Control: reopen -1

On Sat, Jul 20, 2024 at 8:45 AM Jonas Smedegaard <d...@jones.dk> wrote:

> Quoting Reinhard Tartler (2024-07-20 08:10:13)
> > with the patch above, autopkgtest now passes for me. Find the buildlog
> > attached.
> >
> > If it helps you, I could upload as an NMU maybe later today or tomorrow.
> > Either by disabling the tests or applying the patch above. Let me know
> your
> > preference on how to proceed with this RC bug.
>
> Thanks a lot for your help here!
>
> No need for an NMU, I can issue a release now.
>
> NB! If you can find time to look at bug#1076602, then that'll allow
> enabling features in axum that are needed for a use case of mine.
>
>
done, uploaded to unstable.

I'm taking the liberty to reopen this bug again, because
it seems there are additional autopkgtest failures in the rust-axum crate:

cf.
https://ci.debian.net/data/autopkgtest/testing/amd64/r/rust-axum/49298334/log.gz

3549s autopkgtest [04:14:16]: test rust-axum-macros-0.3:default:
-----------------------]
3549s autopkgtest [04:14:16]: test rust-axum-macros-0.3:default:  - - - - -
- - - - - results - - - - - - - - - -
3549s rust-axum-macros-0.3:default PASS
3550s autopkgtest [04:14:17]: @@@@@@@@@@@@@@@@@@@@ summary
3550s rust-axum-0.6:@      PASS
3550s rust-axum-0.6:       PASS
3550s rust-axum-0.6:default PASS
3550s rust-axum-0.6:form   PASS
3550s rust-axum-0.6:headers PASS
3550s rust-axum-0.6:http1  PASS
3550s rust-axum-0.6:http2  PASS
3550s rust-axum-0.6:json   PASS
3550s rust-axum-0.6:macros PASS
3550s rust-axum-0.6:matched-path PASS
3550s rust-axum-0.6:multipart PASS
3550s rust-axum-0.6:original-uri PASS
3550s rust-axum-0.6:query  PASS
3550s rust-axum-0.6:tokio  PASS
3550s rust-axum-0.6:tower-log PASS
3550s rust-axum-0.6:tracing PASS
3550s rust-axum-0.6:ws     PASS
3550s rust-axum-core-0.3:@ PASS
3550s rust-axum-core-0.3:  PASS
3550s rust-axum-core-0.3:default PASS
3550s rust-axum-core-0.3:tracing PASS
3550s rust-axum-extra-0.7:@ FAIL non-zero exit status 101
3550s rust-axum-extra-0.7: FAIL non-zero exit status 101
3550s rust-axum-extra-0.7:default FAIL non-zero exit status 101
3550s rust-axum-extra-0.7:async-read-body FAIL non-zero exit status 101
3550s rust-axum-extra-0.7:cookie FAIL non-zero exit status 101
3550s rust-axum-extra-0.7:cookie-key-expansion FAIL non-zero exit status 101
3550s rust-axum-extra-0.7:cookie-private FAIL non-zero exit status 101
3550s rust-axum-extra-0.7:cookie-signed FAIL non-zero exit status 101
3550s rust-axum-extra-0.7:erased-json FAIL non-zero exit status 101
3550s rust-axum-extra-0.7:json-lines FAIL non-zero exit status 101
3550s rust-axum-extra-0.7:multipart FAIL non-zero exit status 101
3550s rust-axum-extra-0.7:protobuf FAIL non-zero exit status 101
3550s rust-axum-macros-0.3:@ PASS
3550s rust-axum-macros-0.3: PASS
3550s rust-axum-macros-0.3:default PASS


The first one can be reproduced with:

$ ( cd axum-extra &&  cargo test --all-targets --all-features )

This fails in a similar way as above, apparently the test code was never
written in a way
to work without the 'tonic' feature present.

The other tests fail with:

siretart@x1:/srv/scratch/packages/rust/rust-axum $ ( cd axum-extra &&
 cargo test  --all-targets --no-default-features --features no-workspace )
warning: /srv/scratch/packages/rust/rust-axum/axum-extra/Cargo.toml:
`default_features` is deprecated in favor of `default-features` and will
not work in the 2024 edition
(in the `tower` dependency)
error: none of the selected packages contains these features: no-workspace,
did you mean: in-workspace?

Jonas, given upstream's stance on rejecting the notion of even attempting to
allow non-git checkouts to run the test-suites, in particular from the
crates
as published from crates.io (cf.
https://github.com/tokio-rs/axum/issues/2599#issuecomment-1970630060), I
wonder why you are investing so much time and aggravation into
getting the tests to work in Debian?

May I kindly suggest and ask you to defer that work until after we have
rust-axum
in testing to allow dependent packages to migrate, and then consider
getting back to
fixing those autopkgtests?


-- 
regards,
    Reinhard

Reply via email to