[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH] Add Elixir
From: |
Ricardo Wurmus |
Subject: |
Re: [PATCH] Add Elixir |
Date: |
Tue, 2 Aug 2016 23:25:30 +0200 |
User-agent: |
mu4e 0.9.16; emacs 25.1.1 |
Ludovic Courtès <address@hidden> writes:
> Hello!
>
> Ricardo Wurmus <address@hidden> skribis:
>
>> I documented the most common test errors, but there don’t seem to be any
>> quick fixes. Attached is a new patch based on yours. If this looks
>> good to you I’ll push it with you as the commit author and with a
>> “Co-authored-by: Ricardo Wurmus” line in the commit summary.
>
> Woow, congratulation, and a big thanks for taking the time to
> investigate the issues!
>
>> +++ b/gnu/packages/patches/elixir-disable-failing-tests.patch
>> @@ -0,0 +1,261 @@
>> +Most of these tests fail for unknown reasons when run in the chroot
>> +environment of a Guix build process.
>> +
>> +Common failures are:
>> +
>> + * Mix.Shell.cmd() fails with error 130
>
> Could it be that this method relies on /bin/sh? We have
> patches/substitutions for similar functions in libc, gawk, and Guile,
> for instance.
It uses “sh -c” and even after replacing “sh” with the result of “(which
"sh")” the errors remained, so I removed the change again.
>> + * The git_repo fixture cannot be found
>> + * Communication with spawned processes fails with EPIPE
>
> It might be that processes are spawned with /bin/sh too, and thus fail
> instantly.
I’m not familiar enough with Erlang/Elixir to understand exactly what’s
going on. As far as I could see there are no explicit references to
“/bin/sh” in the code.
I pushed the commit as 0a4ebe01481d8f899bfc8849a27c1cf33c271c24.
Thanks, Pjotr, for the work!
~~ Ricardo