HEADS UP: <utmp.h> gone. All welcome <utmpx.h>.
in [Ports]
|
From: Gary Jennejohn on 27 Jan 2010 04:52 On Tue, 26 Jan 2010 21:05:37 +0100 Ed Schouten <ed(a)80386.nl> wrote: > * Dirk Meyer <dirk.meyer(a)dinoex.sub.org> wrote: > > 1. POLA > > > > http://www.freebsd.org/doc/en_US.ISO8859-1/articles/committers-guide/rules.html > > > > $ man utmp > > does not mention that this interafce is deprectaed (see 11.4) > > First of all, utmp(5) should no longer exist. Second of all, you clearly > seem to misread section 11.4, which says "whenever possible." > > Instead of just screaming POLA, could you please explain how I could > have introduced utmpx into FreeBSD without causing any breakage or > confusion among our users. As someone else already quoted earlier, our > old utmp interface was a landmine we buried ourselves. > My understanding has always been that POLA doesn't apply to HEAD. If you want to use HEAD then learn to live with that. --- Gary Jennejohn _______________________________________________ freebsd-ports(a)freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscribe(a)freebsd.org"
From: Vinicius Abrahao on 1 Feb 2010 14:17 I found another port that shows a little error when replace utmp.h by utmpx.h. Is security/ipsec-tools Wno-unused -MT isakmp_cfg.o -MD -MP -MF .deps/isakmp_cfg.Tpo -c -o isakmp_cfg.o isakmp_cfg.c isakmp_cfg.c: In function 'isakmp_cfg_accounting_system': isakmp_cfg.c:1655: error: storage size of 'ut' isn't known isakmp_cfg.c:1656: error: 'UT_LINESIZE' undeclared (first use in this function) isakmp_cfg.c:1656: error: (Each undeclared identifier is reported only once isakmp_cfg.c:1656: error: for each function it appears in.) isakmp_cfg.c:1669: error: 'UT_NAMESIZE' undeclared (first use in this function) isakmp_cfg.c:1676: error: 'UT_HOSTSIZE' undeclared (first use in this function) cc1: warnings being treated as errors isakmp_cfg.c:1685: warning: implicit declaration of function 'login' isakmp_cfg.c:1694: warning: implicit declaration of function 'logout' *** Error code 1 Stop in /usr/ports/security/ipsec-tools/work/ipsec-tools-0.7.3/src/racoon. *** Error code 1 _______________________________________________ freebsd-ports(a)freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscribe(a)freebsd.org"
From: Ed Schouten on 2 Feb 2010 05:49 Hello Yvan, * VANHULLEBUS Yvan <vanhu(a)FreeBSD.org> wrote: > We're working on this, but as we have now to work with both utmp.h and > utmpx.h (at least for FreeBSD releases and FreeBSD CURRENT), we're > trying to find a clean way to solve the issue. I think the cleanest solution would be to split them off completely. Most projects do things like: #define utmp utmpx but this really makes the code harder to interpret. Just moving all the utmp stuff into a small set of routines that you port per operating system is probably the best way to go. Greetings, -- Ed Schouten <ed(a)80386.nl> WWW: http://80386.nl/
From: VANHULLEBUS Yvan on 2 Feb 2010 05:44 Hi. On Mon, Feb 01, 2010 at 05:17:18PM -0200, Vinicius Abrahao wrote: > I found another port that shows a little error when replace utmp.h by utmpx.h. > Is security/ipsec-tools Yes, this is already known. A workaround may be to compile without XAUTH/Mode Config, which is the only part of the code which uses utmp.h afaik. We're working on this, but as we have now to work with both utmp.h and utmpx.h (at least for FreeBSD releases and FreeBSD CURRENT), we're trying to find a clean way to solve the issue. Yvan. _______________________________________________ freebsd-ports(a)freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscribe(a)freebsd.org"
From: VANHULLEBUS Yvan on 2 Feb 2010 07:22
On Tue, Feb 02, 2010 at 11:49:06AM +0100, Ed Schouten wrote: > Hello Yvan, > > * VANHULLEBUS Yvan <vanhu(a)FreeBSD.org> wrote: > > We're working on this, but as we have now to work with both utmp.h and > > utmpx.h (at least for FreeBSD releases and FreeBSD CURRENT), we're > > trying to find a clean way to solve the issue. > > I think the cleanest solution would be to split them off completely. > Most projects do things like: Looks like all other implementations ipsec-tools runs on have both utmp.h and utmpx.h, so I am actually considering y just switch from utmp.h to utmpx.h (for example using the patchset sent yesterday on this list), and keep a reverse patch for FreeBSD's port if OS version is "old enough". As we just use utmp.h in a single part of the code, any other solution may be very complex for just that..... Yvan. _______________________________________________ freebsd-ports(a)freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscribe(a)freebsd.org" |