1. 13 Oct, 2023 1 commit
    • Todd Volkert's avatar
      Make constraints a covariant argument in RenderBox.computeDryLayout() (#136432) · bc688cf0
      Todd Volkert authored
      Some render box subclasses have a specific layout contract that is tightly coupled with other render box subclasses (e.g. two private classes in a local project file). In these cases, it is also possible that they use a constraints object that is a subclass of `BoxConstraints`. To allow for this, this change makes the `constraints` argument to `RenderBox.computeDryLayout()` a covariant argument.
      
      For completeness' sake, this updates the other render objects in the rendering package to also use the covariant keyword for this argument.
      bc688cf0
  2. 20 Sep, 2023 1 commit
  3. 07 Aug, 2023 1 commit
  4. 16 Jun, 2023 1 commit
  5. 22 Mar, 2023 1 commit
  6. 08 Feb, 2023 1 commit
    • Greg Price's avatar
      Add test for RenderProxyBoxMixin; clarify doc, resolve TODO (#117664) · cd125e1f
      Greg Price authored
      * Add test for RenderProxyBoxMixin; clarify doc, resolve TODO
      
      The TODO comment suggested this mixin would no longer be needed once
      a Dart issue on inherited constructors was fixed:
        https://github.com/dart-lang/sdk/issues/31543
      That issue is now long since fixed, so I went to go carry out the TODO.
      
      But in doing so, I realized that the mixin's documentation was more
      right than the TODO comment: even with that issue fixed, there is a
      legitimate use case for this mixin, namely to reuse the implementation
      of RenderProxyBox in a class that also inherits from some other base
      class.  Moreover, searching GitHub I found an example of a library
      that makes real use of that capability.
      
      So I think the right resolution is to accept that this separation
      is useful and delete the TODO.
      
      Then, add a test with an extremely simplified sketch of that
      real-world example.  In case someone in the future attempts to
      simplify this mixin away, the test will point us at the use case
      that would be broken by such a change.
      
      Also remove the only in-tree use of the mixin, which was redundant;
      and expand the mixin's documentation to advise about that case.
      
      * Tweak formatting
      Co-authored-by: 's avatarMichael Goderbauer <goderbauer@google.com>
      
      * Cut comments
      
      ---------
      Co-authored-by: 's avatarMichael Goderbauer <goderbauer@google.com>
      cd125e1f
  7. 08 Sep, 2022 1 commit
  8. 29 Aug, 2022 1 commit
  9. 27 Jul, 2022 1 commit
  10. 02 Jul, 2022 1 commit
  11. 21 Jun, 2022 1 commit
  12. 17 Jun, 2022 1 commit
  13. 08 Jun, 2022 1 commit
  14. 07 Jun, 2022 1 commit
  15. 06 Jun, 2022 1 commit
  16. 20 May, 2022 1 commit
  17. 19 May, 2022 1 commit
  18. 16 May, 2022 1 commit
  19. 14 May, 2022 1 commit
  20. 27 Apr, 2022 1 commit
  21. 26 Apr, 2022 2 commits
  22. 14 Apr, 2022 1 commit
  23. 03 Feb, 2022 1 commit
  24. 08 Oct, 2021 3 commits
  25. 14 Sep, 2021 1 commit
  26. 13 Sep, 2021 2 commits
  27. 05 Aug, 2021 1 commit
  28. 08 Jun, 2021 1 commit
  29. 04 Jun, 2021 1 commit
  30. 23 Apr, 2021 1 commit
  31. 16 Mar, 2021 2 commits
  32. 15 Mar, 2021 1 commit
  33. 27 Jan, 2021 1 commit
  34. 08 Oct, 2020 1 commit
  35. 07 Oct, 2020 1 commit