-
Notifications
You must be signed in to change notification settings - Fork 155
11.x: fix CI #835
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: release-11.x
Are you sure you want to change the base?
11.x: fix CI #835
Conversation
e340d80
to
a3b212e
Compare
This is working but I will need to rebase it after #832 because the lockfile has the old version number in it. |
a3b212e
to
f05624d
Compare
As in rust-bitcoin#835, I changed a test function that used 'A'..'Z' to use an inclusive range, which is behavior-changing, and changed the impl of OrdF64::partial_cmp to be correct. Everything else is just syntactic. I apologize for structuring this PR very differently from rust-bitcoin#835 so you can't really range-diff. I tried rebasing and there were so many conflicts that I simply redid the whole thing, and since I had some experience I made some different decisions about what order to take things in.
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.
tACK f05624d
Aside from being more future-proof, this also works with my local CI.
It may be that ubuntu-latest breaks us, but we're not doing anything particularly exotic so it seems unlikely. Meanwhile, our use of ubuntu-20.04 HAS broken us, because that runner is no longer supported. Meanwhile also update a bunch of our action versions to v4, since some of the v2s have become deprecated and removed. (Hopefully this won't happen very often.)
f05624d
to
d2cdeb4
Compare
Rebased. |
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.
On d2cdeb4 successfully ran local tests
Weird that this only has an effect on a couple files, which aren't checked in the Github CI. Same story with 12.x.
This, and the next commit, are not organized in any way except that I did a new commit partway to try to make review a bit easier. There were too many lints for me to split them all up and do them in a nice PR. This changes `OrdF64::partial_cmp` to be consistent with `OrdF64::cmp` which is a bugfix and an observable behavior change (but not an easy one to observe). All other changes are just syntax stuff.
In src/test_utils.rs I changed 'A'..'Z' to 'A'..='Z', which was a legitimate bugfix. The rest of these are just syntax stuff.
d2cdeb4
to
5e6e870
Compare
Regarding the above "successfully ran local tests" commints -- I currently have various checks disabled on backport branches. I hope to re-enable most of them but need to get these PRs in first. |
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.
On 5e6e870 successfully ran local tests
Can you re-ACK Sanket? |
As in rust-bitcoin#835, I changed a test function that used 'A'..'Z' to use an inclusive range, which is behavior-changing, and changed the impl of OrdF64::partial_cmp to be correct. Everything else is just syntactic. I apologize for structuring this PR very differently from rust-bitcoin#835 so you can't really range-diff. I tried rebasing and there were so many conflicts that I simply redid the whole thing, and since I had some experience I made some different decisions about what order to take things in.
As in rust-bitcoin#835, I changed a test function that used 'A'..'Z' to use an inclusive range, which is behavior-changing, and changed the impl of OrdF64::partial_cmp to be correct. Everything else is just syntactic. I apologize for structuring this PR very differently from rust-bitcoin#835 so you can't really range-diff. I tried rebasing and there were so many conflicts that I simply redid the whole thing, and since I had some experience I made some different decisions about what order to take things in.
As in rust-bitcoin#835, I changed a test function that used 'A'..'Z' to use an inclusive range, which is behavior-changing, and changed the impl of OrdF64::partial_cmp to be correct. Everything else is just syntactic. I apologize for structuring this PR very differently from rust-bitcoin#835 so you can't really range-diff. I tried rebasing and there were so many conflicts that I simply redid the whole thing, and since I had some experience I made some different decisions about what order to take things in.
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.
reACK 5e6e870.
Reviewed range-diff from previous ACK
Should get us green CI, and pins the nightly compiler version so that it stays green.