Re: syslogd: Adjust to readutmp changes in Gnulib.

[Simon Josefsson]

Collin Funk <collin.funk1@gmail.com> writes:

> That reminds me though, Simon is there any specific reason Inetutils
> 'make check' doesn't also run Gnulib's checks? I think it is better to
> include them to better verify the programs work correctly. And hopefully
> catch Gnulib bugs when possible.

In my experience, adding gnulib self-tests to projects lead to a
maintainance cost to debug gnulib self-tests failures, which are rarely
of importance to the core project.  Most usages of gnulib consumes a
very tiny fraction of the API, but the gnulib self-test excercise all of
the APIs.  A fine-grained mechanism to add gnulib workarounds on demand
would be nice, but I don't think it will happen anytime soon.

The above is my old experience, so things may have changed these days.
Shall we enable it for inetutils to see how things play out?  I find
myself with little time to look into such problems, but I'm happy that
there is now a number of people providing patches which allows me to
simply ignore some e-mails I cannot find time to reply to.  Sometimes
simply having reports without answers is actually better than nothing.

Btw, if enable it would be nice to add a simple instruction to the
README saying how to run 'make check' with gnulib self-tests disabled.
Is there any agreed on practice to achieve that?  Primary to allow
people to bypass gnulib self-checks in case they get stuck in an
infinite loop (which used to happen not too rarely a couple of years ago
at least).

/Simon

signature.asc
Description: PGP signature