*: Exchange partition, fix LIST COLUMNs validation as well as NULL validation

[mjonss]

What problem does this PR solve?

Issue Number: close #46492

Problem Summary:
The main issue was that the values are store in single quoted strings, and when adding them to the parser as parameters, they are quoted again, which issues (ignored) internal warnings, and no matching rows will be found.

Also the check did not handle NULL, so I added proper null-safe comparison.

What is changed and how it works?

Check List

Tests

• Unit test
• Integration test
• Manual test (add detailed scripts or steps below)
• No code

Side effects

• Performance regression: Consumes more CPU
• Performance regression: Consumes more Memory
• Breaking backward compatibility

Documentation

• Affects user behaviors
• Contains syntax changes
• Contains variable changes
• Contains experimental features
• Changes MySQL compatibility

Release note

Please refer to Release Notes Language Style Guide to write a quality release note.

None

[tiprow]

Hi @mjonss. Thanks for your PR.

PRs from untrusted users cannot be marked as trusted with /ok-to-test in this repo meaning untrusted PR authors can never trigger tests themselves. Collaborators can still trigger tests on the PR using /test all.

I understand the commands that are listed here.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[codecov]

Codecov Report

Merging #46533 (213d9e2) into master (923d98f) will decrease coverage by 0.6772%.
Report is 3 commits behind head on master.
The diff coverage is 88.8888%.

Additional details and impacted files

@@               Coverage Diff                @@
##             master     #46533        +/-   ##
================================================
- Coverage   73.3649%   72.6878%   -0.6772%     
================================================
  Files          1318       1339        +21     
  Lines        395802     402216      +6414     
================================================
+ Hits         290380     292362      +1982     
- Misses        86948      91335      +4387     
- Partials      18474      18519        +45     

Flag	Coverage Δ
integration	25.4197% <0.0000%> (?)
unit	73.3750% <88.8888%> (+0.0100%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Components	Coverage Δ
dumpling	54.0444% <ø> (ø)
parser	84.9581% <ø> (-0.0109%)	⬇️
br	48.1632% <ø> (-4.3043%)	⬇️

[mjonss]

I wonder if it would be better to just run the partition pruning on each row, and see if they match or not? (but then we would not be able to use any indexes etc.)

[YangKeao]

I wonder if it would be better to just run the partition pruning on each row, and see if they match or not? (but then we would not be able to use any indexes etc.)

It sounds not scalable? Even without index, the SQL can distribute the seletions to multiple TiKVs and make it faster.

[YangKeao]

Rest LGTM

[YangKeao]

It's a little complex, but maybe helpful. See /parser/format/format.go: (*RestoreContext).WriteName. You can use a ast.ColumnName to Restore the string.

It's also possible to "Restore" the whole SQL from ast, and maybe building an AST is more intuitive than concating strings for a complex statement, (but I'm not sure).

[mjonss]

OK, that is the function I saw as well, but the overhead of using it was bigger than the benefit, so I propose to use this direct quoting. Also generating the ast tree instead of sql string is probably more complex than the benefits would be.

[tiancaiamao]

This should work for most of the cases, I would suggest double check the internal sql not using ANSI/ANSI_QUOTE sql_mode

[mjonss]

I wonder if it would be better to just run the partition pruning on each row, and see if they match or not? (but then we would not be able to use any indexes etc.)

It sounds not scalable? Even without index, the SQL can distribute the seletions to multiple TiKVs and make it faster.

You are right, I realised that as well afterwards, since this would not allow to push down the filtering to TiKV/TiFlash, so all rows would need to go to TiDB, which would scale poorly.
So let us just keep it as is.
The only technical issue is that we need have this code complexity of generating the WHERE clause as in this PR, instead of reusing existing code for partitioning.

[YangKeao]

LGTM

[ti-chi-bot]

[LGTM Timeline notifier]

Timeline:

• 2023-09-05 08:35:43.745963583 +0000 UTC m=+2434508.294979567: ☑️ agreed by YangKeao.
• 2023-09-05 09:08:27.31250614 +0000 UTC m=+2436471.861522126: ☑️ agreed by xhebox.

[tiancaiamao]

This should work for most of the cases, I would suggest double check the internal sql not using ANSI/ANSI_QUOTE sql_mode

[ti-chi-bot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: tiancaiamao, xhebox, YangKeao

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [YangKeao,tiancaiamao,xhebox]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[ti-chi-bot]

[FORMAT CHECKER NOTIFICATION]

Notice: To remove the do-not-merge/needs-tests-checked label, please finished the test for the pull request and check the finished items in description.

For example:

Tests

• Unit test
• Integration test
• Manual test (add detailed scripts or steps below)
• No code

_{📖 For more info, you can check the "Contribute Code" section in the development guide.}

[ti-chi-bot]

In response to a cherrypick label: new pull request created to branch release-7.1: #46697.

[ti-chi-bot]

In response to a cherrypick label: new pull request created to branch release-6.5: #46698.