Re: optionsng and tinderbox?
On Tue, Jun 19, 2012 at 10:02:21PM -0400, Michael Scheidell wrote:
I have my favorite options set in tinderbox with
/tb/scripts/etc/env/build*
things like:
export NOPORTDOCS=true
export NOPORTEXAMPLES=true
(which seems to translate over just fine)
but, on clamav I need: WITH_ICONV=yes
Now that we have options ng WITHOUT_/WITH_ are gone because inconsistent.
User needs to do:
OPTIONS_SET=ICONV
To get iconv activated everywhere
OPTIONS_UNSET= ICONV
To get iconv deactivated everywhere
nginx_UNSET=ICONV
will deactivate ICONV for nginx only
and on nginx I need WITHOUT_ICONV=yes
(it used to be WITHOUT_ICONV_MODULE, which didn't conflict, but now does)
anyone understand how this works in the ../script/etc/env/build* files?
(if you don't, please don't speculate!)
Q:
pretend this is a /etc/make.conf (that doesn't seem to honor if
${.CURDIR:M*www/nginx} )
Because it doesn't exists anymore. yes it honors it but nginx has been converted
to optionsng so it doesn't check anymore for the old inconsistent WITH_/WITHOUT
so, other than if ${.CURDIR..) how can I set options in 'make.conf'?
if I do it this way, and the maintainer has not converted to optionsng,
are they honored?
http://wiki.freebsd.org/Ports/Options/OptionsNG#What_users_need_to_know
(I need WITH_MYSQL=yes in postfix/barnyard2, but don't need it in
snort.. I put up with it in snort, but, since WITH_ICONV means
something different in clamav and nginx, I want to set them this way)
oh, this is a shared tinderbox, I can't see these globally, but have to
do them by build* files.
--
Michael Scheidell, CTO
*| * SECNAP Network Security Corporation
d: +1.561.948.2259
w: http://people.freebsd.org/~scheidell
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
pgp72qeN5MZLI.pgp
Description: PGP signature
Re: Unable to build graphics/ImageMagick port
On Tue, Jun 19, 2012 at 9:59 PM, Simon Wright simon.wri...@gmx.net wrote: On 20/06/12 06:22, Kevin Oberman wrote: For this to work, install ImageMagick WITH HDRI, The test that fails is for HDRI and attempts to use files that don't yet exist. Thus analyze.sh failes. To install ImageMagick with all default options, turn on HDRI (High Dynamic Range), a very desirable capability for many but turn off TESTS. Build and INSTALL ImageMagick. That puts all of the needed files in place. Then turn on TESTS and re-buildand ImageMagick.It will then pass all tests. On i386 with a ports tree from this morning, with TESTS disabled ImageMagick builds and installs. Re-enabling TESTS and rebuilding fails the test averageImages.sh. YMMV of course! Sorry.I have never had averageImages.sh fail on any install on i386 or amd64. It has always worked out of the box. All builds using gcc. I have built it with both threaded and unthreaded perl. Perl 5.12 and 5.14 have both worked. you can take a look at the test and the log of its run in work/ImageMagick-6.7.7-7/Magick++/tests. averageImages.log si the output and averageImages.cpp is the test source. It's about 73 lines of CPP -- R. Kevin Oberman, Network Engineer E-mail: kob6...@gmail.com ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: optionsng and tinderbox?
On 6/20/12 2:11 AM, Baptiste Daroussin wrote:
nginx_UNSET=ICONV
will deactivate ICONV for nginx only
lowercase?
wiki seems to indicate UPPERCASE:
${UNIQUENAME:U}_SET=OPT1
${UNIQUENAME:U}_UNSET= OPT2
and, it seems that uppercase didn't work:
export NGINX_SET=FILE_AIO
but, in tinderlog:
grep FILE_AIO
NGINX_SET=FILE_AIO
FILE_AIO=off: Enable file aio
so, I edit /tb/scripts/etc/env/build* to:
export nginx_SET=FILE_AIO
but, tinderlog still shows UPPERCASE:
grep FILE_AIO
NGINX_SET=FILE_AIO
FILE_AIO=off: Enable file aio
old way still works (but potential conflicts as you know,)
(if in /tb/scripts/etc/env/build* I have this:
export WITH_FILE_AIO=yes
I get this in tinderlog:
WITH_FILE_AIO=yes
FILE_AIO=on: Enable file aio
So, is it supposed to be export NGINX_SET or nginx_SET?
and is it a bug in tinderbox that forces uppercase? (which seems strange
it would do that for env vars)
http://people.freebsd.org/~scheidell/nginx_SET.log
(there are lowercase env's in this log, so I don't think its tinderbox)
--
Michael Scheidell, CTO
*| * SECNAP Network Security Corporation
d: +1.561.948.2259
w: http://people.freebsd.org/~scheidell
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Horde4 eats all memory
Hi, Just checking if anyone on the list is experiencing a similar problem. We have a 9-STABLE amd64 test system with H4 that since sometime in May started to exhaust all memory, it *probably* coincides with upgrade to PHP 5.4 but not 100% sure. We are in the process of learning how to debug PHP and if someone has a good advice on this please share. I suspect that the problem is BSD-specific because no one on the Horde lists seem to have a similar issue on Linux. Have continuously updated the system and ports since problem started, right now we're on 9.0-STABLE #0: Wed Jun 13 05:23:35 CEST 2012 horde-4.0.15 php5-5.4.3 Thanks, ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: clang, dbus, firefox+thunderbird
Hi All, Andrew, thanks for your time and help! 19.06.2012 21:14, Andrew W. Nosenko написал: On Tue, Jun 19, 2012 at 6:16 PM, Boris Samorodovb...@passap.ru wrote: Hi List, I use clang compiled world (WITH_CLANG_IS_CC=YES and WITH_LIBCPLUSPLUS=YES at /etc/make.conf). The system is current: - % uname -a FreeBSD bsam.wart.ru 10.0-CURRENT FreeBSD 10.0-CURRENT #16 r237055: Thu Jun 14 17:16:43 SAMT 2012 b...@bsam.wart.ru:/usr/obj/usr/src/sys/BBX i386 - While updating to the latest firefox I get an error: - /usr/local/include/dbus-1.0/dbus/dbus-protocol.h:459:77: error: invalid suffix on literal; C++11 requires a space between literal and identifier [-Wreserved-user-defined-literal] #define DBUS_INTROSPECT_1_0_XML_DOCTYPE_DECL_NODE !DOCTYPE node PUBLIC \DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER\\n\DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER\\n ^ /usr/local/include/dbus-1.0/dbus/dbus-protocol.h:459:126: error: invalid suffix on literal; C++11 requires a space between literal and identifier [-Wreserved-user-defined-literal] #define DBUS_INTROSPECT_1_0_XML_DOCTYPE_DECL_NODE !DOCTYPE node PUBLIC \DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER\\n\DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER\\n ^ 2 errors generated. - OK, I did make config and deselect DBUS. Great, firefox compiles and installs (the latest one, 13.0.1). Then I try to upgrade thunderbird and get the same error. However there is no option DBUS to disable. What am I to do? Thanks! It's because of User defined literals (the new c++11 feature) and your clang++ is in c++11 (or gnu++11?) mode by default. As workaround, you may try either: downgrade to something like gnu++ 98 (-std=gnu++98), or edit dbus-protocol.h header for make it c++11 compatible. It's easy: just add spaces around DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER and DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER literals in the line 459 (just as compiler suggested). I.e. the line 459 #define DBUS_INTROSPECT_1_0_XML_DOCTYPE_DECL_NODE !DOCTYPE node PUBLIC \DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER\\n\DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER\\n should be written as #define DBUS_INTROSPECT_1_0_XML_DOCTYPE_DECL_NODE !DOCTYPE node PUBLIC \ DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER \\n\ DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER \\n You are right. And AFAIK a similar fix was committed to the ports tree while I was testing your suggestion. The fix helped clang to build firefox with DBUS on. However, thunderbird stopped building with another error. I'll create a new thread on the matter after some testing. -- WBR, Boris Samorodov (bsam) FreeBSD Committer, http://www.FreeBSD.org The Power To Serve ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: clang, dbus, firefox+thunderbird
2012/6/20 Boris Samorodov b...@passap.ru: Hi All, Andrew, thanks for your time and help! 19.06.2012 21:14, Andrew W. Nosenko написал: On Tue, Jun 19, 2012 at 6:16 PM, Boris Samorodovb...@passap.ru wrote: Hi List, I use clang compiled world (WITH_CLANG_IS_CC=YES and WITH_LIBCPLUSPLUS=YES at /etc/make.conf). The system is current: - % uname -a FreeBSD bsam.wart.ru 10.0-CURRENT FreeBSD 10.0-CURRENT #16 r237055: Thu Jun 14 17:16:43 SAMT 2012 b...@bsam.wart.ru:/usr/obj/usr/src/sys/BBX i386 - While updating to the latest firefox I get an error: - /usr/local/include/dbus-1.0/dbus/dbus-protocol.h:459:77: error: invalid suffix on literal; C++11 requires a space between literal and identifier [-Wreserved-user-defined-literal] #define DBUS_INTROSPECT_1_0_XML_DOCTYPE_DECL_NODE !DOCTYPE node PUBLIC \DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER\\n\DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER\\n ^ /usr/local/include/dbus-1.0/dbus/dbus-protocol.h:459:126: error: invalid suffix on literal; C++11 requires a space between literal and identifier [-Wreserved-user-defined-literal] #define DBUS_INTROSPECT_1_0_XML_DOCTYPE_DECL_NODE !DOCTYPE node PUBLIC \DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER\\n\DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER\\n ^ 2 errors generated. - OK, I did make config and deselect DBUS. Great, firefox compiles and installs (the latest one, 13.0.1). Then I try to upgrade thunderbird and get the same error. However there is no option DBUS to disable. What am I to do? Thanks! It's because of User defined literals (the new c++11 feature) and your clang++ is in c++11 (or gnu++11?) mode by default. As workaround, you may try either: downgrade to something like gnu++ 98 (-std=gnu++98), or edit dbus-protocol.h header for make it c++11 compatible. It's easy: just add spaces around DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER and DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER literals in the line 459 (just as compiler suggested). I.e. the line 459 #define DBUS_INTROSPECT_1_0_XML_DOCTYPE_DECL_NODE !DOCTYPE node PUBLIC \DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER\\n\DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER\\n should be written as #define DBUS_INTROSPECT_1_0_XML_DOCTYPE_DECL_NODE !DOCTYPE node PUBLIC \ DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER \\n\ DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER \\n You are right. And AFAIK a similar fix was committed to the ports tree while I was testing your suggestion. The fix helped clang to build firefox with DBUS on. However, thunderbird stopped building with another error. I'll create a new thread on the matter after some testing. There are patches to build Thunderbird with clang: http://lists.freebsd.org/pipermail/freebsd-gecko/2012-June/002273.html Maybe you could give them a try. Beat ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: clang, dbus, firefox+thunderbird
On 2012-06-19 19:14, Andrew W. Nosenko wrote: On Tue, Jun 19, 2012 at 6:16 PM, Boris Samorodov b...@passap.ru wrote: ... /usr/local/include/dbus-1.0/dbus/dbus-protocol.h:459:126: error: invalid suffix on literal; C++11 requires a space between literal and identifier [-Wreserved-user-defined-literal] #define DBUS_INTROSPECT_1_0_XML_DOCTYPE_DECL_NODE !DOCTYPE node PUBLIC \DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER\\n\DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER\\n ... It's because of User defined literals (the new c++11 feature) and your clang++ is in c++11 (or gnu++11?) mode by default. Note, the version of clang in base defaults to gnu++98, *not* c++11 or gnu++11. Maybe this is added by the configure script, after detection of c++11 support? ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: png-1.5.11 status?
Unless the shareable version number is bumped, there is no need to re-build dependent ports. Most updates to shareables do not change the API or ABI, so dependent packages simply pull in the updated shareable when they load. You should only have to do the mass update once, whether for 1.5.10 or 1.5.11. I would suggest using the '-w' option if you use portmaster as the old shareable was libpng and the new is libpng15. If you keep a copy of the old one around, many things that have not been re-linked will still run OK. Note that when you start re-building stuff, you will certainly hit cases when you link to a library linked to one libpng and either the port being installed or another dependency is linked to libpng15. I'm unsure exactly what impact that might have. It really should have none, but I am not really sure. When you are finished re-building (or think that you are), I strongly recommend running pkg_libchk -o | grep png to make sure that nothing has been missed. pkg_libchk is part of the sysutils/bsdadminscripts port. R. Kevin Oberman, Network Engineer Thanks for information. Since the portmaster.rc.sample recommends it and I don't want to worry about forgetting -w, my /usr/local/etc/portmaster.rc reads MAKE_PACKAGE=gopt SAVE_SHARED=wopt PM_LOG=/var/log/portm.log I hope I have all the correct syntax. I never (yet) installed sysutils/bsdadminscripts but have been persuaded by reading list messages that it is too useful to let go by. Tom ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: optionsng and tinderbox?
On Wed, Jun 20, 2012 at 03:58:21AM -0400, Michael Scheidell wrote:
On 6/20/12 2:11 AM, Baptiste Daroussin wrote:
nginx_UNSET=ICONV
will deactivate ICONV for nginx only
lowercase?
Yes lower case, the wiki is wrong.
make -VUNIQUENAME should show you the right name
wiki seems to indicate UPPERCASE:
${UNIQUENAME:U}_SET=OPT1
${UNIQUENAME:U}_UNSET= OPT2
and, it seems that uppercase didn't work:
export NGINX_SET=FILE_AIO
but, in tinderlog:
grep FILE_AIO
NGINX_SET=FILE_AIO
FILE_AIO=off: Enable file aio
so, I edit /tb/scripts/etc/env/build* to:
export nginx_SET=FILE_AIO
but, tinderlog still shows UPPERCASE:
grep FILE_AIO
NGINX_SET=FILE_AIO
FILE_AIO=off: Enable file aio
old way still works (but potential conflicts as you know,)
(if in /tb/scripts/etc/env/build* I have this:
export WITH_FILE_AIO=yes
I get this in tinderlog:
WITH_FILE_AIO=yes
FILE_AIO=on: Enable file aio
So, is it supposed to be export NGINX_SET or nginx_SET?
nginx_SET
Which is see correctly in your log.
regards,
Bapt
pgpk5v6Snd0wM.pgp
Description: PGP signature
Re: Unable to build graphics/ImageMagick port
On Tue, 19 Jun 2012 21:22:51 -0700 Kevin Oberman articulated: To install ImageMagick with all default options, turn on HDRI (High Dynamic Range), a very desirable capability for many but turn off TESTS. Build and INSTALL ImageMagick. That puts all of the needed files in place. Then turn on TESTS and re-buildand ImageMagick.It will then pass all tests. Sorry, been there -- done that -- didn't work. Please reread my previous post. This is on a FreeBSD-8.3 STABLE amd64 machine. -- Carmel ✌ carmel...@hotmail.com ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: clang, dbus, firefox+thunderbird
On Wed, Jun 20, 2012 at 11:54 AM, Dimitry Andric d...@freebsd.org wrote: On 2012-06-19 19:14, Andrew W. Nosenko wrote: On Tue, Jun 19, 2012 at 6:16 PM, Boris Samorodov b...@passap.ru wrote: ... /usr/local/include/dbus-1.0/dbus/dbus-protocol.h:459:126: error: invalid suffix on literal; C++11 requires a space between literal and identifier [-Wreserved-user-defined-literal] #define DBUS_INTROSPECT_1_0_XML_DOCTYPE_DECL_NODE !DOCTYPE node PUBLIC \DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER\\n\DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER\\n ... It's because of User defined literals (the new c++11 feature) and your clang++ is in c++11 (or gnu++11?) mode by default. Note, the version of clang in base defaults to gnu++98, *not* c++11 or gnu++11. Maybe this is added by the configure script, after detection of c++11 support? Another posibility: message about C++11 user defined literals is just a warning intended for forward compatibility, but promoted to error because of -Werror. But it's just a guess, I'm didn't tried to build firefox or thunderbird by self. -- Andrew W. Nosenko andrew.w.nose...@gmail.com ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: optionsng and tinderbox?
On 6/20/12 5:13 AM, Baptiste Daroussin wrote: nginx_SET Which is see correctly in your log. regards, Bapt strange! when I first looked at log, it was all uppercase. Ok thanks. (browser cache?) -- Michael Scheidell, CTO *| * SECNAP Network Security Corporation d: +1.561.948.2259 w: http://people.freebsd.org/~scheidell ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: clang, dbus, firefox+thunderbird
On 2012-06-19 19:14, Andrew W. Nosenko wrote: On Tue, Jun 19, 2012 at 6:16 PM, Boris Samorodov b...@passap.ru wrote: ... /usr/local/include/dbus-1.0/dbus/dbus-protocol.h:459:77: error: invalid suffix on literal; C++11 requires a space between literal and identifier [-Wreserved-user-defined-literal] #define DBUS_INTROSPECT_1_0_XML_DOCTYPE_DECL_NODE !DOCTYPE node PUBLIC \DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER\\n\DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER\\n ... As workaround, you may try either: downgrade to something like gnu++ 98 (-std=gnu++98), or edit dbus-protocol.h header for make it c++11 compatible. It's easy: just add spaces around DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER and DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER literals in the line 459 (just as compiler suggested). Yes, this is exactly how upstream has fixed it: http://cgit.freedesktop.org/dbus/dbus/commit/dbus/dbus-protocol.h?id=51b88b4c7919487290c0862b013cd8e7cd2de34b ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: optionsng and tinderbox?
On 6/20/12 5:13 AM, Baptiste Daroussin wrote:
Yes lower case, the wiki is wrong.
make -VUNIQUENAME should show you the right name
wiki seems to indicate UPPERCASE:
${UNIQUENAME:U}_SET=OPT1
${UNIQUENAME:U}_UNSET= OPT2
and, lastly:
nginx_SET=FILE_AIO HTTP_GZIP_STATIC HTTP HTTP_CACHE
(for wiki:
${UNIQUENAME:M}_SET OPT1 OPT2
or
${UNIQUENAME:M}_SET OPT1,OPT2
spaces or commas?
(and for googlers, if using tinderbox:
export nginx_SET=FILE_AIO\ HTTP_GZIP_STATIC\ HTTP HTTP_CACHE
(escape the space with \ to force it from being mucked up)
--
Michael Scheidell, CTO
*| * SECNAP Network Security Corporation
d: +1.561.948.2259
w: http://people.freebsd.org/~scheidell
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: optionsng and tinderbox?
On Wed, 20 Jun 2012 11:13:21 +0200, Baptiste Daroussin wrote Hi Baptiste, make -VUNIQUENAME should show you the right name I have a port where UNIQUENAME is *not* the right prefix to use before '_SET' for positioning this specific port's options : $ pwd /usr/ports/net/openldap24-client $ make -V UNIQUENAME openldap24-client $ make -V PORT_OPTIONS DOCS EXAMPLES NLS $ make openldap24-client_SET=FETCH -V PORT_OPTIONS DOCS EXAMPLES NLS but : $ make openldap24-server_SET=FETCH -V PORT_OPTIONS DOCS EXAMPLES FETCH NLS So, here, I have to set the following in make.conf : openldap24-server_SET+=FETCH and not : openldap24-client_SET+=FETCH which would not work. Are you aware of that specific problem ? Best regards, -- Ganael LAPLANCHE ganael.laplan...@martymac.org http://www.martymac.org | http://contribs.martymac.org FreeBSD: martymac marty...@freebsd.org, http://www.FreeBSD.org ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: clang, dbus, firefox+thunderbird
20.06.2012 12:45, Beat Gaetzi пишет: 2012/6/20 Boris Samorodov b...@passap.ru: Hi All, Andrew, thanks for your time and help! 19.06.2012 21:14, Andrew W. Nosenko написал: On Tue, Jun 19, 2012 at 6:16 PM, Boris Samorodovb...@passap.ru wrote: Hi List, I use clang compiled world (WITH_CLANG_IS_CC=YES and WITH_LIBCPLUSPLUS=YES at /etc/make.conf). The system is current: - % uname -a FreeBSD bsam.wart.ru 10.0-CURRENT FreeBSD 10.0-CURRENT #16 r237055: Thu Jun 14 17:16:43 SAMT 2012 b...@bsam.wart.ru:/usr/obj/usr/src/sys/BBX i386 - While updating to the latest firefox I get an error: - /usr/local/include/dbus-1.0/dbus/dbus-protocol.h:459:77: error: invalid suffix on literal; C++11 requires a space between literal and identifier [-Wreserved-user-defined-literal] #define DBUS_INTROSPECT_1_0_XML_DOCTYPE_DECL_NODE !DOCTYPE node PUBLIC \DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER\\n\DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER\\n ^ /usr/local/include/dbus-1.0/dbus/dbus-protocol.h:459:126: error: invalid suffix on literal; C++11 requires a space between literal and identifier [-Wreserved-user-defined-literal] #define DBUS_INTROSPECT_1_0_XML_DOCTYPE_DECL_NODE !DOCTYPE node PUBLIC \DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER\\n\DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER\\n ^ 2 errors generated. - OK, I did make config and deselect DBUS. Great, firefox compiles and installs (the latest one, 13.0.1). Then I try to upgrade thunderbird and get the same error. However there is no option DBUS to disable. What am I to do? Thanks! It's because of User defined literals (the new c++11 feature) and your clang++ is in c++11 (or gnu++11?) mode by default. As workaround, you may try either: downgrade to something like gnu++ 98 (-std=gnu++98), or edit dbus-protocol.h header for make it c++11 compatible. It's easy: just add spaces around DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER and DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER literals in the line 459 (just as compiler suggested). I.e. the line 459 #define DBUS_INTROSPECT_1_0_XML_DOCTYPE_DECL_NODE !DOCTYPE node PUBLIC \DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER\\n\DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER\\n should be written as #define DBUS_INTROSPECT_1_0_XML_DOCTYPE_DECL_NODE !DOCTYPE node PUBLIC \ DBUS_INTROSPECT_1_0_XML_PUBLIC_IDENTIFIER \\n\ DBUS_INTROSPECT_1_0_XML_SYSTEM_IDENTIFIER \\n You are right. And AFAIK a similar fix was committed to the ports tree while I was testing your suggestion. The fix helped clang to build firefox with DBUS on. However, thunderbird stopped building with another error. I'll create a new thread on the matter after some testing. There are patches to build Thunderbird with clang: http://lists.freebsd.org/pipermail/freebsd-gecko/2012-June/002273.html Maybe you could give them a try. Confirmed. It works for thunderbird 13.0.1. Thanks! -- WBR, Boris Samorodov (bsam) FreeBSD Committer, http://www.FreeBSD.org The Power To Serve ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: optionsng and tinderbox?
On Wed, Jun 20, 2012 at 09:54:45AM +, Ganael LAPLANCHE wrote: On Wed, 20 Jun 2012 11:13:21 +0200, Baptiste Daroussin wrote Hi Baptiste, make -VUNIQUENAME should show you the right name I have a port where UNIQUENAME is *not* the right prefix to use before '_SET' for positioning this specific port's options : $ pwd /usr/ports/net/openldap24-client $ make -V UNIQUENAME openldap24-client $ make -V PORT_OPTIONS DOCS EXAMPLES NLS $ make openldap24-client_SET=FETCH -V PORT_OPTIONS DOCS EXAMPLES NLS but : $ make openldap24-server_SET=FETCH -V PORT_OPTIONS DOCS EXAMPLES FETCH NLS So, here, I have to set the following in make.conf : openldap24-server_SET+=FETCH and not : openldap24-client_SET+=FETCH which would not work. Are you aware of that specific problem ? Best regards, -- Ganael LAPLANCHE ganael.laplan...@martymac.org http://www.martymac.org | http://contribs.martymac.org FreeBSD: martymac marty...@freebsd.org, http://www.FreeBSD.org This is about to be fixed by the the uniquename being really unique (see matthew@) CFT. regards, Bapt pgprOlT5LCSEW.pgp Description: PGP signature
print/ghostscript9 build fail : vga.h not found
sudo make MAKE_ARGS= cc -DHAVE_MKSTEMP -DHAVE_FONTCONFIG -DHAVE_LIBIDN -DHAVE_SETLOCALE -DHAVE_SSE2 -DHAVE_DBUS -O2 -pipe -fno-strict-aliasing -fPIC -DUPD_SIGNAL=0 -I. -I/usr/ports/print/ghostscript9/work/ghostscript-9.05/lcms/include -I/usr/local/include/libpng -I/usr/ports/print/ghostscript9/work/ghostscript-9.05/openjpeg/libopenjpeg -I/usr/ports/print/ghostscript9/work/ghostscript-9.05/openjpeg -I/usr/local/include -Wall -Wstrict-prototypes -Wundef -Wmissing-declarations -Wmissing-prototypes -Wwrite-strings -Wno-strict-aliasing -Wdeclaration-after-statement -fno-builtin -fno-common -DHAVE_STDINT_H=1 -DHAVE_SYS_TIME_H=1 -DGX_COLOR_INDEX_TYPE=unsigned long int -O2 -pipe -fno-strict-aliasing -DUSE_LIBICONV_GNU -DUSE_LIBPAPER -I/usr/local/include -DGS_DEVS_SHARED -DGS_DEVS_SHARED_DIR=\/usr/local/lib/ghostscript/9.05\ -fPIC -I./soobj -I./base -DWHICH_CMS=lcms2 -o ./soobj/gdevl256.o -c ./base/gdevl256.c ./base/gdevl256.c:33:17: error: vga.h: No such file or directory ./base/gdevl256.c:34:19: error: vgagl.h: No such file or directory ./base/gdevl256.c: In function 'lvga256_open': ./base/gdevl256.c:104: warning: implicit declaration of function 'vga_init' ./base/gdevl256.c:105: warning: implicit declaration of function 'vga_getdefaultmode' ./base/gdevl256.c:107: error: 'G320x200x256' undeclared (first use in this function) ./base/gdevl256.c:107: error: (Each undeclared identifier is reported only once ./base/gdevl256.c:107: error: for each function it appears in.) ./base/gdevl256.c:108: warning: implicit declaration of function 'vga_setmode' ./base/gdevl256.c:109: warning: implicit declaration of function 'gl_setcontextvga' ./base/gdevl256.c:110: warning: implicit declaration of function 'vga_getxdim' ./base/gdevl256.c:111: warning: implicit declaration of function 'vga_getydim' ./base/gdevl256.c:122: warning: implicit declaration of function 'gl_setpalettecolor' ./base/gdevl256.c: In function 'lvga256_close': ./base/gdevl256.c:136: error: 'TEXT' undeclared (first use in this function) ./base/gdevl256.c: In function 'lvga256_map_rgb_color': ./base/gdevl256.c:150: warning: ISO C90 forbids mixed declarations and code ./base/gdevl256.c: In function 'lvga256_copy_mono': ./base/gdevl256.c:227: warning: implicit declaration of function 'gl_fillbox' ./base/gdevl256.c:238: warning: implicit declaration of function 'gl_setpixel' ./base/gdevl256.c: In function 'lvga256_copy_color': ./base/gdevl256.c:261: warning: implicit declaration of function 'gl_putbox' ./base/gdevl256.c: In function 'lvga256_draw_line': ./base/gdevl256.c:301: warning: implicit declaration of function 'gl_line' gmake[2]: *** [soobj/gdevl256.o] Erreur 1 gmake[2]: *** Attente des tâches non terminées gmake[2] : on quitte le répertoire « /usr/ports/print/ghostscript9/work/ghostscript-9.05 » gmake[1]: *** [so-subtarget] Erreur 2 gmake[1] : on quitte le répertoire « /usr/ports/print/ghostscript9/work/ghostscript-9.05 » gmake: *** [so] Erreur 2 *** Error code 1 Stop in /usr/ports/print/ghostscript9. *** Error code 1 Stop in /usr/ports/print/ghostscript9. Problem : no vga.h in my system, even in source tree. ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: optionsng and tinderbox?
On 20/06/2012 11:34, Baptiste Daroussin wrote:
On Wed, Jun 20, 2012 at 09:54:45AM +, Ganael LAPLANCHE wrote:
On Wed, 20 Jun 2012 11:13:21 +0200, Baptiste Daroussin wrote
Hi Baptiste,
make -VUNIQUENAME should show you the right name
I have a port where UNIQUENAME is *not* the right prefix to use before
'_SET' for positioning this specific port's options :
$ pwd
/usr/ports/net/openldap24-client
$ make -V UNIQUENAME
openldap24-client
$ make -V PORT_OPTIONS
DOCS EXAMPLES NLS
$ make openldap24-client_SET=FETCH -V PORT_OPTIONS
DOCS EXAMPLES NLS
but :
$ make openldap24-server_SET=FETCH -V PORT_OPTIONS
DOCS EXAMPLES FETCH NLS
So, here, I have to set the following in make.conf :
openldap24-server_SET+=FETCH
and not :
openldap24-client_SET+=FETCH
which would not work.
This is about to be fixed by the the uniquename being really unique (see
matthew@) CFT.
Hmmm... Except my patches make no difference here:
% cd /usr/ports/net/openldap24-client
% make -V UNIQUENAME
openldap24-client
% make -V PORT_OPTIONS
DOCS EXAMPLES NLS
% make openldap24-client_SET=FETCH -V PORT_OPTIONS
DOCS EXAMPLES NLS
After looking into this a bit more deeply, there are two different
problems here.
1) Setting ${${UNIQUENAME}_SET} on the command line to override the
settings in OPTIONSFILE -- looks like a logic problem. bsd.options.mk
sets all the foo_SET options from make.conf or the command line around
line 106:
. for opt in ${${UNIQUENAME}_SET}
PORT_OPTIONS+= ${opt}
. endfor
but then around line 124 it has:
.for opt in ${OPTIONS_FILE_UNSET}
PORT_OPTIONS:= ${PORT_OPTIONS:N${opt}}
.endfor
and if the OPTIONSFILE exists from a previous 'make config' all the
different options are listed in it either in OPTIONS_FILE_UNSET or
OPTIONS_FILE_SET. Which means OPTIONSFILE overrides make.conf for
*setting* variables.
Shouldn't make.conf / commandline settings override OPTIONSFILE rather
than the other way round? Seems there's not much point in being able to
set options from make.conf unless that is so, as OPTIONSFILE would be
created more often than not whenever make(1) was invoked in the port's
directory.
2) There is a port-specific problem: the port overrides the OPTIONSFILE
setting, before my patches as a side-effect of setting LATEST_LINK and
specifically preserved in the patch by using the new OPTIONS_DIR variable.
Before my patches the port had:
% make -V UNIQUENAME -V LATEST_LINK -V OPTIONSFILE
openldap24
openldap24-client
/var/db/ports/openldap24/options
% cd ../openldap24-server
% make -V UNIQUENAME -V LATEST_LINK -V OPTIONSFILE
openldap24
openldap24-server
/var/db/ports/openldap24/options
After patching, it has:
% cd ../openldap24-client
% make -V UNIQUENAME -V LATEST_LINK -V OPTIONSFILE
openldap24-client
openldap24-client
/var/db/ports/openldap24/options
% cd ../openldap24-server
% make -V UNIQUENAME -V LATEST_LINK -V OPTIONSFILE
openldap24-server
openldap24-server
/var/db/ports/openldap24/options
So only the value of UNIQUENAME differs before and after.
However on inspection, openldap24-server and openldap24-client take
different sets of options: -server takes a super-set of -client. So
having a shared OPTIONSFILE doesn't make sense to me: either the
OPTIONSFILEs have to be different, or -client has to present exactly the
same options as -server, even if most of the settings would not affect
the -client port.
Cheers,
Matthew
--
Dr Matthew J Seaman MA, D.Phil.
PGP: http://www.infracaninophile.co.uk/pgpkey
signature.asc
Description: OpenPGP digital signature
Re: Horde4 eats all memory
Are there any stable Linux distros with php 5.4 yet? I'm not aware of any right now. ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
samba36 port unable to build PAM backends?
Hi, I'm getting these errors while trying to install the samba36 port with the PAM modules: === Installing for samba36-3.6.5 === samba36-3.6.5 depends on package: tdb=1.2.6 - found === samba36-3.6.5 depends on package: talloc=2.0.1 - found === samba36-3.6.5 depends on shared library: execinfo.1 - found === samba36-3.6.5 depends on shared library: cups.2 - found === samba36-3.6.5 depends on shared library: popt.0 - found === samba36-3.6.5 depends on shared library: ldap-2.4.8 - found === samba36-3.6.5 depends on shared library: iconv.3 - found === Generating temporary packing list === Checking if net/samba36 already installed Using CFLAGS = -O2 -pipe -DLDAP_DEPRECATED -fno-strict-aliasing -I/usr/local/include -I. -I/usr/ports/net/samba36/work/samba-3.6.5/source3 -I/usr/ports/net/samba36/work/samba-3.6.5/source3/../lib/iniparser/src -Iinclude -I./include -I. -I. -I./../lib/replace -I./../lib/tevent -I./librpc -I./.. -DHAVE_CONFIG_H -Iinclude -I./include -I. -I. -I./../lib/replace -I./../lib/tevent -I./librpc -I./.. -I./../lib/popt -DLDAP_DEPRECATED -I/usr/local/include -I/usr/ports/net/samba36/work/samba-3.6.5/source3/lib -I.. -D_SAMBA_BUILD_=3 -D_SAMBA_BUILD_=3 PICFLAG= -fPIC -DPIC LIBS = -lexecinfo -liconv LDFLAGS= -pie -Wl,-z,relro -L/usr/local/lib -L./bin -L/usr/local/lib DYNEXP = -Wl,--export-dynamic LDSHFLAGS = -fPIC -DPIC -shared -Wl,-z,relro -L/usr/local/lib -L./bin -L/usr/local/lib -lc -Wl,-z,defs SHLIBEXT = so SONAMEFLAG = -Wl,-soname, Installing bin/smbd as ///usr/local/sbin/smbd Installing bin/nmbd as ///usr/local/sbin/nmbd Installing bin/winbindd as ///usr/local/sbin/winbindd == The binaries are installed. You may restore the old binaries (if there were any) using the command make revert. You may uninstall the binaries using the command make uninstallbin or make uninstall to uninstall binaries, man pages and shell scripts. == Installing bin/wbinfo as ///usr/local/bin/wbinfo Installing bin/smbclient as ///usr/local/bin/smbclient Installing bin/net as ///usr/local/bin/net Installing bin/smbspool as ///usr/local/bin/smbspool Installing bin/testparm as ///usr/local/bin/testparm Installing bin/smbstatus as ///usr/local/bin/smbstatus Installing bin/smbget as ///usr/local/bin/smbget Installing bin/smbta-util as ///usr/local/bin/smbta-util Installing bin/smbcontrol as ///usr/local/bin/smbcontrol Installing bin/smbtree as ///usr/local/bin/smbtree Installing bin/nmblookup as ///usr/local/bin/nmblookup Installing bin/pdbedit as ///usr/local/bin/pdbedit Installing bin/smbpasswd as ///usr/local/bin/smbpasswd Installing bin/rpcclient as ///usr/local/bin/rpcclient Installing bin/smbcacls as ///usr/local/bin/smbcacls Installing bin/profiles as ///usr/local/bin/profiles Installing bin/ntlm_auth as ///usr/local/bin/ntlm_auth Installing bin/sharesec as ///usr/local/bin/sharesec Installing bin/smbcquotas as ///usr/local/bin/smbcquotas Installing bin/eventlogadm as ///usr/local/bin/eventlogadm == The binaries are installed. You may restore the old binaries (if there were any) using the command make revert. You may uninstall the binaries using the command make uninstallbin or make uninstall to uninstall binaries, man pages and shell scripts. == Installing scripts in /usr/local/bin Installing /usr/local/bin/smbtar Installing /usr/local/bin/findsmb == The scripts have been installed. You may uninstall them using the command make uninstallscripts or make install to install binaries, man pages and shell scripts. You may recover the previous version (if any by make revert. == Installing /usr/ports/net/samba36/work/samba-3.6.5/source3//../codepages/lowcase.dat as ///usr/local/lib/samba/lowcase.dat Installing /usr/ports/net/samba36/work/samba-3.6.5/source3//../codepages/upcase.dat as ///usr/local/lib/samba/upcase.dat Installing /usr/ports/net/samba36/work/samba-3.6.5/source3//../codepages/valid.dat as ///usr/local/lib/samba/valid.dat == The dat files have been installed. You may uninstall the dat files using the command make uninstalldat or make uninstall to uninstall binaries, man pages, dat files, and shell scripts. == Preserving old module as ///usr/local/lib/samba/vfs/recycle.so.old Installing bin/recycle.so as ///usr/local/lib/samba/vfs/recycle.so Preserving old module as ///usr/local/lib/samba/vfs/audit.so.old Installing bin/audit.so as ///usr/local/lib/samba/vfs/audit.so Preserving old
Re: Unable to build graphics/ImageMagick port
On Wed, Jun 20, 2012 at 2:33 AM, Carmel carmel...@hotmail.com wrote: On Tue, 19 Jun 2012 21:22:51 -0700 Kevin Oberman articulated: To install ImageMagick with all default options, turn on HDRI (High Dynamic Range), a very desirable capability for many but turn off TESTS. Build and INSTALL ImageMagick. That puts all of the needed files in place. Then turn on TESTS and re-buildand ImageMagick.It will then pass all tests. Sorry, been there -- done that -- didn't work. Please reread my previous post. This is on a FreeBSD-8.3 STABLE amd64 machine. I re-read the prior posts and I am confused. First, I have updated multiple systems, i386 and amd64, version 8.3 and 9.0 with no problems. I simply has worked every time. Second, in your earlier report, you stated that I installed it sans Tests and HDRI without problems. I then attempted to install it with both options and it bombed out. I finally just installed it with HDRI sans Tests and it installed fine. Yes. As long as you do not run tests,it installs fine whether HDRI is selected or not. The only failure is in running tests, not building HDRI. But,it it is built and installed with HDRI, if tests are then run, they (or at least analyze.sh) will pass, or at least it always has for me. If you just want to confirm that ImageMagick did build correctly, you can always install but don't clean and 'cd /usr/ports/graphics/ImageMagick/work/ImageMagick-6.7.7-7 make check-TESTS check-local'. I don't recommend this as you have now deleted your working version and, if there is an issue with the build, it's harder to go back. That's why I suggest 'portmaster -b ImageMagick' as that will build and test the new version while saving the old one so it can be easily re-installed if there is a problem running the tests. With HDRI support built, here is what I get: cd work/ImageMagick-6.7.7-7/ rogue# make check-TESTS check-local PASS: tests/validate-compare.sh PASS: tests/validate-composite.sh PASS: tests/validate-convert.sh PASS: tests/validate-identify.sh PASS: tests/validate-import.sh PASS: tests/validate-montage.sh PASS: tests/validate-pipe.sh PASS: tests/validate-colorspace.sh PASS: tests/validate-stream.sh PASS: tests/validate-formats-in-memory.sh PASS: tests/validate-formats-on-disk.sh PASS: Magick++/tests/exceptions.sh PASS: Magick++/tests/appendImages.sh PASS: Magick++/tests/attributes.sh PASS: Magick++/tests/averageImages.sh PASS: Magick++/tests/coalesceImages.sh PASS: Magick++/tests/coderInfo.sh PASS: Magick++/tests/colorHistogram.sh PASS: Magick++/tests/color.sh PASS: Magick++/tests/montageImages.sh PASS: Magick++/tests/morphImages.sh PASS: Magick++/tests/readWriteBlob.sh PASS: Magick++/tests/readWriteImages.sh PASS: Magick++/demo/analyze.sh PASS: Magick++/demo/button.sh PASS: Magick++/demo/demo.sh PASS: Magick++/demo/flip.sh PASS: Magick++/demo/gravity.sh PASS: Magick++/demo/piddle.sh PASS: Magick++/demo/shapes.sh PASS: Magick++/demo/zoom_bessel.sh PASS: Magick++/demo/zoom_blackman.sh PASS: Magick++/demo/zoom_box.sh PASS: Magick++/demo/zoom_catrom.sh PASS: Magick++/demo/zoom_cubic.sh PASS: Magick++/demo/zoom_gaussian.sh PASS: Magick++/demo/zoom_hamming.sh PASS: Magick++/demo/zoom_hanning.sh PASS: Magick++/demo/zoom_hermite.sh PASS: Magick++/demo/zoom_lanczos.sh PASS: Magick++/demo/zoom_mitchell.sh PASS: Magick++/demo/zoom_point.sh PASS: Magick++/demo/zoom_quadratic.sh PASS: Magick++/demo/zoom_sample.sh PASS: Magick++/demo/zoom_scale.sh PASS: Magick++/demo/zoom_sinc.sh PASS: Magick++/demo/zoom_triangle.sh PASS: wand/drawtest.sh PASS: wand/wandtest.sh === All 49 tests passed === -- R. Kevin Oberman, Network Engineer E-mail: kob6...@gmail.com ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: Unable to build graphics/ImageMagick port
On Wed, 20 Jun 2012 09:10:50 -0700 Kevin Oberman articulated: On Wed, Jun 20, 2012 at 2:33 AM, Carmel carmel...@hotmail.com wrote: On Tue, 19 Jun 2012 21:22:51 -0700 Kevin Oberman articulated: To install ImageMagick with all default options, turn on HDRI (High Dynamic Range), a very desirable capability for many but turn off TESTS. Build and INSTALL ImageMagick. That puts all of the needed files in place. Then turn on TESTS and re-buildand ImageMagick.It will then pass all tests. Sorry, been there -- done that -- didn't work. Please reread my previous post. This is on a FreeBSD-8.3 STABLE amd64 machine. I re-read the prior posts and I am confused. First, I have updated multiple systems, i386 and amd64, version 8.3 and 9.0 with no problems. I simply has worked every time. Second, in your earlier report, you stated that I installed it sans Tests and HDRI without problems. I then attempted to install it with both options and it bombed out. I finally just installed it with HDRI sans Tests and it installed fine. Yes. As long as you do not run tests,it installs fine whether HDRI is selected or not. The only failure is in running tests, not building HDRI. But,it it is built and installed with HDRI, if tests are then run, they (or at least analyze.sh) will pass, or at least it always has for me. Now I am confused. If I build sans Tests but with HDRI, it works fine. However, if I then attempt to rebuild and include Tests, it bombs out. The only way I can get HDRI included is to exclude Tests. If you want, I will rebuild it with both options enabled and send you a copy of the log. Your call ... -- Carmel ✌ carmel...@hotmail.com ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: optionsng and tinderbox?
On 20 June 2012 10:13, Baptiste Daroussin b...@freebsd.org wrote: On Wed, Jun 20, 2012 at 03:58:21AM -0400, Michael Scheidell wrote: On 6/20/12 2:11 AM, Baptiste Daroussin wrote: nginx_UNSET= ICONV will deactivate ICONV for nginx only lowercase? Yes lower case, the wiki is wrong. My fault, someone convinced me otherwise a while ago :/ Fixed. Chris ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: optionsng and tinderbox?
On 6/20/12 2:26 PM, Chris Rees wrote: My fault, someone convinced me otherwise a while ago :/ Fixed. Chris 29 lashes with a wet noodle! -- Michael Scheidell, CTO *| * SECNAP Network Security Corporation d: +1.561.948.2259 w: http://people.freebsd.org/~scheidell ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: optionsng and tinderbox?
On 6/20/12 2:26 PM, Chris Rees wrote: On 20 June 2012 10:13, Baptiste Daroussinb...@freebsd.org wrote: On Wed, Jun 20, 2012 at 03:58:21AM -0400, Michael Scheidell wrote: On 6/20/12 2:11 AM, Baptiste Daroussin wrote: nginx_UNSET=ICONV will deactivate ICONV for nginx only lowercase? Yes lower case, the wiki is wrong. My fault, someone convinced me otherwise a while ago :/ also, while at it, document how you do multiple options? MixedCase_SET=OPT1 OPT2 OPT3? or MixedCase_SET=OPT1,OPT2,OPT3 ? (I assume mixed case, not lowercase, in CASE make -V UNIQUENAME is Mixed CaSe? Fixed. Chris -- Michael Scheidell, CTO *| * SECNAP Network Security Corporation d: +1.561.948.2259 w: http://people.freebsd.org/~scheidell ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: optionsng and tinderbox?
On 20 June 2012 19:33, Michael Scheidell scheid...@freebsd.org wrote: On 6/20/12 2:26 PM, Chris Rees wrote: On 20 June 2012 10:13, Baptiste Daroussinb...@freebsd.org wrote: On Wed, Jun 20, 2012 at 03:58:21AM -0400, Michael Scheidell wrote: On 6/20/12 2:11 AM, Baptiste Daroussin wrote: nginx_UNSET= ICONV will deactivate ICONV for nginx only lowercase? Yes lower case, the wiki is wrong. My fault, someone convinced me otherwise a while ago :/ also, while at it, document how you do multiple options? MixedCase_SET=OPT1 OPT2 OPT3? or MixedCase_SET=OPT1,OPT2,OPT3 ? Don't forget, makevars are lists of words, separated by whitespace. This particular case is dealt with using .for, so definitely the former. Chris ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
USE_ and WANT_ for the Porter's Handbook
Here's a bare-bones pass at documenting the difference between USE_ and WANT_ variables. WITH_ and WITHOUT_ have been removed because of the optionsng rework, and there is a Knobs section later in the PH that already describes them. This section has been added to the end of the Dependencies section, currently 5.7. It could be moved elsewhere. http://www.wonkity.com/~wblock/porter/book.html http://www.wonkity.com/~wblock/porter/usewant.diff ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: USE_ and WANT_ for the Porter's Handbook
On 6/20/12 7:58 PM, Warren Block wrote: WITH_ and WITHOUT_ have been removed because of the optionsng rework, maybe leave them in for the legacy ports? ports that haven't gotten around to it yet? -- Michael Scheidell, CTO *| * SECNAP Network Security Corporation d: +1.561.948.2259 w: http://people.freebsd.org/~scheidell ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: paths of ruby, python, perl...
You can use portupgrade -r[R] your_package. Or portupgrade -f `pkg_glob 'p5-*'` after upgrade your main package or something else (perl-after-upgrade). On Thu, 2012-06-07 at 11:09 -0700, jsb...@speakeasy.net wrote: Each time I've upgraded ruby, python, and perl I've encountered lengthy = reinstalls of the port files (p5-...) as well as higher-level programs which depend upon them. (Many pkg_which oneliners to parse. ..) This could be simplified if those /lang/ ports were all-in-= one-dir (like /perl/ rather than perl site_perl/5,10 site= _perl/5.12 /5.10 /5.12 etc... Maybe someone knows i= f such a path reconfiguration would break some upstream non-BSD standard... Thanks. J. Bouquet ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: Unable to build graphics/ImageMagick port
My head hurts! On Wed, Jun 20, 2012 at 9:43 AM, Carmel carmel...@hotmail.com wrote: On Wed, 20 Jun 2012 09:10:50 -0700 Kevin Oberman articulated: On Wed, Jun 20, 2012 at 2:33 AM, Carmel carmel...@hotmail.com wrote: On Tue, 19 Jun 2012 21:22:51 -0700 Kevin Oberman articulated: To install ImageMagick with all default options, turn on HDRI (High Dynamic Range), a very desirable capability for many but turn off TESTS. Build and INSTALL ImageMagick. That puts all of the needed files in place. Then turn on TESTS and re-buildand ImageMagick.It will then pass all tests. Sorry, been there -- done that -- didn't work. Please reread my previous post. This is on a FreeBSD-8.3 STABLE amd64 machine. I re-read the prior posts and I am confused. First, I have updated multiple systems, i386 and amd64, version 8.3 and 9.0 with no problems. I simply has worked every time. Second, in your earlier report, you stated that I installed it sans Tests and HDRI without problems. I then attempted to install it with both options and it bombed out. I finally just installed it with HDRI sans Tests and it installed fine. Yes. As long as you do not run tests,it installs fine whether HDRI is selected or not. The only failure is in running tests, not building HDRI. But,it it is built and installed with HDRI, if tests are then run, they (or at least analyze.sh) will pass, or at least it always has for me. Now I am confused. If I build sans Tests but with HDRI, it works fine. However, if I then attempt to rebuild and include Tests, it bombs out. The only way I can get HDRI included is to exclude Tests. If you want, I will rebuild it with both options enabled and send you a copy of the log. Your call ... To be really, really clear, did you build and install ImageMagick with HDRI (but not running tests)? Did you then try building again, but have the analyze.sh test bombed out? After building and installing with HDRI, did you try just running the tests and, if so, did you look at analyze.log? -- R. Kevin Oberman, Network Engineer E-mail: kob6...@gmail.com ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: USE_ and WANT_ for the Porter's Handbook
On Wed, 20 Jun 2012, Michael Scheidell wrote: On 6/20/12 7:58 PM, Warren Block wrote: WITH_ and WITHOUT_ have been removed because of the optionsng rework, maybe leave them in for the legacy ports? ports that haven't gotten around to it yet? I should have been more clear about that: there was a portion of this new section that described WITH_ and WITHOUT_. That section was removed, both because of optionsng and because of the existing Knobs section of the Porter's Handbook. That Knobs section is still present, and still describes the WITH_ variables. ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: ports/169280: [Maintainer Update] Minor bug fix to sysutils/bsdconfig
Synopsis: [Maintainer Update] Minor bug fix to sysutils/bsdconfig Responsible-Changed-From-To: dteske-ports Responsible-Changed-By: dteske Responsible-Changed-When: Thu Jun 21 03:05:37 UTC 2012 Responsible-Changed-Why: Back to pool. http://www.freebsd.org/cgi/query-pr.cgi?pr=169280 ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: ports/169280: [Maintainer Update] Minor bug fix to sysutils/bsdconfig
Synopsis: [Maintainer Update] Minor bug fix to sysutils/bsdconfig Responsible-Changed-From-To: ports-wxs Responsible-Changed-By: wxs Responsible-Changed-When: Thu Jun 21 03:11:34 UTC 2012 Responsible-Changed-Why: I'll take it. http://www.freebsd.org/cgi/query-pr.cgi?pr=169280 ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Need advice on a problems with the Django CMS port
Hi all,
I maintain the Django CMS port (www/py-django-cms). I could use some
advice on the following problem both from other maintainers and from
other users of the ports tree:
Django CMS requires a database backend, which it accesses through the
Django web app framework. The Django port (www/py-django) doesn't
include database support by default. If someone naively runs cd
/usr/ports/www/py-django-cms; make install, Django CMS won't work
properly because of the missing database drivers. The Django port does
have knobs for PostgreSQL, MySQL, and SQLite, but those knobs don't
affect what bits of Django get installed. All the knobs do is add
databases/{py-psycopg2,py-MySQLdb,py-sqlite3} to the Django package's
RUN_DEPENDS. As I see it, I have the following options:
(a) I could add a warning to www/py-django-cms/pkg-message that says
something to the effect of make sure you build Django with database
support. This is the easiest thing for me to do, but it isn't very
user friendly.
(b) I could try to come up with some pre-installation test code that
checks for a working database backend in Django and then prompt the user
to re-build Django with database support. I have no idea how to do this
in a reliable way, but this is no less user friendly than having a port
set BROKEN or IGNORE when certain knobs are frobbed.
(c) I could add knobs to the Django CMS port similar to those found in
the Django port (i.e., add to RUN_DEPENDS if knob is set) -
functionally, it doesn't matter which port pulls in the required
database drivers. This is probably the most user friendly, in that a
single run of make install will result in a working version of Django
CMS.
I think option (c) is the best, but I'd love to hear what the community
thinks.
Best wishes,
Matthew
P.S. If you want to try out Django CMS, make sure you install Django
1.3. I'm going to update the Django CMS port to reflect this
requirement RSN.
--
I FIGHT FOR THE USERS
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: Need advice on a problems with the Django CMS port
On Wed, Jun 20, 2012 at 11:38 PM, xenophon\+freebsd
xenophon+free...@irtnog.org wrote:
Hi all,
I maintain the Django CMS port (www/py-django-cms). I could use some
advice on the following problem both from other maintainers and from
other users of the ports tree:
Django CMS requires a database backend, which it accesses through the
Django web app framework. The Django port (www/py-django) doesn't
include database support by default. If someone naively runs cd
/usr/ports/www/py-django-cms; make install, Django CMS won't work
properly because of the missing database drivers. The Django port does
have knobs for PostgreSQL, MySQL, and SQLite, but those knobs don't
affect what bits of Django get installed. All the knobs do is add
databases/{py-psycopg2,py-MySQLdb,py-sqlite3} to the Django package's
RUN_DEPENDS. As I see it, I have the following options:
:
(c) I could add knobs to the Django CMS port similar to those found in
the Django port (i.e., add to RUN_DEPENDS if knob is set) -
functionally, it doesn't matter which port pulls in the required
database drivers. This is probably the most user friendly, in that a
single run of make install will result in a working version of Django
CMS.
I think option (c) is the best, but I'd love to hear what the community
thinks.
Option C would be the best, as the port depends on a database (MySQL,
PostgreSQL or SQLite).
I believe the following would work with the new options framework:
OPTIONS_MULTI= DATABASE
OPTIONS_MULTI_DATABASE= MYSQL PGSQL SQLITE
OPTIONS_DEFAULT=SQLITE
.include bsd.port.options.mk
.if ${PORT_OPTIONS:MMYSQL}
RUN_DEPENDS+=
${PYTHON_PKGNAMEPREFIX}MySQLdb=1.2.2:${PORTSDIR}/databases/py-MySQLdb
.endif
.if ${PORT_OPTIONS:MPGSQL}
RUN_DEPENDS+=
${PYTHON_SITELIBDIR}/psycopg2/_psycopg.so:${PORTSDIR}/databases/py-psycopg2
.endif
.if ${PORT_OPTIONS:MSQLITE}
RUN_DEPENDS+=
${PYTHON_SITELIBDIR}/_sqlite3.so:${PORTSDIR}/databases/py-sqlite3
.endif
:
:
.include bsd.port.mk
Scot
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: Need advice on a problems with the Django CMS port
On Thu, Jun 21, 2012 at 12:14 AM, Scot Hetzel swhet...@gmail.com wrote:
On Wed, Jun 20, 2012 at 11:38 PM, xenophon\+freebsd
xenophon+free...@irtnog.org wrote:
Hi all,
I maintain the Django CMS port (www/py-django-cms). I could use some
advice on the following problem both from other maintainers and from
other users of the ports tree:
Django CMS requires a database backend, which it accesses through the
Django web app framework. The Django port (www/py-django) doesn't
include database support by default. If someone naively runs cd
/usr/ports/www/py-django-cms; make install, Django CMS won't work
properly because of the missing database drivers. The Django port does
have knobs for PostgreSQL, MySQL, and SQLite, but those knobs don't
affect what bits of Django get installed. All the knobs do is add
databases/{py-psycopg2,py-MySQLdb,py-sqlite3} to the Django package's
RUN_DEPENDS. As I see it, I have the following options:
:
(c) I could add knobs to the Django CMS port similar to those found in
the Django port (i.e., add to RUN_DEPENDS if knob is set) -
functionally, it doesn't matter which port pulls in the required
database drivers. This is probably the most user friendly, in that a
single run of make install will result in a working version of Django
CMS.
I think option (c) is the best, but I'd love to hear what the community
thinks.
Option C would be the best, as the port depends on a database (MySQL,
PostgreSQL or SQLite).
I believe the following would work with the new options framework:
OPTIONS_MULTI= DATABASE
OPTIONS_MULTI_DATABASE= MYSQL PGSQL SQLITE
This should be OPTIONS_SINGLE and OPTIONS_SINGLE_DATABASE, as you only
need one of these.
OPTIONS_DEFAULT= SQLITE
.include bsd.port.options.mk
.if ${PORT_OPTIONS:MMYSQL}
RUN_DEPENDS+=
${PYTHON_PKGNAMEPREFIX}MySQLdb=1.2.2:${PORTSDIR}/databases/py-MySQLdb
.endif
.if ${PORT_OPTIONS:MPGSQL}
RUN_DEPENDS+=
${PYTHON_SITELIBDIR}/psycopg2/_psycopg.so:${PORTSDIR}/databases/py-psycopg2
.endif
.if ${PORT_OPTIONS:MSQLITE}
RUN_DEPENDS+=
${PYTHON_SITELIBDIR}/_sqlite3.so:${PORTSDIR}/databases/py-sqlite3
.endif
:
:
.include bsd.port.mk
Scot
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
Re: Need advice on a problems with the Django CMS port
On Thu, Jun 21, 2012 at 12:14:12AM -0500, Scot Hetzel thus spake:
On Wed, Jun 20, 2012 at 11:38 PM, xenophon\+freebsd
xenophon+free...@irtnog.org wrote:
Hi all,
I maintain the Django CMS port (www/py-django-cms). I could use some
advice on the following problem both from other maintainers and from
other users of the ports tree:
Django CMS requires a database backend, which it accesses through the
Django web app framework. The Django port (www/py-django) doesn't
include database support by default. If someone naively runs cd
/usr/ports/www/py-django-cms; make install, Django CMS won't work
properly because of the missing database drivers. The Django port does
have knobs for PostgreSQL, MySQL, and SQLite, but those knobs don't
affect what bits of Django get installed. All the knobs do is add
databases/{py-psycopg2,py-MySQLdb,py-sqlite3} to the Django package's
RUN_DEPENDS. As I see it, I have the following options:
:
(c) I could add knobs to the Django CMS port similar to those found in
the Django port (i.e., add to RUN_DEPENDS if knob is set) -
functionally, it doesn't matter which port pulls in the required
database drivers. This is probably the most user friendly, in that a
single run of make install will result in a working version of Django
CMS.
I think option (c) is the best, but I'd love to hear what the community
thinks.
Option C would be the best, as the port depends on a database (MySQL,
PostgreSQL or SQLite).
I believe the following would work with the new options framework:
OPTIONS_MULTI= DATABASE
OPTIONS_MULTI_DATABASE= MYSQL PGSQL SQLITE
OPTIONS_DEFAULT=SQLITE
.include bsd.port.options.mk
.if ${PORT_OPTIONS:MMYSQL}
RUN_DEPENDS+=
${PYTHON_PKGNAMEPREFIX}MySQLdb=1.2.2:${PORTSDIR}/databases/py-MySQLdb
.endif
.if ${PORT_OPTIONS:MPGSQL}
RUN_DEPENDS+=
${PYTHON_SITELIBDIR}/psycopg2/_psycopg.so:${PORTSDIR}/databases/py-psycopg2
.endif
.if ${PORT_OPTIONS:MSQLITE}
RUN_DEPENDS+=
${PYTHON_SITELIBDIR}/_sqlite3.so:${PORTSDIR}/databases/py-sqlite3
.endif
:
:
.include bsd.port.mk
Scot
At a minimum, I would suggest to also include an applicable client in each
case as well, that is, if they aren't included in the software or the
dependency tree.
-jgh
--
Jason Helfman
System Administrator
experts-exchange.com
http://www.experts-exchange.com/M_4830110.html
E4AD 7CF1 1396 27F6 79DD 4342 5E92 AD66 8C8C FBA5
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to freebsd-ports-unsubscr...@freebsd.org
