o Some fix (NULL != NUL, void main -> int main).
o Update archive md5.
some bug fix in new archive(this was released after 19 hours than old one).
Submitted by: Yoshihiko - he is in seat next mine in jammed
Chuou Highway - Sarumaru <mistral@imasy.or.jp> (maintainer)
Committed at: jammed Chuou Highway in Japan
* Add code for formal master/slave checking rules as defined in ports/64393 [1]:
1. A port is a slave port if and only if MASTERDIR != .CURDIR
2. Slave ports should define MASTERDIR using one of the following lines:
MASTERDIR= ${.CURDIR}/../../<category>/<port>
MASTERDIR= ${.CURDIR}/../<port>
3. Non-slave ports shouldn't define MASTERDIR at all
4. The last line of a slave port's Makefile has to be
.include "${MASTERDIR}/Makefile"
5. The last line of a non-slave ports Makefile must be one of:
.include <bsd.port.mk>
.include <bsd.port.post.mk>
6. slave ports may not include bsd.port(.pre).mk
* Check to make sure MACHINE_ARCH is not defined and make sure it is never
tested directly [2]
* Fix some grammar nits [3]
PR: 64420 [1]
Submitted by: eik [1]
krion [3]
Requested by: kris [2]
In fact, the message was killed in old startup script. While converting
to the script for RCng era, I just forget to do that. I don't know that
"command_args" variable is ok in this case , but /etc/rc.d/ipxrouted
and /etc/rc.d/ntpdate already did it so it seems there's no problem.
Since the script is changed, bump PORTREVISION.
Noticed by: Alan Barber <abarber _at_ alanbarber.org>
Pointy hat to: matusita
2) Add WITHOUT_POSTPROCESS knob for backward compatibility. However,
using might break ports that require postprocess to be enabled
in ffmpeg
3) Bump PORTREVISION due to [1]
4) Remove PKGMESSAGE since it is no longer necessary to warn about
ffmpeg045 port. It has been removed since the reasons for
maintaining an old parallel version of ffmpeg no longer exist
PR: 64257 [1,2,3]
Submitted by: dinoex
variable for replacing includes as well. Users who replaced the base
BIND and then attempted to compile other software with -D_REENTRANT
were having problems because netdb.h has a different definition for
h_errno.
The idea of replacing includes originated with users who wanted to
compile apps against the libbind installed by the port, however those
"power users" can either add the switch to replace includes, or modify
their own includes slightly to use the "safe" location for the port
include files (PREFIX/include/bind).
The PR mentions the problem, along with a cast of thousands. :)
Other tuneups for the port are just portlint appeasements.
PR: ports/64449
Submitted by: eik
