On Sat, Feb 22, 2003 at 01:57:24AM +0200, Tuomo Valkonen wrote:
> On Fri, Feb 21, 2003 at 11:37:49PM +0100, Per Olofsson wrote:
> > * libc is full of functions. A library covering all of them will not
> >   necessary be minimal. (also large)
> 
> Most libc functions are rather standardized, AFAIK. Correct me if I'm
> wrong. The *nix stuff is the problem.

It is libc that contains the *nix stuff (not always though, sockets
are in libsocket on Solaris), and that's what I meant.

> The problem is that autoconf was written instead of a better solution
> that would remove most of the problems instead just hiding them from
> the user. The programmer's task isn't any easier than it was, on the
> contrary. Instead of worrying of differences between *nix variants and
> such, he now has to worry of how to check for those too and if the script
> will work on all platforms.
>
> > Still, I don't see any reasons to blame Autoconf for everything. I
> > don't think Autoconf is unreliable, I would say it is the input it
> > gets that is bad then. You should blame the packages that use
> > Autoconf, not Autoconf itself.
> 
> Of course. There can be bad Makefile-based configurations too. They're
> just easier to fix and less things that can go wrong than with autoconf.
> 

That's not my opinion. I don't think it's much harder to fix autoconf
scripts. And if I would compare two packages, who both needs to
collect a large amount of data for the build process, but where one
uses autoconf and the other has a manual makefile configuration, I
would prefer the package that uses autoconf. I don't want to enter all
that data manually. And I don't think the amount of data increased
with autoconf. Of course, if the makefile-style package had all the
defaults tuned for Linux + glibc, it would be much easier to build
that package, if you happen to be using Linux + glibc.

But I agree that using a library would be a much better solution. The
build times would be shorter, it would be easier to write portable
code, there's a lower risk of problems occuring, etc. But does that
remove the need for configure scripts? Not really, the library would
still need to check for everything, even if only done once. And each
package that uses the library would have to check if the library
exists, and also for any other libraries the package depends on. Of
course, you could have a system.mk like Ion, but I prefer having
things checked automatically.

/Pelle

Reply via email to