Skip to content

branch-4.2: [fix](regression) Stabilize variant schema change and SNII cases - #70354

Open
eldenmoon wants to merge 2 commits into
apache:branch-4.2from
eldenmoon:branch-pick-70196-4.2-20261106
Open

eldenmoon wants to merge 2 commits into
apache:branch-4.2from
eldenmoon:branch-pick-70196-4.2-20261106

Conversation

@eldenmoon

Copy link
Copy Markdown
Member

cherry-pick #70196

Conflicts resolved in var_index_snii.groovy (took parse_to_variant multi-rowset inserts from the pick).

…che#70196)

Two VARIANT regression cases can fail despite correct results:

- The schema-change double-write case compares JSON booleans against
numeric values in its expected output. The input contains booleans,
which are preserved before and after schema change. Regenerate the
expected output and add a tie breaker to the limited query.
- The SNII range case checks a cumulative filtered-row counter during
each segment's initialization. When a segment that filters no rows runs
first, it fails before the segment containing the filtered row is
processed. Check the completed query profile instead, and write the test
data in separate rowsets to cover this ordering.

None

- Test: ASAN BE and FE build; `double_write_schema_change_with_variant`
regression (generated expected output and ordinary comparison);
`regression_test_variant_var_index_snii` regression (five consecutive
runs). The original schema-change mismatch and a multi-rowset counter
failure were reproduced before the fixes.
- Behavior changed: No product behavior change.
- Does this need documentation: No.

(cherry picked from commit 6969d59)
@eldenmoon
eldenmoon requested a review from yiguolei as a code owner November 6, 2026 11:49
@eldenmoon

Copy link
Copy Markdown
Member Author

run buildall

@hello-stephen

Copy link
Copy Markdown
Contributor

Thank you for your contribution to Apache Doris.
Don't know what should be done next? See How to process your PR.

Please clearly describe your PR:

  1. What problem was fixed (it's best to include specific error reporting information). How it was fixed.
  2. Which behaviors were modified. What was the previous behavior, what is it now, why was it modified, and what possible impacts might there be.
  3. What features were added. Why was this function added?
  4. Which code was refactored and why was this part of the code refactored?
  5. Which functions were optimized and what is the difference before and after the optimization?

@eldenmoon

Copy link
Copy Markdown
Member Author

run cloud_p0

1 similar comment
@eldenmoon

Copy link
Copy Markdown
Member Author

run cloud_p0

…e .out

Cherry-pick of apache#70196 brought master's expected JSON (public:true,
author-first commit keys). branch-4.x still serializes public as 1/0 and
sha-first commit object keys, so NonConcurrent
test_double_write_when_schema_change failed qt_two_phase_off/on.
Keep the groovy order-by stabilization; rewrite .out to 4.x format.
@eldenmoon

Copy link
Copy Markdown
Member Author

run buildall

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