[m-rev.] for review: configure.in improvements
Fergus Henderson
fjh at cs.mu.OZ.AU
Sat Nov 16 18:08:14 AEDT 2002
On 16-Nov-2002, Simon Taylor <stayl at cs.mu.OZ.AU> wrote:
> > Several points:
> > - Could you please add a newline to the error message?
>
> I can change it to something like:
>
> checking whether to enable the GCC back-end interface... configure: error:
> --enable-gcc-back-end requires an already installed Mercury compiler.
>
> The "configure: error:" part is generated by AC_MSG_ERROR.
Hmm. I guess that is OK.
But it would IMHO be nicer to call AC_MSG_RESULT(...) before AC_MSG_ERROR(...),
e.g.
AC_MSG_RESULT(yes)
AC_MSG_ERROR(--enable-gcc-back-end requires an already
installed Mercury compiler.)
so that the output comes out as
checking whether to enable the GCC back-end interface... yes
configure: error: --enable-gcc-back-end requires an already
installed Mercury compiler.
Technically this is required by the specification of autoconf, which says
Macro: AC_MSG_CHECKING (FEATURE-DESCRIPTION)
...
This macro ... must be followed by a call to `AC_MSG_RESULT'
Another alternative would be to use AC_MSG_RESULT() with an empty argument.
> configure.in:
> @@ -242,7 +245,8 @@
> case $enable_aditi_back_end in
> no)
> if test "$BOOTSTRAP_MC" = ""; then
> - AC_MSG_WARN(--disable-aditi-back-end requires an already installed Mercury compiler.)
> + AC_MSG_WARN([
> +--disable-aditi-back-end requires an already installed Mercury compiler])
> enable_aditi_back_end=yes
> else
> # This will regenerate compiler/rl_out.m and
Here the output still isn't going to look good:
checking for whether to enable the Aditi back-end...
configure: warning: --disable-aditi-back-end requires an already installed Mercury compiler
no
It would be better to move the call to AC_MSG_WARN earlier,
and add a newline (and a tab) so that it comes out as
checking for whether to enable the Aditi back-end... no
configure: warning: --disable-aditi-back-end requires an already
installed Mercury compiler
> @@ -304,7 +308,8 @@
> fi
> if test "$MKFIFO" = ""; then
> - AC_MSG_WARN(cannot find a working \`mkfifo' or \`mknod')
> + AC_MSG_WARN([
> +cannot find a working \`mkfifo' or \`mknod'])
Why is a newline needed there?
> @@ -328,7 +333,8 @@
> esac
> fi
> if test "$MKTEMP" = ""; then
> - AC_MSG_WARN(cannot find a working \`mktemp', using \`mkdir || exit'')
> + AC_MSG_WARN([
> +cannot find a working \`mktemp', using \`mkdir || exit''])
Likewise here?
> @@ -1869,7 +1875,8 @@
>
> if test $mercury_cv_can_enable_deep_profiler = no; then
> echo
> - AC_MSG_ERROR(--enable-deep-profiler specified but system does not support it)
> + AC_MSG_ERROR([
> +--enable-deep-profiler specified but system does not support it])
> exit 1
> fi
The "echo" there should be AC_MSG_RESULT(), I think.
Then you wouldn't need a newline in the AC_MSG_ERROR(...) call.
> @@ -1882,7 +1889,8 @@
> true
> else
> echo
> - AC_MSG_ERROR(CGI script directory $mercury_cv_cgi_dir does not exist)
> + AC_MSG_ERROR([
> +CGI script directory $mercury_cv_cgi_dir does not exist])
> exit 1
Likewise here.
> @@ -3047,8 +3055,8 @@
> AC_DEFINE(MR_USE_EXTERNAL_DEBUGGER)
> else
> AC_MSG_RESULT(no)
> - AC_MSG_ERROR(Cannot enable external debugger)
> - AC_MSG_ERROR(because sockets do not work)
> + AC_MSG_ERROR([
> +Cannot enable external debugger because sockets do not work])
> fi
I don't think you need a newline there, because AC_MSG_RESULT should
output a newline.
This is all cosmetic, of course. If you want to commit your change as is
I'd be happy to tweak these details.
--
Fergus Henderson <fjh at cs.mu.oz.au> | "I have always known that the pursuit
The University of Melbourne | of excellence is a lethal habit"
WWW: <http://www.cs.mu.oz.au/~fjh> | -- the last words of T. S. Garp.
--------------------------------------------------------------------------
mercury-reviews mailing list
post: mercury-reviews at cs.mu.oz.au
administrative address: owner-mercury-reviews at cs.mu.oz.au
unsubscribe: Address: mercury-reviews-request at cs.mu.oz.au Message: unsubscribe
subscribe: Address: mercury-reviews-request at cs.mu.oz.au Message: subscribe
--------------------------------------------------------------------------
More information about the reviews
mailing list