Skip to content

proposal: deprecate (and remove?) lints for 3.0 #58947

Closed
@pq

Description

@pq

3.0 would be a good time to deprecate a host of lints that don't make sense in a null-safe world, are backed by stale advice or a have poor user experience (e.g., too many false positives).

  • ✅ deprecate: always_require_non_null_named_parameters [recommended]
  • ✅ deprecate: avoid_returning_null_for_future
  • ✅ deprecate: avoid_returning_null
  • ✅ deprecate prefer_equal_for_default_values [recommended] (analyzer now produces a diagnostic)
  • Deprecate iterable_contains_unrelated_type and list_remove_unrelated_type #58958 [core] (DELAYED)
  • 🤔 deprecate avoid_annotating_with_dynamic
  • deprecate: null_closures [recommended] (DECISION: still valuable)

We might also take this opportunity to deprecate lints that we no longer want to support.

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

    Milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions