date:20120817

CVSROOT:/cvs
Module name:ports
Changes by: ajacou...@cvs.openbsd.org   2012/08/17 05:41:34

Modified files:
x11/polkit-gnome: Makefile 

Log message:
Remove uneeded CONFIGURE_ARGS and BUILD_DEPENDS.

CVS: cvs.openbsd.org: ports

2012-08-17 Thread Stuart Henderson

CVSROOT:/cvs
Module name:ports
Changes by: st...@cvs.openbsd.org   2012/08/17 07:30:16

Modified files:
textproc/mupdf : Makefile distinfo 
textproc/mupdf/patches: patch-Makerules 
textproc/mupdf/pkg: PLIST 
Removed files:
textproc/mupdf/patches: patch-apps_mupdfclean_c 
patch-apps_mupdfextract_c 
patch-apps_mupdfinfo_c 
patch-apps_mupdfshow_c 

Log message:
update to MuPDF 1.1

CVS: cvs.openbsd.org: ports

CVSROOT:/cvs
Module name:ports
Changes by: ajacou...@cvs.openbsd.org   2012/08/17 08:08:25

Modified files:
sysutils/consolekit: Makefile 

Log message:
Fix WANTLIB.

CVS: cvs.openbsd.org: ports

2012-08-17 Thread Gonzalo L. Rodriguez

CVSROOT:/cvs
Module name:ports
Changes by: gonz...@cvs.openbsd.org 2012/08/17 07:57:59

Modified files:
textproc/diffstat: Makefile distinfo 

Log message:
Little update for diffstat to 1.55.

Tested on i386.

Ok sthen@

CVS: cvs.openbsd.org: ports

CVSROOT:/cvs
Module name:ports
Changes by: ajacou...@cvs.openbsd.org   2012/08/17 07:54:26

Modified files:
print/foo2zjs  : Makefile distinfo 
print/foo2zjs/patches: patch-Makefile patch-getweb_in 
print/foo2zjs/pkg: PLIST-main 

Log message:
Update to foo2zjs-20120601.

CVS: cvs.openbsd.org: ports

2012-08-17 Thread Stuart Cassoff

CVSROOT:/cvs
Module name:ports
Changes by: s...@cvs.openbsd.org2012/08/17 08:48:34

Modified files:
emulators/frodo: Makefile distinfo 

Log message:
Use MODULES for Tcl/Tk.

Ok espie@ (maintainer)

CVS: cvs.openbsd.org: ports

2012-08-17 Thread Stefan Sperling

CVSROOT:/cvs
Module name:ports
Changes by: s...@cvs.openbsd.org2012/08/17 09:33:33

Log message:
Import ardour-2.8.14, with added support for sndio midi.

Ardour is a full-featured, free and open-source hard disk recorder and
digital audio workstation program suitable for professional use. It
features unlimited audio tracks and buses, non-destructive, non-linear
editing with unlimited undo, and anything-to-anywhere signal routing. It
supports standard file formats, such as BWF, WAV, WAV64, AIFF and CAF,
and it can use LADSPA, LV2, VST and AudioUnit plugin formats.

with and ok ajacoutot@

Status:

Vendor Tag: stsp
Release Tags:   stsp_20120817

N ports/audio/ardour/distinfo
N ports/audio/ardour/Makefile
N ports/audio/ardour/pkg/PLIST
N ports/audio/ardour/pkg/DESCR
N ports/audio/ardour/patches/patch-gtk2_ardour_ardour_ui_cc
N ports/audio/ardour/patches/patch-gtk2_ardour_engine_dialog_cc
N ports/audio/ardour/patches/patch-libs_midi++2_midiport_cc
N ports/audio/ardour/patches/patch-libs_pbd_SConscript
N ports/audio/ardour/patches/patch-libs_ardour_source_cc
N ports/audio/ardour/patches/patch-gtk2_ardour_SConscript
N ports/audio/ardour/patches/patch-libs_midi++2_midi++_port_h
N ports/audio/ardour/patches/patch-gtk2_ardour_ardour_sh_in
N ports/audio/ardour/patches/patch-libs_midi++2_SConscript
N ports/audio/ardour/patches/patch-gtk2_ardour_engine_dialog_h
N ports/audio/ardour/patches/patch-gtk2_ardour_new_session_dialog_cc
N ports/audio/ardour/patches/patch-SConstruct
N ports/audio/ardour/patches/patch-libs_midi++2_midifactory_cc
N ports/audio/ardour/patches/patch-libs_ardour_SConscript
N ports/audio/ardour/files/sndio_midiport.cc
N ports/audio/ardour/files/sndio_midiport.h

No conflicts created by this import

CVS: cvs.openbsd.org: ports

2012-08-17 Thread Stefan Sperling

CVSROOT:/cvs
Module name:ports
Changes by: s...@cvs.openbsd.org2012/08/17 09:36:03

Modified files:
audio  : Makefile 

Log message:
+= ardour

CVS: cvs.openbsd.org: ports

2012-08-17 Thread Stuart Cassoff

CVSROOT:/cvs
Module name:ports
Changes by: s...@cvs.openbsd.org2012/08/17 10:05:03

Modified files:
databases/sqlite: Makefile distinfo 

Log message:
Use MODULES for Tcl/Tk.

Ok espie@

CVS: cvs.openbsd.org: ports

2012-08-17 Thread Landry Breuil

CVSROOT:/cvs
Module name:ports
Changes by: lan...@cvs.openbsd.org  2012/08/17 15:50:28

ports/devel/py-greenlet/patches

Update of /cvs/ports/devel/py-greenlet/patches
In directory cvs.openbsd.org:/tmp/cvs-serv16082/patches

Log Message:
Directory /cvs/ports/devel/py-greenlet/patches added to the repository

CVS: cvs.openbsd.org: ports

2012-08-17 Thread Landry Breuil

CVSROOT:/cvs
Module name:ports
Changes by: lan...@cvs.openbsd.org  2012/08/17 15:51:49

Modified files:
devel/py-greenlet: Makefile 
Added files:
devel/py-greenlet/patches: patch-slp_platformselect_h 

Log message:
Unbreak on ppc  give it a chance to build on mips*. Set ONLY_FOR_ARCHS
since it has platform-dependent code in slp_platformdetect.h..

breakage reported by ajacoutot@

CVS: cvs.openbsd.org: ports

CVSROOT:/cvs
Module name:ports
Changes by: es...@cvs.openbsd.org   2012/08/17 16:32:29

Modified files:
infrastructure/mk: bsd.port.mk 

Log message:
fix subpackage bugs.
- correct syntax for variable (Vadim Zhukov)
- both _DO_LOCK and _cache_fragment want to use traps.
Since that's the only place where the problem occurs, simply put the second
trap in a subshell...

CVS: cvs.openbsd.org: ports

CVSROOT:/cvs
Module name:ports
Changes by: jas...@cvs.openbsd.org  2012/08/17 01:11:39

Modified files:
audio/liblo: Makefile 

Log message:
- tweak COMMENT/SHARED_LIBS
- fix license marker

CVS: cvs.openbsd.org: ports

CVSROOT:/cvs
Module name:ports
Changes by: jas...@cvs.openbsd.org  2012/08/17 01:20:08

Modified files:
audio/aubio: Makefile 

Log message:
- add missing build dependency
- tweak MASTER_SITES

CVS: cvs.openbsd.org: ports

CVSROOT:/cvs
Module name:ports
Changes by: ajacou...@cvs.openbsd.org   2012/08/17 01:58:51

Modified files:
graphics/cairo : Makefile 
graphics/cairo/pkg: PLIST 

Log message:
Comment static archive and la file for the loadable module.
Add CONFIGURE_SHARED.
Remove REGRESS_DEPENDS as gs is actually needed at build time for the
tests (along with rsvg...).

ok eric@ (maintainer)

CVS: cvs.openbsd.org: ports

CVSROOT:/cvs
Module name:ports
Changes by: ajacou...@cvs.openbsd.org   2012/08/17 02:24:39

Modified files:
graphics/geeqie: Makefile distinfo 
graphics/geeqie/pkg: PLIST 

Log message:
Update to geeqie-1.1.

CVS: cvs.openbsd.org: ports

CVSROOT:/cvs
Module name:ports
Changes by: jas...@cvs.openbsd.org  2012/08/17 02:58:46

Modified files:
www/phplist: Makefile distinfo 
www/phplist/pkg: PLIST 
Added files:
www/phplist/pkg: README 

Log message:
security update to 2.10.19, fixes CVE-2012-3952
phplist unconfirmed Cross-Site Scripting Vulnerability

CVS: cvs.openbsd.org: ports

CVSROOT:/cvs
Module name:ports
Changes by: ajacou...@cvs.openbsd.org   2012/08/17 03:00:10

Modified files:
infrastructure/db: user.list 

Log message:
Reserve a _polkitd user/group for upcoming sysutils/polkit update.

CVS: cvs.openbsd.org: ports

CVSROOT:/cvs
Module name:ports
Changes by: jas...@cvs.openbsd.org  2012/08/17 03:09:01

Modified files:
editors/emacs23: Makefile 
Added files:
editors/emacs23/patches: patch-lisp_files_el 

Log message:
Security fix for CVE-2012-3479
GNU Emacs enable-local-variables Variable Processing Vulnerability

patch from upstream git; earlier releases are not affected

Ìàðêåòèíã â ñîöèàëüíûõ ñåòÿõ. ÌÎÄ (ìåõàíèçì îáðàçîâàíèÿ äîõîäà)
18-19 àâãóñòà â Êèåâå ïðîéäåò 12-÷àñîâûé òðåíèíã äèðåêòîðà ïåðâîãî â Ðîññèè
PR-àãåíñòâà GREENPR, ñïåöèàëèçèðîâàííîãî íà ìàðêåòèíãå â ñîöèàëüíûõ-ñåòÿõ
Äàìèðà Õàëèëîâà.
Äàìèð Õàëèëîâ - SMM - ñïåöèàëèñò ¹1 â ÑÍÃ.
Êëèåíòû Äàìèðà Õàëèëîâà: Pepsi, Sony Pictures, Universal, BenQ, Renault, ISIC,
Stockholm School Of Economics, Åâðîñåòü, Áàëòèêà.
Óíèêàëüíûé êóðñ, ïîñâÿùåííûé âñåì àñïåêòàì ïðîäâèæåíèÿ â ñîöèàëüíûõ ñåòÿõ
(áëîããèíã, Òâèòòåð, ïðîäâèæåíèå âî ÂÊîíòàêòå, ñêðûòûé ìàðêåòèíã, âèðóñíûé
ìàðêåòèíã è ò.ä.).
Â ïðîãðàììå:
1. Âîçìîæíîñòè ìàðêåòèíãà â ñîöèàëüíûõ ñåòÿõ (SMM)
2. Ïðîäâèæåíèå âî ÂÊîíòàêòå
3. Ïðîäâèæåíèå íà Facebook
4. Êîðïîðàòèâíûå áëîãè
5. Ïðîäâèæåíèå â Òâèòòåðå
6. Ðàáîòà ñ ëèäåðàìè ìíåíèé
7. Âèðóñíûé ìàðêåòèíã
8. Ñêðûòûé ìàðêåòèíã
9. Íåéòðàëèçàöèÿ íåãàòèâà
10. Ìîíèòîðèíã è îöåíêà ýôôåêòèâíîñòè
Áèçíåñ-òðåíåð:
Äàìèð Õàëèëîâ - ãåíåðàëüíûé äèðåêòîð àãåíòñòâà GreenPR. Êóðñ îñíîâàí íà
ìíîãîëåòíåé ïðàêòèêå è âìåùàåò â ñåáÿ âåñü îáúåì ñîâðåìåííîãî SMM (Social
Media Marketing).

Ñ óâàæåíèåì,
© 2007 Áèçíåñ-Öåíòð Íàöèîíàëüíûé
Telephone: (044) 221-23-85/221-23-87 | Mailto: off...@send.seminar1.net.ua
| Www: seminar1.org.ua

Ýòî ïèñüìî ÿâëÿåòñÿ èíôîðìàöèåé äëÿ îçíàêîìëåíèÿ ïîëüçîâàòåëÿ. Â ñëó÷àå, åñëè
Âû õîòèòå ïðèîñòàíîâèòü ïîëó÷åíèå êîììåð÷åñêèõ ïðåäëîæåíèé, ïðîñèì ñîîáùèòü
íàì http://seminar1.org.ua/delete.php c óêàçàíèåì ñðîêîâ, íà êîòîðûå
íåîáõîäèìî îñòàíîâèòü îòïðàâêó ïî÷òû.

The poppler,-qt4 problem

2012-08-17 Thread Christian Weisgerber

Recent changes have led to a new chain of dependencies where gtk+[23] now
depends on qt4:

  x11/gtk+[23]
print/cups,-libs
  print/cups-filters
print/poppler
  x11/qt4

For one thing, this is annoying for casual ports building.  You are
building something that requires gtk+2 and suddenly your build takes
a detour of several hours, if not longer, into qt4.

But we don't care about this, says the official line, the tree
is optimized for bulk package builds, yada yada.

Well, in bulk builds, the libreoffice build also can't start until
qt4 has been built.  And on the amd64*.ports setup, with eight build
slots in dpb, the accumulative time of sequential qt4 and libreoffice
builds is starting to get close to the total time required for a
bulk build.  If I had another build machine or more CPU cores, these
would run idle at the end of the build, with everything else already
finished but libreoffice still chugging along.

We need to come up with something involving the no_qt4 flavor or a
bootstrap flavor to break this dependency chain.  If that means
building poppler twice in bulk builds, then that's fine.

-- 
Christian naddy Weisgerber  na...@mips.inka.de

Re: UPDATE: Zile-2.4.8

2012-08-17 Thread Christian Weisgerber

Gonzalo L. R. gonz...@x61.com.ar wrote:

 Trivial update for Zile to 2.4.8 with some bug fixed.
 Tested on amd64.

Doesn't work for me.  When I modify a buffer and try to exit with
^X^C zile starts to behave strangely.  I guess it prompts me if I
want to exit, except that there is no visible prompt.

 Comments?

patch-configure can now be removed completely.

-- 
Christian naddy Weisgerber  na...@mips.inka.de

Re: The poppler,-qt4 problem

On Fri, Aug 17, 2012 at 05:30:26PM +0200, Christian Weisgerber wrote:
 Recent changes have led to a new chain of dependencies where gtk+[23] now
 depends on qt4:
 
   x11/gtk+[23]
 print/cups,-libs
   print/cups-filters
 print/poppler
   x11/qt4
 
 For one thing, this is annoying for casual ports building.  You are
 building something that requires gtk+2 and suddenly your build takes
 a detour of several hours, if not longer, into qt4.
 
 But we don't care about this, says the official line, the tree
 is optimized for bulk package builds, yada yada.
 
 Well, in bulk builds, the libreoffice build also can't start until
 qt4 has been built.  And on the amd64*.ports setup, with eight build
 slots in dpb, the accumulative time of sequential qt4 and libreoffice
 builds is starting to get close to the total time required for a
 bulk build.  If I had another build machine or more CPU cores, these
 would run idle at the end of the build, with everything else already
 finished but libreoffice still chugging along.
 
 We need to come up with something involving the no_qt4 flavor or a
 bootstrap flavor to break this dependency chain.  If that means
 building poppler twice in bulk builds, then that's fine.

Can't we just set the default FLAVOR to no_* ?

-- 
Antoine

hello

2012-08-17 Thread Edgar Joyce

How are the things.

Re: The poppler,-qt4 problem

On Fri, Aug 17, 2012 at 06:05:06PM +0200, Antoine Jacoutot wrote:
 On Fri, Aug 17, 2012 at 05:30:26PM +0200, Christian Weisgerber wrote:
  Recent changes have led to a new chain of dependencies where gtk+[23] now
  depends on qt4:
  
x11/gtk+[23]
  print/cups,-libs
print/cups-filters
  print/poppler
x11/qt4
  
  For one thing, this is annoying for casual ports building.  You are
  building something that requires gtk+2 and suddenly your build takes
  a detour of several hours, if not longer, into qt4.
  
  But we don't care about this, says the official line, the tree
  is optimized for bulk package builds, yada yada.
  
  Well, in bulk builds, the libreoffice build also can't start until
  qt4 has been built.  And on the amd64*.ports setup, with eight build
  slots in dpb, the accumulative time of sequential qt4 and libreoffice
  builds is starting to get close to the total time required for a
  bulk build.  If I had another build machine or more CPU cores, these
  would run idle at the end of the build, with everything else already
  finished but libreoffice still chugging along.
  
  We need to come up with something involving the no_qt4 flavor or a
  bootstrap flavor to break this dependency chain.  If that means
  building poppler twice in bulk builds, then that's fine.
 
 Can't we just set the default FLAVOR to no_* ?

that will solve things for casual ports builders. Which is already
half the problem.

Re: The poppler,-qt4 problem

On Fri, Aug 17, 2012 at 06:05:06PM +0200, Antoine Jacoutot wrote:
 On Fri, Aug 17, 2012 at 05:30:26PM +0200, Christian Weisgerber wrote:
  Recent changes have led to a new chain of dependencies where gtk+[23] now
  depends on qt4:
  
x11/gtk+[23]
  print/cups,-libs
print/cups-filters
  print/poppler
x11/qt4
  
  For one thing, this is annoying for casual ports building.  You are
  building something that requires gtk+2 and suddenly your build takes
  a detour of several hours, if not longer, into qt4.
  
  But we don't care about this, says the official line, the tree
  is optimized for bulk package builds, yada yada.
  
  Well, in bulk builds, the libreoffice build also can't start until
  qt4 has been built.  And on the amd64*.ports setup, with eight build
  slots in dpb, the accumulative time of sequential qt4 and libreoffice
  builds is starting to get close to the total time required for a
  bulk build.  If I had another build machine or more CPU cores, these
  would run idle at the end of the build, with everything else already
  finished but libreoffice still chugging along.
  
  We need to come up with something involving the no_qt4 flavor or a
  bootstrap flavor to break this dependency chain.  If that means
  building poppler twice in bulk builds, then that's fine.
 
 Can't we just set the default FLAVOR to no_* ?


Actually, you're shifting the problem away from the actual offender.

cups-filter that depends on print/poppler.

So cups-filter can switch to print/poppler,no_qt4

and if naddy wants, you can add a bootstrap there as well, but it needs
documenting that this is NOT a build loop, just a hint to avoid a critical
path.

In case you don't remember, full bulk builds do zap  no_* pseudo-flavors 
unless there's a bootstrap pseudo flavor involved.

Re: The poppler,-qt4 problem

2012-08-17 Thread Kenneth R Westerback

On Fri, Aug 17, 2012 at 05:30:26PM +0200, Christian Weisgerber wrote:
 Recent changes have led to a new chain of dependencies where gtk+[23] now
 depends on qt4:
 
   x11/gtk+[23]
 print/cups,-libs
   print/cups-filters
 print/poppler
   x11/qt4
 
 For one thing, this is annoying for casual ports building.  You are
 building something that requires gtk+2 and suddenly your build takes
 a detour of several hours, if not longer, into qt4.
 
 But we don't care about this, says the official line, the tree
 is optimized for bulk package builds, yada yada.
 
 Well, in bulk builds, the libreoffice build also can't start until
 qt4 has been built.  And on the amd64*.ports setup, with eight build
 slots in dpb, the accumulative time of sequential qt4 and libreoffice
 builds is starting to get close to the total time required for a
 bulk build.  If I had another build machine or more CPU cores, these
 would run idle at the end of the build, with everything else already
 finished but libreoffice still chugging along.
 
 We need to come up with something involving the no_qt4 flavor or a
 bootstrap flavor to break this dependency chain.  If that means
 building poppler twice in bulk builds, then that's fine.
 
 -- 
 Christian naddy Weisgerber  na...@mips.inka.de
 

Just had my normal macppc ports build jump from 12 hours to 36 hours, in
no small part because qt4 took a day or so to build. So I'd be interested
in cutting this back to 12 hours. :-)

 Ken

switch GNUstep based ports to use libobjc2

2012-08-17 Thread Sebastian Reitenbach

Hi,

the patch below tweaks x11/gnustep/make and x11/gnustep/base to use the new 
Objective-C runtime provided by x11/gnustep/libobjc2 instead of the old 
libobjc runtime from gcc. gnustep-base had a compatibility layer on top of the 
old runtime, which
now more or less moved to libobjc2, therefore it required a major library bump 
there.
The rest is just bumping the REVISION of all the ports in x11/gnustep and the 
tree ports depending on it, not in the x11/gnustep
tree which are: games/oolite, www/sope and www/sogo.

tested on all GNUstep supported platforms (i386, amd64, macppc). Whereas the 
GUI stuff was mostly tested on i386 and macppc, and on amd64 the server stuff, 
i.e. www/sogo in conjunction with www/sope. I started testing and working on it 
probably a year ago,
so had different versions of gnustep and libobjc2 running before. I think all 
the flaws encountered while those testing is gone, and 
its now fine to do the switch.

But, still required is the import of x11/gnustep/libobjc2, which I sent here a 
couple of days ago.


Index: games/oolite/Makefile
===
RCS file: /cvs/ports/games/oolite/Makefile,v
retrieving revision 1.6
diff -u -r1.6 Makefile
--- games/oolite/Makefile   8 Jul 2012 14:15:46 -   1.6
+++ games/oolite/Makefile   17 Aug 2012 17:20:26 -
@@ -3,7 +3,7 @@
 COMMENT=   space combat and trading game in the style of Elite
 
 VERSION=   1.73.4
-REVISION = 2
+REVISION = 3
 DISTNAME=  oolite-dev-source-${VERSION}
 PKGNAME=   oolite-${VERSION}
 CATEGORIES=games
@@ -27,7 +27,7 @@
 LIB_DEPENDS=   devel/sdl-mixer \
graphics/png
 
-WANTLIB += GL GLU SDL c m pthread SDL_mixer objc gnustep-base png
+WANTLIB += GL GLU SDL c m pthread SDL_mixer objc2 gnustep-base png
 
 MAKE_FILE =Makefile
 ALL_TARGET=release
Index: www/sope/Makefile
===
RCS file: /cvs/ports/www/sope/Makefile,v
retrieving revision 1.22
diff -u -r1.22 Makefile
--- www/sope/Makefile   17 Aug 2012 09:57:30 -  1.22
+++ www/sope/Makefile   17 Aug 2012 17:20:26 -
@@ -5,6 +5,7 @@
 COMMENT-postgres=  SOPE PostgreSQL adaptor
 
 VERSION =  1.3.17
+REVISION = 0
 DISTNAME = SOPE-${VERSION}
 PKGNAME-main = sope-${VERSION}
 PKGNAME-mysql =sope-mysql-${VERSION}
Index: www/sogo/Makefile
===
RCS file: /cvs/ports/www/sogo/Makefile,v
retrieving revision 1.20
diff -u -r1.20 Makefile
--- www/sogo/Makefile   17 Aug 2012 09:59:08 -  1.20
+++ www/sogo/Makefile   17 Aug 2012 17:20:26 -
@@ -3,6 +3,7 @@
 COMMENT =  web based groupware server
 
 VERSION =  1.3.17
+REVISION = 0
 DISTNAME = SOGo-${VERSION}
 PKGNAME =  sogo-${VERSION}
 
Index: x11/gnustep/gnustep.port.mk
===
RCS file: /cvs/ports/x11/gnustep/gnustep.port.mk,v
retrieving revision 1.16
diff -u -r1.16 gnustep.port.mk
--- x11/gnustep/gnustep.port.mk 19 Jul 2011 06:19:49 -  1.16
+++ x11/gnustep/gnustep.port.mk 17 Aug 2012 17:20:26 -
@@ -25,7 +25,7 @@
 MODGNUSTEP_NEEDS_BACK ?=   Yes
 
 .if ${MODGNUSTEP_NEEDS_GUI:L} == yes 
-MODGNUSTEP_WANTLIB +=  objc gnustep-base gnustep-gui
+MODGNUSTEP_WANTLIB +=  objc2 gnustep-base gnustep-gui
 MODGNUSTEP_LIB_DEPENDS +=  x11/gnustep/gui
 .  if ${MODGNUSTEP_NEEDS_BACK:L} == yes
 MODGNUSTEP_RUN_DEPENDS +=  x11/gnustep/back
Index: x11/gnustep/aclock/Makefile
===
RCS file: /cvs/ports/x11/gnustep/aclock/Makefile,v
retrieving revision 1.5
diff -u -r1.5 Makefile
--- x11/gnustep/aclock/Makefile 8 Jul 2012 14:15:11 -   1.5
+++ x11/gnustep/aclock/Makefile 17 Aug 2012 17:20:26 -
@@ -3,7 +3,7 @@
 COMMENT =  analog clock for the GNUstep desktop
 
 DISTNAME = AClock-0.4.0
-REVISION = 0
+REVISION = 1
 
 HOMEPAGE = http://gap.nongnu.org/aclock/
 MASTER_SITES = ${MASTER_SITE_SAVANNAH:=gap/}
Index: x11/gnustep/addresses/Makefile
===
RCS file: /cvs/ports/x11/gnustep/addresses/Makefile,v
retrieving revision 1.13
diff -u -r1.13 Makefile
--- x11/gnustep/addresses/Makefile  6 Aug 2012 11:07:09 -   1.13
+++ x11/gnustep/addresses/Makefile  17 Aug 2012 17:20:27 -
@@ -3,6 +3,7 @@
 COMMENT=   GNUstep address book
 
 DISTNAME=  Addresses-0.4.8
+REVISION = 0
 
 SHARED_LIBS+=  Addresses 0.1   # .0.1
 SHARED_LIBS+=  AddressView 0.1 # .0.1
Index: x11/gnustep/affiche/Makefile
===
RCS file: /cvs/ports/x11/gnustep/affiche/Makefile,v
retrieving revision 1.2
diff -u -r1.2 Makefile
--- x11/gnustep/affiche/Makefile8 Jul 2012 14:15:11 -   1.2
+++

Re: UPDATE: Zile-2.4.8

2012-08-17 Thread Gonzalo L. R.

new diff without the patch, but I can't reproduce what your scenario.

On Fri, Aug 17, 2012 at 03:44:37PM +, Christian Weisgerber wrote:
; Gonzalo L. R. gonz...@x61.com.ar wrote:
; 
;  Trivial update for Zile to 2.4.8 with some bug fixed.
;  Tested on amd64.
; 
; Doesn't work for me.  When I modify a buffer and try to exit with
; ^X^C zile starts to behave strangely.  I guess it prompts me if I
; want to exit, except that there is no visible prompt.
; 
;  Comments?
; 
; patch-configure can now be removed completely.
; 
; -- 
; Christian naddy Weisgerber  na...@mips.inka.de
; 

-- 
Sending from my VCR...
? bla
Index: Makefile
===
RCS file: /cvs/ports/editors/zile/Makefile,v
retrieving revision 1.39
diff -u -p -r1.39 Makefile
--- Makefile27 Mar 2012 22:01:50 -  1.39
+++ Makefile17 Aug 2012 18:24:05 -
@@ -1,11 +1,11 @@
 # $OpenBSD: Makefile,v 1.39 2012/03/27 22:01:50 ajacoutot Exp $
 
 COMMENT=   lightweight Emacs clone
-DISTNAME=  zile-2.4.7
+DISTNAME=  zile-2.4.8
 CATEGORIES=editors
 HOMEPAGE=  http://www.gnu.org/software/zile/
 
-MAINTAINER=Gonzalo L. R. gonz...@x61.com.ar
+MAINTAINER=Gonzalo L. R. gonz...@openbsd.org
 
 # GPLv3
 PERMIT_PACKAGE_CDROM=  Yes
Index: distinfo
===
RCS file: /cvs/ports/editors/zile/distinfo,v
retrieving revision 1.33
diff -u -p -r1.33 distinfo
--- distinfo27 Mar 2012 22:01:50 -  1.33
+++ distinfo17 Aug 2012 18:24:05 -
@@ -1,5 +1,2 @@
-MD5 (zile-2.4.7.tar.gz) = lTiwIQx/z1bPhSyxoCu2yQ==
-RMD160 (zile-2.4.7.tar.gz) = aIN5c42Na73Un57CCP42Y0DxGxE=
-SHA1 (zile-2.4.7.tar.gz) = MMR6OZuUtdzmiheP6YgH+GcZpGY=
-SHA256 (zile-2.4.7.tar.gz) = jPUy/ki1ygniz2er96vjNkLrevTnzYvR65ATfZbEZ6o=
-SIZE (zile-2.4.7.tar.gz) = 1254385
+SHA256 (zile-2.4.8.tar.gz) = t5P04GTjzcrC9ypCsWjFzA2m99B4uKrEWLfTy6z3vEk=
+SIZE (zile-2.4.8.tar.gz) = 1184855
Index: patches/patch-configure
===
RCS file: patches/patch-configure
diff -N patches/patch-configure
--- patches/patch-configure 27 Mar 2012 22:01:50 -  1.4
+++ /dev/null   1 Jan 1970 00:00:00 -
@@ -1,12 +0,0 @@
-$OpenBSD: patch-configure,v 1.4 2012/03/27 22:01:50 ajacoutot Exp $
 configure.orig Tue Mar 20 16:43:43 2012
-+++ configure  Sun Mar 25 13:45:47 2012
-@@ -30805,7 +30805,7 @@ else
-   gc_ok=no
- fi
- 
--gc_libs=-lpthread
-+gc_libs=-pthread
- { $as_echo $as_me:${as_lineno-$LINENO}: checking for GC_malloc in -lgc 
5
- $as_echo_n checking for GC_malloc in -lgc...  6; }
- if ${ac_cv_lib_gc_GC_malloc+:} false; then :

[UPDATE] emulators/spim 8.0

2012-08-17 Thread Azwaw OUSADOU

Hi !
This is an update to the latest version of spim, a MIPS simulator.
It works fine on amd64 and i386 with many programs.

test2.asm is proving that spim works well.

No replies of the maintainer.
Please test it.

Azwaw OUSADOU


spim2.diff
Description: Binary data


test2.asm
Description: Binary data

Re: The poppler,-qt4 problem

2012-08-17 Thread Brian Callahan


On 8/17/2012 1:22 PM, Kenneth R Westerback wrote:

On Fri, Aug 17, 2012 at 05:30:26PM +0200, Christian Weisgerber wrote:

Recent changes have led to a new chain of dependencies where gtk+[23] now
depends on qt4:

   x11/gtk+[23]
 print/cups,-libs
   print/cups-filters
 print/poppler
   x11/qt4

For one thing, this is annoying for casual ports building.  You are
building something that requires gtk+2 and suddenly your build takes
a detour of several hours, if not longer, into qt4.

But we don't care about this, says the official line, the tree
is optimized for bulk package builds, yada yada.

Well, in bulk builds, the libreoffice build also can't start until
qt4 has been built.  And on the amd64*.ports setup, with eight build
slots in dpb, the accumulative time of sequential qt4 and libreoffice
builds is starting to get close to the total time required for a
bulk build.  If I had another build machine or more CPU cores, these
would run idle at the end of the build, with everything else already
finished but libreoffice still chugging along.

We need to come up with something involving the no_qt4 flavor or a
bootstrap flavor to break this dependency chain.  If that means
building poppler twice in bulk builds, then that's fine.

--
Christian naddy Weisgerber  na...@mips.inka.de



Just had my normal macppc ports build jump from 12 hours to 36 hours, in
no small part because qt4 took a day or so to build. So I'd be interested
in cutting this back to 12 hours. :-)

 Ken



Also, some arches that can build gtk+[2,3] cannot build qt4 or qt3, like 
mips64el. So it would be quite devastating to the packages for those 
arches. Unless the bulk builds have some way of compensating for this (I 
don't know).


Alternatively, NOT_FOR_ARCHES-qt4 = mips64el
and NOT_FOR_ARCHES-qt = mips64el
should be added to poppler's port Makefile. It looks like this has 
already been done for arm.


Thanks.

~Brian

Re: The poppler,-qt4 problem