Fedora Account System
Red Hat Associate
Red Hat Customer
SPEC: https://pbrobinson.fedorapeople.org/rust-yasna.spec SRPM: https://pbrobinson.fedorapeople.org/rust-yasna-0.5.2-1.fc39.src.rpm Description: Generic session middleware for the warp HTTP framework FAS: pbrobinson koji: https://koji.fedoraproject.org/koji/taskinfo?taskID=115678137
1. unaddressed FIXME: > # FIXME: no license files detected This needs to be addressed in one way or another, I see you already filed an issue - thanks! https://github.com/qnighy/yasna.rs/issues/76 Ideally, upstream should be asked to include "LICENSE-*" files here: https://github.com/qnighy/yasna.rs/blob/master/Cargo.toml#L14-L17 And publish new versions. Can you submit a PR for this? Until that is fixed upstream, you will need to include the license files from https://github.com/qnighy/yasna.rs manually as separate source files. 2. Why do tests fail on 32-bit x86? Is this a real problem in the code or problems with the tests? It looks like the Debian packagers also found this problem and submitted a PR to fix something: https://github.com/qnighy/yasna.rs/pull/73 Not sure if this addresses *all* failures on i686, but it would be great if that commit could be included as a patch and the tests enabled unconditionally.
(In reply to Fabio Valentini from comment #1) > 1. unaddressed FIXME: > > > # FIXME: no license files detected > > This needs to be addressed in one way or another, I see you already filed an > issue - thanks! > https://github.com/qnighy/yasna.rs/issues/76 > > Ideally, upstream should be asked to include "LICENSE-*" files here: > https://github.com/qnighy/yasna.rs/blob/master/Cargo.toml#L14-L17 > > And publish new versions. Can you submit a PR for this? It's on my backlog but I don't see upstream fix should block it, yes I'll update locally to add the LICENSE file. > Until that is fixed upstream, you will need to include the license files from > https://github.com/qnighy/yasna.rs manually as separate source files. > > 2. Why do tests fail on 32-bit x86? Is this a real problem in the code or > problems with the tests? TBH I don't have time to look into it and I don't personally care about 32 bit > Not sure if this addresses *all* failures on i686, but it would be great if > that commit could be included as a patch and the tests enabled > unconditionally. I would generally just err on disabling 32 bit in general, can you explain why it matters?
Updated with LICENSE-APACHE and LICENSE-MIT from upstream git: SPEC: https://pbrobinson.fedorapeople.org/rust-yasna.spec SRPM: https://pbrobinson.fedorapeople.org/rust-yasna-0.5.2-1.fc39.src.rpm https://koji.fedoraproject.org/koji/taskinfo?taskID=116002393
Sorry for the delay. I just checked, and the patch that was proposed by debian packagers indeed fixes the test failures. Can you include it make tests run unconditionally? Ideally, the license files would also come from upstream URLs instead of being plain files, to indicate that they are, indeed, from upstream. Something like this should do it: """ Source1: https://github.com/qnighy/yasna.rs/raw/yasna-0.5.0/LICENSE-APACHE Source2: https://github.com/qnighy/yasna.rs/raw/yasna-0.5.0/LICENSE-MIT # backport upstream patch to fix test failures on 32-bit architectures Patch: https://github.com/qnighy/yasna.rs/commit/bf5afbe.patch """
Are you still interested in getting this crate packaged?
(In reply to Fabio Valentini from comment #5) > Are you still interested in getting this crate packaged? Yes, need it for the newer parsec-tool
I also need this packaged. fwiw, it builds fine on i686 for me when using the following rust2rpm.toml [package] extra-sources = [ { number = 2, file = "https://github.com/qnighy/yasna.rs/raw/yasna-0.5.0/LICENSE-APACHE", comments = ["https://github.com/qnighy/yasna.rs/issues/76"]}, { number = 3, file = "https://github.com/qnighy/yasna.rs/raw/yasna-0.5.0/LICENSE-MIT", comments = ["https://github.com/qnighy/yasna.rs/issues/76"]}, ] extra-patches = [ { number = 2, file = "https://github.com/qnighy/yasna.rs/pull/73.patch", comments = ["Fix test failure on i686"]}, ] [scripts] prep.pre = ["cp -p %SOURCE2 %SOURCE3 ."]
Turns out I will need rcgen -> yasna too for the rustls / quinn stack soon. Peter, if you don't have the cycles to continue this package review, would it be OK if Davide or me submitted a new ticket to supersede yours?
*** This bug has been marked as a duplicate of bug 2312030 ***
(In reply to Fabio Valentini from comment #8) > Turns out I will need rcgen -> yasna too for the rustls / quinn stack soon. > Peter, if you don't have the cycles to continue this package review, would > it be OK if Davide or me submitted a new ticket to supersede yours? Yup, sorry, I completely missed that update.