Paul Eggert wrote:
> the getlogin_r.m4 change (adding
> AC_REQUIRE([AC_GNU_SOURCE])) looks a bit GNU-specific. Wouldn't it be
> better to depend on the 'extensions' module instead? That would catch
> similar problems in non-GNU systems.
>
> (Come to think of it, doesn't a similar argument apply
Ralf Wildenhues <[EMAIL PROTECTED]> writes:
> * getlogin_r.m4 (gl_PREREQ_GETLOGIN_R): Require AC_GNU_SOURCE.
> * host-os.m4 (gl_HOST_OS): Require AC_CANONICAL_HOST.
The host-os.m4 change looks fine, but the getlogin_r.m4 change (adding
AC_REQUIRE([AC_GNU_SOURCE])) looks a bit GNU-spec
> (b) comparing (1) and (2):
> getlogin_r declared in the glob test, but not in the getlogin_r test.
> (c) HOST_OPERATING_SYSTEM is set to empty in (2).
OK to apply?
Cheers,
Ralf
* getlogin_r.m4 (gl_PREREQ_GETLOGIN_R): Require AC_GNU_SOURCE.
* host-os.m4 (gl_HOST_OS): Require AC
