('no_match', arrow_cast(7, 'Decimal128(10, 2)'), INTERVAL '7 months'), | ||
| ('nulls', NULL, NULL) | ||
| ) AS t(label, d128, imdn); | ||
|
|
||
| query TT | ||
| SELECT arrow_typeof(d128), arrow_typeof(imdn) FROM in_list_wide LIMIT 1 | ||
| ---- | ||
| Decimal128(10, 2) Interval(MonthDayNano) |
Contributor
There was a problem hiding this comment.
Should Decimal128 and INTERVAL maybe be split into different tests?
- Align the AS aliases in the in_list_temporal CREATE TABLE for readability. - Drop the unneeded intermediate arrow_cast to Int32/Int64 for all temporal types except Time32(Second), which requires an Int32 input. - Split the combined Decimal128/Interval table and tests into separate sections, one per type.
Contributor
Author
|
Thank you @nuno-faria -- I have addressed your comments |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Which issue does this PR close?
Rationale for this change
IN lists over temporal,
Decimal128andIntervalcolumns had no SQL level coverage at all, and the specializations being added in #23014 route those types onto new code paths.What changes are included in this PR?
Adds
in_list.sltcoverage forDate32,Date64,Time32(Second),Time64(Nanosecond),Timestamp(Nanosecond, None),Timestamp(Second, "UTC"),Duration(Second),Decimal128andInterval(MonthDayNano)Are these changes tested?
This PR is only tests; they pass on
mainand against the head of #23014.Are there any user-facing changes?
No.