Re: Don't punish non-DJGPP/non-Windows platforms..., rev. 3 [PATCH]

[Alexandre Duret-Lutz]

>>> "Richard" == Richard Dawe <address@hidden> writes:

[...]

 Richard> Let me know if this is OK and I will commit it, when I
 Richard> get a CVS account at sources.redhat.com.

OK, with the change below

[...]

 Richard> +# The amount we should wait after modifying files depends on the 
platform.
 Richard> +# On Windows '95, '98 and ME, files modifications have 2-seconds
 Richard> +# granularity and can be up to 3 seconds in the future w.r.t. the
 Richard> +# system clock.  When it is important to ensure one file is older
 Richard> +# than another we wait at least 5 seconds between creations.
 Richard> +AC_MSG_CHECKING([how many seconds we should wait after modifying a 
file])
 Richard> +case $build in
 Richard> +  *-pc-msdosdjgpp) result=5;;
 Richard> +  *)               result=2;;
 Richard> +esac
 Richard> +AC_SUBST([MODIFICATION_DELAY], [$result])
 Richard> AC_MSG_RESULT([$result])

It's less armored, but it's still autotankish.  Please strip out
the AC_MSG_CHECKING and AC_MSG_RESULT for the reason I gave in
my previous message.  This is not a check.  The check was done
AC_CANONICAL_BUILD and you are just exploiting its result.
You might as well rename $result as $MODIFICATION_DELAY, BTW.

[...]

Sorry for the slow throughput and for being a pest.  I expect to
have more time next week.  (And I'm happy to see people willing
to help even if that might not sound so from my wording.)
-- 
Alexandre Duret-Lutz