omit_local_variable_types conflicts with --no-implicit-dynamic

[natebosch]

Specifically this conflicts with strong_mode_implicit_dynamic_variable.

[pq]

/cc @bwilkerson

[srawlins]

Linter rules and analyzer modes can conflict. What is the bug and what would you like the fix to be?

[natebosch]

At the time I wanted dynamic to be an allowed exception to omit_local_variable_types. I've hit other blockers to --no-implicit-dynamic so it hasn't been something I've pushed on.

[srawlins]

I see. So in this example:

dynamic x = 0;
dynamic y = x;

you'd like omit_local_variable_types to not complain. Seems reasonable.

[pq]

Is the proposal to make this conditional on whether no-implicit-dynamic is enabled?

[natebosch]

In my opinion it doesn't need to be conditional. dynamic is dangerous - if you're using it and want to make it explicit I think the lint should allow that regardless of other configuration.

[pq]

@bwilkerson: any thoughts/reservations?

[bwilkerson]

The purpose of this rule is to enforce the style guide. I don't know whether the style guide has changed since I last looked at it, but it seems to me that making dynamic an exception is consistent with the guide.