>>>> "Colin" == Colin Rafferty
<craffert(a)ms.com> writes:
Colin> Then it's just a warning.
Colin> Of course, I hate warnings, and I try to rid them from all code that I
Colin> see. However, sometimes I just feel as though it's a losing battle
Colin> against entropy. :-(
I have that feeling too sometimes, but the XEmacs sources are
extremely clean now, warning-wise.
It would be good for beta testers to try to understand the warnings,
and submit a patch if it is clear what the problem is, and why they
would get the warning while I don't.
As of 21.2.5, I have checked in all the warning-removal patches, and
there are ZERO warnings in src/ using these compilers[1]:
egcs-1.1.1 -c -g -Wall -Wpointer-arith -Winline -Wmissing-prototypes
-Wmissing-declarations -Wunused -Wformat -Wno-switch -Wno-sign-compare -Wshadow
Sun cc -v
Martin
P.S. Alright, I admit there are still these warnings in png.h that
cannot be fixed (by us):
In file included from /home/martin/x/dev/src/glyphs-eimage.c:66:
/usr/include/png.h:1204: warning: declaration of `error' shadows global declaration
/usr/include/png.h:1208: warning: declaration of `error' shadows global declaration
/usr/include/png.h:1212: warning: declaration of `message' shadows global declaration
/usr/include/png.h:1216: warning: declaration of `message' shadows global declaration
P.P.S. There's also this bug in egcs-1.1.1 that erroneously prints:
/home/martin/x/dev/src/events.c: In function `print_event':
/home/martin/x/dev/src/events.c:182: warning: long int format, int arg (arg 3)
/home/martin/x/dev/src/events.c:182: warning: long int format, int arg (arg 4)