Specifically, newer autoconf (> 2.13) has different semantic of the
configure target. In short, one should use --build=CONFIGURE_TARGET
instead of CONFIGURE_TARGET directly. Otherwise, you will get a warning
and the old semantic may be removed in later autoconf releases.
To workaround this issue, many ports hack the CONFIGURE_TARGET variable
so that it contains the ``--build='' prefix.
To solve this issue, under the fact that some ports still have
configure script generated by the old autoconf, we use runtime detection
in the do-configure target so that the proper argument can be used.
Changes to Mk/*:
- Add runtime detection magic in bsd.port.mk
- Remove CONFIGURE_TARGET hack in various bsd.*.mk
- USE_GNOME=gnometarget is now an no-op
Changes to individual ports, other than removing the CONFIGURE_TARGET hack:
= pkg-plist changed (due to the ugly CONFIGURE_TARGET prefix in * executables)
- comms/gnuradio
- science/abinit
- science/elmer-fem
- science/elmer-matc
- science/elmer-meshgen2d
- science/elmerfront
- science/elmerpost
= use x86_64 as ARCH
- devel/g-wrap
= other changes
- print/magicfilter
GNU_CONFIGURE -> HAS_CONFIGURE since it's not generated by autoconf
Total # of ports modified: 1,027
Total # of ports affected: ~7,000 (set GNU_CONFIGURE to yes)
PR: 126524 (obsoletes 52917)
Submitted by: rafan
Tested on: two pointyhat 7-amd64 exp runs (by pav)
Approved by: portmgr (pav)
We have not checked for this KEYWORD for a long time now, so this
is a complete noop, and thus no PORTREVISION bump. Removing it at
this point is mostly for pedantic reasons, and partly to avoid
perpetuating this anachronism by copy and paste to future scripts.
DNRD will always bind to all interfaces, even if the user specifically
says it should bind to only one, using the -a option.
PR: ports/89774
Submitted by: Natanael Copa <natanael.copa@gmail.com>
Version bump. This release fixes segfault in argument
parsing. Cleaned -h output (remved longopts when they are
unavailable) Improved signal handling and blacklist feature
added. Proper handling of paths in documentation.
I implemeted a pselect in lib.c for BSD 4.x support. Please
let me know if this could be handeled in a better way.
PR: ports/76060
Submitted by: Natanael Copa <ncopa@users.sourceforge.net>
This version introduce random source ports to
reduce the risk of getting the cache poisoned.
Assign maintainership to submitter/author.
PR: ports/73420
Submitted by: Natanael Copa
to prevent cache poisoning.
PR: ports/72522
Submitted by: Natanael Copa <ncopa@users.sourceforge.net>
While I'm here:
- Rework installation mechanism - use do-install target
- Respect NOPORTDOCS
- Install samples into EXAMPLESDIR
- Use DOCSDIR and EXAMPLESDIR macros in pkg-plist
Update of dnrd. The TCP support is totally broken in previous
versions. Even if it should work in this verion it is
disabled by default. (The code was so ugly). It can be
enabled with WITH_TCP.
In this version it is also possible to disable the MASTER
file support. So dnrd can be a compiled as proxy only.
WITHOUT_MASTER.
PR: ports/71770
Submitted by: Natanael Copa <ncopa@users.sourceforge.net>
There are a few bugs in dnrd that should probably be fixed
by the author but could be made to work on bsd:
1. Initialization in common.h of recv_addr is broken,
causing at least the '-a' switch not to work. Instead of
assuming positions of fields in the struct across platforms
I thought it safer to do a standard initialization in main().
2. The buffer overflow code in udp.c:dnsrecv() is off by
one, rejecting messages where the size exactly fills the
available buffer. I also changed to the calls to dnsrecv
to pass 512 as the max size instead of the buffers that
seem to be padded by 4 bytes for a reason I don't understand.
This causes a calling named to resend packets. Eventually
one seems to get through but response times can be painfully
slow.
PR: ports/41128
Submitted by: Michael C. Adler <mad1@tapil.com>
