variable which is de-facto standard for MUAs.
Teach bomail to generate an in-reply-to header so threading MUAs and
mail->news gateways won't lose context.
While i was at it, removed two gratuitous standard violations for
functions starting with an underscore.
counted.
* re-word parts of the man page which I felt were badly worded
or ambiguous.
* change the behaviour of argument processing so that when more
than one of the -P, -H and -L options are specified it will
print an error message, rather than choosing the last option
specified, this behaviour is more logical and consistent with
other utilities.
* change the behaviour of argument processing so that negative
arguments to the -d option are not allowed.
PR: 5388
Submitted by: Niall Smart <rotel@indigo.ie>
Don't override the (correct) default for SRCS or MAN1.
Replaced bogus Lite1 (?) Id by $Id$. The key server files were
mis-imported so it is hard to tell where they came from, but at
least in Lite2 they don't have useful Makefiles, and I rewrote
most of the Makefile.
linkage rule is overridden. The -L option must be in ${DPADD} so that
`make checkdpadd' works. Actually use ${DPADD}.
FIxed missing dependencies for doscmd. Use ${LIBCRT0} instead of a
hard-coded path in the rule for doscmd.
Added comments about the kludges used to build 2 binaries and 2 data
files in one directory. It shouldn't be done this way. The dependencies
on sources took extra work to get right, and the dependencies on objects
are still broken (one set is missing and the other has the wrong libs).
Fixed some style bugs while I'm here:
- don't override the (correct) default for MAN1.
- use `beforeinstall', not `afterinstall' to install auxiliary files.
`afterinstall' is for fixing messes made by `install'.
change from
ioctl(fd, PIOC<foo>, &i);
to
ioctl(fd, PIOC<foo>, i);
This is going from the _IOW to _IO ioctl macro. The kernel, procctl, and
truss must be in synch for it all to work (not doing so will get errors about
inappropriate ioctl's, fortunately). Hopefully I didn't forget anything :).
said process will not have its event mask cleared (and be restarted) on
the last close of a procfs/mem file for that pid. This reduces the chance
that a truss-monitored process will be left hanging with these bits set
and nobody looking for it.
This is the least-tested change of all of these, I'm afraid.
