-
Notifications
You must be signed in to change notification settings - Fork 97
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
Enable -Z stubbing
and error out instead of ignore stub
#2678
Conversation
Enable `-Z stubbing`, deprecate `--enable-stubbing`, and emit an error if the user tries to verify a harness that has stubs without explicitly enabling stubs. This used to be a warning before which was easy to miss. It was also hard to debug when the harness would take a while to run.
There are a number of tests that have |
They will emit a warning but the test shouldn't fail because of that. I can modify them in this PR If you prefer |
If we are deprecating a feature then, yes, I think we should completely remove its use from our repository. |
Replace the deprecated `enable-unstable` table.
@tautschnig just checking if you are planning on reviewing this PR or should I ping anyone else? Thanks! |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We should probably do a similar thing for the compiler. It also tracks stubbing differently from unstable features for some reason and it's quite confusing.
Also we could consider factoring the whole "unstable features" stuff out into a crate that is common between driver and compiler so we don't have to reimplement it twice for every feature.
Ah, do you mean replace the
I do think having a crate just for the "unstable features" stuff is a bit overkill, but we have been talking about creating a |
I actually already wrote the code for that and then realized it'd work best if we refactored the compiler args to use clap deriving also so I opened #2689 |
Ok. So I won't do that in this PR. Thanks @JustusAdam |
Thanks. I work fast when I get annoyed by something 😂 Looks good to me otherwise |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This makes sense to me, thanks @celinval !
One question: Are we tracking the removal of this feature somewhere? I don't think we have a mechanism in place to keep track of deprecated feature and when their removal is estimated to occur.
Co-authored-by: Adrian Palacios <73246657+adpaco-aws@users.noreply.github.com>
Description of changes:
Enable
-Z stubbing
, deprecate--enable-stubbing
, and emit an error if the user tries to verify a harness that has stubs without explicitly enabling stubs. This used to be a warning before which was easy to miss. It was also hard to debug when the harness would take a while to run.Resolved issues:
Resolves #ISSUE-NUMBER
Related RFC:
Optional #ISSUE-NUMBER.
Call-outs:
Testing:
How is this change tested?
Is this a refactor change?
Checklist
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 and MIT licenses.