This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH] Write errors to stdout and not stderr in wcsmbs/tst-mbrtowc.c
- From: Arjun Shankar <arjun dot is at lostca dot se>
- To: Roland McGrath <roland at hack dot frob dot com>
- Cc: libc-alpha at sourceware dot org
- Date: Fri, 3 Oct 2014 23:05:01 +0200
- Subject: Re: [PATCH] Write errors to stdout and not stderr in wcsmbs/tst-mbrtowc.c
- Authentication-results: sourceware.org; auth=none
- References: <20141003205833 dot 6deb3569 at zion> <20141003193628 dot 9B2BD2C3AB2 at topped-with-meat dot com>
> The dup2 hack is either cute or horrible (likely both). I don't think we
> should do anything like that in one random test.
Okay.
> Rather, we should settle
> on a uniform policy for tests and using assert. It could be that we never
> use assert. Or it could be that we use assert freely and just make
> test-skeleton.c always do the dup2 hack itself. Or it could be something
> else I haven't thought of off hand. But let's get a plan.
One thing we cannot do is use `dup2' in the skeleton. It might confuse tests
that specifically test for output written to stdout/stderr. One such example
is 'posix/tst-getopt_long1.c'.
> If you'd like to convert these tests to use test-skeleton.c first as a
> separate change before we figure out how to address the assert issue,
> that could go in right away.
Okay. And I see that this itself is a task in the master TODO [1]. I will
get on it more thoroughly than just these two tests. In the meanwhile,
there are also a lot of tests writing to stderr [2]. If we have the policy
on that wrapped up, I will work on it next.
I am sort of new here, by the way. Hello, glibc hackers!
Cheers,
Arjun
[1] https://sourceware.org/glibc/wiki/Development_Todo/Master#Use_test-skeleton.c
[2] http://fpaste.org/139008/