Possibly unbuildable ports reminder

2006-10-28 Thread Bill Fenner
Dear porters,

  This is just a reminder to please periodically check the list of
unbuildable ports at http://pointyhat.freebsd.org/errorlogs/ .
A list by MAINTAINER is

http://people.freebsd.org/~fenner/errorlogs/

so you can easily check the status of ports that you maintain.  In
addition, the list of ports with no MAINTAINER with build problems is

http://people.freebsd.org/~fenner/errorlogs/[EMAIL PROTECTED]

Since no one is responsible for these ports, the problem won't get
fixed unless someone on this list takes the initiative.

Thanks for your help!

Bill "annoying port email" Fenner
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "[EMAIL PROTECTED]"


festival+OGI brokenness: reason found

2006-10-28 Thread Andriy Gapon

I think I found a reason why build of audio/festival+OGI touches
LOCALBASE fs before installation (during regular build).

I think it is the following line in speech_tools/config/rules/install.mak:
PROJECT_HOME_PATH := $(shell mkdir -p $($(PROJECT_PREFIX)_HOME); cd
$($(PROJECT_PREFIX)_HOME); pwd)

Here $(PROJECT_PREFIX) is "FESTIVAL" and $(FESTIVAL_HOME) is assigned
from $(PKG_FESTIVAL_HOME) [in festival/config/config] which passed to
gmake through environment from the port's Makefile and it is
${PREFIX}/share/festival.

That file, tools/config/rules/install.mak, is included into
speech_tools/Makefile and thus mkdir+cd are executed by mere makefile
inclusion and variable assignment.

I think that this way "they" were trying to do some realpath-ish thing,
but I believe that this is totally unnecessary. So, I think, that we can
fix the port by changing the offending assignment to a simple:
PROJECT_HOME_PATH := $($(PROJECT_PREFIX)_HOME)


P.S. I see that "regular" festival is at version 1.95, maybe
festival+OGI could be updated to that too ?
Or is OGI stuff already integrated into festival 1.95 and this
festival+OGI port is obsolete ?
I am not sure that I understand how OpenBSD ports are organized, but
they might have done festival+OGI 1.9.5:
http://www.openbsd.org/cgi-bin/cvsweb/ports/audio/festival/

Personally I am using festival+OGI because of festvox-mwm.
-- 
Andriy Gapon
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "[EMAIL PROTECTED]"


Re: festival+OGI brokenness: reason found

2006-10-28 Thread Mikhail Teterin
On Saturday 28 October 2006 09:04, Andriy Gapon wrote:
= P.S. I see that "regular" festival is at version 1.95, maybe
= festival+OGI could be updated to that too?

Very likely...

= Or is OGI stuff already integrated into festival 1.95 and this
= festival+OGI port is obsolete ?
= I am not sure that I understand how OpenBSD ports are organized, but
= they might have done festival+OGI 1.9.5:
= http://www.openbsd.org/cgi-bin/cvsweb/ports/audio/festival/
= 
= Personally I am using festival+OGI because of festvox-mwm.

The audio/fest* ports are in dire need of an active maintainer...

-mi

P.S. And of Ukrainian-speech datafiles...
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "[EMAIL PROTECTED]"


Re: port_history - track port commit message comments

2006-10-28 Thread Hans Lambermont
Ion-Mihai IOnut Tetcu wrote:

...
 On 10/27/06, Christopher Boumenot wrote:
> Every time I upgrade a port I am usually left wondering what
> changed.
...
> The problem with the OP's work is that some submitters/commiters think
> that the CVS commit message should document the _port_ changes while
> others think it should also document (or at least provide pointers) to
> the changes in the software itself (I fall in the second category).
> Since we don't have a standard for this, relaying only on the CVS logs
> might drive one into nasty problems.

But it is much better than what we have now (nothing) and I think that
having a system like what Christopher proposed probably leads to
establishing a standard (preferably of the second category :) .
If the information is easily available I think it will be used.
 
regards,
   Hans Lambermont
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "[EMAIL PROTECTED]"


Re: FreeBSD Port: clamav-0.88.5

2006-10-28 Thread Hans Lambermont
Wiebe Pestman wrote:

...
> This mail is about a possible bug in the package ClamAV. I tried it
> out on a test computer. But when running "clamscan -r /path/to/stuff",
> systematically the next message was returned:  segmentation fault
> (core dumped)

Can you please send the output of the following to help us pinpoint the
problem :

ldd `which clamscan`

pkg_info -g clamav-0.88.5

And most important a backtrace of the core dump :

gdb clamscan clamscan.core
bt

Succes !

groetjes,
   Hans Lambermont
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "[EMAIL PROTECTED]"


Re: experimental qemu port update, please test

2006-10-28 Thread Juergen Lock
On Thu, Oct 26, 2006 at 10:17:52PM +0200, Juergen Lock wrote:
> While ports are still in freeze, I'd like to take the opportunity
> to let you test an experimental port update of today's cvs snapshot
> with any guests you have.  qemu cvs now uses aio, so make sure you
> kldload that (or put it into your kernel) before running the
> updated qemu.  ...

Ok, I just added WITHOUT_SDL and WITHOUT_CDROM_DMA knobs (the latter
being for FreeSBIE which still has atapicam in the kernel which for
some reason qemu cvs doesn't like), and since the port has now grown
a whopping six knobs I decided I should probably convert it to
OPTIONS.  Please test, and tell me how you like it... :)  (no new
commits to qemu cvs apparently so I reused the older snapshot.)

New files: files/patch-vl.h files/patch-Makefile files/patch-vl.c-serial
files/cdrom-dma-patch

Index: Makefile
===
RCS file: /home/ncvs/ports/emulators/qemu/Makefile,v
retrieving revision 1.57
diff -u -r1.57 Makefile
--- Makefile20 Sep 2006 11:20:47 -  1.57
+++ Makefile28 Oct 2006 14:56:48 -
@@ -6,19 +6,16 @@
 #
 
 PORTNAME=  qemu
-PORTVERSION=   0.8.2
-PORTREVISION=  2
+PORTVERSION=   0.8.2s.20061026
 CATEGORIES=emulators
 MASTER_SITES=  http://www.qemu.org/:release \
+   http://qemu-forum.ipi.fi/qemu-snapshots/:snapshot \
http://people.fruitsalad.org/nox/qemu/:snapshot \
http://www.volny.cz/xnavara/qemu/:snapshot \
-   http://qemu.dad-answers.com/download/qemu/:snapshot \
http://people.brandeis.edu/~jcoiner/qemu_idedma/:idedma \
http://people.freebsd.org/~maho/qemu/:misc
-DISTFILES= ${DISTNAME}${EXTRACT_SUFX}:release
-.if defined (WITH_HACKS_CIRRUS) || defined (WITH_HACKS)
-DISTFILES+=patch3_cirrus:misc
-.endif
+DISTNAME=  ${PORTNAME}-snapshot-2006-10-26_05
+DISTFILES= ${DISTNAME}${EXTRACT_SUFX}:snapshot
 DIST_SUBDIR=   qemu
 EXTRACT_ONLY=  ${DISTNAME}${EXTRACT_SUFX}
 
@@ -26,63 +23,54 @@
 COMMENT=   QEMU CPU Emulator
 
 BUILD_DEPENDS+=texi2html:${PORTSDIR}/textproc/texi2html
-.if defined(WITH_SAMBA)
-RUN_DEPENDS+=  ${LOCALBASE}/sbin/smbd:${PORTSDIR}/net/samba3
-.endif
-.if defined(WITH_KQEMU)
-BUILD_DEPENDS+=kqemu-kmod>=1.3.0pre5:${PORTSDIR}/emulators/kqemu-kmod
-.endif
 
 HAS_CONFIGURE= yes
+USE_BZIP2= yes
 USE_GMAKE= yes
 USE_GETOPT_LONG=   yes
-USE_SDL=   sdl
 USE_PERL5= yes
+USE_GCC=   3.4
 PATCH_STRIP=   -p1
 CONFIGURE_ARGS+=   --prefix=${PREFIX} --cc=${CC}
 MAKE_ENV+= BSD_MAKE="${MAKE}" CFLAGS="${CFLAGS}" LDFLAGS="${LDFLAGS}"
 MAN1=  qemu.1 qemu-img.1
 ONLY_FOR_ARCHS=amd64 i386
+
+OPTIONS=   KQEMU "Build with (alpha!) accelerator module" Off \
+   HACKS_CIRRUS "Large display speedup (buggy!)" Off \
+   RTL8139_TIMER "allow use of re(4) nic with FreeBSD guests" Off \
+   SAMBA "samba dependency (for -smb)" Off \
+   SDL "SDL/X dependency (graphical output)" On \
+   CDROM_DMA "IDE CDROM DMA" On
+
+.include 
+
 WITHOUT_CPU_CFLAGS=yes #to avoid problems with register allocation
 
+.if defined(WITHOUT_SDL)
+CONFIGURE_ARGS+=   --disable-sdl --disable-gfx-check
+.else
+USE_SDL=   sdl
+.endif
+
+.if defined (WITH_HACKS_CIRRUS) || defined (WITH_HACKS)
+DISTFILES+=patch3_cirrus:misc
+.endif
+
+.if defined(WITH_SAMBA)
+RUN_DEPENDS+=  ${LOCALBASE}/sbin/smbd:${PORTSDIR}/net/samba3
+.endif
+
 .if defined(WITH_KQEMU)
+BUILD_DEPENDS+=kqemu-kmod>=1.3.0pre5:${PORTSDIR}/emulators/kqemu-kmod
 CONFIGURE_ARGS+=   --enable-kqemu
 .else
 CONFIGURE_ARGS+=   --disable-kqemu
 .endif
 
-.include 
-
 .if ${ARCH} == "amd64"
 MAKE_ARGS+=ARCH=x86_64
 .endif
-USE_GCC=   3.4
-
-pre-everything::
-.if !defined(WITH_KQEMU)
-   @${ECHO_MSG} "Notice: you can build qemu with the (alpha!) kqemu 
accelerator kernel module"
-   @${ECHO_MSG} "by defining WITH_KQEMU."
-.endif
-.if !defined(WITH_HACKS_CIRRUS) && !defined(WITH_HACKS)
-   @${ECHO_MSG} "You can build qemu with some hacks (esp. for speedup)"
-   @${ECHO_MSG} "by defining WITH_HACKS, or specifically:"
-   @${ECHO_MSG} "2. WITH_HACKS_CIRRUS: higher speed on large display 
(cirrus_vga)"
-   @${ECHO_MSG} "by Juergen Pfennig"
-   @${ECHO_MSG} 
"http://lists.gnu.org/archive/html/qemu-devel/2006-01/msg00208.html";
-   @${ECHO_MSG} "Note: this second patch is known to cause mouse problems 
with some(?)"
-   @${ECHO_MSG} "versions of XP, and also minor redraw bugs with some 
Linux guests."
-.endif
-.if !defined(WITH_SAMBA) && !exists(${LOCALBASE}/sbin/smbd)
-   @${ECHO_MSG} "Notice: if you need qemu's -smb option (smb-export local 
dir to guest)"
-   @${ECHO_MSG} "then you also need samba, you can have this port install 
it by defining"
-   @${ECHO_MSG} "WITH_SAMBA."
-.endif
-.if !defined(WITH_RTL8139_TIMER)
-   @${ECHO_MSG} "Notic

Re: port_history - track port commit message comments

2006-10-28 Thread IOnut
On Sat, 28 Oct 2006 20:16:47 +0200
[EMAIL PROTECTED] (Hans Lambermont) wrote:

> Ion-Mihai IOnut Tetcu wrote:
> 
> ...
>  On 10/27/06, Christopher Boumenot wrote:
> > Every time I upgrade a port I am usually left wondering what
> > changed.
> ...
> > The problem with the OP's work is that some submitters/commiters
> > think that the CVS commit message should document the _port_
> > changes while others think it should also document (or at least
> > provide pointers) to the changes in the software itself (I fall in
> > the second category). Since we don't have a standard for this,
> > relaying only on the CVS logs might drive one into nasty problems.
> 
> But it is much better than what we have now (nothing) and I think that
> having a system like what Christopher proposed probably leads to
> establishing a standard (preferably of the second category :) .

Christopher's work is indeed valuable and I should have mentioned this
in my first post :(

I'm afraid of a partial solution because one tends to relay (only) on
it and, since it partial, tends nt to check the vendor site, README,
CHANGES or UPDATING included in the tarbal, etc.

A similar thing drives me mad each time I install a new machine and I
discover I have to recurse in the dependency tree of the ports I need
because some of them don't have OPTIONS.

> If the information is easily available I think it will be used.

One has to be optimistic about it, albeit the opposite example provided
by people not reading UPDATING ;-)


-- 
IOnut - Un^d^dregistered ;) FreeBSD "user"
  "Intellectual Property" is   nowhere near as valuable   as "Intellect"

BOFH excuse #138:
BNC (brain not connected)




signature.asc
Description: PGP signature


Re: qt-3.3/php 5.1.6 does not compile FreeBSD 6.2-PRE/AMD64

2006-10-28 Thread cpghost
On Thu, Oct 26, 2006 at 10:18:44PM +0200, O. Hartmann wrote:
> Hello.
> Before sending a PR, I would like to ask you and maybe someone reveals
> my faults in this subject.
> 
> Since a couple of weeks I can not build PHP 5.1.6 anymore. I can do it
> by hand, but not with the ports system. cups and qt are dependend on PHP
> and I would like to fix that problem. It seems that several ports do not
> build anymore especially on AMD64 arch, since I have a i386 box (Pentium
> 4) with nearly the same software status and configuration and I can
> check that there is everything o.k.
> 
> This is the last error when trying to build Qt library, it dies in PHP.
> PHP dies with an error of a non working xml.

I've had a similar show-stopper, until I realized that I had
WITH_THREADS set in /etc/make.conf (to get a threaded perl),
which was then picked up by the libxml2 port. After recompiling
and reinstalling libxml2 without WITH_THREADS, php 5.1.6 port
compiled without any problems.

Regards,
-cpghost.

-- 
Cordula's Web. http://www.cordula.ws/
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "[EMAIL PROTECTED]"