[beta] [also fit for beta rollup] [HOTFIX] Don't delay a bug on malformed meta items involving interpolated tokens

[fmease]

Directly fixes the 3 crater regressions reported in #137687 (comment) (NB: The containing issue #137687 is in fact not an instance of these regressions, see #137687 (comment)).

Why is this a separate PR for beta? Well, the crater regressions were already fixed on master albeit unintentionally so, namely by PR #124141 which we certainly don't want to backport! So this is simply a hotfix. PR #140584 will then provide the regression test for master, too, so it doesn't get 'lost'.

[@]T-release, if/once accepted by T-compiler, this PR will be fit for beta rollup and can be cherry-picked.

FYI, we may also want to (separately) backport PR #140584 to [fix] #137687 (which, again, is not an instance of the crater regressions) but it's unclear if it's really necessary (since it's fuzzer-generated and I don't know of any real users who are impacted).

cc @jdonszelmann
r? fmease (waiting on T-compiler for acceptance)

[rustbot]

Some changes occurred in compiler/rustc_attr_parsing

cc @jdonszelmann

[fmease]

Ah right, PR #140397 hasn't landed yet (async backport nomination process for T-compiler). I guess waiting for the next T-compiler meeting (May 8) may still be fine (current beta is going to become stable May 15).

[jdonszelmann]

I'm not sure if this is normal, but we don't back port the tests? Apart from that, looks good. Pretty much the changes we made together this afternoon. Thanks for helping out <3

[fmease]

The test isn't really necessary but I've kept it as a CI sanity check and for the T-compiler-meetings people to have a quick reference to see what exactly is getting fixed

Edit: Or did you mean why didn't I backport the other tests, too? AFAIK, the others are concerning different issues that still reproduce / behave the same on stable, beta and nightly:

• tests/ui/invalid-compile-flags/print-file-names-request-malformed-crate-name-3.rs passes on all channels
• tests/ui/attributes/crate_name_on_other_items.rs (if split into two): The second #[crate_name] leads to an error on all channels and the first #[crate_name] passes on stable but ICEs on beta and nightly (as we know)

[nnethercote]

r=me if you need it, thanks!

[jieyouxu]

Ah right, PR #140397 hasn't landed yet (async backport nomination process for T-compiler). I guess waiting for the next T-compiler meeting (May 8) may still be fine (current beta is going to become stable May 15).

If you feel like its necessary/safer, manual async backport nomination is always an option.

[jieyouxu]

@fmease I pulled out the crater regressions out to a separate issue #140612 to prevent #137687 from having two issues in the same trenchcoat and be way too confusing. You may wish to adjust the PR description to repoint to that issue instead.