Busybox awk on 64-bit architectures?

[Mike Frysinger]

On Monday 26 September 2005 11:53 pm, Rob Landley wrote:
> On Monday 26 September 2005 22:15, Mike Frysinger wrote:
> > > C) awk is full of other "unsigned long"
> > > instances, D) this patch takes an assignment between "unsigned long"
> > > and "unsigned long" and sticks a typedef in there to something else.
> > > (There's no _way_ that's the correct fix to whatever the problem is...)
> >
> > the correct fix i believe would be to basically run like:
> > sed -i "s:unsigned long:uint32_t:" editors/awk.c
> >
> > /me tries it ...
> >
> > yep, provided test case starts working on my amd64
>
> Woot.  I assume you're checking the patch in yourself. :)

sure, i'll check it in if people feel it's correct ... it shouldnt break 32bit 
builds since sizeof(long) is 4 on 32bit hosts, and it's not like it could 
break anymore with 64bit hosts eh ? :)

> Note that it would be nice to have a beat-awk-to-death regression test set
> in the test suite, but I haven't got the skills to do more than poke at it,
> and have a to-do list of doom in other areas anyway.  Any clues where we
> might find one?

well i just ran `make check` using the testsuite from gawk-3.1.5 against a 
32bit and 64bit build with and without the uint32 change ... 32bit showed no 
change in passes/fails, but the new 64bit one showed failed less tests ;)

other than that, i dont know of any test suites out there ... we could always 
use the gawk as a base though; copy out all the tests that pass and figure 
out which of the failures are ones we care about ...
-mike