Skip to content

chore(c/validation): allow >1 partition in TestSqlPartitionedInts#4493

Merged
lidavidm merged 1 commit into
apache:mainfrom
fornwall:validation/partition-count-ge
Jul 9, 2026
Merged

chore(c/validation): allow >1 partition in TestSqlPartitionedInts#4493
lidavidm merged 1 commit into
apache:mainfrom
fornwall:validation/partition-count-ge

Conversation

@fornwall

@fornwall fornwall commented Jul 9, 2026

Copy link
Copy Markdown
Contributor

TestSqlPartitionedInts hardcoded ASSERT_EQ(1, num_partitions), which is not true for all engines (Spanner reports 2 here - one of them being empty).

Accept >=1 partitions and assert on the the union of all partitions.

@fornwall fornwall requested a review from lidavidm as a code owner July 9, 2026 19:03
@fornwall fornwall changed the title validation: allow >1 partition in TestSqlPartitionedInts chore(c/validation): allow >1 partition in TestSqlPartitionedInts Jul 9, 2026
TestSqlPartitionedInts hardcoded ASSERT_EQ(1, num_partitions), which is
not true for all engines (Spanner reports 2 here - one of them being empty).

Accept >=1 partitions and assert on the the union of all partitions.

Signed-off-by: Fredrik Fornwall <fredrik@fornwall.net>
@fornwall fornwall force-pushed the validation/partition-count-ge branch from e2317f5 to bcf52e2 Compare July 9, 2026 19:11
@lidavidm lidavidm merged commit f9886dd into apache:main Jul 9, 2026
86 of 87 checks passed
fornwall added a commit to fornwall/adbc-spanner that referenced this pull request Jul 9, 2026
…onedInts (#213)

apache/arrow-adbc#4493 relaxed TestSqlPartitionedInts from a hardcoded
ASSERT_EQ(1, num_partitions) to allowing >=1 partitions and asserting on
the union of all of them, so Spanner (whose partitionQuery may return more
than one — the emulator returns 2) now passes it.

Bump ARROW_ADBC_TAG in adbc-validation/CMakeLists.txt from the
apache-arrow-adbc-23 release tag to the main revision that carries the fix
(f9886ddca13a2ea2f044a453216b3973b66cd98f, current main HEAD), and drop
SpannerStatementTest.SqlPartitionedInts from the EXCLUDED list so the gate
enforces it. Docs updated to match (README bucket note + gate count 44->45).

Verified against a throwaway emulator: gate 47 passed / 1 self-skip / 0 fail,
expected-failure guard clean (52 excluded still fail-or-skip), stale guard
clean, and SqlPartitionedInts passes in isolation.


Claude-Session: https://claude.ai/code/session_01EMPx5dY5oYwNEP6fsViqu5

Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
@fornwall

fornwall commented Jul 9, 2026

Copy link
Copy Markdown
Contributor Author

Thanks!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants