Status of -STABLE maintainer?
Hey all, I was just curious, has anyone been found that was willing and able to take over the job of maintaining the -STABLE ports tree? -- ((name Aaron Hsu) (email/xmpp [EMAIL PROTECTED]) (site http://www.aaronhsu.com;))
Re: Status of -STABLE maintainer?
On Fri, 16 Nov 2007, Aaron Hsu wrote: I was just curious, has anyone been found that was willing and able to take over the job of maintaining the -STABLE ports tree? No, -stable ports should be considerer unmaintained. -- Antoine
[update] opencdk-0.6.5 and gnutls-2.0.3
Hi, this submission is based on a previous work from Giovanni Bechis, thanks to him. These two diffs updates opencdk from 0.5.5 to 0.6.5 and gnutls from 1.0.25 to 2.0.3 (recently released). make regress fails (at least on i386) on opencdk at t-key test: t-key.c:516 expired key test FAILED t-key.c:522 missing key test FAILED FAIL: t-key It didn't fail with 2.0.1 and 2.0.2, i have to figure out what changed. make regress fails too for gnutls on pathlen test: make check-TESTS --- ./no-ca-or-pathlen.pem Sat Nov 10 18:17:51 2007 +++ new-no-ca-or-pathlen.pemFri Nov 16 11:36:54 2007 @@ -22,7 +22,7 @@ Basic Constraints (not critical): Certificate Authority (CA): FALSE Unknown extension 2.5.29.32 (not critical): - ASCII: 0;09..`.H...E0*0(..+.https://www.verisign.com/rpa + ASCII: 0;09..`.H..øE0*0(..+.https://www.verisign.com/rpa Hexdump: 303b3039060b6086480186f84501070108302a302806082b06010505070201161c68747470733a2f2f772e766572697369676e2e636f6d2f727061 Unknown extension 2.16.840.1.113730.1.1 (not critical): ASCII: FAIL: pathlen Apparently this one is not important, it's related to our isprint() implem vs linux one. Ports depending on gnutls have been tested (mostly at p2k7) with this upgrade and still works fine : thanks sthen@ for testing iksemel and asterisk, thanks aanriot@ for testing prelude and friends. liferea still builds and run fine, wmbiff has already been updated to support it. I'd really like to commit this soon, so please report any other error messages you get on other archs. diffs : http://gcu.info/~gaston/patches/opencdk-0.6.5.diff http://gcu.info/~gaston/patches/gnutls-2.0.3.diff Thanks for testing and feedback. Landry
Re: Could anyone get Neils Provos's SpyBye compiled successfully on 4.2/amd64
On Nov 14, 2007 2:19 PM, Stuart Henderson [EMAIL PROTECTED] wrote: from a point-of-view of porting this (rather than hand-building), does anyone have a clue what to do with libevent? (spybye uses the http parts which are disabled in src/) So finally I got SpyBye compiled like this http://www.bsd-india.org/pipermail/bsd-india/2007-November/001496.html http://www.bsd-india.org/pipermail/bsd-india/2007-November/001496.html Can this idea be adopted to port spybye? Many thanks to Raja Subramanian in BSD-INDIA list and to all who helped here :-) Kind Regards Siju
NEW: databases/mysqltcl
COMMENT=Tcl interface to MySQL MySQLTcl is a Tcl library which makes possible to do SQL queries (select, insert, delete...) to a MySQL database server from Tcl scripts. http://secure.lv/~nikns/stuff/ports/mysqltcl-3.02.tar mysqltcl-3.02.tar Description: Unix tar archive
UPDATE: KTorrent 2.2.3
On November 11, 2007 07:23:15 pm Brad wrote:
Here is an update to KTorrent 2.2.2.
Here is an update to KTorrent 2.2.3.
Index: Makefile
===
RCS file: /cvs/ports/net/ktorrent/Makefile,v
retrieving revision 1.3
diff -u -p -r1.3 Makefile
--- Makefile8 Nov 2007 21:13:10 - 1.3
+++ Makefile16 Nov 2007 09:39:24 -
@@ -4,9 +4,8 @@ SHARED_ONLY=Yes
COMMENT= BitTorrent client for KDE
-VERSION= 2.1.4
+VERSION= 2.2.3
DISTNAME= ktorrent-${VERSION}
-PKGNAME= ${DISTNAME}p0
CATEGORIES=net x11 x11/kde
MASTER_SITES= ${HOMEPAGE}downloads/${VERSION}/
@@ -20,9 +19,9 @@ PERMIT_PACKAGE_FTP= Yes
PERMIT_DISTFILES_CDROM=Yes
PERMIT_DISTFILES_FTP= Yes
WANTLIB= art_lgpl_2 jpeg ICE SM X11 Xau Xdmcp Xext Xrender c m
pcre \
- pcreposix png pthread stdc++ util z idn iconv fam rpcsvc
+ pcreposix png pthread stdc++ util z idn iconv intl fam
rpcsvc
-LIB_DEPENDS=
kdecore.=6,DCOP,kdefx,kdeprint,kdesu,kparts,kdeui,khtml,kio,kjs,kmdi,kutils,kwalletclient::x11/kde/libs3
\
+LIB_DEPENDS=
kdecore.=6,DCOP,kdefx,kdeprint,kdesu,kparts,kdeui,khtml,kio,kjs,kutils,kwalletclient::x11/kde/libs3
\
gmp.=4.::devel/gmp
MODULES= x11/kde
Index: distinfo
===
RCS file: /cvs/ports/net/ktorrent/distinfo,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 distinfo
--- distinfo19 Jul 2007 02:59:44 - 1.1.1.1
+++ distinfo16 Nov 2007 09:39:39 -
@@ -1,5 +1,5 @@
-MD5 (ktorrent-2.1.4.tar.gz) = adPX32IAcQGBWU2hmrjCSA==
-RMD160 (ktorrent-2.1.4.tar.gz) = goYBLl+8C9gn1ctwx4leVGDyGgs=
-SHA1 (ktorrent-2.1.4.tar.gz) = RUmS7F8LT2fmJHTPhjotZtvXfyg=
-SHA256 (ktorrent-2.1.4.tar.gz) = 1JL3WAHBgCAumcqcnN/wtU4vXY5K4t3cy/GqwhgnbEg=
-SIZE (ktorrent-2.1.4.tar.gz) = 3631999
+MD5 (ktorrent-2.2.3.tar.gz) = f/Rn7pRD+fpjdpzYpkCTLg==
+RMD160 (ktorrent-2.2.3.tar.gz) = zopuFv2NSCgZTvYNpF/sUh9Pma0=
+SHA1 (ktorrent-2.2.3.tar.gz) = N6p1TV30Cc6YfktwDTxptGF11Kg=
+SHA256 (ktorrent-2.2.3.tar.gz) = JBrM+glMRkBjzA8X4n880ry49vRt+S9MqOrF+QZOj8I=
+SIZE (ktorrent-2.2.3.tar.gz) = 3842249
Index: patches/patch-libktorrent_net_address_cpp
===
RCS file: patches/patch-libktorrent_net_address_cpp
diff -N patches/patch-libktorrent_net_address_cpp
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-libktorrent_net_address_cpp 8 Nov 2007 23:09:08 -
@@ -0,0 +1,11 @@
+$OpenBSD$
+--- libktorrent/net/address.cpp.orig Sat Sep 29 01:31:27 2007
libktorrent/net/address.cppSat Sep 29 01:31:39 2007
+@@ -17,6 +17,7 @@
+ * Free Software Foundation, Inc., *
+ * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. *
+ ***/
++#include sys/types.h
+ #include netinet/in.h
+ #include arpa/inet.h
+ #include address.h
Index: patches/patch-libktorrent_net_socket_cpp
===
RCS file: /cvs/ports/net/ktorrent/patches/patch-libktorrent_net_socket_cpp,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 patch-libktorrent_net_socket_cpp
--- patches/patch-libktorrent_net_socket_cpp19 Jul 2007 02:59:44 -
1.1.1.1
+++ patches/patch-libktorrent_net_socket_cpp16 Nov 2007 09:43:01 -
@@ -1,6 +1,6 @@
$OpenBSD: patch-libktorrent_net_socket_cpp,v 1.1.1.1 2007/07/19 02:59:44
deanna Exp $
libktorrent/net/socket.cpp.origSun Apr 22 07:25:46 2007
-+++ libktorrent/net/socket.cpp Mon May 7 17:45:09 2007
+--- libktorrent/net/socket.cpp.origTue Nov 13 14:36:49 2007
libktorrent/net/socket.cpp Fri Nov 16 04:42:46 2007
@@ -38,7 +38,7 @@
#include sys/filio.h
#endif
@@ -10,12 +10,12 @@ $OpenBSD: patch-libktorrent_net_socket_c
#define MSG_NOSIGNAL 0
#endif
-@@ -258,7 +258,7 @@ namespace net
+@@ -260,7 +260,7 @@ namespace net
- bool Socket::setTOS(char type_of_service)
+ bool Socket::setTOS(unsigned char type_of_service)
{
-#if defined(Q_OS_MACX) || defined(Q_OS_DARWIN) || (defined(Q_OS_FREEBSD)
__FreeBSD_version 600020) || defined(Q_OS_NETBSD) || defined(Q_OS_BSD4)
+#if defined(Q_OS_MACX) || defined(Q_OS_DARWIN) || (defined(Q_OS_FREEBSD)
__FreeBSD_version 600020) || defined(Q_OS_NETBSD) || defined(Q_OS_OPENBSD) ||
defined(Q_OS_BSD4)
- int c = type_of_service;
+ unsigned int c = type_of_service;
#else
- char c = type_of_service;
+ unsigned char c = type_of_service;
Index: patches/patch-libktorrent_net_speed_cpp
===
RCS file: patches/patch-libktorrent_net_speed_cpp
diff -N patches/patch-libktorrent_net_speed_cpp
--- patches/patch-libktorrent_net_speed_cpp 8
Re: www/php5, what's going on?
On 2007/11/16 11:58, Alexey Suslikov wrote: I see neither /var/www/conf/modules.sample/, nor /var/www/conf/modules/. What can I do to fix it? Install a new base snapshot.
Status of Xplore?
Hey all, I was wondering what the status was of importing the Xplore port which I submitted a bit ago. I know sometimes ports fall through the cracks, and I just wanted to see if there was any reason other than that for it not being imported? If so, I'll try to get it fixed, but if it just slipped through the cracks, well . . . :-). Thanks! -- ((name Aaron Hsu) (email/xmpp [EMAIL PROTECTED]) (site http://www.aaronhsu.com;))
Re: NEW: net/p3scan
On 2007/11/16 15:24, Soner Tari wrote:
On Thu, 2007-11-15 at 22:00 +, Stuart Henderson wrote:
On 2007/11/15 23:04, Soner Tari wrote:
I've checked the other options like the one you are mentioning, but
handling this in a separate shell script seems better to me (for example
it's more readable and manageable in my opinion). Also note that I took
postfix files as the basis for my p3scan-config. Therefore, if there is
no harm, I'd like to keep p3scan-config.
This is the wrong way to go, the pkg_* tools can handle this, so
let them. @exec is useful for cases where you can't use the normal
mechanisms, it's available but it should almost never be necessary.
You don't see it used very much in our ports tree.
Postfix has to populate the chroot with files from the installed
system, and has its own mechanism to upgrade config files between
versions via /etc/postfix/post-install. This goes to make it a
bit of a special case... but even so, it does use the standard
@extra mechanism so that it doesn't break pkg_delete -c.
Please find attached the last version based on your input. (My nice
custom messages to the user are gone now, but anyway. Also, @sample
cannot handle symlinks correctly, so I had to use @exec for it, hope
that's ok.)
Btw, net category is back, because the proxy software I see in the tree
include net among their categories too, see pop3gwd for an example.
Thanks for incorporating that; I don't think there's a need to put
all the language versions of p3scan.mail into /etc/p3scan though;
wouldn't it be simpler to just
share/examples/p3scan/
@sample ${SYSCONFDIR}/p3scan/
share/examples/p3scan/p3scan-en.mail
@sample ${SYSCONFDIR}/p3scan/p3scan.mail
then don't @sample the rest; users can always copy the file from
examples if they'd like another language, and then there's no need
to worry about symlinks. (Actually you can install a symlink
under examples and then @sample that, but this way is cleaner).
Btw, please don't be disheartened by the backwards-and-forwards
tweaking things until they fit nicely; it can seem annoying at
times but, without it, the ports tree would be much less pleasant
to work with. It's hard work to keep things simple but I think
the results speak for themselves (-:
Re: NEW: net/p3scan
On Fri, 2007-11-16 at 13:52 +, Stuart Henderson wrote:
Thanks for incorporating that; I don't think there's a need to put
all the language versions of p3scan.mail into /etc/p3scan though;
wouldn't it be simpler to just
share/examples/p3scan/
@sample ${SYSCONFDIR}/p3scan/
share/examples/p3scan/p3scan-en.mail
@sample ${SYSCONFDIR}/p3scan/p3scan.mail
then don't @sample the rest; users can always copy the file from
examples if they'd like another language, and then there's no need
to worry about symlinks. (Actually you can install a symlink
under examples and then @sample that, but this way is cleaner).
Done. I'll have to copy those internationalized files to /etc/p3scan/ in
my project's install.site script (I need them there for the web
interface), but that's ok. (In my experience, @sample'ing a symlink
copies the linked file, not the symlink file, i.e. the link is followed,
which was not my intention as was obvious on the @exec line. I confirmed
by trying again. In any case, we don't need that symlink anymore.)
Please find the new ports package with those changes. This one has many
other improvements to the patch files too.
Btw, please don't be disheartened by the backwards-and-forwards
tweaking things until they fit nicely; it can seem annoying at
times but, without it, the ports tree would be much less pleasant
to work with. It's hard work to keep things simple but I think
the results speak for themselves (-:
I appreciate your input and encouragement, and agree with you.
p3scan-2.3.2.tar.gz
Description: application/compressed-tar
Re: Menpower/testing needed to update Ports for 4.2?
On Fri, Nov 16, 2007 at 02:00:05PM +0100, [EMAIL PROTECTED] wrote: Hi everybody, Just wanna know if I could help out to get some ports updated. Sorry, but developers run -current, so you are out of luck. Releases are just for selling CD's anyway. If you are running desktop, I recommend you to use -current. I just wanna name some Ports who are propably in use by many users out there (all Ports I list here do have sec. problems): - flac (1.2.1 is avaiable for current and fixes the Bufferoverflows) - xvid (well 1.0.3 is in the Ports, has a buffer overflow if I remember correctly, 1.1.2 is avaiable at the website) - mozilla-firefox (2.0.0.9 is avaiable for current) - xpdf (update is avaiable for current) And if you own a Webserver you're propably more interested into pcre or php updates. During the media libaries propably also mplayer and others are affected. Because the fixes where applied to current I guess more testing is propably needed. So that's a more general request how to help the maintainers. Kind regards, Sebastian p.s. Openoffice is something special I guess :-/
Re: Menpower/testing needed to update Ports for 4.2?
AFAIK there was 0 response to Nikolay's mail about stable ports, so apparently nobody is interested. On 2007/11/16 14:00, [EMAIL PROTECTED] wrote: Hi everybody, Just wanna know if I could help out to get some ports updated. I just wanna name some Ports who are propably in use by many users out there (all Ports I list here do have sec. problems): - flac (1.2.1 is avaiable for current and fixes the Bufferoverflows) - xvid (well 1.0.3 is in the Ports, has a buffer overflow if I remember correctly, 1.1.2 is avaiable at the website) - mozilla-firefox (2.0.0.9 is avaiable for current) - xpdf (update is avaiable for current) And if you own a Webserver you're propably more interested into pcre or php updates. During the media libaries propably also mplayer and others are affected. Because the fixes where applied to current I guess more testing is propably needed. So that's a more general request how to help the maintainers. Kind regards, Sebastian p.s. Openoffice is something special I guess :-/
Re: Ion3 port is obsolete
On 2007-11-16, Stuart Henderson [EMAIL PROTECTED] wrote: The version in tree is before the license change; the additional restrictions on the newer code are a problem. They are not a problem for reasonable distributors that care to pay a bit of respect towards the author's time and work. Of course, reason, literacy, and respect towards authors and persoanl choice are something seldom seen among the FOSS herd, rather replacing them with blind ideology and monocultures. It is a popular myth that you have to provide the new release within 28 days, and although I encourage that, it is not true and what the license says. Alternatively, you must after those 28 days prominently notify the user installing the software that the release is likely to be antiquated, not representative of the project's present state, and the author will not provide support for it. Not much asked, in my opinion. You could even base this notification on a dead-man switch, which would be quite nice even generally, considering package maintainers often going MIA. Boy, that's a lot of must's in that paragraph. Sure sounds free. It's free, but you MUST list of things What's great about me jumping into this conversation is that you talking about respect of the author is so interesting. I don't use your software, but I am sure you use OpenSSH. And now you are telling me what I (who distribute OpenBSD with all the things) must do. You say Not much asked, in my opinion. But you did not ask. You demanded, and everyone can see that.
Menpower/testing needed to update Ports for 4.2?
Hi everybody, Just wanna know if I could help out to get some ports updated. I just wanna name some Ports who are propably in use by many users out there (all Ports I list here do have sec. problems): - flac (1.2.1 is avaiable for current and fixes the Bufferoverflows) - xvid (well 1.0.3 is in the Ports, has a buffer overflow if I remember correctly, 1.1.2 is avaiable at the website) - mozilla-firefox (2.0.0.9 is avaiable for current) - xpdf (update is avaiable for current) And if you own a Webserver you're propably more interested into pcre or php updates. During the media libaries propably also mplayer and others are affected. Because the fixes where applied to current I guess more testing is propably needed. So that's a more general request how to help the maintainers. Kind regards, Sebastian p.s. Openoffice is something special I guess :-/
Re: NEW: net/p3scan
On Thu, 2007-11-15 at 22:00 +, Stuart Henderson wrote: On 2007/11/15 23:04, Soner Tari wrote: I've checked the other options like the one you are mentioning, but handling this in a separate shell script seems better to me (for example it's more readable and manageable in my opinion). Also note that I took postfix files as the basis for my p3scan-config. Therefore, if there is no harm, I'd like to keep p3scan-config. This is the wrong way to go, the pkg_* tools can handle this, so let them. @exec is useful for cases where you can't use the normal mechanisms, it's available but it should almost never be necessary. You don't see it used very much in our ports tree. Postfix has to populate the chroot with files from the installed system, and has its own mechanism to upgrade config files between versions via /etc/postfix/post-install. This goes to make it a bit of a special case... but even so, it does use the standard @extra mechanism so that it doesn't break pkg_delete -c. Please find attached the last version based on your input. (My nice custom messages to the user are gone now, but anyway. Also, @sample cannot handle symlinks correctly, so I had to use @exec for it, hope that's ok.) Btw, net category is back, because the proxy software I see in the tree include net among their categories too, see pop3gwd for an example. p3scan-2.3.2.tar.gz Description: application/compressed-tar
Re: [update] opencdk-0.6.5 and gnutls-2.0.3
On Fri, Nov 16, 2007 at 11:51:01AM +0100, Landry Breuil wrote: Hi, this submission is based on a previous work from Giovanni Bechis, thanks to him. These two diffs updates opencdk from 0.5.5 to 0.6.5 and gnutls from 1.0.25 to 2.0.3 (recently released). make regress fails (at least on i386) on opencdk at t-key test: t-key.c:516 expired key test FAILED t-key.c:522 missing key test FAILED FAIL: t-key t-key test fails on amd64 too. liferea, wireshark, filezilla seems to work fine with this update ;] === Regression check for opencdk-0.6.5 Making check in src Making check in doc Making check in tests make t-sign t-key t-encr t-stream t-keydb t-misc basic `t-sign' is up to date. `t-key' is up to date. `t-encr' is up to date. `t-stream' is up to date. `t-keydb' is up to date. `t-misc' is up to date. `basic' is up to date. make check-TESTS PASS: t-stream PASS: t-sign t-key.c:516 expired key test FAILED t-key.c:522 missing key test FAILED FAIL: t-key PASS: t-encr PASS: t-keydb PASS: t-misc OpenCDK header version 0.6.5. OpenCDK library version 0.6.5. PASS: basic 1 of 7 tests failed Please report to [EMAIL PROTECTED] *** Error code 1 It didn't fail with 2.0.1 and 2.0.2, i have to figure out what changed. make regress fails too for gnutls on pathlen test: make check-TESTS --- ./no-ca-or-pathlen.pem Sat Nov 10 18:17:51 2007 +++ new-no-ca-or-pathlen.pemFri Nov 16 11:36:54 2007 @@ -22,7 +22,7 @@ Basic Constraints (not critical): Certificate Authority (CA): FALSE Unknown extension 2.5.29.32 (not critical): - ASCII: 0;09..`.H...E0*0(..+.https://www.verisign.com/rpa + ASCII: 0;09..`.H..?E0*0(..+.https://www.verisign.com/rpa Hexdump: 303b3039060b6086480186f84501070108302a302806082b06010505070201161c68747470733a2f2f772e766572697369676e2e636f6d2f727061 Unknown extension 2.16.840.1.113730.1.1 (not critical): ASCII: FAIL: pathlen Apparently this one is not important, it's related to our isprint() implem vs linux one. Ports depending on gnutls have been tested (mostly at p2k7) with this upgrade and still works fine : thanks sthen@ for testing iksemel and asterisk, thanks aanriot@ for testing prelude and friends. liferea still builds and run fine, wmbiff has already been updated to support it. I'd really like to commit this soon, so please report any other error messages you get on other archs. diffs : http://gcu.info/~gaston/patches/opencdk-0.6.5.diff http://gcu.info/~gaston/patches/gnutls-2.0.3.diff Thanks for testing and feedback. Landry
Re: Ion3 port is obsolete
On 2007-11-16, Craig Brozefsky [EMAIL PROTECTED] wrote: Yah, sucks to write free software, perhaps you should just stop. Indeed, Ion3 is my final gift to the FOSS herd, that it can never hope to repay. After that any software I might create, will come without any license at all (the djb way). With or without source, I have not yet decided. Probably without, since FOSS is degrading into a pile of steaming shit so fast, that I'm likely to be switching to Windows within a few years time, and binaries will work just fine there. -- Tuomo
Re: Ion3 port is obsolete
On 2007-11-16, Stuart Henderson [EMAIL PROTECTED] wrote: The version in tree is before the license change; the additional restrictions on the newer code are a problem. They are not a problem for reasonable distributors that care to pay a bit of respect towards the author's time and work. Of course, reason, literacy, and respect towards authors and persoanl choice are something seldom seen among the FOSS herd, rather replacing them with blind ideology and monocultures. It is a popular myth that you have to provide the new release within 28 days, and although I encourage that, it is not true and what the license says. Alternatively, you must after those 28 days prominently notify the user installing the software that the release is likely to be antiquated, not representative of the project's present state, and the author will not provide support for it. Not much asked, in my opinion. You could even base this notification on a dead-man switch, which would be quite nice even generally, considering package maintainers often going MIA. -- Tuomo
Re: Ion3 port is obsolete
On 2007/11/16 17:08, Tuomo Valkonen wrote: The Ion3 port at [1] is obsolete and should be upgraded, or at least users be made very sure that they don't come asking me for help. It is also misnamed: there's no such things as Ion 20070318. It's Ion3, __development snapshot__ 20070318. Read that emphasised portion again, and think for a moment. Users should be made _highly_ aware of that fact, especially when distributing such antiquated releases. [1]: http://www.openbsd.org/4.2_packages/i386/ion-20070318p1.tgz-long.html The version in tree is before the license change; the additional restrictions on the newer code are a problem. Your new license requires that all old OS releases with Ion(tm) packages have upgrades made available within 28 days, even if that OS release is no longer supported. This is somewhat counter-productive imho, since the change encourages people to continue distributing the obsolete code that came with the standard LGPL (or just remove the package), rather than update...
ICE when compiling texlive on i386/current
Hi, I just noticed this while performing building print/texlive/base (in the install target), 100% reproducable. Has anyone else seen it? cc -DHAVE_CONFIG_H -I. -I/usr/ports/print/texlive/base/w-texlive_base-2007p2/texlive_base-2007/texk/web2c -I.. -I/usr/ports/print/texlive/base/w-texlive_base-2007p2/texlive_base-2007/texk/web2c/.. -I/usr/X11R6/include -I../../libs/obsdcompat -I../../libs/obsdcompat/.. -I/usr/ports/print/texlive/base/w-texlive_base-2007p2/texlive_base-2007/texk/web2c/../../libs/obsdcompat -I/usr/ports/print/texlive/base/w-texlive_base-2007p2/texlive_base-2007/texk/web2c/../../libs/obsdcompat/.. -O2 -pipe -c mp2.c -o mp2.o major internal compiler error: mp2.c, line 2954 gmake[2]: *** [mp2.o] Error 1 gmake[2]: Leaving directory `/usr/ports/print/texlive/base/w-texlive_base-2007p2/build-i386/texk/web2c' gmake[1]: *** [install] Error 1 gmake[1]: Leaving directory `/usr/ports/print/texlive/base/w-texlive_base-2007p2/build-i386/texk' gmake: *** [install] Error 1 *** Error code 2 Stop in /usr/ports/print/texlive/base (line 2117 of /usr/ports/infrastructure/mk/bsd.port.mk). *** Error code 1 Stop in /usr/ports/print/texlive/base (line 1373 of /usr/ports/infrastructure/mk/bsd.port.mk). *** Error code 1 Stop in /usr/ports/print/texlive/base (line 1861 of /usr/ports/infrastructure/mk/bsd.port.mk). *** Error code 1 Stop in /usr/ports/print/texlive/base (line 1403 of /usr/ports/infrastructure/mk/bsd.port.mk).
Re: www/php5, what's going on?
On Fri, Nov 16, 2007 at 11:58:05AM +0200, Alexey Suslikov wrote: === Installing php5-core-5.2.5p0 from /usr/ports/packages/i386/all/ File /var/www/conf/modules.sample/php5.conf could not be installed:** | 72% No such file or directory php5-core-5.2.5p0: complete --- php5-core-5.2.5p0 --- To enable the php5 module please create a symbolic link from /var/www/conf/modules.sample/php5.conf to /var/www/conf/modules/php5.conf. ln -s /var/www/conf/modules.sample/php5.conf \ /var/www/conf/modules The recommended php configuration has been installed to /var/www/conf/php.ini. $ ls -l /var/www/conf total 376 -rw-r--r-- 1 root wheel 88 Nov 13 22:31 bgplg.css -rw-r--r-- 1 root wheel 13 Nov 13 22:31 bgplg.foot -rw-r--r-- 1 root wheel 194 Nov 13 22:31 bgplg.head -r--r--r-- 1 root bin 37634 Nov 13 22:30 httpd.conf -r--r--r-- 1 root bin 43451 Nov 13 22:30 httpd.conf-dist -r--r--r-- 1 root bin 12965 Nov 13 22:30 magic -r--r--r-- 1 root bin 15151 Nov 13 22:30 mime.types -rw-r--r-- 1 root daemon 65970 Nov 16 11:47 php.ini drwxr-xr-x 2 root daemon512 Nov 16 11:47 php5 drwxr-xr-x 2 root daemon512 Nov 16 11:47 php5.sample I see neither /var/www/conf/modules.sample/, nor /var/www/conf/modules/. What can I do to fix it? Try a newer snapshot. At least the november 1st snapshot for sparc64 did include those directories. Regards, Markus
Re: Ion3 port is obsolete
On Fri, Nov 16, 2007 at 04:26:05PM +, Tuomo Valkonen wrote: On 2007-11-16, Stuart Henderson [EMAIL PROTECTED] wrote: The version in tree is before the license change; the additional restrictions on the newer code are a problem. They are not a problem for reasonable distributors that care to pay a bit of respect towards the author's time and work. Of course, reason, literacy, and respect towards authors and persoanl choice are something seldom seen among the FOSS herd, rather replacing them with blind ideology and monocultures. It is a popular myth that you have to provide the new release within 28 days, and although I encourage that, it is not true and what the license says. Alternatively, you must after those 28 days prominently notify the user installing the software that the release is likely to be antiquated, not representative of the project's present state, and the author will not provide support for it. Not much asked, in my opinion. You could even base this notification on a dead-man switch, which would be quite nice even generally, considering package maintainers often going MIA. Package is no longer maintained due to your license change. I fail to see the relevance of trying to retroactively impose its new terms. -- Tuomo
Re: Status of Xplore?
On 2007/11/16 03:52, Aaron Hsu wrote: Hey all, I was wondering what the status was of importing the Xplore port which I submitted a bit ago. I know sometimes ports fall through the cracks, and I just wanted to see if there was any reason other than that for it not being imported? If so, I'll try to get it fixed, but if it just slipped through the cracks, well . . . :-). Needs some WANTLIB and s/// in COMMENT; build fails on sparc64 and amd64, this log is from sparc64 but amd64 does the same: +++ Fri Nov 16 10:54:46 GMT 2007 === Checking files for xplore-1.2a `/usr/ports/distfiles/xplore-1.2a.tar.gz' is up to date. (SHA256) xplore-1.2a.tar.gz: OK === xplore-1.2a depends on: openmotif-* - found === Verifying specs: Xm.=2 Xm.=2 ICE SM Xext Xmu Xt X11 Xpm Xau Xdmcp c ICE SM Xext Xmu Xt X11 Xpm Xau Xdmcp c === found Xm.4.0 ICE.8.1 SM.8.0 Xext.10.0 Xmu.10.0 Xt.10.0 X11.11.0 Xpm.8.0 Xau.9.0 Xdmcp.9.0 c.42.0 === Extracting for xplore-1.2a === Patching for xplore-1.2a === Configuring for xplore-1.2a mv -f Makefile Makefile.bak imake -DPorts -DUseInstalled -I/usr/X11R6/lib/X11/config make Makefiles making Makefiles in xg... mv -f Makefile Makefile.bak making Makefiles in regexp... mv -f Makefile Makefile.bak making Makefiles in scripts... mv -f Makefile Makefile.bak make includes make depend depending in ./xg... rm -f .depend gccmakedep -f- -- -DUSE_XM_MWMUTIL_H -DUSE_XGPM -I./include -I/usr/local/include -I/usr/X11R6/include -DCSRG_BASED -DFUNCPROTO=15 -DNARROWPROTO -- ClockLabel.c ComboBox.c ExtPushB.c FastLabel.c NumEntry.c Row.c Tabs.c TimeEntry.c ToolBar.c ToolPushB.c XgUtils.c grabarea.c x_rotate.c xgpm.c .depend depending in ./regexp... rm -f .depend gccmakedep -f- -- -I. -I/usr/local/include -I/usr/X11R6/include -DCSRG_BASED -DFUNCPROTO=15 -DNARROWPROTO -- regexp.c regsub.c regerror.c .depend depending in ./scripts... sed -e '[EMAIL PROTECTED]@?/usr/local/lib/X11/xplore?g' Xplore.sed Xplore.ad ./ad2c Xplore.ad defaults.h rm -f .depend gccmakedep -f- -- -I./regexp -I./icons -I./xg/include -I/usr/local/include -I/usr/X11R6/include -DCSRG_BASED -DFUNCPROTO=15 -DNARROWPROTO -DXPLORELIBDIR=\/usr/local/lib/X11/xplore\ -DCPP=\/usr/libexec/cpp -traditional \ -- callbacks.c curdir.c devmount.c dialogs.c dirs.c error.c fileops.c ftype.c icons.c interface.c magic.c menus.c mnttable.c signals.cutil.c xplore.c xploretype.c .depend === Building for xplore-1.2a making all in ./xg... rm -f ClockLabel.o cc -c -O2 -DUSE_XM_MWMUTIL_H -DUSE_XGPM -I./include -I/usr/local/include -I/usr/X11R6/include -DCSRG_BASED -DFUNCPROTO=15 -DNARROWPROTO ClockLabel.c rm -f ComboBox.o cc -c -O2 -DUSE_XM_MWMUTIL_H -DUSE_XGPM -I./include -I/usr/local/include -I/usr/X11R6/include -DCSRG_BASED -DFUNCPROTO=15 -DNARROWPROTO ComboBox.c ComboBox.c: In function `CvtStringToStringTable': ComboBox.c:355: warning: cast to pointer from integer of different size ComboBox.c:378: warning: cast to pointer from integer of different size rm -f ExtPushB.o cc -c -O2 -DUSE_XM_MWMUTIL_H -DUSE_XGPM -I./include -I/usr/local/include -I/usr/X11R6/include -DCSRG_BASED -DFUNCPROTO=15 -DNARROWPROTO ExtPushB.c ExtPushB.c: In function `Initialize': ExtPushB.c:361: warning: cast to pointer from integer of different size ExtPushB.c:361: warning: cast to pointer from integer of different size ExtPushB.c:361: warning: cast to pointer from integer of different size rm -f FastLabel.o cc -c -O2 -DUSE_XM_MWMUTIL_H -DUSE_XGPM -I./include -I/usr/local/include -I/usr/X11R6/include -DCSRG_BASED -DFUNCPROTO=15 -DNARROWPROTO FastLabel.c rm -f NumEntry.o cc -c -O2 -DUSE_XM_MWMUTIL_H -DUSE_XGPM -I./include -I/usr/local/include -I/usr/X11R6/include -DCSRG_BASED -DFUNCPROTO=15 -DNARROWPROTO NumEntry.c rm -f Row.o cc -c -O2 -DUSE_XM_MWMUTIL_H -DUSE_XGPM -I./include -I/usr/local/include -I/usr/X11R6/include -DCSRG_BASED -DFUNCPROTO=15 -DNARROWPROTO Row.c rm -f Tabs.o cc -c -O2 -DUSE_XM_MWMUTIL_H -DUSE_XGPM -I./include -I/usr/local/include -I/usr/X11R6/include -DCSRG_BASED -DFUNCPROTO=15 -DNARROWPROTO Tabs.c rm -f TimeEntry.o cc -c -O2 -DUSE_XM_MWMUTIL_H -DUSE_XGPM -I./include -I/usr/local/include -I/usr/X11R6/include -DCSRG_BASED -DFUNCPROTO=15 -DNARROWPROTO TimeEntry.c rm -f ToolBar.o cc -c -O2 -DUSE_XM_MWMUTIL_H -DUSE_XGPM -I./include -I/usr/local/include -I/usr/X11R6/include -DCSRG_BASED -DFUNCPROTO=15 -DNARROWPROTO ToolBar.c rm -f ToolPushB.o cc -c -O2 -DUSE_XM_MWMUTIL_H -DUSE_XGPM -I./include -I/usr/local/include -I/usr/X11R6/include -DCSRG_BASED -DFUNCPROTO=15 -DNARROWPROTO ToolPushB.c rm -f XgUtils.o cc -c -O2 -DUSE_XM_MWMUTIL_H -DUSE_XGPM -I./include -I/usr/local/include -I/usr/X11R6/include -DCSRG_BASED -DFUNCPROTO=15 -DNARROWPROTO XgUtils.c
Re: Ion3 port is obsolete
On Fri, 16 Nov 2007, Marco Peereboom wrote: [snipped] Meanwhile I'll use Ion3, __development snapshot__ 20070318 until something that suits me better comes along. Wasn't there a fork already that was based on the last version with GPL? -- Antti Harri
Re: Ion3 port is obsolete
On 2007-11-16 11:05 -0600, Marco Peereboom wrote: Package is no longer maintained due to your license change. So remove it. Speaks loads of the so-called quality of the OpenBSD distribution when it distributes ancient unmaintained software with various bugs. -- Tuomo
Re: Ion3 port is obsolete
Tuomo Valkonen [EMAIL PROTECTED] writes: On 2007-11-16 10:13 -0700, Theo de Raadt wrote: Boy, that's a lot of must's in that paragraph. Sure sounds free. Typically free means: free for the herd to do anything, including fucking the author in the arse. Straitjacket and pain in the arse for the author who has to bear with the herd. Yah, sucks to write free software, perhaps you should just stop. So, please, spare me of your ideology. And spare us your gripes. plonk -- Sincerely, Craig Brozefsky [EMAIL PROTECTED] what a klon - neko http://www.red-bean.com/~craig Less matter, more form! - Bruno Schulz ignazz, I am truly korrupted by yore sinful tzourceware. -jb
www/php5, what's going on?
=== Installing php5-core-5.2.5p0 from /usr/ports/packages/i386/all/ File /var/www/conf/modules.sample/php5.conf could not be installed:** | 72% No such file or directory php5-core-5.2.5p0: complete --- php5-core-5.2.5p0 --- To enable the php5 module please create a symbolic link from /var/www/conf/modules.sample/php5.conf to /var/www/conf/modules/php5.conf. ln -s /var/www/conf/modules.sample/php5.conf \ /var/www/conf/modules The recommended php configuration has been installed to /var/www/conf/php.ini. $ ls -l /var/www/conf total 376 -rw-r--r-- 1 root wheel 88 Nov 13 22:31 bgplg.css -rw-r--r-- 1 root wheel 13 Nov 13 22:31 bgplg.foot -rw-r--r-- 1 root wheel 194 Nov 13 22:31 bgplg.head -r--r--r-- 1 root bin 37634 Nov 13 22:30 httpd.conf -r--r--r-- 1 root bin 43451 Nov 13 22:30 httpd.conf-dist -r--r--r-- 1 root bin 12965 Nov 13 22:30 magic -r--r--r-- 1 root bin 15151 Nov 13 22:30 mime.types -rw-r--r-- 1 root daemon 65970 Nov 16 11:47 php.ini drwxr-xr-x 2 root daemon512 Nov 16 11:47 php5 drwxr-xr-x 2 root daemon512 Nov 16 11:47 php5.sample I see neither /var/www/conf/modules.sample/, nor /var/www/conf/modules/. What can I do to fix it? - Alexey.
Re: Ion3 port is obsolete
On 2007-11-16 10:13 -0700, Theo de Raadt wrote: Boy, that's a lot of must's in that paragraph. Sure sounds free. Typically free means: free for the herd to do anything, including fucking the author in the arse. Straitjacket and pain in the arse for the author who has to bear with the herd. So, please, spare me of your ideology. It's free, but you MUST list of things Hmm... I guess (L)GPL isn't very free. It isn't, in fact; and I do consider the BSD license more free. In fact, the name use terms in my license are basically all that I care about; the LGPL is just baggage. -- Tuomo
Re: Ion3 port is obsolete
Licenses can not be retro actively imposed. This package was done before your license change and therefore it'll remain just like it is. I am a pre-ion user and can't even begin to tell you how retarded your new license is. You got what you wanted, you rendered your open source developments useless. Enjoy. Meanwhile I'll use Ion3, __development snapshot__ 20070318 until something that suits me better comes along. On Fri, Nov 16, 2007 at 05:08:07PM +0200, Tuomo Valkonen wrote: The Ion3 port at [1] is obsolete and should be upgraded, or at least users be made very sure that they don't come asking me for help. It is also misnamed: there's no such things as Ion 20070318. It's Ion3, __development snapshot__ 20070318. Read that emphasised portion again, and think for a moment. Users should be made _highly_ aware of that fact, especially when distributing such antiquated releases. [1]: http://www.openbsd.org/4.2_packages/i386/ion-20070318p1.tgz-long.html -- Tuomo
Ion3 port is obsolete
The Ion3 port at [1] is obsolete and should be upgraded, or at least users be made very sure that they don't come asking me for help. It is also misnamed: there's no such things as Ion 20070318. It's Ion3, __development snapshot__ 20070318. Read that emphasised portion again, and think for a moment. Users should be made _highly_ aware of that fact, especially when distributing such antiquated releases. [1]: http://www.openbsd.org/4.2_packages/i386/ion-20070318p1.tgz-long.html -- Tuomo
broken wget build
Hi, I'm trying out 4.2(stable) on this i386 test box i have and there seems to be a broken lib on building libiconv via ports tree, (i need it to build wget package), looks like something is missing on the expat lib side, I've read that 4.2 is putting expat together with the base build instead of adding it as a port package and all but just wondering if there is anyway for me to work around this? I've used cvsup to pull the latest stable for both src and ports tree from anoncvs1.usa.openbsd.org. below is the output where it broke. Link to /usr/ports/packages/i386/cdrom/libiconv-1.9.2p3.tgz === Verifying specs: c === found c.41.0 === Installing libiconv-1.9.2p3 from /usr/ports/packages/i386/all/ libiconv-1.9.2p3: complete === Returning to build of gettext-0.14.6p0 === gettext-0.14.6p0 depends on: libiconv-* - found === Verifying specs: iconv.=2 iconv.=2 c expat c expat Missing library for expat Fatal error *** Error code 1 any help would be awesomely appreciated. -b
Re: broken wget build
On Fri 2007.11.16 at 13:52 -0600, Beavis wrote: Hi, I'm trying out 4.2(stable) on this i386 test box i have and there seems to be a broken lib on building libiconv via ports tree, (i need it to build wget package), looks like something is missing on the expat lib side, I've read that 4.2 is putting expat together with the base build instead of adding it as a port package and all but just wondering if there is anyway for me to work around this? I've used cvsup to pull the latest stable for both src and ports tree from anoncvs1.usa.openbsd.org. in the archives and undeadly: libexpat is in base in -current, and for 4.3
Re: Ion3 port is obsolete
On 2007-11-16 20:13 +0100, Marc Balmer wrote: but windows does not need a window manager... Indeed, Ion is my only remaining umblical cord to FOSS crap, and no thanks to the FOSS herd, but vestiges of software from the age before the FOSS craze, from the age before the WIMP desktop model became hegemonic. If Windows could provide something like Ion, I wouldn't think twice of switching to it. -- Tuomo
Re: Ion3 port is obsolete
Who is talking about using windows apps? I just said I ported it work in cygwin so that I don't have to use windows at work. GNU userland beats even MS cli commands. On Fri, Nov 16, 2007 at 10:27:37PM +0200, Tuomo Valkonen wrote: On 2007-11-16 13:45 -0600, Marco Peereboom wrote: Man you are in luck. I happened to make ion work on windows to make it more bearable. It's on my site; enjoy! Yeah, right. Actually, with the compositing manager now in Vista (which they call the Desktop Window manager, heh), it might be possible to hack a sorry emulation of a WM by replacing it. Even a tabbing and tiling one by scaling the applications' backbuffers, which would of course make things look like shit. And then you'd have to hack around the application-drawn (AFAIK) window frames. -- Tuomo
Re: Ion3 port is obsolete
On 2007-11-16 13:45 -0600, Marco Peereboom wrote: Man you are in luck. I happened to make ion work on windows to make it more bearable. It's on my site; enjoy! Yeah, right. Actually, with the compositing manager now in Vista (which they call the Desktop Window manager, heh), it might be possible to hack a sorry emulation of a WM by replacing it. Even a tabbing and tiling one by scaling the applications' backbuffers, which would of course make things look like shit. And then you'd have to hack around the application-drawn (AFAIK) window frames. -- Tuomo
Re: Ion3 port is obsolete
On Fri, Nov 16, 2007 at 08:07:12PM +0200, Tuomo Valkonen wrote: On 2007-11-16 11:40 -0600, Marco Peereboom wrote: I have a suggestion for you; why don't you rename your software to something else instead of ion, it'll make all the evil people using what-used-to-be-called-ion go away. Problem solved. The whole point is that Ion is name that is associated with me, and the distros are tarnishing it by distributing significantly modified and out-dated software as Ion, without prominently mentioning this. You knew that when you distributed the original under its original license. If you didn't that is your fault. Don't blame me for using free software under a relatively well understood license. I'm sure OpenBSD would like very much for me to distribute some ancient and significantly modified release as the latest. Except, of course, I don't have the resources for such to have much of an effect, unlike The Party, i.e. the big distros. The only thing I'd like (I am not cocky enough to pretend I know what the entire OpenBSD community wants) is for the software to get its original license back so that it could be maintained like it was. Oh and FWIW, OpenBSD left ion almost identical to your specifications. The only thing that was modified was to have the large menu when pressing F12. The rest was 100% identical to what you did. OpenBSD stayed within the spirit of your developments however you chose to sever the ties to the OpenBSD project the second you changed that license. OpenBSD fulfilled all legal requirements as per your original license. -- Tuomo
Re: Ion3 port is obsolete
On 2007-11-16 12:25 -0600, Marco Peereboom wrote: You knew that when you distributed the original under its original license. If you didn't that is your fault. Don't blame me for using free software under a relatively well understood license. No, I just didn't think back then about the license so much; I did not realise what kind of dickheads and fuckwits the FOSS herd is composed of. Call me naive if you want. But as the project started to gain popularity, you get a lot of people complaining and asking help for ancient versions distributed by the distributions that also have in the meanwhile become more powerful, and gained more central control over conveniently installable software. And at the same time FOSS detoriorates by forcing the anti-aliasing fascist fontconfig/Xft nearly everywhere, and now the herd modifies the version of Ion their distros carry to use that crap, which I will have nothing to do with until my demands [1] are met. When I first started out on Ion, I had hope in FOSS. All that has been lost now. Most people are dickheads and fuckwits everywhere, no matter their proclaimed ideals of so-called freedom. FOSS licenses are only for naive people and those who go with the herd. As should be apparent, I don't. [1]: http://iki.fi/tuomov/ion/faq/entries/Blurred_fonts.html -- Tuomo
Re: Ion3 port is obsolete
Tuomo Valkonen wrote: On 2007-11-16, Craig Brozefsky [EMAIL PROTECTED] wrote: Yah, sucks to write free software, perhaps you should just stop. Indeed, Ion3 is my final gift to the FOSS herd, that it can never hope to repay. After that any software I might create, will come without any license at all (the djb way). With or without source, I have not yet decided. Probably without, since FOSS is degrading into a pile of steaming shit so fast, that I'm likely to be switching to Windows within a few years time, and binaries will work just fine there. but windows does not need a window manager...
Re: Ion3 port is obsolete
On 2007-11-16 11:40 -0600, Marco Peereboom wrote: I have a suggestion for you; why don't you rename your software to something else instead of ion, it'll make all the evil people using what-used-to-be-called-ion go away. Problem solved. The whole point is that Ion is name that is associated with me, and the distros are tarnishing it by distributing significantly modified and out-dated software as Ion, without prominently mentioning this. I'm sure OpenBSD would like very much for me to distribute some ancient and significantly modified release as the latest. Except, of course, I don't have the resources for such to have much of an effect, unlike The Party, i.e. the big distros. -- Tuomo
Re: Ion3 port is obsolete
It works fine as it is and it won't be removed. You should have thought of the legal repercussions of writing free software. You gave it away back then so you can't take it back. As they say, you can't put the shit back in the horse. You can cry us a river all day long using strong profanity. It will not change the law. I have a suggestion for you; why don't you rename your software to something else instead of ion, it'll make all the evil people using what-used-to-be-called-ion go away. Problem solved. On Fri, Nov 16, 2007 at 07:27:53PM +0200, Tuomo Valkonen wrote: On 2007-11-16 11:05 -0600, Marco Peereboom wrote: Package is no longer maintained due to your license change. So remove it. Speaks loads of the so-called quality of the OpenBSD distribution when it distributes ancient unmaintained software with various bugs. -- Tuomo
Re: Ion3 port is obsolete
On Fri, Nov 16, 2007 at 06:01:12PM +, Tuomo Valkonen wrote: On 2007-11-16, Craig Brozefsky [EMAIL PROTECTED] wrote: Yah, sucks to write free software, perhaps you should just stop. Indeed, Ion3 is my final gift to the FOSS herd, that it can never hope to repay. After that any software I might create, will come without any license at all (the djb way). With or without source, I have not yet decided. Probably without, since FOSS is degrading into a pile of steaming shit so fast, that I'm likely to be switching to Windows within a few years time, and binaries will work just fine there. good for you, and what did you have for breakfast ? -- Gilles Chehade http://www.evilkittens.org/ http://www.evilkittens.org/blog/gilles/
Re: Ion3 port is obsolete
On 2007-11-16 13:38 -0600, Marco Peereboom wrote: You are naive. The open source community is harsh and does not tend to cater to someone's feelings. Kind of like the real world. In the harsh real world the companies sue you for distributing their software. I might just as well adopt their licenses and practices: as you have admitted, the FOSS herd is harsh and no different from them. I fail to see how that is your problem. Its free software, they change it they deal with it. They don't deal with it: they don't rename the software and tell users to not bug the original author. As long the software clearly points to the original author, users will come asking support for the distro's version. That is the case if the software has a face and has not become such generic software that just is there, and of which there are known to be various implementations (such as the basic *nix tools). Why do you care if someone else is starring at blurry fonts? I care when they make it purposefully difficult for me to personally use unblurry fonts, or some particular font (such as the beautiful X Helvetica bitmap font, which is often blocked). I will not have my software support such software that takes away or makes personal choice for me very difficult. Dealing with Linux people tends to anger people. Maybe you should try to leave the linuxers behind and work in a more constructive community. And that is? While indeed *BSD (of which only FreeBSD is likely to have the driver support I'd need) don't suffer from such utter and total crap as udev, and other recent idiot box idiocies in the Linux kernel, they still unfortunately rely on the same luserland (sic) that tends to be designed for the monoculturist desktop projects these days. Your bitterness stems from getting involved with people you are incompatible with. That's about 99% of people. As witnessed in this thread. And if you are that pissed off, why don't you just quit? You don't owe anyone anything. I like to finish what I've started. Then I'll quit. -- Tuomo
Re: Ion3 port is obsolete
Tuomo Valkonen wrote: On 2007-11-16 12:25 -0600, Marco Peereboom wrote: You knew that when you distributed the original under its original license. If you didn't that is your fault. Don't blame me for using free software under a relatively well understood license. No, I just didn't think back then about the license so much; I did not realise what kind of dickheads and fuckwits the FOSS herd is composed of. Call me naive if you want. But as the project started to gain popularity, you get a lot of people complaining and asking help for ancient versions distributed by the distributions that also have in the meanwhile become more powerful, and gained more central control over conveniently installable software. And at the same time FOSS detoriorates by forcing the anti-aliasing fascist fontconfig/Xft nearly everywhere, and now the herd modifies the version of Ion their distros carry to use that crap, which I will have nothing to do with until my demands [1] are met. When I first started out on Ion, I had hope in FOSS. All that has been lost now. Most people are dickheads and fuckwits everywhere, no matter their proclaimed ideals of so-called freedom. FOSS licenses are only for naive people and those who go with the herd. As should be apparent, I don't. no tiene cojones, as the spaniard would say... [1]: http://iki.fi/tuomov/ion/faq/entries/Blurred_fonts.html
Re: broken wget build
http://www.openbsd.org/faq/upgrade42.html#libexpat On Nov 16, 2007 12:52 PM, Beavis [EMAIL PROTECTED] wrote: Hi, I'm trying out 4.2(stable) on this i386 test box i have and there seems to be a broken lib on building libiconv via ports tree, (i need it to build wget package), looks like something is missing on the expat lib side, I've read that 4.2 is putting expat together with the base build instead of adding it as a port package and all but just wondering if there is anyway for me to work around this? I've used cvsup to pull the latest stable for both src and ports tree from anoncvs1.usa.openbsd.org. below is the output where it broke. Link to /usr/ports/packages/i386/cdrom/libiconv-1.9.2p3.tgz === Verifying specs: c === found c.41.0 === Installing libiconv-1.9.2p3 from /usr/ports/packages/i386/all/ libiconv-1.9.2p3: complete === Returning to build of gettext-0.14.6p0 === gettext-0.14.6p0 depends on: libiconv-* - found === Verifying specs: iconv.=2 iconv.=2 c expat c expat Missing library for expat Fatal error *** Error code 1 any help would be awesomely appreciated. -b
Re: Ion3 port is obsolete
On Fri, Nov 16, 2007 at 09:06:21PM +0200, Tuomo Valkonen wrote: On 2007-11-16 12:25 -0600, Marco Peereboom wrote: You knew that when you distributed the original under its original license. If you didn't that is your fault. Don't blame me for using free software under a relatively well understood license. No, I just didn't think back then about the license so much; I did not realise what kind of dickheads and fuckwits the FOSS herd is composed of. Call me naive if you want. But as the project started You are naive. The open source community is harsh and does not tend to cater to someone's feelings. Kind of like the real world. to gain popularity, you get a lot of people complaining and asking help for ancient versions distributed by the distributions that also have in the meanwhile become more powerful, and gained more central control over conveniently installable software. And at the same time FOSS detoriorates by forcing the anti-aliasing fascist fontconfig/Xft nearly everywhere, and now the herd modifies the version of Ion their distros carry to use that crap, which I will have nothing to do with until my demands [1] are met. I fail to see how that is your problem. Its free software, they change it they deal with it. The only choice you have to make is how much fun to poke at those people. Why do you care if someone else is starring at blurry fonts? Why do you care if someone else is watching those fonts scroll by at a snails pace slowing down their overall machine? Really, why do you care about someone's stupidity? When I first started out on Ion, I had hope in FOSS. All that has been lost now. Most people are dickheads and fuckwits everywhere, no matter their proclaimed ideals of so-called freedom. FOSS licenses are only for naive people and those who go with the herd. As should be apparent, I don't. You should take up some drama classes and put that anger to use. Dealing with Linux people tends to anger people. Maybe you should try to leave the linuxers behind and work in a more constructive community. I for one truly appreciate ion and its intentions. Knowing other folks that use ion, they agree and do the same. Your bitterness stems from getting involved with people you are incompatible with. I'll be the first to admit that getting involved in the Linux community is frustrating and boring. I therefore quit the community and found one that I am compatible with. And if you are that pissed off, why don't you just quit? You don't owe anyone anything. [1]: http://iki.fi/tuomov/ion/faq/entries/Blurred_fonts.html -- Tuomo
Re: Ion3 port is obsolete
On Fri, Nov 16, 2007 at 09:20:34PM +0200, Tuomo Valkonen wrote: On 2007-11-16 20:13 +0100, Marc Balmer wrote: but windows does not need a window manager... Indeed, Ion is my only remaining umblical cord to FOSS crap, and no thanks to the FOSS herd, but vestiges of software from the age before the FOSS craze, from the age before the WIMP desktop model became hegemonic. If Windows could provide something like Ion, I wouldn't think twice of switching to it. Man you are in luck. I happened to make ion work on windows to make it more bearable. It's on my site; enjoy! -- Tuomo
Re: ImageMagic distfile
On Thu 2007.11.15 at 10:16 +0100, Giovanni Bechis wrote: On Wed, Nov 14, 2007 at 11:02:33AM -0500, Okan Demirmen wrote: MASTER_SITES needs an update, as well as the whole port. are you working on that? Imagemagick updated to latest version, ports tested @i386. Comments ? Ok ? Giovanni thanks. regress on p5-PerlMagick fails - need to add to REGRESS_DEPENDS.
Re: Ion3 port is obsolete
On Fri, Nov 16, 2007 at 10:23:02PM +0200, Tuomo Valkonen wrote: On 2007-11-16 13:38 -0600, Marco Peereboom wrote: You are naive. The open source community is harsh and does not tend to cater to someone's feelings. Kind of like the real world. In the harsh real world the companies sue you for distributing their software. I might just as well adopt their licenses and practices: as you have admitted, the FOSS herd is harsh and no different from them. So why are you acting all surprised? I fail to see how that is your problem. Its free software, they change it they deal with it. They don't deal with it: they don't rename the software and tell users to not bug the original author. As long the software clearly points to the original author, users will come asking support for the distro's version. That is the case if the software has a face and has not become such generic software that just is there, and of which there are known to be various implementations (such as the basic *nix tools). And you say: go away or nothing at all. Wow, thats really hard. Why do you care if someone else is starring at blurry fonts? I care when they make it purposefully difficult for me to personally use unblurry fonts, or some particular font (such as the beautiful X Helvetica bitmap font, which is often blocked). I will not have my software support such software that takes away or makes personal choice for me very difficult. How do they make it difficult for you? What you are saying is: I use their stuff and I don't like it. How about not using their stuff? Dealing with Linux people tends to anger people. Maybe you should try to leave the linuxers behind and work in a more constructive community. And that is? While indeed *BSD (of which only FreeBSD is likely to have the driver support I'd need) don't suffer from such utter and total crap as udev, and other recent idiot box idiocies in the Linux kernel, they still unfortunately rely on the same luserland (sic) that tends to be designed for the monoculturist desktop projects these days. The best community for you seems to be the Tuomo one. You know, your own world where everything is just like you want it. You know what, you could even control who gets a passport. Your bitterness stems from getting involved with people you are incompatible with. That's about 99% of people. As witnessed in this thread. Then why do you keep talking? And if you are that pissed off, why don't you just quit? You don't owe anyone anything. I like to finish what I've started. Then I'll quit. You don't need to release your code and put up with it. You can just keep it all to yourself. You are apparently a masochist that keeps asking for more. -- Tuomo
Re: Ion3 port is obsolete
On 2007-11-16, Marco Peereboom [EMAIL PROTECTED] wrote: So why are you acting all surprised? What surprise? I said I used to have some hope in FOSS ages ago, but gradually it has worn off, to the extent that I no longer care for the cause at all. And you say: go away or nothing at all. Wow, thats really hard. It's still users bugging you, and a few more turned off from your software because the ancient or modified versions provided by the distros fail on them. How do they make it difficult for you? What you are saying is: I use their stuff and I don't like it. How about not using their stuff? That means not using a lot of software: it basically means limiting yourself to xterm (and a few other odd utilities, such as xdvi and gv, for now anyway). Soon maybe not even that, once the obsolete the old font system completely. Already e.g. Ubuntu at some point did not come with any fonts for the X core font system. Certainly you'd have no graphical browsers to use, none that can access any more pages than a text-mode one anyway. And unfortunately, while the Web indeed is crap, it has some useful and interesting information within it. Even on pages that refuse to work fine in text-mode browsers. The best community for you seems to be the Tuomo one. You know, your own world where everything is just like you want it. At least a world where not everything is polar to where you want it -- as things seem to be heading -- and where people are more open to your ideas. -- Tuomo
[UPDATE] xchat-2.8.4p1
Hi All,
I was corrected xchat ports :
* Fix WANTLIB
* Fix PLIST
* Add libsexy depend
The diff:
###
--- Makefile.oldSun Nov 11 20:08:41 2007
+++ MakefileFri Nov 16 22:24:39 2007
@@ -5,7 +5,7 @@
MAJOR= 2.8
VERSION= ${MAJOR}.4
DISTNAME= xchat-${VERSION}
-PKGNAME= ${DISTNAME}p0
+PKGNAME= ${DISTNAME}p1
CATEGORIES=net x11
HOMEPAGE= http://www.xchat.org/
@@ -21,14 +21,16 @@
crypto expat fontconfig freetype glib-2.0 \
glitz gmodule-2.0 gobject-2.0 m \
pango-1.0 pangocairo-1.0 \
- pangoft2-1.0 png ssl z
+ pangoft2-1.0 png ssl z \
+ Xcomposite Xdamage expat pcre xml2
MASTER_SITES= ${HOMEPAGE}files/source/${MAJOR}/
MASTER_SITES0= ${MASTER_SITE_SOURCEFORGE:=xchat/}
EXTRACT_SUFX= .tar.bz2
-LIB_DEPENDS= gtk-x11-2.0,gdk-x11-2.0,gdk_pixbuf-2.0::x11/gtk+2
+LIB_DEPENDS= gtk-x11-2.0,gdk-x11-2.0,gdk_pixbuf-2.0::x11/gtk+2 \
+ sexy::x11/libsexy
MODULES= devel/gettext
--- pkg/PLIST.orig Sat Sep 15 22:01:00 2007
+++ pkg/PLIST Fri Nov 16 22:16:20 2007
@@ -1,5 +1,7 @@
@comment $OpenBSD: PLIST,v 1.22 2007/07/07 12:51:10 couderc Exp $
bin/xchat
+man/cat3f/
+man/man3f/
share/applications/
share/applications/xchat.desktop
share/locale/be/LC_MESSAGES/xchat.mo
@@ -36,5 +38,6 @@
share/locale/vi/LC_MESSAGES/xchat.mo
share/locale/zh_CN/LC_MESSAGES/xchat.mo
share/locale/zh_TW/LC_MESSAGES/xchat.mo
+share/nls/pl_PL.ISO_8859-2/
share/pixmaps/
share/pixmaps/xchat.png
###
Work fine on i386.
BSDManiak
Re: NEW: net/p3scan
On Fri 2007.11.16 at 19:08 +0200, Soner Tari wrote:
Please find the new ports package with those changes. This one has many
other improvements to the patch files too.
great, thanks.
i think the one thing i'd ask is replacing the hardcoded /usr/local bits
in the patches with !!LOCALBASE!! and do something like the following in
pre-configure:
perl -pi -e 's,!!LOCALBASE!!,${LOCALBASE},g'
(shameless stolen from sthen@'s mail earlier on...)
cheers,
okan
Re: NEW: net/p3scan
On Fri 2007.11.16 at 17:27 -0500, Okan Demirmen wrote:
On Fri 2007.11.16 at 19:08 +0200, Soner Tari wrote:
Please find the new ports package with those changes. This one has many
other improvements to the patch files too.
great, thanks.
i think the one thing i'd ask is replacing the hardcoded /usr/local bits
in the patches with !!LOCALBASE!! and do something like the following in
pre-configure:
perl -pi -e 's,!!LOCALBASE!!,${LOCALBASE},g'
(shameless stolen from sthen@'s mail earlier on...)
cheers,
okan
and i forgot to mention, openbsd doesn't gz man pages.
Re: [update] opencdk-0.6.5 and gnutls-2.0.3
On Fri, Nov 16, 2007 at 02:18:28PM +0200, Nikns Siankin wrote: On Fri, Nov 16, 2007 at 11:51:01AM +0100, Landry Breuil wrote: Hi, this submission is based on a previous work from Giovanni Bechis, thanks to him. These two diffs updates opencdk from 0.5.5 to 0.6.5 and gnutls from 1.0.25 to 2.0.3 (recently released). make regress fails (at least on i386) on opencdk at t-key test: t-key.c:516 expired key test FAILED t-key.c:522 missing key test FAILED FAIL: t-key t-key test fails on amd64 too. fails too on sparc64. make regress fails too for gnutls on pathlen test: fails too on sparc64... Landry
Re: draft port of security/pwman
Stuart Henderson wrote: On 2007/11/15 14:07, Aaron S. [EMAIL PROTECTED] wrote: Okan Demirmen wrote: fyi, we've a minor formating fix on DESCR, but the larger issue is that pwman doesn't seem to work on at least macppc and possibly other big endian archs. Unfortunately, I don't have access to none of big endian machines. Do we add this for now? ONLY_FOR_ARCHS= amd64 i386 Well, there are more LE than these. I'm not sure whether they take too much care about alignment issues either, though, I didn't get time to test it much yet... Is there anything else needed for it to be committed?
Re: draft port of security/pwman
On 2007/11/16 16:56, Aaron S. [EMAIL PROTECTED] wrote: Stuart Henderson wrote: On 2007/11/15 14:07, Aaron S. [EMAIL PROTECTED] wrote: Okan Demirmen wrote: fyi, we've a minor formating fix on DESCR, but the larger issue is that pwman doesn't seem to work on at least macppc and possibly other big endian archs. Unfortunately, I don't have access to none of big endian machines. Do we add this for now? ONLY_FOR_ARCHS= amd64 i386 In general we try to fix these if possible. Well, there are more LE than these. I'm not sure whether they take too much care about alignment issues either, though, I didn't get time to test it much yet... Is there anything else needed for it to be committed? Someone to look at macppc and try and fix it, a committer who thinks it's ready, another to ok it. On sparc64, it dumps core (unaligned memory access) if it can't find the gpg binary: (gdb) bt #0 vfscanf (fp=0xe73394da24fe62cb, fmt0=0x20cce0 gpg (GnuPG) %d.%d.%d, ap=0xfffd4f70) at /usr/src/lib/libc/stdio/vfscanf.c:120 #1 0x4ccd2104 in fscanf (fp=0xe73394da24fe62cb, fmt=0x20cce0 gpg (GnuPG) %d.%d.%d) at /usr/src/lib/libc/stdio/fscanf.c:44 #2 0x0010a19c in gnupg_check_executable () at gnupg.c:339 #3 0x0010a2bc in gnupg_write_many (doc=0x47486600, ids=0xfffd5478, num_ids=1, filename=0x47486200 /home/sthen/.pwman.db) at gnupg.c:369 #4 0x0010a25c in gnupg_write (doc=0x47486600, id=0x47486000 [EMAIL PROTECTED], filename=0x47486200 /home/sthen/.pwman.db) at gnupg.c:356 #5 0x00103860 in pwlist_write_file () at pwlist.c:486 #6 0x0010756c in pwman_quit () at pwman.c:140 #7 0x001075c4 in main (argc=0, argv=0xfffd5788) at pwman.c:157 but the main part of the app does seem to work there (just tried saving one password, so not an extensive test...). also I just spotted this; COMMENT=console password manager based on gpgme the current version is based on gpg (they did switch to gpgme for a while but switched back again).
Re: draft port of security/pwman
Stuart Henderson wrote: On 2007/11/16 16:56, Aaron S. [EMAIL PROTECTED] wrote: Stuart Henderson wrote: On 2007/11/15 14:07, Aaron S. [EMAIL PROTECTED] wrote: Okan Demirmen wrote: fyi, we've a minor formating fix on DESCR, but the larger issue is that pwman doesn't seem to work on at least macppc and possibly other big endian archs. Unfortunately, I don't have access to none of big endian machines. Do we add this for now? ONLY_FOR_ARCHS= amd64 i386 In general we try to fix these if possible. Well, there are more LE than these. I'm not sure whether they take too much care about alignment issues either, though, I didn't get time to test it much yet... Is there anything else needed for it to be committed? Someone to look at macppc and try and fix it, a committer who thinks it's ready, another to ok it. On sparc64, it dumps core (unaligned memory access) if it can't find the gpg binary: (gdb) bt #0 vfscanf (fp=0xe73394da24fe62cb, fmt0=0x20cce0 gpg (GnuPG) %d.%d.%d, ap=0xfffd4f70) at /usr/src/lib/libc/stdio/vfscanf.c:120 #1 0x4ccd2104 in fscanf (fp=0xe73394da24fe62cb, fmt=0x20cce0 gpg (GnuPG) %d.%d.%d) at /usr/src/lib/libc/stdio/fscanf.c:44 #2 0x0010a19c in gnupg_check_executable () at gnupg.c:339 #3 0x0010a2bc in gnupg_write_many (doc=0x47486600, ids=0xfffd5478, num_ids=1, filename=0x47486200 /home/sthen/.pwman.db) at gnupg.c:369 #4 0x0010a25c in gnupg_write (doc=0x47486600, id=0x47486000 [EMAIL PROTECTED], filename=0x47486200 /home/sthen/.pwman.db) at gnupg.c:356 #5 0x00103860 in pwlist_write_file () at pwlist.c:486 #6 0x0010756c in pwman_quit () at pwman.c:140 #7 0x001075c4 in main (argc=0, argv=0xfffd5788) at pwman.c:157 but the main part of the app does seem to work there (just tried saving one password, so not an extensive test...). also I just spotted this; COMMENT=console password manager based on gpgme the current version is based on gpg (they did switch to gpgme for a while but switched back again). Fixed COMMENT, - http://www.x96.org/pwman.tar.gz
Re: New: games/lincity-ng
On Thu, Nov 15, 2007 at 03:30:13PM +0200, Antti Harri wrote: On Thu, 15 Nov 2007, Landry Breuil wrote: Builds fine too.. but i get a crash too with this message (after adding only some houses and waiting some time) : No Component specified in cell 2, 1 Yeah exactly the same bug I'm getting.. Maybe we should report this upstream instead of looking further? I'm not stopping anyone if someone wants to hack 'n patch, but at least I'm currently a bit busy :-) And i can't reproduce on amd64 and i386... anyone else? Simon
