Update MSRV to 1.56.1 #639
Conversation
Nothing out of the ordinary. The most alarming was the non-inclusive range in `test_utils.rs`. In some test functions I allowed myself to use `#[allow(clippy::type_complexity)]` because clippy was asking to create a type instead of using the big tuple. Since these were used just once, I thought it was overkill. Note, depends on: - rust-bitcoin#639 (we can also merge this into it instead of `master`) - rust-bitcoin#638 (has some allows on dead_code)
Following `rust-bitcoin` MSRV of 1.56.1, this pins MSRV to 1.56.1. Quoting @tcharding in rust-bitcoin/rust-bitcoin@d9cc724 > Rust version 1.56.0 introduced edition 2021. Shortly afterwards, on > October 21 2021 Rust version 1.56.1 was released. > > Debian stable is currently shipping `rustc 1.63.0`. > > Our stated MSRV policy is: In Debian stable and at least 2 years old. > > Therefore our MSRV policy is met by Rust version 1.56.1 and we can strat > to bump our MSRV org wide. > > Links: > - https://blog.rust-lang.org/2021/11/01/Rust-1.56.1.html > - https://blog.rust-lang.org/2021/10/21/Rust-1.56.0.html > - https://packages.debian.org/stable/rust/rustc Removing the pins in contrib/test.sh since they are not necessary in 1.56.1.
storopoli
force-pushed
the
js/msrv-1.56.1
branch
from
ad446c2
to
c301ce3
Compare
|
I ACKed because my local tests pass, but the real CI fails (an unsurprising outcome for a PR that's trying to change CI and not much else). |
|
Hey @storopoli I've sent you on a wild goose chase, I forgot we do not run the linter in CI in this repo. I think you should throw the You could put this on the attribute: |
|
Actually -- if the CI failures are because of missing parts from #638, then I can merge this. But I probably won't have time to check this until tomorrow. |
|
kludge and merge! |
|
What! I don't have overlord powers to change peoples PR descriptions - outrageous. @storopoli can you please remove the mention in the PR description, it makes github spam my notifications. Thanks man. |
Done, sorry about the spam. |
Nothing out of the ordinary. The most alarming was the non-inclusive range in `test_utils.rs`. In some test functions I allowed myself to use `#[allow(clippy::type_complexity)]` because clippy was asking to create a type instead of using the big tuple. Since these were used just once, I thought it was overkill. Note, depends on: - rust-bitcoin#639 (we can also merge this into it instead of `master`) - rust-bitcoin#638 (has some allows on dead_code)
Just so you know, the merge script warns about this (and if you can't edit descriptions, you can't merge either, I guess you're not a maintainer here) and I always edit out |
|
Lol, CI won't run the MSRV job unless the nightly job also passes. That's a dumb ordering because nightly breaks on its own and MSRV only breaks if the code is wrong. Anyway I'll just run contrib.sh locally and see how things work. |
|
Ok, looks good. On the off chance that the pins are actually needed we can reinstate them when we're done other the other CI fixes. |
Nothing out of the ordinary. The most alarming was the non-inclusive range in `test_utils.rs`. In some test functions I allowed myself to use `#[allow(clippy::type_complexity)]` because clippy was asking to create a type instead of using the big tuple. Since these were used just once, I thought it was overkill. Note, depends on: - rust-bitcoin#639 (we can also merge this into it instead of `master`) - rust-bitcoin#638 (has some allows on dead_code)
e2636f3 Clippy lints MSRV 1.56.1 (Jose Storopoli) Pull request description: Nothing out of the ordinary. The most alarming was the non-inclusive range in `test_utils.rs`. In some test functions I allowed myself to use `#[allow(clippy::type_complexity)]` because clippy was asking to create a type instead of using the big tuple. Since these were used just once, I thought it was overkill. Note, depends on: - #639 (we can also merge this into it instead of `master`) - #638 (has some allows on dead_code) ACKs for top commit: apoelstra: ACK e2636f3 Tree-SHA512: d1f38d1e2a1415da816898806377c90986072fe0bc9a9dbc3e5476026c05c53743453cbd0a237a70aa27762f1be86e44c19422575b7e62f686140accc412177c
Following
rust-bitcoinMSRV of 1.56.1, this pins MSRV to 1.56.1.Quoting a recent merge in
rust-bitcoinRemoving the pins in
contrib/test.shsince they are not necessaryin 1.56.1.
EDIT: CI fails because of the
#[allow(dead_code)]implemented in #638 on some Psbt traits.Do I port these to here? Conceptually they belong to #638 and not here.