There's no promise that imap-2004g will work in a 64-bit build. I
generally recommend that people build it in 32-bit mode until further
notice. It certainly will NOT behave reasonably if off_t and/or size_t is
a different size than long.
In particular, compilers which use "long long" for 64-bit values will
probably not build the code reasonably in 64-bit mode. It may work if the
compiler makes "long" and "unsigned long" be 64 bit along with off_t and
size_t.
I've never encountered an HP-UX 11i, and our last HP-UX system was shut
down some time ago. So I have little/no idea what may have been done
after HP-UX 10.
The non-standard location of the openssl files is best dealt with in the
top-level makefile, rather than changing the low-level makefile. Take a
look at how it is done for lrh, oxp, etc (search for "Knotheads"...)
The removal of -lnet and -lV3 is probably best done with spawning a
different port.
Many of these warnings are complaining about pointer argument passing or
assignment with different signs. In none of these cases does it actually
matter. There should be a C compiler flag to tell it to stop checking.
In gcc4 (which added a similar check), it is -Wno-pointer-sign.
That compiler warning is quite annoying, since it forces one to:
. write double the number of string routines, with one version that does
"char *" and the other than does "unsigned char *"
or
. sprinkle casts throughout the code for the majority of occurences where
it absolutely does not matter what the case is
or
. find the compiler option to turn the behavior off.
-- Mark --
http://staff.washington.edu/mrc
Science does not emerge from voting, party politics, or public debate.
Si vis pacem, para bellum.
_______________________________________________
Imap-uw mailing list
[email protected]
https://mailman1.u.washington.edu/mailman/listinfo/imap-uw
