#5014 Fix MaxChainedCallsOnSameLine false positives #5020
Conversation
Fix issues where MaxChainedCallsOnSameLine would report long import and package directives, where a warning certainly isn't necessary.
dzirbel
changed the title
|
I've now run on the detekt codebase (with a max of 7 since there are some violations longer than 5) to verify that there aren't any more obvious false positives. There are still interesting cases (example) with fully-qualified class names which still violate this rule; thoughts on what to do there? I haven't looked into it yet, but I'm not certain whether it will be possible to differentiate those from normal qualified expressions. |
|
I think that we will need to use type solving. |
How would that rule benefit from type solving? Could you please highlight an example? |
This was in response to this example of a fully-qualified class name which this rule currently flags. Probably this rule should ignore such cases (although this one can just be imported), but I agree that type resolution is probably necessary to determine whether a chained calls are from a fully qualified class name. I'll open a separate issue to track/discuss that. |
No description provided.