[Zhuo Ming (Pen-NineCat)]

Resolves https://cplusplus.github.io/LWG/issue4158

Implement packaged_task& operator=(packaged_task&&) with

packaged_task(_STD move(_Other)).swap(*this);

Close #6315

Maybe an other problem:

The self-assignment test (not in current PR)

{
     std::packaged_task<int()> task([] { return 42; });
     task = std::move(task);
     assert(task.valid());
}

passes on most configurations but fails on 5/46 (all /std:c++latest, mixed /MT//MDd).

Failed Tests (5):
  std :: tests/LWG4158_packaged_task_move_assignment:41
  std :: tests/LWG4158_packaged_task_move_assignment:42
  std :: tests/LWG4158_packaged_task_move_assignment:43
  std :: tests/LWG4158_packaged_task_move_assignment:44
  std :: tests/LWG4158_packaged_task_move_assignment:45

I don't know why only these configurations fail. My implementation follows the proposed resolution packaged_task(std::move(rhs)).swap(*this), so I expected task to stay valid after self-move-assignment. Could this be a separate issue, or is my test wrong? I'm not sure if self-move-assignment is supposed to be covered by this test at all.

[Zhuo Ming (Pen-NineCat)]

@microsoft-github-policy-service agree

[Copilot]

🟢 Approval recommended

The focused implementation reuses existing abandonment behavior, and the regression test covers the reported defect.

0 open findings

🧠 Review effort: Balanced

[Stephan T. Lavavej (StephanTLavavej)]

Can you please edit your PR description to properly auto-close our tracking issue? See https://docs.github.com/en/get-started/writing-on-github/working-with-advanced-formatting/using-keywords-in-issues-and-pull-requests . You need to literally say Close #6315 in your PR description to activate this functionality. Instead, your PR description has a much more complicated link Close https://github.com/orgs/microsoft/projects/1152/views/1?pane=issue&itemId=201549613&issue=microsoft%7CSTL%7C6315 even though it happens to be displayed the same way.

[Zhuo Ming (Pen-NineCat)]

Sorry for the confusion — I've updated the PR description to use Close #6315. Thanks for the clarification!