[pelovett]

Resolves: #2466

First time contributing so please let me know if I missed anything!

[pelovett]

@LucaCappelletti94 if you have some time, could you take another look? I tried to address the issues you identified.

[LucaCappelletti94]

Sure, currently I am trying to ultimate the fuzzing CI, once that is done reviews might get more swift given it should provide a stronger claim of consistency

[codecov-commenter]

Codecov Report

❌ Patch coverage is 78.18182% with 12 lines in your changes missing coverage. Please review.
✅ Project coverage is 80.96%. Comparing base (9ae00e7) to head (f292e1a).

Files with missing lines	Patch %	Lines
src/ast/table_constraints.rs	71.42%	4 Missing ⚠️
src/parser/mod.rs	85.71%	0 Missing and 4 partials ⚠️
src/dialect/mod.rs	50.00%	3 Missing ⚠️
src/ast/spans.rs	0.00%	1 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2467   +/-   ##
=======================================
  Coverage   80.96%   80.96%           
=======================================
  Files          42       42           
  Lines       33359    33412   +53     
  Branches    33359    33412   +53     
=======================================
+ Hits        27009    27053   +44     
- Misses       2790     2798    +8     
- Partials     3560     3561    +1     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[LucaCappelletti94]

The unparenthesized CHECK path has the same defect. CONSTRAINT c CHECK (a + 1) > 0 fails here but ClickHouse accepts it. Add an appropriate test for it, and correct this parser.

[pelovett]

Hey @LucaCappelletti94 , sorry for the delay in responding to your comments after requesting them. I've updated the code with your suggestions and added a fix for check constraints without parentheses. My solution adds a field to the constraint structs tracking if they were created with parentheses. This would be a breaking change for these structs.

The other option I could think of was storing the parenthesis state by wrapping the expressions in Expr::Nested but this could break code that creates structs manually rather than with the parser.

Not sure the best option here but would appreciate your opinion.