Re: Android port

[Eli Zaretskii]

> From: Po Lu <luangruo@yahoo.com>
> Cc: eggert@cs.ucla.edu,  bruno@clisp.org,  angelo.g0@libero.it,
>   emacs-devel@gnu.org
> Date: Sun, 06 Aug 2023 18:10:32 +0800
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> > Then I don't think I understand this, since (a) MinGW64 which Angelo
> > uses does have asprintf, and (b) we could use the configure variable
> > that records the existence of asprintf to override that, like we do
> > with other ac_cv_func_* variables.  But for some reason you didn't use
> > that variable (or any similar one) in your proposed patch, why?
> 
> That's because my patch was an addendum to changes installed on the
> feature/android branch months ago.  The diff between mingw-cfg.site on
> the branch and on master is more illustrative:
> 
> diff --git a/nt/mingw-cfg.site b/nt/mingw-cfg.site
> index 425eaace30d..f78ee525bf1 100644
> --- a/nt/mingw-cfg.site
> +++ b/nt/mingw-cfg.site
> @@ -173,3 +173,21 @@ gl_cv_func_nanosleep=yes
>  # Suppress configure-time diagnostic from unnecessary libxattr check,
>  # as xattr will not be supported here.
>  enable_xattr=no
> +# Don't build gnulib printf either.
> +gl_cv_func_printf_sizes_c99=yes
> +gl_cv_func_printf_sizes_c23=yes
> +gl_cv_func_printf_long_double=yes
> +gl_cv_func_printf_infinite_long_double=yes
> +gl_cv_func_printf_directive_a=yes
> +gl_cv_func_printf_directive_b=yes
> +gl_cv_func_printf_directive_f=yes
> +gl_cv_func_printf_directive_n=yes
> +gl_cv_func_printf_directive_ls=yes
> +gl_cv_func_printf_directive_lc=yes
> +gl_cv_func_printf_positions=yes
> +gl_cv_func_printf_flag_grouping=yes
> +gl_cv_func_printf_flag_leftadjust=yes
> +gl_cv_func_printf_flag_zero=yes
> +gl_cv_func_printf_precision=yes
> +gl_cv_func_printf_enomem=yes
> +ac_cv_func_vasprintf=yes

Isn't there some "summary" variable for printf, like
gl_cv_func_printf_works or some such, which is set if all of the above
variables are set?  I'd prefer to override summary if it exists,
rather than each one of the above function-testing variables
individually.

> Both the check for vasprintf (Gnulib eschews testing for asprintf, since
> asprintf is never present where vasprintf is not) and the checks for
> printf features are overridden, because Gnulib also tries to replace
> vasprintf if it discovers that the conventional printf functions will
> need to be replaced.

Perhaps we should ask Gnulib to allow disabling these in a more
convenient manner.