Re: Android port

[Paul Eggert]

On 2023-08-05 22:24, Eli Zaretskii wrote:
> Paul didn't chime in yet, so I'd like to wait for him to comment on
> this.  I see no reason why we would be unable to omit these modules
> like we do with others.

I don't either, but I hope we don't have to worry about it.

As I understand it the Android port uses Gnulib printf-posix and 
vasprintf-posix modules only because Android printf lacks support for 
"%td", "%jd" and "%ju". If this understanding is correct, how about if 
we go through the printf formats in the Emacs C source code, and replace 
all uses of "%jd" and "%ju" with "%"PRIdMAX and "%"PRIuMAX, and all uses 
of "%td" with "%"pT"d" where pT is an Emacs invention defined like this:

 #ifdef __ANDROID__
 # define pT "z"
 #else
 # define pT "t"
 #endif

That way, the Android branch wouldn't need to use the printf-posix and 
vasprintf-posix modules, and we wouldn't have to worry about the hassle 
of porting them into the Emacs world.

Admittedly this hack is not the Gnulib Way, but Emacs departs so far 
from the Gnulib Way that this one extra little thing shouldn't be that 
big of a deal.