feat: add use-setstate-synchronously rule

[Desdaemon]

What is the purpose of this pull request? (put an "X" next to an item)

• Documentation update
• Bug fix
• New rule
• Changes an existing rule
• Add autofixing to a rule
• Add a CLI option
• Add something to the core
• Other, please explain:

#1117

What changes did you make? (Give an overview)

• Basic implementation
• Rule customization
• Flow analysis
  • if checks
  • while checks
  • Ternaries
  • And-chains (mounted && ..)
  • Or-chains (!mounted || ..)

Is there anything you'd like reviewers to focus on?

Should this rule try to be as comprehensive as possible and check for all cases, or should the current heuristics for determining mounted-ness be sufficient?

[incendial]

@Desdaemon could you please update the changelog?

Should this rule try to be as comprehensive as possible and check for all cases, or should the current heuristics for determining mounted-ness be sufficient?

The most important that the rule shouldn't produce false-positives. So, if I'm not mistaken, covering And-chains (mounted && ..) is a must.

[incendial]

Great result! Left a few comments. Haven't seen @pragma('vm:prefer-inline') in action before 🙂

[incendial]

Amazing! The only question I have - will || create false positives? I'll later run this rule on the Flutter codebase (or you can, if you want to) to know for sure, but it'd be better if we avoid as many false positives as possible

[Desdaemon]

There's one I haven't found a way to cleanly implement:

if (!mounted || foo) return;
setState(); // should be mounted here

if (!mounted && bar || foo) return;
setState(); // LINT

When the first or-chain diverges, it should also have the effect of guaranteeing the widget is mounted, but currently it does not do so. If there are other false positives I'll prioritize fixing them as well.

[codecov]

Codecov Report

Merging #1120 (2111ddd) into master (a352607) will increase coverage by 0.17%.
The diff coverage is 95.48%.

@@            Coverage Diff             @@
##           master    #1120      +/-   ##
==========================================
+ Coverage   85.19%   85.37%   +0.17%     
==========================================
  Files         342      347       +5     
  Lines        7540     7673     +133     
==========================================
+ Hits         6424     6551     +127     
- Misses       1116     1122       +6     

Impacted Files	Coverage Δ
...c/analyzers/lint_analyzer/rules/rules_factory.dart	75.00% <ø> (ø)
...es/rules_list/use_setstate_synchronously/fact.dart	87.50% <87.50%> (ø)
...rules_list/use_setstate_synchronously/visitor.dart	93.22% <93.22%> (ø)
.../rules_list/use_setstate_synchronously/config.dart	100.00% <100.00%> (ø)
...rules_list/use_setstate_synchronously/helpers.dart	100.00% <100.00%> (ø)
...synchronously/use_setstate_synchronously_rule.dart	100.00% <100.00%> (ø)

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[Desdaemon]

I'll later run this rule on the Flutter codebase (or you can, if you want to) to know for sure, but it'd be better if we avoid as many false positives as possible

I ran this on the Flutter package on the stable branch, the only violation it detected was this situation:

https://github.com/flutter/flutter/blob/135454af32477f815a7525073027a3ff9eff1bfd/packages/flutter/lib/src/material/refresh_indicator.dart#L432-L439

I think this is too specific for this rule to handle, but otherwise no false positives yet. I'll run this on other codebases as well to cover as many edge cases as possible.

[incendial]

I think this is too specific for this rule to handle

Yeah, looks questionable, tbh (I mean, the code).

I'll run this on other codebases as well to cover as many edge cases as possible.

I've noticed a fix for a false positive, if you can also add tests - that'd be great. And after that I'm ready to merge, so tell me when you're ready.

[Desdaemon]

@incendial That should be my last batch of changes, should be ready to review and merge.

[incendial]

@Desdaemon please add tests for the config (example) and I'll merge

[incendial]

And please check the failing tests

[sonarqubecloud]

Kudos, SonarCloud Quality Gate passed!   

0 Bugs
0 Vulnerabilities
0 Security Hotspots
0 Code Smells

No Coverage information
0.0% Duplication

[incendial]

@Desdaemon thank you for the contribution! 🚀