Re: API levels and functions on Android

[Bruno Haible]

Paul Eggert wrote:
> > Defining REPLACE_GLOB=1 despite HAVE_GLOB=0 will be the way to make Gnulib
> > define 'rpl_glob' instead of 'glob', so that there is no conflict with 
> > libc.so.
> > 
> > I still need to work out the details how this would/will work. So as to
> > do things correctly on Android, while OTOH not adding lots of configure 
> > tests
> > for the other platforms...
> 
> Thanks for looking into this.
> 
> On Android, should we always define REPLACE_WHATEVER when 'whatever' 
> doesn't exist, at least for all the REPLACE_WHATEVER instances that 
> currently exist in Gnulib? After all, 'whatever' might exist in some 
> future Android version.

I think this would be overkill. Remember, for debuggability, when Gnulib
defines a function 'foo', we want gdb to know it by the name 'foo', if
there is no imperative to name it differently.

The two known imperatives are
  - call it 'rpl_foo', to avoid a clash with libc,
  - when shipping a shared library libgaga, call it 'libgaga_foo' or 'gaga_foo',
    to avoid a clash with other shared libraries.
If none of these two imperatives apply, the name 'foo' without any prefix is
the natural choice.

> Would it make sense to wait until you've worked out the details, before 
> installing patches like the REPLACE_UTIMENSAT=1 patch?

Yes, definitely.

Bruno