Re: git daemon rc.d script

[Joerg Jung]

> Am 29.05.2016 um 22:45 schrieb Stuart Henderson :
> 
>> On 2016/05/29 20:16, Joerg Jung wrote:
>> Hi,
>> 
>> similar to svnserve from yesterday, please find below a diff which adds
>> a git daemon rc.d script and _gitdaemon user to devel/git.
>> 
>> Comments, OKs?
>> 
>> Regards,
>> Joerg
> 
> I think it wants an rc_reload=NO in the rc script.

Yes I'll add.

> Otherwise it looks
> fairly similar to my local rc.d script for this so with that added I'm
> ok with it - I'll continue to use my own script though because
> I certainly want to use chroot(8) for this :)

chroot how exactly, I guess via $daemon or $rcexec?
I think a daemon_chroot in rc.subr would be nice to simplify this.

Isn't using --base-path and --strict-path options not enough in this specific 
case?

Stumbled over this patch, but it seems never committed:
http://www.gelato.unsw.edu.au/archives/git/0509/9000.html

Re: [new] devel/cargo

[Sebastien Marie]

On Sun, May 29, 2016 at 07:25:44PM +0100, Edd Barrett wrote:
> On Sun, May 29, 2016 at 07:56:34PM +0200, Sebastien Marie wrote:
> > I prefer to keep it as it for now: a more "automatic" way will not be
> > able to copte with -MMDDnn format if I need to build more than one
> > archive some day.
> 
> Alright.
> 
> So I can't vouch for the magic in post-extract, but I don't see anything
> completely bonkers.

the magic is requires to make cargo (bootstrapper) to think it doesn't
need to do any network access. there is no automatic way build in
cargo to make it depend only on local filesystem. so I rebuild a
registry (cache, src, index) as cargo would have build it alone.

but I agree it is not a very clean method... but I don't think it could
be better for now.

> The last few comments I have:
> 
> > - Invokes rustc or another build tool with the correct parameters to build 
> > your
> 
> Lines in DESCR needs to be no more than 72 wide. Usually you can use
> `fmt -72`, but I think that will break your bullet list.

portcheck makes the verification on 80 wide. I will reformat before
commiting.

> > PERMIT_DISTFILES_FTP = No as hamcrest has no license information
> 
> After you import this, can you follow this up? Not having a package on
> ftp is going to be pretty annoying.

there is already one issue opened for proper licensing (dual MIT/Apache 2.0).
but as there are many contributors, each of them has to relicensing his
contributions: https://github.com/carllerche/hamcrest-rust/issues/22
and some previous contributors seems also inactive now :-/

but as hamcrest isn't build with cargo (it is only used in the test
suite), the cargo package is ok for distribution: it is only the
distfiles that can't be mirrored.
-- 
Sebastien Marie

Inkscape core dump

[Predrag Punosevac]

Running amd54 5.9 stable (upgraded from 5.8)

predrag@oko$ uname -a
OpenBSD oko.bagdala2.net 5.9 GENERIC.MP#4 amd64

predrag@oko$ pkg_info inkscape
Information for inst:inkscape-0.91p7

Comment:
SVG vector drawing application

Description:
Inkscape is a vector graphics editor, with capabilities similar
to Illustrator, CorelDraw, or Xara X, using the W3C standard Scalable
Vector Graphics (SVG) file format.

Maintainer: The OpenBSD ports mailing-list 

WWW: http://www.inkscape.org/


Trying to export SVG image as eps or ps will cause InkScape to crash

predrag@oko$ egdb
GNU gdb (GDB) 7.10.1
Copyright (C) 2015 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later

This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show
copying"
and "show warranty" for details.
This GDB was configured as "x86_64-unknown-openbsd5.9".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
.
Find the GDB manual and other documentation resources online at:
.
For help, type "help".
Type "apropos word" to search for commands related to "word".
(gdb) core-file inkscape.core
[New process 9413]
[New process 20111]
[New process 8182]
[New process 21072]
Core was generated by `inkscape'.
Program terminated with signal SIGABRT, Aborted.
#0  0x0b579bd0887a in ?? ()
[Current thread is 1 (process 9413)]


Thanks!
Predrag

Re: Version Bump Request for uGet Download Manager

[Juan Francisco Cantero Hurtado]

On Sun, May 29, 2016 at 03:01:37PM +0100, Stuart Henderson wrote:
> On 2016/05/29 05:39, Juan Francisco Cantero Hurtado wrote:
> > On Sat, May 28, 2016 at 08:41:51PM +0100, Stuart Henderson wrote:
> > > On 2016/05/28 11:20, Michael Tunnell wrote:
> > > > I'm not sure if this is the proper method for making these requests so I
> > > > apologize if it's not.
> > > > 
> > > > I'd like to request that uGet be version bump from 2.0.4 to the latest
> > > > version of 2.0.8, released on 16 May, 2016.
> > > 
> > > Ports diff below, but I don't use this so will need someone to test.
> > 
> > Sorry, I didn't see your patch before of I committed the update :( . I
> > forgot to remove the patches.
> 
> I've committed those.
> 
> What should we do about the --check-certificate=false nonsense?
> There's some --disable-ipv6 in there as well. It's 2016, it's
> ridiculous to use these flags.

OK? I've tested the program without the options downloading an iso from
a https URL. Everything works fine.

Index: Makefile
===
RCS file: /cvs/ports/www/uget/Makefile,v
retrieving revision 1.13
diff -u -p -r1.13 Makefile
--- Makefile29 May 2016 13:59:01 -  1.13
+++ Makefile29 May 2016 21:30:14 -
@@ -2,7 +2,7 @@
 
 COMMENT =  lightweight graphical download manager
 DISTNAME = uget-2.0.8
-REVISION = 0
+REVISION = 1
 CATEGORIES =   www net x11
 
 HOMEPAGE = http://ugetdm.com/
Index: patches/patch-tests_test-plugin+app_c
===
RCS file: patches/patch-tests_test-plugin+app_c
diff -N patches/patch-tests_test-plugin+app_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-tests_test-plugin+app_c   29 May 2016 21:30:14 -
@@ -0,0 +1,12 @@
+$OpenBSD$
+--- tests/test-plugin+app.c.orig   Sun May 29 23:23:28 2016
 tests/test-plugin+app.cSun May 29 23:23:41 2016
+@@ -159,7 +159,7 @@ void  test_setup_plugin_aria2 (void)
+   "C:\\Program Files\\uGet\\bin\\aria2c.exe");
+ #endif
+   uget_plugin_set (pinfo, UGET_PLUGIN_ARIA2_ARGUMENT,
+-  "--enable-rpc=true -D --check-certificate=false");
++  "--enable-rpc=true -D");
+   uget_plugin_set (pinfo, UGET_PLUGIN_ARIA2_LAUNCH, (void*) TRUE);
+   uget_plugin_set (pinfo, UGET_PLUGIN_ARIA2_SHUTDOWN, (void*) TRUE);
+   ug_sleep (1000);
Index: patches/patch-uget_UgetAria2_c
===
RCS file: /cvs/ports/www/uget/patches/patch-uget_UgetAria2_c,v
retrieving revision 1.2
diff -u -p -r1.2 patch-uget_UgetAria2_c
--- patches/patch-uget_UgetAria2_c  29 May 2016 13:59:01 -  1.2
+++ patches/patch-uget_UgetAria2_c  29 May 2016 21:30:14 -
@@ -1,12 +1,14 @@
 $OpenBSD: patch-uget_UgetAria2_c,v 1.2 2016/05/29 13:59:01 sthen Exp $
 uget/UgetAria2.c.orig  Sun May 15 19:28:25 2016
-+++ uget/UgetAria2.c   Sat May 28 20:30:26 2016
-@@ -53,7 +53,7 @@
+--- uget/UgetAria2.c.orig  Sun May 15 20:28:25 2016
 uget/UgetAria2.c   Sun May 29 23:21:36 2016
+@@ -53,8 +53,8 @@
  #define RPC_URI  "http://localhost:6800/jsonrpc";
  #define RPC_BATCH_LEN5
  #define RPC_INTERVAL 500
 -#define ARIA2_PATH   "aria2c"
+-#define ARIA2_ARGS   "--enable-rpc=true -D --check-certificate=false"
 +#define ARIA2_PATH   "${TRUEPREFIX}/bin/aria2c"
- #define ARIA2_ARGS   "--enable-rpc=true -D --check-certificate=false"
++#define ARIA2_ARGS   "--enable-rpc=true -D"
  
  #if defined _WIN32 || defined _WIN64
+ #define  ug_sleep Sleep
Index: patches/patch-uget_UgetAria2_c.orig
===
RCS file: patches/patch-uget_UgetAria2_c.orig
diff -N patches/patch-uget_UgetAria2_c.orig
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-uget_UgetAria2_c.orig 29 May 2016 21:30:14 -
@@ -0,0 +1,12 @@
+$OpenBSD: patch-uget_UgetAria2_c,v 1.2 2016/05/29 13:59:01 sthen Exp $
+--- uget/UgetAria2.c.orig  Sun May 15 19:28:25 2016
 uget/UgetAria2.c   Sat May 28 20:30:26 2016
+@@ -53,7 +53,7 @@
+ #define RPC_URI  "http://localhost:6800/jsonrpc";
+ #define RPC_BATCH_LEN5
+ #define RPC_INTERVAL 500
+-#define ARIA2_PATH   "aria2c"
++#define ARIA2_PATH   "${TRUEPREFIX}/bin/aria2c"
+ #define ARIA2_ARGS   "--enable-rpc=true -D --check-certificate=false"
+ 
+ #if defined _WIN32 || defined _WIN64
Index: patches/patch-ui-gtk-1to2_UgSetting_c
===
RCS file: patches/patch-ui-gtk-1to2_UgSetting_c
diff -N patches/patch-ui-gtk-1to2_UgSetting_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-ui-gtk-1to2_UgSetting_c   29 May 2016 21:30:14 -
@@ -0,0 +1,12 @@
+$OpenBSD$
+--- ui-gtk-1to2/UgSetting.c.orig   Sun May 29 23:24:17 2016
 ui-gtk-1to2/UgSett

Re: Update: devel/subversion

[Dmitrij D. Czarkoff]

Stefan Sperling said:
> On Sun, May 29, 2016 at 10:02:14PM +0200, Dmitrij D. Czarkoff wrote:
> > The port is in openbsd-wip.  Feel free to modify it.
> 
> You don't want to send an updated diff after a commit to CVS happened?
> If so, why not?
> 
> Do you generally expect to get contributions to your diffs on github?

No.  I just don't have the updated diff right now.  I'll integrate the
change when I'll have time.  But I won't mind if someone else does it,
thus mention of openbsd-wip.

-- 
Dmitrij D. Czarkoff

Re: [new] devel/cargo

[Stuart Henderson]

On 2016/05/29 19:25, Edd Barrett wrote:
> On Sun, May 29, 2016 at 07:56:34PM +0200, Sebastien Marie wrote:
> > I prefer to keep it as it for now: a more "automatic" way will not be
> > able to copte with -MMDDnn format if I need to build more than one
> > archive some day.
> 
> Alright.
> 
> So I can't vouch for the magic in post-extract, but I don't see anything
> completely bonkers.
> 
> The last few comments I have:
> 
> > - Invokes rustc or another build tool with the correct parameters to build 
> > your
> 
> Lines in DESCR needs to be no more than 72 wide. Usually you can use
> `fmt -72`, but I think that will break your bullet list.

up to 80 is ok here, a rigid 72 means more awkward linebreaks.
making it look nice is more important as long as it fits in a normal
terminal width :)

> > PERMIT_DISTFILES_FTP = No as hamcrest has no license information
> 
> After you import this, can you follow this up? Not having a package on
> ftp is going to be pretty annoying.

it's only DISTFILES.

Update sourceforge.jp mirrors

[Anthony J. Bentley]

These mirrors are dead.

ok?


Index: infrastructure/templates/network.conf.template
===
RCS file: /cvs/ports/infrastructure/templates/network.conf.template,v
retrieving revision 1.205
diff -u -p -r1.205 network.conf.template
--- infrastructure/templates/network.conf.template  28 May 2016 13:51:18 
-  1.205
+++ infrastructure/templates/network.conf.template  29 May 2016 21:02:15 
-
@@ -118,9 +118,7 @@ MASTER_SITE_SOURCEFORGE+=   \
http://downloads.sourceforge.net/sourceforge/
 
 MASTER_SITE_SOURCEFORGE_JP+=   \
-   http://osdn.dl.sourceforge.jp/ \
-   http://downloads.sourceforge.jp/ \
-   http://prdownloads.sourceforge.jp/
+   http://osdn.dl.sourceforge.jp/
 
 MASTER_SITE_GNOME+=\
http://ftp.acc.umu.se/pub/GNOME/ \

Re: [NEW] net/geoipupdate

[Frederic Cambus]

On Thu, May 26, 2016 at 11:39:13PM +0100, Stuart Henderson wrote:
> > 
> > The GeoIP Update program performs automatic updates of GeoIP2 and GeoIP
> > Legacy binary databases.
>
> here are some bits on top;
> 
> - @sample the config file
> - patch config to use the directory net/GeoIP is already using
> (these "updated at runtime" things shouldn't update in /usr/local)
> - patch to make it easier for free users (commercial users have an
> online config generator page anyway)

I checked your patch and everything looks good to me, thanks for your
input.

> but I think maybe we should also add shepherd users towards running it
> as a non-root user by previding a separate uid to own /var/db/GeoIP files ..

That makes sense. So if I picture things correctly, that would mean
something in those lines:

1) Adding a _geoip user/group in /usr/ports/infrastructure/db/user.list

2) Tweak net/GeoIP and net/libmaxminddb to add @newgroup and @newuser 
directives in PLIST to effectively create user/group, set @user and
@group to _geoip, and lastly use @mode 660 so any users in the group 
_geoip can run the geoipupdate program

Is that what you had in mind?

Re: Update: devel/subversion

[Stefan Sperling]

On Sun, May 29, 2016 at 10:02:14PM +0200, Dmitrij D. Czarkoff wrote:
> The port is in openbsd-wip.  Feel free to modify it.

You don't want to send an updated diff after a commit to CVS happened?
If so, why not?

Do you generally expect to get contributions to your diffs on github?

You can use whatever tooling you like internally to your private process,
of course. But before starting this thread, you also asked me to look at
openbsd-wip, and I asked you to please post the patch here. The reason I
asked you to do this is that I'd like us to stick to the established process
for collaboration (i.e. mail out a diff, ask for ok, if ok then commit to cvs,
rinse repeat). I can barely keep track of all that's happening on the mailing
lists, so I'm not going to waste additional time on github or wherever else.

Re: git daemon rc.d script

[Stuart Henderson]

On 2016/05/29 20:16, Joerg Jung wrote:
> Hi,
> 
> similar to svnserve from yesterday, please find below a diff which adds
> a git daemon rc.d script and _gitdaemon user to devel/git.
> 
> Comments, OKs?
> 
> Regards,
> Joerg

I think it wants an rc_reload=NO in the rc script. Otherwise it looks
fairly similar to my local rc.d script for this so with that added I'm
ok with it - I'll continue to use my own script though because
I certainly want to use chroot(8) for this :)

Re: [NEW] net/libmaxminddb

[Frederic Cambus]

On Thu, May 26, 2016 at 10:12:02AM +0100, Stuart Henderson wrote:
> > 
> > The libmaxminddb library provides a C library for reading MaxMind DB files,
> > including the GeoIP2 databases from MaxMind. This is a custom binary format
> > designed to facilitate fast lookups of IP adresses while allowing for great
> > flexibility in the type of data associated with an address.
> > 
> 
> We could fit a bit more information into COMMENT, how about this?

> library for MaxMind GeoIP2/GeoLite2 IP geolocation databases

Sounds better indeed, used this as COMMENT-main.

> Do you fancy adding GeoLite2 dbs as subpackages, something like
> net/GeoIP does? (They could go in a separate port but I think it
> makes it easier for people to find them if it's in the main one).

Sure, here is a new tarball implementing the suggestion, I tried to
stay as close as possible to what net/GeoIP is doing.

The DB files are installed in ${LOCALSTATEDIR}/db/GeoIP/, same path
as net/GeoIP. I'm not sure if it's a good idea or not, but in case
geoipupdate gets imported, it would allow to update both GeoLite and 
GeoLite2 databases with only one configuration file.



libmaxminddb.tar.gz
Description: application/tar-gz

Re: WIP: tex live 2015

[Ingo Schwarze]

Hi Edd,

Edd Barrett wrote on Sun, May 29, 2016 at 04:04:05PM +0100:
> On Tue, May 24, 2016 at 06:54:33PM +0100, Edd Barrett wrote:

>> Comments? OK? We need to think carefully if this goes in pre or post
>> 6.0.

> To keep people in the loop, I now have an OK for this, but I will wait
> until after 6.0 to commit.

Why?  We are not anywhere near slowdown yet.
Even if something breaks, you have plenty of time
to get reports and fix it.

Perhaps your mean because -current says -beta?
Since a few releases, Theo bumps to -beta early in order
to keep the associated churn far away from release time.
It doesn't imply there is a need to hold back any improvements.

Yours,
  Ingo

Re: Update: devel/subversion

[Dmitrij D. Czarkoff]

Gleydson Soares said:
> 
> Could you merge it [1] on top of your diff?
> 
> It adds a rcscript and _svnserver user to db/user.list
> (Sent by jung@ yesterday) [1]

The port is in openbsd-wip.  Feel free to modify it.

-- 
Dmitrij D. Czarkoff

Re: [new] devel/cargo

[Edd Barrett]

On Sun, May 29, 2016 at 07:56:34PM +0200, Sebastien Marie wrote:
> I prefer to keep it as it for now: a more "automatic" way will not be
> able to copte with -MMDDnn format if I need to build more than one
> archive some day.

Alright.

So I can't vouch for the magic in post-extract, but I don't see anything
completely bonkers.

The last few comments I have:

> - Invokes rustc or another build tool with the correct parameters to build 
> your

Lines in DESCR needs to be no more than 72 wide. Usually you can use
`fmt -72`, but I think that will break your bullet list.

> PERMIT_DISTFILES_FTP = No as hamcrest has no license information

After you import this, can you follow this up? Not having a package on
ftp is going to be pretty annoying.

-- 
Best Regards
Edd Barrett

http://www.theunixzoo.co.uk

add py3 flavor to devel/py-click (and configure the tests)

[Daniel Jakots]

Hi,

It looks like it will be needed by www/py-flask 0.11 (I began to work
on it, though I'm waiting for the distfile to be available
through pypi).

Comments? OK?

Cheers,
Daniel

Index: Makefile
===
RCS file: /cvs/ports/devel/py-click/Makefile,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 Makefile
--- Makefile7 Jan 2016 11:35:16 -   1.1.1.1
+++ Makefile29 May 2016 18:01:06 -
@@ -5,6 +5,7 @@ COMMENT=command line library for Python
 MODPY_EGG_VERSION=6.2
 DISTNAME=  click-${MODPY_EGG_VERSION}
 PKGNAME=   py-click-${MODPY_EGG_VERSION}
+REVISION=  0
 
 CATEGORIES=devel
 
@@ -20,5 +21,13 @@ MODPY_PI =   Yes
 MODULES=   lang/python
 
 MODPY_SETUPTOOLS= Yes
+
+FLAVORS=   python3
+FLAVOR?=
+
+TEST_DEPENDS=  devel/py-test${MODPY_FLAVOR}
+
+do-test:
+   @cd ${WRKSRC} && env PYTHONPATH=${WRKINST}/${MODPY_SITEPKG} 
${MODPY_BIN} -m pytest
 
 .include 
Index: pkg/PLIST
===
RCS file: /cvs/ports/devel/py-click/pkg/PLIST,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 PLIST
--- pkg/PLIST   7 Jan 2016 11:35:16 -   1.1.1.1
+++ pkg/PLIST   29 May 2016 18:01:06 -
@@ -6,36 +6,37 @@ lib/python${MODPY_VERSION}/site-packages
 
lib/python${MODPY_VERSION}/site-packages/click-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/dependency_links.txt
 
lib/python${MODPY_VERSION}/site-packages/click-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/top_level.txt
 lib/python${MODPY_VERSION}/site-packages/click/__init__.py
-lib/python${MODPY_VERSION}/site-packages/click/__init__.pyc
+${MODPY_COMMENT}lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}/
+lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}__init__.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}_bashcomplete.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}_compat.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}_termui_impl.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}_textwrap.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}_unicodefun.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}_winconsole.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}core.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}decorators.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}exceptions.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}formatting.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}globals.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}parser.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}termui.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}testing.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}types.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/click/${MODPY_PYCACHE}utils.${MODPY_PYC_MAGIC_TAG}pyc
 lib/python${MODPY_VERSION}/site-packages/click/_bashcomplete.py
-lib/python${MODPY_VERSION}/site-packages/click/_bashcomplete.pyc
 lib/python${MODPY_VERSION}/site-packages/click/_compat.py
-lib/python${MODPY_VERSION}/site-packages/click/_compat.pyc
 lib/python${MODPY_VERSION}/site-packages/click/_termui_impl.py
-lib/python${MODPY_VERSION}/site-packages/click/_termui_impl.pyc
 lib/python${MODPY_VERSION}/site-packages/click/_textwrap.py
-lib/python${MODPY_VERSION}/site-packages/click/_textwrap.pyc
 lib/python${MODPY_VERSION}/site-packages/click/_unicodefun.py
-lib/python${MODPY_VERSION}/site-packages/click/_unicodefun.pyc
 lib/python${MODPY_VERSION}/site-packages/click/_winconsole.py
-lib/python${MODPY_VERSION}/site-packages/click/_winconsole.pyc
 lib/python${MODPY_VERSION}/site-packages/click/core.py
-lib/python${MODPY_VERSION}/site-packages/click/core.pyc
 lib/python${MODPY_VERSION}/site-packages/click/decorators.py
-lib/python${MODPY_VERSION}/site-packages/click/decorators.pyc
 lib/python${MODPY_VERSION}/site-packages/click/exceptions.py
-lib/python${MODPY_VERSION}/site-packages/click/exceptions.pyc
 lib/python${MODPY_VERSION}/site-packages/click/formatting.py
-lib/python${MODPY_VERSION}/site-packages/click/formatting.pyc
 lib/python${MODPY_VERSION}/site-packages/click/globals.py
-lib/python${MODPY_VERSION}/site-packages/click/globals.pyc
 lib/python${MODPY_VERSION}/site-packages/click/parser.py
-lib/python${MODPY_VERSION}/site-packages/click/parser.pyc
 lib/python${MODPY_VERSION}/site-packages/click/termui.py
-lib/python${MO

git daemon rc.d script

[Joerg Jung]

Hi,

similar to svnserve from yesterday, please find below a diff which adds
a git daemon rc.d script and _gitdaemon user to devel/git.

Comments, OKs?

Regards,
Joerg


Index: infrastructure/db/user.list
===
RCS file: /cvs/ports/infrastructure/db/user.list,v
retrieving revision 1.274
diff -u -p -r1.274 user.list
--- infrastructure/db/user.list 29 May 2016 17:38:38 -  1.274
+++ infrastructure/db/user.list 29 May 2016 18:11:12 -
@@ -286,3 +286,4 @@ id  usergroup   port options
 775 _letskencrypt  _letskencrypt   security/letskencrypt
 776 _ioq3  _ioq3   games/ioquake3
 777 _svnserve  _svnserve   devel/subversion
+778 _gitdaemon _gitdaemon  devel/git
Index: devel/git/Makefile
===
RCS file: /cvs/ports/devel/git/Makefile,v
retrieving revision 1.154
diff -u -p -r1.154 Makefile
--- devel/git/Makefile  25 May 2016 07:13:05 -  1.154
+++ devel/git/Makefile  29 May 2016 18:11:13 -
@@ -7,6 +7,7 @@ COMMENT-x11 =   GIT - graphical tools
 V =2.8.3
 DISTNAME = git-${V}
 PKGNAME-main = ${DISTNAME}
+REVISION-main =0
 PKGNAME-svn =  git-svn-${V}
 PKGNAME-x11 =  git-x11-${V}
 CATEGORIES =   devel
Index: devel/git/pkg/PLIST-main
===
RCS file: /cvs/ports/devel/git/pkg/PLIST-main,v
retrieving revision 1.70
diff -u -p -r1.70 PLIST-main
--- devel/git/pkg/PLIST-main17 May 2016 11:54:58 -  1.70
+++ devel/git/pkg/PLIST-main29 May 2016 18:11:13 -
@@ -417,3 +417,4 @@ share/locale/ru/LC_MESSAGES/git.mo
 share/locale/sv/LC_MESSAGES/git.mo
 share/locale/vi/LC_MESSAGES/git.mo
 share/locale/zh_CN/LC_MESSAGES/git.mo
+@rcscript ${RCDIR}/gitdaemon
Index: devel/git/pkg/gitdaemon.rc
===
RCS file: devel/git/pkg/gitdaemon.rc
diff -N devel/git/pkg/gitdaemon.rc
--- /dev/null   1 Jan 1970 00:00:00 -
+++ devel/git/pkg/gitdaemon.rc  29 May 2016 18:11:13 -
@@ -0,0 +1,13 @@
+#!/bin/sh
+#
+# $OpenBSD$
+
+daemon="${TRUEPREFIX}/bin/git daemon --detach"
+daemon_flags="--user=_gitdaemon"
+daemon_user="_gitdaemon"
+
+. /etc/rc.d/rc.subr
+
+pexp="git-daemon --detach ${daemon_flags:+ ${daemon_flags}}"
+
+rc_cmd $1

Re: maintainer update to py-cares-2.0.0

[Daniel Jakots]

On Sun, 29 May 2016 16:09:01 +0500, Alexandr Shadchin
 wrote:

> On Sun, May 29, 2016 at 11:52:52AM +0200, Daniel Jakots wrote:
> > On Fri, 27 May 2016 22:42:30 +0200, Daniel Jakots
> >  wrote:
> >   
> > > Hi,
> > > 
> > > Here's a diff to update py-cares to latest release.  
> > 
> > Just after that they did a bugfix release.
> >
> > > Comments? OK?  
> >   
> 
> ok shadchin@
> 
> I would have included tests:
> do-test: fake
>cd ${WRKSRC} && env PYTHONPATH=${WRKINST}/${MODPY_SITEPKG}
> ${MODPY_BIN} tests/tests.py

Great, thanks!

Index: Makefile
===
RCS file: /cvs/ports/net/py-cares/Makefile,v
retrieving revision 1.2
diff -u -p -r1.2 Makefile
--- Makefile17 May 2016 05:12:02 -  1.2
+++ Makefile29 May 2016 18:20:48 -
@@ -3,7 +3,7 @@
 # using bundled c-ares which is patched for TTL support
 COMMENT =  python interface for c-ares
 
-MODPY_EGG_VERSION =1.0.0
+MODPY_EGG_VERSION =2.0.1
 DISTNAME = pycares-${MODPY_EGG_VERSION}
 PKGNAME =  py-cares-${MODPY_EGG_VERSION}
 CATEGORIES =   net
@@ -24,5 +24,9 @@ USE_GMAKE =   Yes
 
 FLAVORS =  python3
 FLAVOR ?=
+
+do-test: fake
+   cd ${WRKSRC} && env PYTHONPATH=${WRKINST}/${MODPY_SITEPKG} \
+   ${MODPY_BIN} tests/tests.py
 
 .include 
Index: distinfo
===
RCS file: /cvs/ports/net/py-cares/distinfo,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 distinfo
--- distinfo7 May 2016 14:46:08 -   1.1.1.1
+++ distinfo29 May 2016 18:20:48 -
@@ -1,2 +1,2 @@
-SHA256 (pycares-1.0.0.tar.gz) = oYNB6gMOLMB0Os30qnIwK99rggk4s2zkvXbkP6oidqM=
-SIZE (pycares-1.0.0.tar.gz) = 220165
+SHA256 (pycares-2.0.1.tar.gz) = OmhMCbdT4UPYYZC1fXG9gtJEkgYtEdaraswhNJ0to00=
+SIZE (pycares-2.0.1.tar.gz) = 225788
Index: pkg/PLIST
===
RCS file: /cvs/ports/net/py-cares/pkg/PLIST,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 PLIST
--- pkg/PLIST   7 May 2016 14:46:08 -   1.1.1.1
+++ pkg/PLIST   29 May 2016 18:20:48 -
@@ -1,7 +1,15 @@
 @comment $OpenBSD: PLIST,v 1.1.1.1 2016/05/07 14:46:08 danj Exp $
+lib/python${MODPY_VERSION}/site-packages/pycares/
 
lib/python${MODPY_VERSION}/site-packages/pycares-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/
 
lib/python${MODPY_VERSION}/site-packages/pycares-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/PKG-INFO
 
lib/python${MODPY_VERSION}/site-packages/pycares-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/SOURCES.txt
 
lib/python${MODPY_VERSION}/site-packages/pycares-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/dependency_links.txt
 
lib/python${MODPY_VERSION}/site-packages/pycares-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/top_level.txt
-lib/python${MODPY_VERSION}/site-packages/pycares.so
+lib/python${MODPY_VERSION}/site-packages/pycares/__init__.py
+${MODPY_COMMENT}lib/python${MODPY_VERSION}/site-packages/pycares/${MODPY_PYCACHE}/
+lib/python${MODPY_VERSION}/site-packages/pycares/${MODPY_PYCACHE}__init__.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/pycares/${MODPY_PYCACHE}_version.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/pycares/${MODPY_PYCACHE}errno.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/pycares/_core.so
+lib/python${MODPY_VERSION}/site-packages/pycares/_version.py
+lib/python${MODPY_VERSION}/site-packages/pycares/errno.py

Re: [new] devel/cargo

[Sebastien Marie]

On Sun, May 29, 2016 at 06:31:52PM +0100, Edd Barrett wrote:
> On Sun, May 29, 2016 at 04:40:09PM +0100, Edd Barrett wrote:
> > I need to test the bootstrap target. Watch this space.
> 
> Alright, had a look at this.
> 
> > # make a directory suitable for bootstrapping
> > BOOTSTRAP_NVERSION = ${CARGO_VERSION}-new
> > bootstrap: fake
> > ...
> 
> Could we make it so that the 'bootstrap' target makes the bootstrap
> tarball distfile directly? Is that just a matter of tarring up the
> ${CARGO_VERSION}-new directory (maybe with the right directory name)?
> 

the 'bootstrap' target is more a tool for maintener than a "standard"
target for users. and about the current use I have of it, yes it is just
renaming -new to -datelikeformat, make a .tar.gz and upload it.

I prefer to keep it as it for now: a more "automatic" way will not be
able to copte with -MMDDnn format if I need to build more than one
archive some day.

thanks for your review.
-- 
Sebastien Marie

Re: sparc64 gd or libpng coredump

[Tinker]

On 2016-05-30 00:47, Otto Moerbeek wrote:

On Sun, May 29, 2016 at 06:48:53PM +0200, Tinker wrote:


On 2016-05-29 18:24, Otto Moerbeek wrote:
> Hi,
>
> running analog on current sparc64 gives more coredumps coming from
> libpng (via gd). At the moment I do not have time to investigate this,
> so if anyone wants to give it a shot, I'd really appreciate it,
>
> #0  0x000f68e018c0 in ?? () from /usr/local/lib/libpng.so.17.2
> #1  0x000f688f4fa4 in png_create_write_struct ()
>from /usr/local/lib/libpng.so.17.2
> #2  0x000ffc0c1ab0 in gdImagePngCtxEx () from
> /usr/local/lib/libgd.so.21.1
> #3  0x000ffc0c2784 in gdImagePng () from
> /usr/local/lib/libgd.so.21.1
> #4  0x000d4a44c270 in piechart_write () from /usr/local/bin/analog
> #5  0x000d4a44dfa0 in piechart () from /usr/local/bin/analog
> #6  0x000d4a4413d0 in genrep () from /usr/local/bin/analog
> #7  0x000d4a44450c in output () from /usr/local/bin/analog
> #8  0x000d4a426688 in main () from /usr/local/bin/analog
>
>-Otto


Otto,

Does this mean that what you have encountered is the realization that 
libpng

is actually *BROKEN* on sparc64 currently?

(Or maybe gd, we don't know;) How frequent are the crashes?

Thanks!
Tinker


It's 100% reproducable.

-Otto


Ouch. Ok so libpng (maybe also libgd) is broken currently on sparc64.

Did it ever work in the past for you?

Re: sparc64 gd or libpng coredump

[Otto Moerbeek]

On Mon, May 30, 2016 at 12:55:21AM +0700, Tinker wrote:

> On 2016-05-30 00:47, Otto Moerbeek wrote:
> > On Sun, May 29, 2016 at 06:48:53PM +0200, Tinker wrote:
> > 
> > > On 2016-05-29 18:24, Otto Moerbeek wrote:
> > > > Hi,
> > > >
> > > > running analog on current sparc64 gives more coredumps coming from
> > > > libpng (via gd). At the moment I do not have time to investigate this,
> > > > so if anyone wants to give it a shot, I'd really appreciate it,
> > > >
> > > > #0  0x000f68e018c0 in ?? () from /usr/local/lib/libpng.so.17.2
> > > > #1  0x000f688f4fa4 in png_create_write_struct ()
> > > >from /usr/local/lib/libpng.so.17.2
> > > > #2  0x000ffc0c1ab0 in gdImagePngCtxEx () from
> > > > /usr/local/lib/libgd.so.21.1
> > > > #3  0x000ffc0c2784 in gdImagePng () from
> > > > /usr/local/lib/libgd.so.21.1
> > > > #4  0x000d4a44c270 in piechart_write () from /usr/local/bin/analog
> > > > #5  0x000d4a44dfa0 in piechart () from /usr/local/bin/analog
> > > > #6  0x000d4a4413d0 in genrep () from /usr/local/bin/analog
> > > > #7  0x000d4a44450c in output () from /usr/local/bin/analog
> > > > #8  0x000d4a426688 in main () from /usr/local/bin/analog
> > > >
> > > > -Otto
> > > 
> > > 
> > > Otto,
> > > 
> > > Does this mean that what you have encountered is the realization
> > > that libpng
> > > is actually *BROKEN* on sparc64 currently?
> > > 
> > > (Or maybe gd, we don't know;) How frequent are the crashes?
> > > 
> > > Thanks!
> > > Tinker
> > 
> > It's 100% reproducable.
> > 
> > -Otto
> 
> Ouch. Ok so libpng (maybe also libgd) is broken currently on sparc64.
> 
> Did it ever work in the past for you?
> 

Yes, I've run analog on this machine for quite some years.

-Otto

Re: Update: devel/subversion

[Joerg Jung]

On Sun, May 29, 2016 at 10:29:32AM -0300, Gleydson Soares wrote:
> 
> Could you merge it [1] on top of your diff?
> 
> It adds a rcscript and _svnserver user to db/user.list
> (Sent by jung@ yesterday) [1]

I've committed this already.

> [1] http://marc.info/?l=openbsd-ports&m=146447355118340&w=2
>
> Dmitrij D. Czarkoff  writes:
> 
> > Hi!
> >
> > Below is an update for devel/subversion (1.8.15 => 1.9.4).  It is only
> > slightly tested because I am not a subversion user, and I only needed
> > svn as a dependency for another port I am working on.  Please test.
> 

Re: [new] devel/cargo

[Edd Barrett]

On Sun, May 29, 2016 at 04:40:09PM +0100, Edd Barrett wrote:
> I need to test the bootstrap target. Watch this space.

Alright, had a look at this.

> # make a directory suitable for bootstrapping
> BOOTSTRAP_NVERSION = ${CARGO_VERSION}-new
> bootstrap: fake
> ...

Could we make it so that the 'bootstrap' target makes the bootstrap
tarball distfile directly? Is that just a matter of tarring up the
${CARGO_VERSION}-new directory (maybe with the right directory name)?

-- 
Best Regards
Edd Barrett

http://www.theunixzoo.co.uk

Re: Update: devel/subversion

[Stefan Sperling]

On Sun, May 29, 2016 at 03:03:06PM +0100, Stuart Henderson wrote:
> I can't comment on the svn update, other than there is very likely
> to be a good reason why Stefan hasn't updated it yet.

Just for a lack of time. Up to 1.9.3 there was a known memory corruption bug
affecting sparc64, so I hesitated. That's fixed in 1.9.4, so we can update.

[new] mosml 2.10.1

[Yozo TODA]

Recent messages on mlton reminds me about mosml port I submitted 3 years ago.
here is a port stuff updated to the current version 2.10.1.
I confirmed this on amd64.

  [DESCR]
  Moscow ML implements Standard ML (SML), as defined in the 1997
  Definition of Standard ML, including the SML Modules language
  and some extensions.  Moreover, Moscow ML supports most required parts
  of the SML Basis Library.  It supports separate compilation and the
  generation of stand-alone executables.

anyone please check and commit to the ports tree?

-- yozo.



mosml-2.10.1.tar.gz
Description: Binary data

sparc64 gd or libpng coredump

[Otto Moerbeek]

Hi,

running analog on current sparc64 gives more coredumps coming from
libpng (via gd). At the moment I do not have time to investigate this,
so if anyone wants to give it a shot, I'd really appreciate it,

#0  0x000f68e018c0 in ?? () from /usr/local/lib/libpng.so.17.2
#1  0x000f688f4fa4 in png_create_write_struct ()
   from /usr/local/lib/libpng.so.17.2
#2  0x000ffc0c1ab0 in gdImagePngCtxEx () from /usr/local/lib/libgd.so.21.1
#3  0x000ffc0c2784 in gdImagePng () from /usr/local/lib/libgd.so.21.1
#4  0x000d4a44c270 in piechart_write () from /usr/local/bin/analog
#5  0x000d4a44dfa0 in piechart () from /usr/local/bin/analog
#6  0x000d4a4413d0 in genrep () from /usr/local/bin/analog
#7  0x000d4a44450c in output () from /usr/local/bin/analog
#8  0x000d4a426688 in main () from /usr/local/bin/analog

-Otto

Re: [new] devel/cargo

[Edd Barrett]

On Sun, May 29, 2016 at 12:10:25PM +0200, Sebastien Marie wrote:
> ping ?

I was able to initialise a project and define a dependency and use it
within a tiny test program. So looks good from a functionality
standpoint.

This was with the in-tree rust-1.8.0. I have not tried it with 1.9.0.

I need to test the bootstrap target. Watch this space.

-- 
Best Regards
Edd Barrett

http://www.theunixzoo.co.uk

Re: WIP: tex live 2015

[Edd Barrett]

On Sun, May 29, 2016 at 05:13:38PM +0200, Martin Schröder wrote:
> 2016-05-29 17:04 GMT+02:00 Edd Barrett :
> > To keep people in the loop, I now have an OK for this, but I will wait
> > until after 6.0 to commit.
> 
> Any chance to skip TL2015 and directly go to TL2016, which will be
> released shortly?

I would rather use 2015 as an intermediate step, even if it is
relatively short-lived.

I will be looking at 2016 at the general hackathon later this year.

-- 
Best Regards
Edd Barrett

http://www.theunixzoo.co.uk

Re: Version Bump Request for uGet Download Manager

[Michael Tunnell]

Thank you both very much for your assistance!

> "I've updated the package. We have not an official maintainer for the
package, so sometimes we miss the new versions."

That's fine with me, I don't mind sending update requests as long as you
all don't mind receiving them. :)

On Sat, May 28, 2016 at 9:07 PM, Juan Francisco Cantero Hurtado <
i...@juanfra.info> wrote:

> On Sat, May 28, 2016 at 11:20:29AM -0500, Michael Tunnell wrote:
> > I'm not sure if this is the proper method for making these requests so I
> > apologize if it's not.
> >
> > I'd like to request that uGet be version bump from 2.0.4 to the latest
> > version of 2.0.8, released on 16 May, 2016.
>
> I've updated the package. We have not an official maintainer for the
> package, so sometimes we miss the new versions.
>
> --
> Juan Francisco Cantero Hurtado http://juanfra.info
>

Re: WIP: tex live 2015

[Martin Schröder]

2016-05-29 17:04 GMT+02:00 Edd Barrett :
> To keep people in the loop, I now have an OK for this, but I will wait
> until after 6.0 to commit.

Any chance to skip TL2015 and directly go to TL2016, which will be
released shortly?

Best
   Martin

Re: WIP: tex live 2015

[Edd Barrett]

On Tue, May 24, 2016 at 06:54:33PM +0100, Edd Barrett wrote:
> Comments? OK? We need to think carefully if this goes in pre or post
> 6.0.

To keep people in the loop, I now have an OK for this, but I will wait
until after 6.0 to commit.

The pregen tarball has also now been mirrored.

-- 
Best Regards
Edd Barrett

http://www.theunixzoo.co.uk

Re: Update: devel/subversion

[Stuart Henderson]

On 2016/05/29 11:35, Dmitrij D. Czarkoff wrote:
> Hi!
> 
> Below is an update for devel/subversion (1.8.15 => 1.9.4).  It is only
> slightly tested because I am not a subversion user, and I only needed
> svn as a dependency for another port I am working on.  Please test.

I can't comment on the svn update, other than there is very likely
to be a good reason why Stefan hasn't updated it yet. But:

> Index: pkg/PLIST-perl
> ===
> RCS file: /cvs/ports/devel/subversion/pkg/PLIST-perl,v
> retrieving revision 1.8
> diff -u -p -r1.8 PLIST-perl
> --- pkg/PLIST-perl14 Jun 2014 10:34:44 -  1.8
> +++ pkg/PLIST-perl29 May 2016 09:29:28 -
> @@ -3,6 +3,9 @@
>  lib/libsvn_swig_perl-1.a
>  lib/libsvn_swig_perl-1.la
>  @lib lib/libsvn_swig_perl-1.so.${LIBsvn_swig_perl-1_VERSION}
> +libdata/perl5/${MACHINE_ARCH}-openbsd/
> +libdata/perl5/${MACHINE_ARCH}-openbsd/5.20.2/
> +libdata/perl5/${MACHINE_ARCH}-openbsd/5.20.2/perllocal.pod
>  libdata/perl5/site_perl/${MACHINE_ARCH}-openbsd/
>  libdata/perl5/site_perl/${MACHINE_ARCH}-openbsd/SVN/
>  libdata/perl5/site_perl/${MACHINE_ARCH}-openbsd/SVN/Base.pm
> 

This is definitely wrong, that file shouldn't be installed.

Re: Version Bump Request for uGet Download Manager

[Stuart Henderson]

On 2016/05/29 05:39, Juan Francisco Cantero Hurtado wrote:
> On Sat, May 28, 2016 at 08:41:51PM +0100, Stuart Henderson wrote:
> > On 2016/05/28 11:20, Michael Tunnell wrote:
> > > I'm not sure if this is the proper method for making these requests so I
> > > apologize if it's not.
> > > 
> > > I'd like to request that uGet be version bump from 2.0.4 to the latest
> > > version of 2.0.8, released on 16 May, 2016.
> > 
> > Ports diff below, but I don't use this so will need someone to test.
> 
> Sorry, I didn't see your patch before of I committed the update :( . I
> forgot to remove the patches.

I've committed those.

What should we do about the --check-certificate=false nonsense?
There's some --disable-ipv6 in there as well. It's 2016, it's
ridiculous to use these flags.

Re: Update: devel/subversion

[Gleydson Soares]

Could you merge it [1] on top of your diff?

It adds a rcscript and _svnserver user to db/user.list
(Sent by jung@ yesterday) [1]

[1] http://marc.info/?l=openbsd-ports&m=146447355118340&w=2

Dmitrij D. Czarkoff  writes:

> Hi!
>
> Below is an update for devel/subversion (1.8.15 => 1.9.4).  It is only
> slightly tested because I am not a subversion user, and I only needed
> svn as a dependency for another port I am working on.  Please test.

Re: NEW: devel/py-test-mock 1.0

[Alexandr Shadchin]

On Sat, May 21, 2016 at 02:07:35PM +0500, Alexandr Shadchin wrote:
> Hi,
> 
> ok to import devel/py-test-mock ?
> Need for update py-requests.
> 
> DESCR:
> This plugin installs a mocker fixture which is a thin-wrapper around the
> patching API provided by the mock package, but with the benefit of not
> having to worry about undoing patches at the end of a test.
> 

Ping

-- 
Alexandr Shadchin

Re: maintainer update to py-cares-2.0.0

[Alexandr Shadchin]

On Sun, May 29, 2016 at 11:52:52AM +0200, Daniel Jakots wrote:
> On Fri, 27 May 2016 22:42:30 +0200, Daniel Jakots 
> wrote:
> 
> > Hi,
> > 
> > Here's a diff to update py-cares to latest release.
> 
> Just after that they did a bugfix release.
>  
> > Comments? OK?
> 

ok shadchin@

I would have included tests:
do-test: fake
 cd ${WRKSRC} && env PYTHONPATH=${WRKINST}/${MODPY_SITEPKG} 
${MODPY_BIN} tests/tests.py


> Index: Makefile
> ===
> RCS file: /cvs/ports/net/py-cares/Makefile,v
> retrieving revision 1.2
> diff -u -p -r1.2 Makefile
> --- Makefile  17 May 2016 05:12:02 -  1.2
> +++ Makefile  29 May 2016 09:49:33 -
> @@ -3,7 +3,7 @@
>  # using bundled c-ares which is patched for TTL support
>  COMMENT =python interface for c-ares
>  
> -MODPY_EGG_VERSION =  1.0.0
> +MODPY_EGG_VERSION =  2.0.1
>  DISTNAME =   pycares-${MODPY_EGG_VERSION}
>  PKGNAME =py-cares-${MODPY_EGG_VERSION}
>  CATEGORIES = net
> Index: distinfo
> ===
> RCS file: /cvs/ports/net/py-cares/distinfo,v
> retrieving revision 1.1.1.1
> diff -u -p -r1.1.1.1 distinfo
> --- distinfo  7 May 2016 14:46:08 -   1.1.1.1
> +++ distinfo  29 May 2016 09:49:33 -
> @@ -1,2 +1,2 @@
> -SHA256 (pycares-1.0.0.tar.gz) = oYNB6gMOLMB0Os30qnIwK99rggk4s2zkvXbkP6oidqM=
> -SIZE (pycares-1.0.0.tar.gz) = 220165
> +SHA256 (pycares-2.0.1.tar.gz) = OmhMCbdT4UPYYZC1fXG9gtJEkgYtEdaraswhNJ0to00=
> +SIZE (pycares-2.0.1.tar.gz) = 225788
> Index: pkg/PLIST
> ===
> RCS file: /cvs/ports/net/py-cares/pkg/PLIST,v
> retrieving revision 1.1.1.1
> diff -u -p -r1.1.1.1 PLIST
> --- pkg/PLIST 7 May 2016 14:46:08 -   1.1.1.1
> +++ pkg/PLIST 29 May 2016 09:49:33 -
> @@ -1,7 +1,15 @@
>  @comment $OpenBSD: PLIST,v 1.1.1.1 2016/05/07 14:46:08 danj Exp $
> +lib/python${MODPY_VERSION}/site-packages/pycares/
>  
> lib/python${MODPY_VERSION}/site-packages/pycares-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/
>  
> lib/python${MODPY_VERSION}/site-packages/pycares-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/PKG-INFO
>  
> lib/python${MODPY_VERSION}/site-packages/pycares-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/SOURCES.txt
>  
> lib/python${MODPY_VERSION}/site-packages/pycares-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/dependency_links.txt
>  
> lib/python${MODPY_VERSION}/site-packages/pycares-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/top_level.txt
> -lib/python${MODPY_VERSION}/site-packages/pycares.so
> +lib/python${MODPY_VERSION}/site-packages/pycares/__init__.py
> +${MODPY_COMMENT}lib/python${MODPY_VERSION}/site-packages/pycares/${MODPY_PYCACHE}/
> +lib/python${MODPY_VERSION}/site-packages/pycares/${MODPY_PYCACHE}__init__.${MODPY_PYC_MAGIC_TAG}pyc
> +lib/python${MODPY_VERSION}/site-packages/pycares/${MODPY_PYCACHE}_version.${MODPY_PYC_MAGIC_TAG}pyc
> +lib/python${MODPY_VERSION}/site-packages/pycares/${MODPY_PYCACHE}errno.${MODPY_PYC_MAGIC_TAG}pyc
> +lib/python${MODPY_VERSION}/site-packages/pycares/_core.so
> +lib/python${MODPY_VERSION}/site-packages/pycares/_version.py
> +lib/python${MODPY_VERSION}/site-packages/pycares/errno.py
> 

-- 
Alexandr Shadchin

Re: update to www/py-werkzeug-0.11.10

[Alexandr Shadchin]

On Fri, May 27, 2016 at 09:27:39PM +0200, Daniel Jakots wrote:
> Hi,
> 
> Here's a diff to update to latest werkzeug release. They created a
> project to manage it (among others) called Pallets [1] so change the
> GH_ACCOUNT accordingly.
> 
> [1]: https://www.palletsprojects.com/blog/hello/
> 
> Cheers,
> Daniel
> 
> Index: Makefile
> ===
> RCS file: /cvs/ports/www/py-werkzeug/Makefile,v
> retrieving revision 1.24
> diff -u -p -r1.24 Makefile
> --- Makefile  16 Apr 2016 19:06:26 -  1.24
> +++ Makefile  27 May 2016 19:26:15 -
> @@ -2,10 +2,9 @@
>  
>  COMMENT =WSGI utility collection
>  
> -MODPY_EGG_VERSION =  0.11.5
> +MODPY_EGG_VERSION =  0.11.10
>  DISTNAME =   Werkzeug-${MODPY_EGG_VERSION}
>  PKGNAME =py-${DISTNAME:L}
> -REVISION =   0
>  
>  CATEGORIES = www devel
>  
> @@ -14,7 +13,7 @@ HOMEPAGE =  http://werkzeug.pocoo.org/
>  # BSD
>  PERMIT_PACKAGE_CDROM =   Yes
>  
> -GH_ACCOUNT=  mitsuhiko
> +GH_ACCOUNT=  pallets
>  GH_PROJECT=  werkzeug
>  GH_TAGNAME=  ${MODPY_EGG_VERSION}
>  
> Index: distinfo
> ===
> RCS file: /cvs/ports/www/py-werkzeug/distinfo,v
> retrieving revision 1.10
> diff -u -p -r1.10 distinfo
> --- distinfo  28 Mar 2016 12:20:33 -  1.10
> +++ distinfo  27 May 2016 19:26:15 -
> @@ -1,2 +1,2 @@
> -SHA256 (Werkzeug-0.11.5.tar.gz) = 
> 7pj/5TVl7kRtzMSQC7MFLG641wBG9+fUpNXx2PgXt3E=
> -SIZE (Werkzeug-0.11.5.tar.gz) = 1168054
> +SHA256 (Werkzeug-0.11.10.tar.gz) = 
> eYO/KQ8QXawl2oQ3lgh3oV01KNXVv0qylKWohiW1CGc=
> +SIZE (Werkzeug-0.11.10.tar.gz) = 1170919
> 

You missed 'make update-patches'
Otherwise, ok shadchin@

-- 
Alexandr Shadchin

Re: [new] devel/cargo

[Sebastien Marie]

ping ?

On Thu, May 26, 2016 at 06:45:14AM +0200, Sebastien Marie wrote:
> On Fri, May 20, 2016 at 05:51:54AM +0200, Sebastien Marie wrote:
> > Hi,
> > 
> > I would like to import cargo-0.10.0.
> > 
> > The attached tgz is the same as openbsd-wip port.
> > 
> 
> attached is an updated archive (same as openbsd-wip c949090).
> 
> It includes:
>   - add a 'bootstrap' target to make a new bootstrap
>   - permit to build the port with a bootstrap of same version (using
> cargo-0.10.0 to build cargo-0.10.0)
>   - some tricks for building under dpb (place a minimal gzip in registry
> cache instead of linking distfiles)
>   - move bash-completion to share/bash-completion/completions/ directory
> 
> OK to import ?
> -- 
> Sebastien Marie



-- 
Sebastien Marie

Re: maintainer update to py-cares-2.0.0

[Daniel Jakots]

On Fri, 27 May 2016 22:42:30 +0200, Daniel Jakots 
wrote:

> Hi,
> 
> Here's a diff to update py-cares to latest release.

Just after that they did a bugfix release.
 
> Comments? OK?

Index: Makefile
===
RCS file: /cvs/ports/net/py-cares/Makefile,v
retrieving revision 1.2
diff -u -p -r1.2 Makefile
--- Makefile17 May 2016 05:12:02 -  1.2
+++ Makefile29 May 2016 09:49:33 -
@@ -3,7 +3,7 @@
 # using bundled c-ares which is patched for TTL support
 COMMENT =  python interface for c-ares
 
-MODPY_EGG_VERSION =1.0.0
+MODPY_EGG_VERSION =2.0.1
 DISTNAME = pycares-${MODPY_EGG_VERSION}
 PKGNAME =  py-cares-${MODPY_EGG_VERSION}
 CATEGORIES =   net
Index: distinfo
===
RCS file: /cvs/ports/net/py-cares/distinfo,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 distinfo
--- distinfo7 May 2016 14:46:08 -   1.1.1.1
+++ distinfo29 May 2016 09:49:33 -
@@ -1,2 +1,2 @@
-SHA256 (pycares-1.0.0.tar.gz) = oYNB6gMOLMB0Os30qnIwK99rggk4s2zkvXbkP6oidqM=
-SIZE (pycares-1.0.0.tar.gz) = 220165
+SHA256 (pycares-2.0.1.tar.gz) = OmhMCbdT4UPYYZC1fXG9gtJEkgYtEdaraswhNJ0to00=
+SIZE (pycares-2.0.1.tar.gz) = 225788
Index: pkg/PLIST
===
RCS file: /cvs/ports/net/py-cares/pkg/PLIST,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 PLIST
--- pkg/PLIST   7 May 2016 14:46:08 -   1.1.1.1
+++ pkg/PLIST   29 May 2016 09:49:33 -
@@ -1,7 +1,15 @@
 @comment $OpenBSD: PLIST,v 1.1.1.1 2016/05/07 14:46:08 danj Exp $
+lib/python${MODPY_VERSION}/site-packages/pycares/
 
lib/python${MODPY_VERSION}/site-packages/pycares-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/
 
lib/python${MODPY_VERSION}/site-packages/pycares-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/PKG-INFO
 
lib/python${MODPY_VERSION}/site-packages/pycares-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/SOURCES.txt
 
lib/python${MODPY_VERSION}/site-packages/pycares-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/dependency_links.txt
 
lib/python${MODPY_VERSION}/site-packages/pycares-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/top_level.txt
-lib/python${MODPY_VERSION}/site-packages/pycares.so
+lib/python${MODPY_VERSION}/site-packages/pycares/__init__.py
+${MODPY_COMMENT}lib/python${MODPY_VERSION}/site-packages/pycares/${MODPY_PYCACHE}/
+lib/python${MODPY_VERSION}/site-packages/pycares/${MODPY_PYCACHE}__init__.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/pycares/${MODPY_PYCACHE}_version.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/pycares/${MODPY_PYCACHE}errno.${MODPY_PYC_MAGIC_TAG}pyc
+lib/python${MODPY_VERSION}/site-packages/pycares/_core.so
+lib/python${MODPY_VERSION}/site-packages/pycares/_version.py
+lib/python${MODPY_VERSION}/site-packages/pycares/errno.py

Update: devel/subversion

[Dmitrij D. Czarkoff]

Hi!

Below is an update for devel/subversion (1.8.15 => 1.9.4).  It is only
slightly tested because I am not a subversion user, and I only needed
svn as a dependency for another port I am working on.  Please test.

-- 
Dmitrij D. Czarkoff

Index: Makefile
===
RCS file: /cvs/ports/devel/subversion/Makefile,v
retrieving revision 1.147
diff -u -p -r1.147 Makefile
--- Makefile28 Apr 2016 16:03:01 -  1.147
+++ Makefile29 May 2016 09:29:28 -
@@ -7,7 +7,7 @@ COMMENT-ruby=   ruby interface to subvers
 COMMENT-ap2=   apache2 subversion modules
 COMMENT-gnome-keyring= GNOME keyring support for subversion
 
-VERSION=   1.8.15
+VERSION=   1.9.4
 DISTNAME=  subversion-${VERSION}
 PKGNAME-main=  ${DISTNAME}
 FULLPKGNAME-perl=  p5-SVN-${VERSION}
@@ -21,14 +21,10 @@ FULLPKGPATH-ap2=devel/subversion,-ap2
 FULLPKGNAME-gnome-keyring= gnome-keyring-subversion-${VERSION}
 FULLPKGPATH-gnome-keyring= devel/subversion,-gnome-keyring
 
-REVISION-main= 0
-REVISION-ruby= 0
-REVISION-ap2=  0
-
-SO_VERSION=1.4
+SO_VERSION=1.5
 SVN_LIBS=  svn_client-1 svn_delta-1 svn_diff-1 svn_fs-1 \
-   svn_fs_base-1 svn_fs_fs-1 svn_fs_util-1 svn_ra-1 \
-   svn_ra_serf-1 svn_ra_local-1 \
+   svn_fs_base-1 svn_fs_fs-1 svn_fs_util-1 svn_fs_x-1 \
+   svn_ra-1 svn_ra_serf-1 svn_ra_local-1 \
svn_ra_svn-1 svn_repos-1 svn_subr-1 svn_wc-1
 SVN_EXT_LIBS=  svn_swig_perl-1 svn_swig_py-1 svn_swig_ruby-1 \
svn_auth_gnome_keyring-1
@@ -83,7 +79,8 @@ TOOLS_HOOK_SCRIPTS=   commit-access-contro
 # There's also 'mailer', but we handle it separately, see post-install
 
 TOOLS_CLIENT_SIDE_SCRIPTS= bash_completion change-svn-wc-format.py \
-   server-version.py showchange.pl svn-graph.pl
+   detach.py mergeinfo-sanitizer.py server-version.py svn-graph.pl \
+   svn-ssl-fingerprints.sh svn-vendor.py svn-viewspec.py wcfind
 
 WANTLIB-ap2=   ${WANTLIB} apr-1 aprutil-1 db sqlite3
 LIB_DEPENDS-ap2=   ${MODGETTEXT_LIB_DEPENDS} ${MAINSPEC}
Index: distinfo
===
RCS file: /cvs/ports/devel/subversion/distinfo,v
retrieving revision 1.51
diff -u -p -r1.51 distinfo
--- distinfo16 Dec 2015 10:21:52 -  1.51
+++ distinfo29 May 2016 09:29:28 -
@@ -1,2 +1,2 @@
-SHA256 (subversion-1.8.15.tar.bz2) = 
z5ih92hs/WoVy83oN58Vcw/V4Ao1c82wNaZJHbzMyCw=
-SIZE (subversion-1.8.15.tar.bz2) = 6917489
+SHA256 (subversion-1.9.4.tar.bz2) = 
Emf54quYPyYGI77oQebJzEWL9L93YjjtXxAJg/eekpk=
+SIZE (subversion-1.9.4.tar.bz2) = 7911473
Index: patches/patch-Makefile_in
===
RCS file: /cvs/ports/devel/subversion/patches/patch-Makefile_in,v
retrieving revision 1.14
diff -u -p -r1.14 patch-Makefile_in
--- patches/patch-Makefile_in   28 Apr 2016 16:03:01 -  1.14
+++ patches/patch-Makefile_in   29 May 2016 09:29:28 -
@@ -1,7 +1,7 @@
 $OpenBSD: patch-Makefile_in,v 1.14 2016/04/28 16:03:01 stsp Exp $
 Makefile.in.orig   Mon Jul 27 02:23:40 2015
-+++ Makefile.inThu Apr 28 17:16:21 2016
-@@ -81,11 +81,11 @@ swig_pl_libdir = @libdir@
+--- Makefile.in.orig   Thu Dec 10 05:00:15 2015
 Makefile.inThu May  5 15:04:58 2016
+@@ -82,11 +82,11 @@ swig_pl_libdir = @libdir@
  swig_rb_libdir = @libdir@
  
  ### these possibly need further discussion
@@ -14,9 +14,9 @@ $OpenBSD: patch-Makefile_in,v 1.14 2016/
 -toolsdir = @bindir@/svn-tools
 +toolsdir = @bindir@
  
- javahl_javadir = @libdir@/svn-javahl
- javahl_javahdir = @libdir@/svn-javahl/include
-@@ -135,13 +135,13 @@ APACHE_LDFLAGS = @APACHE_LDFLAGS@
+ # where to install pkg-config files
+ pkgconfig_dir = $(datadir)/pkgconfig
+@@ -142,13 +142,13 @@ GMOCK_INCLUDES = -I@GMOCK_SRCDIR@
  SWIG = @SWIG@
  SWIG_PY_INCLUDES = @SWIG_PY_INCLUDES@ -I$(SWIG_SRC_DIR)/python/libsvn_swig_py
  SWIG_PY_COMPILE = @SWIG_PY_COMPILE@
Index: patches/patch-configure_ac
===
RCS file: /cvs/ports/devel/subversion/patches/patch-configure_ac,v
retrieving revision 1.9
diff -u -p -r1.9 patch-configure_ac
--- patches/patch-configure_ac  28 Apr 2016 16:03:01 -  1.9
+++ patches/patch-configure_ac  29 May 2016 09:29:28 -
@@ -1,6 +1,6 @@
 $OpenBSD: patch-configure_ac,v 1.9 2016/04/28 16:03:01 stsp Exp $
 configure.ac.orig  Sat Feb 14 05:00:33 2015
-+++ configure.ac   Thu Apr 28 17:16:21 2016
+--- configure.ac.orig  Mon Jul 27 01:03:10 2015
 configure.ac   Thu May  5 15:04:58 2016
 @@ -23,7 +23,7 @@ AC_PREREQ(2.59)
  dnl Get the version of Subversion, using m4's esyscmd() command to do this
  dnl at m4-time, since AC_INIT() requires it then.
@@ -10,12 +10,12 @@ $OpenBSD: patch-configure_ac,v 1.9 2016/
   [http://subversion.apache.org/])
  
  AC_CONFIG_SRCD

Re: SAMBA crash of smbd(18370) signal 6

[Ian McWilliam]

> On 28 May 2016, at 10:49 PM, Jeremie Courreges-Anglas  wrote:
> 
> johnw  writes:
> 
>> Hi, my samba server crash a lot, after upgrade from 4.1.x to 4.3/4.x
>> 
>> (no problem before, running 4.1.x version)
>> 
>> (my system is running amd64/current with samba package 4.4.3p0)
>> 
>> I found some error message in "/var/log/samba/log.smbd" like this:
>> 
>> [2016/05/28 09:37:30.217939,  0] ../source3/lib/util.c:791(smb_panic_s3)
>>  PANIC (pid 27088): internal error
>> [2016/05/28 09:37:30.218145,  0] ../source3/lib/util.c:902(log_stack_trace)
>>  BACKTRACE: 1 stack frames:
>>   #0 0x18b8b840d51e  at
>> /usr/local/lib/libsmbconf.so.0.0
>> 
>> And "dmesg |grep signal" message like this:
>> 
>> crash of smbd(69427) signal 6
>> crash of smbd(47833) signal 6
>> crash of smbd(79506) signal 6
>> crash of smbd(18370) signal 6
>> 
>> I can not find any core dump file in /var/log/samba/cores/ and else where.
> 
> Yeah I don't think this part of the code works on OpenBSD, but we have
> a generic framework to achieve the same goal.  For the setup see the
> last example in sysctl(8).
> 
>> Attached testparm and /var/log/samba/log.smbd file here.
>> 
>> Any idea, how to solve / debug?
> 
> A gdb backtrace would help.
> 
> doas gdb smbd /var/crash/smbd/something.core, type 'bt'
> 
> BTW testparm is nice but it contains lots of extra data, it is a good
> thing to have its output plus the original smb.conf.


In your smb config file add

panic action = /bin/sleep 9000

Start smbd, do what ever you do to make it crash and attach to the smbd process 
using gdb.