This is the mail archive of the
mailing list for the GCC project.
Re: committed: PR ada/14350 fix
Paolo Bonzini <email@example.com> writes:
> Easier to work on the top of the existing autoconf and make it more
> sensible (less special cases, etc.). But the work is not small and is
> not as necessary as it would seem -- I'm afraid it will not be done
> until someone steps in who has lots of free time (or is paid for it),
> is a shell and m4 guru, and is willing to see his work bashed for a
> couple o' years before it is at least mildly accepted. That's what
> happened for autoconf 2.5x.
I would say that that is because, in my opinion, some poor decisions
were made for autoconf 2.5x, starting with not calling it autoconf 3.
Not that I'm going to spend a lot of time defending autoconf. But my
feeling is that when it was a simple collection of relatively
independent feature test macros, it was OK to write it in sh and m4.
Now that it has become a monster program with all sorts of
dependencies between the macros, sh and m4 are a poor implementation