Warn for outlives lint when gats are enabled for non-gats#91853
Warn for outlives lint when gats are enabled for non-gats#91853jackh726 wants to merge 2 commits intorust-lang:masterfrom
Conversation
|
☔ The latest upstream changes (presumably #91865) made this pull request unmergeable. Please resolve the merge conflicts. |
f237342 to
c018ead
Compare
nikomatsakis
left a comment
There was a problem hiding this comment.
coding wise, r+ from me
|
@jackh726 I'm wondering if we should rebase, make this deny-by-default, and do a crater run |
|
I think that would be the best way to move forward here. It'd be useful to have some data on what's affected and to see how many "false warnings" we get. |
|
I'll do that this weekend. Crater queue is a bit long right now, but I dont think this is time-sensitive at all. |
|
@rustbot author |
|
@bors try |
|
⌛ Trying commit cc51c4e with merge 46e9712624a97b4755100f5e467758cb4d5087e2... |
|
The job Click to see the possible cause of the failure (guessed by this bot) |
|
💔 Test failed - checks-actions |
|
The job Click to see the possible cause of the failure (guessed by this bot) |
|
☔ The latest upstream changes (presumably #93820) made this pull request unmergeable. Please resolve the merge conflicts. |
|
@jackh726 what's the update on this? |
|
I need to rework this a bit. It's on my list of things, but not a priority. |
|
Closing this due to inactivity |
Based on #91849
The basic idea is that having a where clause allows more flexibility, as with GATs. We can't error, because of backwards compatibility.
I think this needs further discussion before merge, but opening this for that discussion.
r? @nikomatsakis