On Fri, Oct 23, 2015 at 01:06:36PM +0200, Lukas Slebodnik wrote:
On (21/10/15 09:20), Sumit Bose wrote:
>On Tue, Oct 20, 2015 at 10:15:06PM +0200, Jakub Hrozek wrote:
>> On Tue, Oct 20, 2015 at 05:17:49PM +0300, Nikolai Kondrashov wrote:
>> > Hi Jakub,
>> >
>> > On 10/19/2015 09:43 PM, Jakub Hrozek wrote:
>> > >I'm working on pam_sss.so tests[1] and I ran into a problem that I
don't
>> > >know how to solve best.
>> > >
>> > >tl;dr, I would like to set different environment variables for
different
>> > >tests in order to set up cwrap libraries differently per-test.
>> > >
>> > >I can't use setenv() in the test itself, because that's too
late, I need
>> > >the variables to be set when __attribute__(constructor) is run, so
normally
>> > >at program startup, when the libraries are loaded.
>> > >
>> > >With cmake it's easy, use set(TEST_ENVIRONMENT). But with
autotools, I
>> > >only found two ways:
>> > > - TESTS_ENVIRONMENT - this is fine, but it's per Makefile.am.
So I
>> > > would have to split the tests more, into pam_wrapper tests that
also
>> > > require uid_wrapper, tests that only require pam_wrapper, ...
>> > > - LOG_COMPILER - this allows to run a wrapper script before a
test
>> > > that receives the test name as argv. So this is pretty much what
I
>> > > want except this is a feature new to automake 1.12, which would
>> > > rule out both RHEL-6 and Ubuntu Trusty (which is used by
Travis)
>> > >
>> > >So I'm really leaning towards creating a
src/tests/cwrap/pwrap/Makefile.am
>> > >and src/tests/cwrap/pwrap_root/Makefile.am. The downside of multiple
>> > >Makefile.am files is that there is some code duplication and the build
>> > >takes longer. But I still think there is enough interest (from us and
from
>> > >our users) to support git master on old platforms. I can file a ticket
to
>> > >remove this and use LOG_COMPILER when we drop support for RHEL-6 and
old
>> > >Ubuntu versions...
>> > >
>> > >If you disagree, please reply, otherwise I'm going to send a patch
with
>> > >per-test Makefile...
>> >
>> > Ah, so these are unit tests, not integration tests?
>>
>> I'm working on both, actually. The first part is more or less an
>> isolated unit test of all the options that pam_sss supports. The reason
>> is that some options (2FA, smart cards, ...) are not really easily
>> testable without a mock back end, at the moment we only have openldap in
>> the integration tests.
>>
>> The next step I will start right after I finish this part is integration
>> tests that will exercise LDAP authentication, password change and maybe
>> authorisation if there's time left.
>>
>> >
>> > I'm not sure I understood everything right, sorry, but perhaps you can
find
>> > something useful in contrib/ci/run, contrib/ci/make-check-wrap and
>> > contrib/ci/valgrind-condense where CI matches and handles particular tests
>> > differently regardless of whether LOG_COMPILER is supported or not. Ping
me
>> > if you need help figuring out what's going on there.
>>
>> So more or less I wanted to have two tests and wanted to run the first
>> as (simplified):
>> PAM_WRAPPER=1 ./src/tests/cwrap/pam_sss_wrapper-tests
>> and other as:
>> PAM_WRAPPER=1 UID_WRAPPER=1 ./src/tests/cwrap/pam_sss_wrapper-root--tests
>>
>> but it occured to me that I can always start with UID wrapper, just drop
>> privileges if I need a strictly non-root test. It's a bit of a hack :-)
>> but since the root is fake anyway, I think it's acceptable.
>>
>> >
>> > Granted this is from outside the build, but maybe you can concoct
something
>> > from inside as well.
>>
>> I think this might work as well; thank you!
>
>I had a short look at libtool. Since we use it during 'make check' not
>the actual binaries are called but a libtool generated wrapper script.
>If it would be possible to set the environment variables here they would
>be visible for the binary at startup.
>
>Libtool has the concept of 'executable wrappers' to support cygwin and
>similar environments but I didn't found an easy way to add own wrapper
>here.
>
>Adding the variables directly in the generated wrapper scripts would be
>quite a hack. But maybe it would be possible if we add our own version
>of build/ltmain.sh where the wrapper scripts are generated in
>func_emit_wrapper()?
>
>So, I'm afraid this is not a direct answer to your question but maybe
>libtool might be useful here.
>
I read this thread after are came up with almost similar solution
as Sumit. With a small difference. I did not decide to inject env
variables to generated script but I decided to write yet another
wrapper on top of and set env variable there.
Here is a POC version. What do you think about such solution.
LS
From 79e8f0e1cdd09ca791755bb03c1e8f38b8078dc3 Mon Sep 17 00:00:00
2001
From: Lukas Slebodnik <lslebodn(a)redhat.com>
Date: Fri, 23 Oct 2015 13:06:00 +0200
Subject: [PATCH] temp
Thank you, this works like a charm!
ACK (but I had to apply with patch(1) in my pwrap branch, not sure if
the patch is applicable atop master...)