[fix] use runtime filter for multi-row prepared insert stmt (cherry-pick to 1.2-dev)

[aunjgr]

to avoid panic

Approved by: @ouyuanning

What type of PR is this?

• API-change
• BUG
• Improvement
• Documentation
• Feature
• Test and CI
• Code Refactoring

Which issue(s) this PR fixes:

issue #16178

What this PR does / why we need it:

to avoid panic

[matrix-meow]

@aunjgr Thanks for your contributions!

Pull Request Review:

Title and Body:

The title of the pull request is clear and concise, indicating that the fix is related to using a runtime filter for multi-row prepared insert statements. The body of the pull request provides additional context by mentioning that the fix is to avoid a panic issue. It also includes relevant information such as the approval and the specific issue number that this PR addresses. However, the explanation could be more detailed to help reviewers understand the specific nature of the panic issue and how the fix resolves it.

Changes:

1. Problem:

   • The added condition if rowsCount > 1 && len(bat.Vecs) > 0 && bat.Vecs[0].AllNull() in the getPkValueExpr function seems to be addressing a specific scenario where the number of rows is greater than 1 and the first vector contains all null values. However, the purpose and implications of this check are not clearly explained in the code comments.

   • Suggestion:

     • Add a comment above the condition to explain why this check is necessary and what specific scenario it is handling. This will help future developers understand the intent behind this logic.

2. Optimization:

   • The condition if rowsCount > 1 && len(bat.Vecs) > 0 && bat.Vecs[0].AllNull() could potentially be optimized for readability and maintainability.

   • Suggestion:

     • Consider extracting this complex condition into a separate descriptive function with a meaningful name that explains the purpose of the check. This can improve code readability and make it easier to understand the intent of the condition.

3. Security Concern:

   • While the change itself does not introduce any obvious security vulnerabilities, it is important to ensure that panic issues are properly handled to prevent potential denial-of-service attacks or unintended system behavior.

   • Suggestion:

     • Ensure that the fix for the panic issue is robust and does not introduce any new vulnerabilities. Consider adding appropriate error handling or logging to provide better visibility into the root cause of the panic and to prevent unexpected system failures.

General Suggestions:

• Documentation:

  • It would be beneficial to update the code comments to explain the logic behind the changes more clearly. This will help future developers understand the codebase and make maintenance easier.

• Testing:

  • Ensure that appropriate tests are added or updated to cover the scenario addressed by this fix. Comprehensive testing can help prevent regressions and validate the correctness of the changes.

• Code Quality:

  • Review the overall code structure and ensure consistency with the existing codebase. Maintain a consistent coding style to improve code readability and maintainability.

By addressing the mentioned points and considering the suggestions provided, the pull request can be enhanced in terms of clarity, maintainability, and security.