Re: Autoconf testing results.

[Pavel Roskin]

Hello, Akim!

> Ash 0.2 is already burned for being too broken.  Does it fail
> gracefully?  Is the user warned properly?

It dumps core gracefully, without saying a single word:

$ ash configure
Segmentation fault

> Could you had this limitation in the documentation?

If we make a patch, the documentation will be changed.

> > Also ifnames cannot be run by ash-0.2 for the same reason.
>
> Well, there is a big AWK program in there, held in a string.  It sure
> is bigger than 1K.

> I don't get it: we reverted the test to use test -f.  How can it fail?
> Arg.  Forgot to update Autoconf's configure.

I'm always doing it, but not this time. Maybe Autoconf needs a dist-hook
to regenerate configure on "make dist"?

> > and fails because it cannot create frozen files. After giving it the
> > right M4, configure and make work. However, all tests involving autoupdate
> > fail because Perl is missing. They shouldn't fail, they should be ignored.
>
> Right.  Should be handled as autoscan is.

We don't have a test for autoscan at all.

> > The issues with ash-0.2 can be worked around by creating temporary files,
> > but I doubt whether we should do it.
>
> Nope, let's not.  This shell is already too dangerous (given the way
> it propagates or does not propagate $?).  But having a nice message
> displayed to warn the user would be a good thing.

How about something like this in AS_SHELL_SANITIZE:

if (`exit 1`; :); then :
else
  AC_MSG_ERROR([Sorry, this shell is not capable of running this script.])
fi

I haven't yet figured out why AC_MSG_ERROR doesn't work as expected at
this point.

Important questions:

1) Do we want it to be in AS_SHELL_SANITIZE?
2) Should we check the other bugs, such as string size limit?

Regards,
Pavel Roskin