modify-by-receiver: PoC of what is not detected yet

[ccoVeille]

Earlier, I looked at the code in the PR

• fix(rule/modifies-value-receiver): miss modification by ++/-- #1175

There was something that was unclear for me.
I wondered how the code would react to more complex cases than the one I saw in the tests.

Tonight, I added the tests, so we could discuss it.

I'm unsure if I detected things revive could report, or things that could lead to false-positive

@denisvmedia @chavacava @alexandear Please take a look and tell me.

Thanks

This follows up :

• modifies-value-receiver: miss modification by ++/-- #1109

Please note I code the test with an increment, but I think it applies to everything modify-by-receiver currently detects

[ccoVeille]

The tests are failing on purpose to show the potential issues we have with the rule now

[denisvmedia]

I agree these should be detected. Do you have a PoC how to detect them?

[ccoVeille]

Not yet, I'm unsure I will be able to work on them anytime soon. My AST is a bit rusty 😅

I wanted to gather feedbacks, first to see if I was wrong or not.

Could you confirm the one you would agree to report? Can you spot them one by one in the code changes provided.

I'm also interested by the one you agree to keep undetected.

When I will some feedbacks I think I will open an issue,because yes, anyone can work on fixing this.

[alexandear]

Maybe it's better to create an issue and continue the discussion there? PRs are usually used for showing work, not for discussions.

[ccoVeille]

In fact, I did a month ago

• modifies-value-receiver: miss modification by setting something in a sub structure  #1117

I have to confess I totally forgot it 😄

@chavacava just replied me about the complexity of such detection.

So let's continue on that issue