Use edition ranges in tests/ui (Batch 1)#151464
Use edition ranges in tests/ui (Batch 1)#151464pvdrz wants to merge 6 commits intorust-lang:mainfrom
tests/ui (Batch 1)#151464Conversation
Co-authored-by: Lukas Wirth <lukastw97@gmail.com>
Co-authored-by: Lukas Wirth <lukastw97@gmail.com>
Co-authored-by: Lukas Wirth <lukastw97@gmail.com>
Co-authored-by: Lukas Wirth <lukastw97@gmail.com>
Co-authored-by: Lukas Wirth <lukastw97@gmail.com>
Co-authored-by: Lukas Wirth <lukastw97@gmail.com> Co-authored-by: Ana Hobden <ana.hobden@ferrous-systems.com>
|
|
|
I'm trying to do this in smaller batches for ease of review and to diminish the chance of the PR becoming stale |
| //@[edition2015]edition:2015 | ||
| //@[edition2018]edition:2018 |
There was a problem hiding this comment.
The stderr files for edition2015 and edition2018 are identical as far as I can tell, for which reason did you split them instead of keeping them combined as 2015..2021?
| //@ revisions: edition2015 edition2018 edition2021 | ||
| //@[edition2015]edition:2015 | ||
| //@[edition2018]edition:2018 | ||
| //@[edition2021]edition:2021.. |
There was a problem hiding this comment.
Strictly speaking, 2021.. isn't "edition2021". I think I'd prefer a different name for that revision, edition2021up or edition2021onwards, idk.
There was a problem hiding this comment.
I find it quite unfortunate and a bit unreasonable that we have to basically duplicate a huge stderr file for a single label change (unexpected argument #2 → unexpected argument #2 of type `!`). Hmm, however splitting out that test cases probably doesn't make sense either due to the large test setup. I don't know what's the call here..
|
☔ The latest upstream changes (presumably #152785) made this pull request unmergeable. Please resolve the merge conflicts. |
r? @fmease