Hello Paul,
* Paul Eggert wrote on Tue, Dec 05, 2006 at 01:40:59AM CET:
> Ralf Wildenhues <[EMAIL PROTECTED]> writes:
>
> > +: ${sleep='sleep 2'} # Command to force different time stamps
>
> Why doesn't 'sleep 1' suffice? That '2' has me worried.
Quoting automake/Makefile.am:maintainer-check:
## Never use `sleep 1' to create files with different timestamps.
## Use `$sleep' instead. Some filesystems (e.g., Windows') have only
## a 2sec resolution.
Quoting automake/configure.ac:
# The amount we should wait after modifying files depends on the platform.
# On Windows '95, '98 and ME, files modifications have 2-seconds
# granularity and can be up to 3 seconds in the future w.r.t. the
# system clock. When it is important to ensure one file is older
# than another we wait at least 5 seconds between creations.
case $build in
*-pc-msdosdjgpp) MODIFICATION_DELAY=5;;
*) MODIFICATION_DELAY=2;;
esac
AC_SUBST([MODIFICATION_DELAY])
> > Do you think we'll need to cater for a 5 second
> > delay when $build is *-pc-msdosdjgpp (as Automake does)?
>
> Naah. Surely we can assume a reasonable build platform for this kind
> of test. If the platform isn't reasonable, we can skip the test.
So how do we determine reasonability? Matching $build (against what)?
*-pc-msdosdjgpp | *cygwin* | *mingw* | *pw32*
?
Cheers,
Ralf
