-
Couldn't load subscription status.
- Fork 13.9k
Gate tests with the right edition #147498
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: master
Are you sure you want to change the base?
Changes from all commits
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,3 +1,4 @@ | ||
| //@ edition:2015..2021 | ||
| fn empty() {} | ||
| fn one_arg<T>(_a: T) {} | ||
| fn two_arg_same(_a: i32, _b: i32) {} | ||
|
|
||
|
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. I'll use this concrete test as an example but this is more of a general question / issue (to be honest, I haven't looked at any of the other test changes so far): With only minor "meta" modifications (i.e., adjusting compiletest directives & annotations only), this test can easily run in Rust 2018, 2021 and 2024, too. So I'm wondering about the greater vision and how we should go about it procedurally: Did you plan on going over all tests again and submitting ~medium-sized follow-up PRs that relax the edition ranges of tests (and require a bit more manual work, namely those "meta" changes)? I mean I would assume so (I can also imagine the possibility that you consider this "done" for now, hence me asking). Alternatively, we could do this now in this PR. ^ This might be of interest for the next T-compiler meeting. There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. E.g., this test can be tweaked to work in Rust 2018 and Rust 2021 via trivial revisions: //@ revisions: e2015 e2018
//@[e2015] edition: 2015
//@[e2018] edition: 2018plus //[e2015]~| ERROR failed to resolve: use of unresolved module or unlinked crate `a`
//[e2018]~| ERROR failed to resolve: could not find `a` in the list of imported cratesOR //~| ERROR failed to resolveAs for Rust 2024, we probably want to migrate the implicitly unsafe That's just for illustration, I get that you plan on doing the Rust 2024 "receptivity" in a follow-up. There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. hey yes, you're right. The idea is that once the whole test suite passes with any edition, I can start submitting patches with smaller changes, more tailored to covering each test with an appropriate range of editions. Submitting those patches before this PR would be very annoying to review and run locally as you'd have to run only the tests that were changed.
This comment was marked as duplicate.
Sorry, something went wrong. There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Yeah, I think we should keep these PRs narrowly focused, mechanical is best. Mixing different kind of changes make it super easy for unexpected changes to be accidentally done among other diffs. There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. yes, this PR was written in a mostly mechanical way, there were some edge cases that required human intervention but most of the changes were done under the "what's the newest edition where this test compiles?" logic, instead of trying to update each test to compile under a reasonable edition. |
Uh oh!
There was an error while loading. Please reload this page.