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
Regression related to AUTOGEN signature #3433
Comments
By the way, Blin points to this commit also for this:
yet the module tests are racey and I am not sure it is the guilty one, but worth a look? |
Hmmm... that's a commit from September, so also in 2019.11 ? |
@lizmat It was part of the big work on roles merged after the 2019.11 release. I'm running bisection on my side to see if it comes out with the same result. The commit seems pretty innocent and hardly related to signature matching at the first glance. |
Ok, the thing is: d89a0c9 isn't guilty. Clean rebuild of the full MoarVM/NQP/Rakudo stack on this commit makes t/00-basic.t test of BTW, my own bisect between 2019.11 and the HEAD came out with even weirder variant of 2c5c013! Anyway, I'm now trying another run between d89a0c9 and the HEAD. Hope to find something more sensible as the bug actually hurts me too. I just didn't realize it's a bug and thought it's me misunderstanding some subtle signature behaviors. |
Ok, got it. The case bisects down to b1e986c. What actually happens is @lizmat, perhaps it makes sense to convert |
With regard to |
It would be really awesome to see the fix for the case found, thus unblocking release. |
As suggested by Vadim Belman. This fixes #3433
After d89a0c9 we can't install a couple of modules with similar logs. For example, Duo:
or Hash::MultiValue
cast @vrurg
Is it a bug we can fix? Or a breakage we must do?
The text was updated successfully, but these errors were encountered: