Re: french/aster ignores PYTHON_VERSION= 2.6 ?

2011-07-20 Thread Ruslan Mahmatkhanov

Anton Shterenlikht wrote on 20.07.2011 17:44:

On Wed, Jul 20, 2011 at 04:55:59PM +0400, Ruslan Mahmatkhanov wrote:



As far i understand PYTHON_VERSION hold value like 'python2.7', so you
should correct this in your Makefile. Or just use USE_PYTHON = 2.6


No, PYTHON_VERSION seems to be overwritten by something else.
I removed all but the PYTHON parts from the Makefile,
and get this:



Yes, i see this in bsd.python.mk:
# Define PYTHON_VERSION to override the
# defaults that USE_PYTHON would give you.

But it seems not true. Here is my test Makefile:
mrk@smeshariki2$ cat Makefile
PORTNAME=   blah
DISTVERSION=1.0
CATEGORIES= misc
COMMENT=Some dumb port for testing

USE_PYTHON= yes

.include 

.if ${ARCH} == "i386"
PYTHON_VERSION= "python2.6"
.endif

.include 

mrk@smeshariki2$ make depends
===>   blah-1.0 depends on file: /usr/local/bin/python2.7 - found
===>   blah-1.0 depends on file: /usr/local/bin/python2.7 - found

mrk@smeshariki2$ uname -p
i386

I added python@ to cc, since it seems like bsd.python.mk bug to me (or 
documentation bug).


--
Regards,
Ruslan
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: french/aster ignores PYTHON_VERSION= 2.6 ?

2011-07-26 Thread Ruslan Mahmatkhanov

Anton Shterenlikht wrote on 26.07.2011 21:07:

On Wed, Jul 20, 2011 at 06:22:31PM +0400, Ruslan Mahmatkhanov wrote:

Anton Shterenlikht wrote on 20.07.2011 17:44:

On Wed, Jul 20, 2011 at 04:55:59PM +0400, Ruslan Mahmatkhanov wrote:



As far i understand PYTHON_VERSION hold value like 'python2.7', so you
should correct this in your Makefile. Or just use USE_PYTHON = 2.6


No, PYTHON_VERSION seems to be overwritten by something else.
I removed all but the PYTHON parts from the Makefile,
and get this:



Yes, i see this in bsd.python.mk:
# Define PYTHON_VERSION to override the
# defaults that USE_PYTHON would give you.

But it seems not true. Here is my test Makefile:
mrk@smeshariki2$ cat Makefile
PORTNAME=   blah
DISTVERSION=1.0
CATEGORIES= misc
COMMENT=Some dumb port for testing

USE_PYTHON= yes

.include

.if ${ARCH} == "i386"
PYTHON_VERSION= "python2.6"
.endif

.include

mrk@smeshariki2$ make depends
===>blah-1.0 depends on file: /usr/local/bin/python2.7 - found
===>blah-1.0 depends on file: /usr/local/bin/python2.7 - found

mrk@smeshariki2$ uname -p
i386

I added python@ to cc, since it seems like bsd.python.mk bug to me (or
documentation bug).

--
Regards,
Ruslan


I've heard nothing back within a week.
Shall I submit a pr?


I think so. Add Mk/bsd.python.mk to description field so it will be 
assigned to portmgr@.


--
Regards,
Ruslan
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: french/aster ignores PYTHON_VERSION= 2.6 ?

2011-07-26 Thread Ruslan Mahmatkhanov

Mark Linimon wrote on 27.07.2011 06:44:

On Tue, Jul 26, 2011 at 11:46:17PM +0400, Ruslan Mahmatkhanov wrote:

Add Mk/bsd.python.mk to description field


a good idea, but ...


so it will be assigned to portmgr@.


portmgr@ doesn't own all the bsd.*.mk files; bsd.python.mk is owned by
python@.  We're happy to put an -exp run in the queue for large changes,
but it's not necessary for something small.  (Note: I only scanned this
email quickly so don't know which one it is :-) )

mcl


Ok, i was confused.

--
Regards,
Ruslan
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Maintainership of py-zopetesting and py-zopeevent

2011-09-05 Thread Ruslan Mahmatkhanov

wen heping wrote on 05.09.2011 18:08:

2011/9/5 Ruslan Mahmatkhanov:

Hi, Wen.

wen heping wrote on 05.09.2011 17:34:


Hi,Ruslan:

Do you have the plan to import Zope-2.13 into portstree?


Yes, i am. But work on Zope2.13 and Plone4 is still underway.



I could run zope-2.13 on FreeBSD correctly with python-2.7, but
many new ports should be created if we import zope-2.13.


Yes, here is all of them:
http://happy-nation.by.ru/ports/zope-deps.tar.gz



Seems we have done many duplicate work.




Please note that this is just zope deps, plone4 will push many of new ports
too.



If you have the plan, give me a message and I am intersting of it.

wen


The plan is:
1. Commit this two guys: ports/159962 and ports/160434.


I shall look into these.



2. I'd prefer to create physical category like `webdevel` and
   move all zope-specific ports there (and all of this django, pyramid,
  pylons, turbogears stuff) to not trash the tree (mainly devel category).



I do not believe portmgr@ shall agree with this. Better discuss it
with miwi@ first.


It's preferable from my POV, but i'm not insist.


3. Rename 4 existing zope ports for consistency.


I agree with this.




4. Push all the dependend ports mentioned before (over 50) to the tree.


Shall we add these new ports one by one?
I remember it is a hard work to deal with the depend relation, so it
is difficult to add so many new ports at one time.


While correct import path may be created, i think that better way - to 
commit them all, but connect them to Makefile's only after all the 
dependend ports will be there. This should not break anything as far i 
understand.




BTW: I forward this mail to miwi@, lwshu@,python@.

wen




5. Import zope 2.13 and plone4 (i think to www/zope and www/plone, because>  
they are the only supported versions atm)
6. Deprecate old zope/plone ports in three months
7. PROFIT

Something like that, but i'm sure there may be some hidden or non-technical
problems. Surely, after that and after handling remaining
elementtree/pysqlite2x pr's, we can finally retire python24.



2011/9/5 Ruslan Mahmatkhanov:


Good day.

I want to grab maintainership of this ports:

devel/py-zopetesting
devel/py-zopeevent
net/py-zopeproxy

But i want them to be renamed to py-zope.testing, py-zope.event and
py-zope.proxy respectively to match upstream projectnames and to
consistency
with other current and upcoming zope ports. Is this reasons sounds enough
to
make a repocopy of them? This change is purely cosmetic but will also
ease
maintainership a bit (Makefile vars like DISTNAME etc will not need to be
redefined), but this will require to fix dependencies in all the ports
that
required it. I'm looking forward to your suggestion before sending pr's.
Thanks.

PS. It would be logical to rename devel/py-zopeInterface to
py-zope.interface too, but this port is maintained and i'm not sure if
maintainer (cc'ed) wishes such change.



--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Fwd: Re: About zope and plone

2011-10-19 Thread Ruslan Mahmatkhanov

Good day, gentlemen.

Please tell what do you think about message bellow and this pr:
http://bugs.freebsd.org/159962

Message bellow is somewhat complements this pr. Please also note that 
Wen agree with this pr, but he is busy for this right now. Also note 
that this pr is a stopper wrt importing of new zope and plone.


Thanks in advance.

 Исходное сообщение 
Тема: Re: About zope and plone
Дата: Sun, 16 Oct 2011 23:03:41 +0400
От: Ruslan Mahmatkhanov 
Кому: wen heping 

wen heping wrote on 14.10.2011 13:15:


Today I tested devel/py-zope.minmax in Tinderbox, it has the same plist error.
Now I need such a port exist in current FreeBSD portstree to be a demo
to python@.

If python@ does not disagree I would ask portmgr@ for an exprun to test it.
Then I would commit it.


Hi Wen,

i'm finally get it.
While seeking an example for you i took some broken/outdated ports, that 
i just can't left in this state :), so i apologize for delay.


You was right, it's a kind of upstream problem, but bsd.python.mk still
contains a problem too, because this error still exists on deinstall of
such silly ports (that may potentially exist in future).

Here is how to reproduce:
Pick any port, that
  a) using setuptools for installing
  b) has non-empty install_requires[] list in it's setup.py
  c) does not define BUILD or RUN DEPENDS for deps, that are listed
 in it's setup.py

Ok, i wasn't able to find such port in the tree. But you can take, say,
devel/py-daemon, remove BUILD/RUN_DEPENDS in it's Makefile and try to
build it in tinderbox - it will builds and installs fine, but you'll get 
extra files installed on deinstall.


The only port, that looked promising, was devel/py-Jinja. It defines
this lines:

PYEASYINSTALL_INSTALLARGS=  -N ${PYTHON_SITELIBDIR}/${PYEASYINSTALL_EGG}
PYEASYINSTALL_UNINSTALLARGS=-q -m -N
${PYTHON_SITELIBDIR}/${PYEASYINSTALL_EGG}

Note adding the -N key on deinstall. But right now they are not needed,
because this port packages successfully without them (it has empty
install_requires[]).

The problem with Products.MailHost, for example, is that if i add zope
as dependency of this port, we well have cyclic deps and broken build.
Resolving this issue with upstream will take much time (i know this
because i already have 3 patches that hanging in zope/plone bugtrackers
more than month).

I can avoid this a different way, by defining
PYEASYINSTALL_UNINSTALLARGS, but i believe that more correct and more
easy is to define this -N in bsd.python.mk, because it's just two bytes
in one file - and this is all, problem solved. Otherwise, i will be
forced to patch many ports with this UNINSTALLARGS line, effectively
littering the tree and making it hard to maintain. We already have it in 
PYEASYINSTALL_INSTALLARGS for some reason anyway, so why to not define 
it in UNINSTALLARGS too for symmetry?


I hope this arguments are sufficient to convince portmgr@ and make this
change.

--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: About zope and plone

2011-10-28 Thread Ruslan Mahmatkhanov

Martin Wilke wrote on 29.10.2011 04:37:

On Wed Oct 19 10:57:41 2011, wen heping wrote:

Yes, I agree with this PR that we should add -N to easy_install's
deinstall argument.

And shall we ask for a exp-run to test it ?

wen

2011/10/19 Ruslan Mahmatkhanov:

Good day, gentlemen.

Please tell what do you think about message bellow and this pr:
http://bugs.freebsd.org/159962

Message bellow is somewhat complements this pr. Please also note that
Wen
agree with this pr, but he is busy for this right now. Also note that
this
pr is a stopper wrt importing of new zope and plone.

Thanks in advance.

 Исходное сообщение 
Тема: Re: About zope and plone
Дата: Sun, 16 Oct 2011 23:03:41 +0400
От: Ruslan Mahmatkhanov
Кому: wen heping

wen heping wrote on 14.10.2011 13:15:


Today I tested devel/py-zope.minmax in Tinderbox, it has the same plist
error.
Now I need such a port exist in current FreeBSD portstree to be a demo
to python@.

If python@ does not disagree I would ask portmgr@ for an exprun to test
it.
Then I would commit it.


Hi Wen,

i'm finally get it.
While seeking an example for you i took some broken/outdated ports,
that i
just can't left in this state :), so i apologize for delay.

You was right, it's a kind of upstream problem, but bsd.python.mk still
contains a problem too, because this error still exists on deinstall of
such silly ports (that may potentially exist in future).

Here is how to reproduce:
Pick any port, that
a) using setuptools for installing
b) has non-empty install_requires[] list in it's setup.py
c) does not define BUILD or RUN DEPENDS for deps, that are listed
in it's setup.py

Ok, i wasn't able to find such port in the tree. But you can take, say,
devel/py-daemon, remove BUILD/RUN_DEPENDS in it's Makefile and try to
build it in tinderbox - it will builds and installs fine, but you'll get
extra files installed on deinstall.

The only port, that looked promising, was devel/py-Jinja. It defines
this lines:

PYEASYINSTALL_INSTALLARGS= -N ${PYTHON_SITELIBDIR}/${PYEASYINSTALL_EGG}
PYEASYINSTALL_UNINSTALLARGS= -q -m -N
${PYTHON_SITELIBDIR}/${PYEASYINSTALL_EGG}

Note adding the -N key on deinstall. But right now they are not needed,
because this port packages successfully without them (it has empty
install_requires[]).

The problem with Products.MailHost, for example, is that if i add zope
as dependency of this port, we well have cyclic deps and broken build.
Resolving this issue with upstream will take much time (i know this
because i already have 3 patches that hanging in zope/plone bugtrackers
more than month).

I can avoid this a different way, by defining
PYEASYINSTALL_UNINSTALLARGS, but i believe that more correct and more
easy is to define this -N in bsd.python.mk, because it's just two bytes
in one file - and this is all, problem solved. Otherwise, i will be
forced to patch many ports with this UNINSTALLARGS line, effectively
littering the tree and making it hard to maintain. We already have it in
PYEASYINSTALL_INSTALLARGS for some reason anyway, so why to not
define it in
UNINSTALLARGS too for symmetry?

I hope this arguments are sufficient to convince portmgr@ and make this
change.

--
Regards,
Ruslan

Tinderboxing kills... the drives.



Hi,

Just a question, all these is still based on python 2.4 right? Because
of security reason we should remove python 2.4 after 9.0 release. I'm
back to the game now and will force that removal.
Please let me know whats about that.

- Martin



Hi, Martin.

No, of course. The whole point is to import zope/plone versions that 
work with 2.6/2.7, and to drop obsoleted 2.4-only versions along with 
python 2.4 an 2.5.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: About zope and plone

2011-10-30 Thread Ruslan Mahmatkhanov

wen heping wrote on 31.10.2011 10:34:

Ruslan:

 Sorry I did not found that you import py-roman too. I imported it
because another my port devel/py-mwlib need it.

wen


No problem. Just one plone port depended on this, so changes was trivial.

--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Expiration of devel/py-[c]elementtree

2011-11-11 Thread Ruslan Mahmatkhanov
Good day!

We have two pr's here:
http://bugs.freebsd.org/155524
http://bugs.freebsd.org/155526

And the last consumer of them in the tree was fixed today.
Since python 2.5 we have xml.etree.ElementTree and
xml.etree.cElementTree in the standard library. This package are drop-in
replacement for aforementioned ports, except of HTMLTreeBuilder class
that is missing. But we don't have ports that use it anyway.

All the modern software uses or bundled xml.etree or lxml
(devel/py-lxml) or html5lib (www/py-html5lib) instead. So i want to ask
is there any objections if I mark them deprecated and remove in one month?


-- 
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Expiration of devel/py-[c]elementtree

2011-11-11 Thread Ruslan Mahmatkhanov
Greg Larkin wrote on 12.11.2011 01:26:
> On 11/11/11 7:30 AM, Ruslan Mahmatkhanov wrote:
>> Good day!
> 
>> We have two pr's here:
>> http://bugs.freebsd.org/155524
>> http://bugs.freebsd.org/155526
> 
>> And the last consumer of them in the tree was fixed today.
>> Since python 2.5 we have xml.etree.ElementTree and
>> xml.etree.cElementTree in the standard library. This package are drop-in
>> replacement for aforementioned ports, except of HTMLTreeBuilder class
>> that is missing. But we don't have ports that use it anyway.
> 
>> All the modern software uses or bundled xml.etree or lxml
>> (devel/py-lxml) or html5lib (www/py-html5lib) instead. So i want to ask
>> is there any objections if I mark them deprecated and remove in one month?
> 
> 
> 
> My sense is these ports should be set to expire at the same time as
> python 2.4. When is that change going to be made to its Makefile?
> 
> Regards,
> Greg

Fair enough. I assumed that nobody actually using python 2.4, while
asking that. I believe Martin knows better when 2.4 will be set to expire :)

-- 
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Obsoleted zope ports deprecation

2011-12-04 Thread Ruslan Mahmatkhanov
Good day.

I wrote to you because you are maintainers of zope ports, listed here:
http://people.freebsd.org/~rm/zope-ports-for-deprecation.html

I'm asking for your approval for this ports deprecation and removing in
one month. Deprecation reasons for each port also listed on link above.

They all rely on long-time unsupported Zope versions. The only supported
versions now is a 2.13 and 2.12. Version 2.13 should be added soon. You
can check latest revision of this port here:
https://github.com/mexicarne/zope/tree/master/www/zope213

or, if you prefer patch against www/zope213:
http://people.freebsd.org/~rm/zope213.repocopy.finished.diff.txt

Thanks in advance for response.

-- 
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Fwd: Re: Obsoleted zope ports deprecation

2011-12-05 Thread Ruslan Mahmatkhanov

Hi, Gerhard.

Please keep python@ and ports@ in cc:.

Gerhard Schmidt wrote on 12/06/11 10:29:
> 
> 
>  Original-Nachricht 
> Betreff: Re: Obsoleted zope ports deprecation
> Datum: Tue, 06 Dec 2011 07:24:46 +0100
> Von: Gerhard Schmidt 
> An: NAKAJI Hiroyuki 
> 
> (11/12/04 20:52), Ruslan Mahmatkhanov wrote:
>> Good day.
>>
>> I wrote to you because you are maintainers of zope ports, listed here:
>> http://people.freebsd.org/~rm/zope-ports-for-deprecation.html
>>
>> I'm asking for your approval for this ports deprecation and removing in
>> one month. Deprecation reasons for each port also listed on link above.
> 
> Is there a reason to remove these Ports. I'm still using Zope2.7 and
> Zope2.11 in our Production Systems. So why don't keep them.

As Martin already stated there is two major reasons for that:
1. Python 2.4 and python 2.5 are EOL and should be moved from the tree
   (this was done in all major linux distros time ago)
2. Zope < 2.12 are EOL and should be moved from the tree too, because:
   1. They have unfixed security issues
   2. They rely on obsoleted python 2.4

As far i understand, you or should try to move to 2.13 or keep required
zope ports locally (with using Makefile.local files), because most
users, imho, doesn't need them now.

> 
> BTW I am the maintainer og the port www/zope too don't know why it's
> listed there as maintained by po...@freebsd.org.

Sorry, it was my fault. Now fixed.

> 
> Having these Ports make my life much more easy.
> 
> Regards
>Estartu
> 


-- 
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/163112: Updates bsd.python.mk to support Zope 2.11, 2.13

2011-12-07 Thread Ruslan Mahmatkhanov
The following reply was made to PR ports/163112; it has been noted by GNATS.

From: Ruslan Mahmatkhanov 
To: Kirk Strauser 
Cc: bug-follo...@freebsd.org
Subject: Re: ports/163112: Updates bsd.python.mk to support Zope 2.11, 2.13
Date: Wed, 07 Dec 2011 21:36:33 +0400

 Hi, Kirk.
 
 Zope 2.7 is in the tree yet - see www/zope.
 Zope 2.13 is not the in tree yet. Yes, www/zope213 directory exists,
 but it contains zope 2.11 actually, because update of this port is
 waiting mentors approval.
 Zope 2.11 is obsoleted too, so i would like to see it deprecated.
 Zope 2.13 will not use SZOPEBASEDIR anymore, so this should not apply to
 it in your patch.
 
 
 Kirk Strauser wrote on 07.12.2011 21:01:
 > ports/Mk/bsd.python.mk only supports Zope versions 2.7, 2.8, 2.9, 2.10, and 
 > 3.2. That's a problem because 2.7 is no longer in ports, 2.8-2.10 are 
 > deprecated, and it doesn't work with the newer, supported Zope 2.11 and 2.13 
 > versions that *are* now in ports.
 > 
 > This patch removes the configuration for ZOPE_VERSION 2.7 and adds versions 
 > 2.11 and 2.13. It also changes ZOPE_PORTBRANCH (the default Zope version) to 
 > 2.11, which is the oldest non-deprecated version in ports.
 > 
 > I marked this as serious/medium because the remaining obsolete Zope ports 
 > are deprecated a month from now, and I'd hate to see a period after that 
 > when there's no supported version of Zope at all.
 >> How-To-Repeat:
 
 -- 
 Regards,
 Ruslan
 
 Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: pyconfig.h and freebsd10: _POSIX_C_SOURCE and _XOPEN_SOURCE

2012-02-12 Thread Ruslan Mahmatkhanov

Hi!

David Schultz wrote on 22.01.2012 23:25:

On Sat, Jan 21, 2012, Andriy Gapon wrote:


It seems that python27's pyconfig.h (artificially?) limits visibility of system
APIs by setting _POSIX_C_SOURCE to 200112L and _XOPEN_SOURCE to 600.
This might not actually change much for earlier FreeBSD versions.


I just submitted http://bugs.freebsd.org/165033 that includes workaround 
from bsam@, that fixes build of devel/py-lxml on 10.0-CURRENT. I hope 
sometime this issue will be resolved properly in FreeBSD base or in 
python distribution. Thanks!


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/128556: net/py-zsi installs with insufficient rights

2012-02-14 Thread Ruslan Mahmatkhanov

cr...@freebsd.org wrote on 09.02.2012 23:54:

Any ideas?  I don't see anything wrong with the port; do we even support
non-standard umasks?


Hi, Chris.

I wasn't able to reproduce it too:

[rm@smeshariki3 ~]> ll /usr/local/lib/python2.7/site-packages/| grep ZSI
-rw-r--r--1 root  wheel   420166 14 фев 19:58 ZSI-2.0-py2.7.egg

And you are right - it looks like submitter's umask setting is the 
culprit, so may be this pr should be just closed.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/128556: net/py-zsi installs with insufficient rights

2012-02-14 Thread Ruslan Mahmatkhanov
The following reply was made to PR ports/128556; it has been noted by GNATS.

From: Ruslan Mahmatkhanov 
To: cr...@freebsd.org
Cc: bug-follo...@freebsd.org, freebsd-python@FreeBSD.org
Subject: Re: ports/128556: net/py-zsi installs with insufficient rights
Date: Tue, 14 Feb 2012 20:02:55 +0400

 cr...@freebsd.org wrote on 09.02.2012 23:54:
 > Any ideas?  I don't see anything wrong with the port; do we even support
 > non-standard umasks?
 
 Hi, Chris.
 
 I wasn't able to reproduce it too:
 
 [rm@smeshariki3 ~]> ll /usr/local/lib/python2.7/site-packages/| grep ZSI
 -rw-r--r--1 root  wheel   420166 14 фев 19:58 ZSI-2.0-py2.7.egg
 
 And you are right - it looks like submitter's umask setting is the 
 culprit, so may be this pr should be just closed.
 
 -- 
 Regards,
 Ruslan
 
 Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Python upgrade to address vulnerability?

2012-02-15 Thread Ruslan Mahmatkhanov

Doug Barton wrote on 15.02.2012 02:20:

So apparently we have a python vulnerability according to
http://portaudit.FreeBSD.org/b4f8be9e-56b2-11e1-9fb7-003067b2972c.html,
but I'm not seeing an upgrade to address it yet. Any idea when that will
happen?


Thanks,

Doug



Patch is there:
http://people.freebsd.org/~rm/python-CVE-2012-0845.diff.txt

Patch for 3.2 is taken there directly:
http://bugs.python.org/file24522/xmlrpc_loop-1.diff

Patch for 2.5, 2.6, 2.7, 3.1 is adopted from this patch:
http://bugs.python.org/file24513/xmlrpc_loop.diff

SimpleXMLRPCServer.py in 2.4 is too different and it is going to die 
anyway so I didn't messed with it.


If noone objects, I can commit it. Please tell me what should i do.

--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Python upgrade to address vulnerability?

2012-02-15 Thread Ruslan Mahmatkhanov

wen heping wrote on 15.02.2012 14:16:

2012/2/15 Ruslan Mahmatkhanov


Doug Barton wrote on 15.02.2012 02:20:


So apparently we have a python vulnerability according to
http://portaudit.FreeBSD.org/**b4f8be9e-56b2-11e1-9fb7-**
003067b2972c.html<http://portaudit.FreeBSD.org/b4f8be9e-56b2-11e1-9fb7-003067b2972c.html>
,
but I'm not seeing an upgrade to address it yet. Any idea when that will
happen?


Thanks,

Doug



Patch is there:
http://people.freebsd.org/~rm/**python-CVE-2012-0845.diff.txt<http://people.freebsd.org/~rm/python-CVE-2012-0845.diff.txt>



Had this patch been committed into upstream? When I found it , it was in
review state.

And CVE-2012-0845 too.

wen


Yes, it is not yet committed, but comments looks promisingly :). And i 
can't reproduce this bug after patching, using procedure described in 
bug report.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: FreeBSD Port: py27-dbutils-1.1

2012-02-19 Thread Ruslan Mahmatkhanov

Hi.

Brian McCann wrote on 08.12.2011 01:14:

Not sure who this is going to, so I'll say "To whom it may concern".

I'm trying to "make package" for py-dbutils, and it's failing.  It's
trying to tar up various files in
"/usr/local/lib/python2.7/site-packages/DBUtils", which doesn't exist
because the make install deleted them when it build the egg file.
Anyone have any ideas on how to make this work?

--Brian


It's because this port looking for setuptools first:
"""
try:
from setuptools import setup
except ImportError:
from distutils.core import setup
"""

So the things will be broken if someone will try to package/install it 
in a system where setuptools is installed. Since nobody depends on this 
port, I just converted it to setuptools, so problem should be gone.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: [patch] Mk/bsd.python.mk: Fix python -c commands for python 3+

2012-03-20 Thread Ruslan Mahmatkhanov

Mel Flynn wrote on 18.03.2012 20:08:

Submitter-Id:   current-users
Originator: Mel Flynn
Organization:   
Confidential:   no
Synopsis:   [patch] Mk/bsd.python.mk: Fix python -c commands for python 3+


Hi!

Python print statement allows to use it with brackets since 2.4 (at 
least, i didn't checked < 2.4) so this change will work with any python 
version, that is in ports at the moment. For 3.x it will fix the issue, 
for 2.x it's just noop. So it's safe to commit as for me.


I rebuilt all my py-stuff with this patch (portmaster py27-) and all was 
fine. I also tested updated commands with python 2.4.3 (on RHEL) and 
they works too. So what to do next? Can I just commit this or exp-run is 
needed? Does portmgr@ approves?


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/166390: devel/py-boto: update to 2.3.0

2012-03-25 Thread Ruslan Mahmatkhanov
The following reply was made to PR ports/166390; it has been noted by GNATS.

From: Ruslan Mahmatkhanov 
To: bug-follo...@freebsd.org
Cc:  
Subject: Re: ports/166390: devel/py-boto: update to 2.3.0
Date: Sun, 25 Mar 2012 17:30:31 +0400

 Ruslan Mahmatkhanov wrote on 25.03.2012 15:11:
 >> Number: 166390
 >> Category:   ports
 >> Synopsis:   devel/py-boto: update to 2.3.0
 >
 >   CONFLICTS= py2[0-9]-boto-devel-[0-9]*
 
 Also, conflict with dns/p5-Net-Amazon-Route53 should be registered, 
 because they both installing bin/route53.
 
 -- 
 Regards,
 Ruslan
 
 Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/166142: [PATCH] www/py-requests: update to 0.10.8

2012-03-25 Thread Ruslan Mahmatkhanov
The following reply was made to PR ports/166142; it has been noted by GNATS.

From: Ruslan Mahmatkhanov 
To: Olivier Duchateau 
Cc: bug-follo...@freebsd.org
Subject: Re: ports/166142: [PATCH] www/py-requests: update to 0.10.8
Date: Sun, 25 Mar 2012 18:34:14 +0400

 Olivier Duchateau wrote on 20.03.2012 10:10:
 > The following reply was made to PR ports/166142; it has been noted by GNATS.
 >
 > From: Olivier Duchateau
 > To: bug-follo...@freebsd.org
 > Cc:
 > Subject: Re: ports/166142: [PATCH] www/py-requests: update to 0.10.8
 > Date: Tue, 20 Mar 2012 07:08:33 +0100
 >
 >   Thank you for your work.
 >
 >   It's ok, but I prefer to use find_packages in setup.py file.
 >   Furthemore, requests work with Python 3.x, so I adjusted Makefile.
 >
 >   Update, with improvement, 
 > http://www.freebsd.org/cgi/query-pr.cgi?pr=3D1662=
 >   52
 
 Please note, that chardet2 is python3 only.
 
 -- 
 Regards,
 Ruslan
 
 Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: conflict between python2 and python3 breaking blender build with pkgng

2012-03-31 Thread Ruslan Mahmatkhanov

Baptiste Daroussin wrote on 31.03.2012 20:59:

Hi,

Running exp run with pkgng, I figured out that somewhere in the dependencies of
blender there is both python2 and python3 required, the problem is that python2
and python3 are conflicting.

pkgng discovers this automatically and fail installing the dependencies needed
to build blender saying:
pkg: python32-3.2.2_3 conflicts with python27-2.7.2_4 (installs files into the
same place).  Problematic file: /usr/local/bin/2to3

Can you fix that?

regards,
Bapt


If I understand correctly, devel/glib20 pulling default python version 
(2.7). It looks like that python is needed at runtime only for 
bin/gtester-report script (some debugging tool). I didn't checked, but 
may be this may be resolved with changing USE_PYTHON=yes to 
USE_PYTHON_BUILD=yes in glib20's Makefile (who needs that gtester-report 
anyway?).


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/166617: textproc/py-pystache: Updated to 0.4.1

2012-04-04 Thread Ruslan Mahmatkhanov
The following reply was made to PR ports/166617; it has been noted by GNATS.

From: Ruslan Mahmatkhanov 
To: Grzegorz Blach 
Cc: bug-follo...@freebsd.org
Subject: Re: ports/166617: textproc/py-pystache: Updated to 0.4.1
Date: Wed, 04 Apr 2012 11:48:59 +0400

 >> Description:
 > textproc/py-pystache: Updated to 0.4.1
 
 Hi!
 
 I'd prefer to skip it, because it affects only python24 that we planning 
 to drop anyway. Would you agree?
 
 -- 
 Regards,
 Ruslan
 
 Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/166617: textproc/py-pystache: Updated to 0.4.1

2012-04-04 Thread Ruslan Mahmatkhanov
The following reply was made to PR ports/166617; it has been noted by GNATS.

From: Ruslan Mahmatkhanov 
To: Grzegorz Blach 
Cc: bug-follo...@freebsd.org
Subject: Re: ports/166617: textproc/py-pystache: Updated to 0.4.1
Date: Wed, 04 Apr 2012 12:14:51 +0400

 Grzegorz Blach wrote on 04.04.2012 11:58:
 > On 04/04/2012 09:48 AM, Ruslan Mahmatkhanov wrote:
 >>>> Description:
 >>> textproc/py-pystache: Updated to 0.4.1
 >>
 >> Hi!
 >>
 >> I'd prefer to skip it, because it affects only python24 that we planning
 >> to drop anyway. Would you agree?
 >>
 >
 > This sounds reasonable, since yesterday 0.5.0-rc1 was released (I expect
 > stable 0.5.0 soon). Skip it.
 
 I closed the PR. Thanks.
 
 -- 
 Regards,
 Ruslan
 
 Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/159215: Mk/bsd.python.mk: PYTHON_VERSION is ignored

2012-04-11 Thread Ruslan Mahmatkhanov

m...@freebsd.org wrote on 11.04.2012 17:17:

Synopsis: Mk/bsd.python.mk: PYTHON_VERSION is ignored

State-Changed-From-To: open->closed
State-Changed-By: miwi
State-Changed-When: Wed Apr 11 13:17:46 UTC 2012
State-Changed-Why:
user mistake, you have to set USE_PYTHON= 2.6.

http://www.freebsd.org/cgi/query-pr.cgi?pr=159215


Martin, I'm not sure about. There is a comment in bsd.python.mk that states:

"""
Define PYTHON_VERSION to override the defaults that USE_PYTHON would 
give you.

"""

In PR I tried to do just that, but it didn't working.

--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: [Fwd: python-doc-html-2.7.3 failed on amd64 8]

2012-04-15 Thread Ruslan Mahmatkhanov


Fixed for default case (python=2.7.3). Updated docs for other python 
versions will come later.


Pav Lucistnik wrote on 15.04.2012 13:32:

 Přeposlaná zpráva 

Od: User Ports-amd64
Komu: c...@oook.cz, ite...@freebsd.org, florent.thou...@gmail.com
Předmět: python-doc-html-2.7.3 failed on amd64 8
Datum: Fri, 13 Apr 2012 04:43:40 GMT

You can also find this build log at

   
http://pointyhat.FreeBSD.org/errorlogs/amd64-errorlogs/a.8.20120412130619/python-doc-html-2.7.3.log

building python-doc-html-2.7.3 on gohan12.freebsd.org
in directory /x/pkgbuild/8/20120412130619/chroot/157
building for: 8.1-RELEASE-p4 amd64
maintained by: pyt...@freebsd.org
port directory: /usr/ports/lang/python-doc-html
Makefile ident: $FreeBSD: ports/lang/python-doc-html/Makefile,v 1.28 2011/10/10 
13:59:57 sunpoet Exp $
build started at Fri Apr 13 04:43:34 UTC 2012
FETCH_DEPENDS=
PATCH_DEPENDS=
EXTRACT_DEPENDS=
BUILD_DEPENDS=python27-2.7.3.tbz
RUN_DEPENDS=
prefixes: LOCALBASE=usr/local
add_pkg


===>   License check disabled, port has not defined LICENSE
=>  python-2.7.3-docs-html.tar.bz2 is not in 
/a/ports/lang/python-doc-html/distinfo.
=>  Either /a/ports/lang/python-doc-html/distinfo is out of date, or
=>  python-2.7.3-docs-html.tar.bz2 is spelled incorrectly.
*** Error code 1

Stop in /a/ports/lang/python-doc-html.

build of /usr/ports/lang/python-doc-html ended at Fri Apr 13 04:43:36 UTC 2012



--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


lang/python-doc-html questions

2012-04-15 Thread Ruslan Mahmatkhanov

Hi!

I'm ready to add updated doc entries to distinfo for other (besides 
2.7.3 that is already updated) python versions. I'll do it right now, 
but there is some questions that I wanna ask in parallel:


1. What's the point to hold doc entries in distinfo for python versions 
that's not in the tree already (like 2.7.2, 2.7.1 etc)? I can't imagine 
the situation when user has updated lang/python-doc-html and outdated 
lang/pythonXX. Yes, user may stick with outadated minor release, but 
it's rare. I believe that this old entries should be just removed. Any 
objections?
2. There are no (and never will be because of EOL) updated docs for 
lang/python24 and lang/python25 for current versions of this ports. So
this ports is broken forever on this python versions. I willing to 
remove 24/25 entries from distinfo and tweak the logic of Makefile to 
inform potential users, who will try to install the docs for this 
versions with IGNORE knob. Any objections?
3. Python 2.6.0 docs logic is no valid for 2.6.8. Tarball no more 
unpacks to ${WRKDIR}/python-docs-${DOCFORMAT}, but into 
${WRKDIR}/${DISTNAME}, as other python versions. So aforementioned tweak 
should be removed because of reasons in item 1. Objections?
4. Remove postscript-docs Makefile logic, because postscript versions of 
documentation were released only for python24/python25 and there is no 
such tarballs for current versions of this ports in the tree, nor there 
is postscript-docs versions for python >=2.6. So this is just 
unnecessary complexity in Makefile. Corresponding slave ports 
(lang/python-doc-postscript-a4 and lang/python-doc-postscript-letter) 
should be removed too, because they were broken for a long time. Any 
objections?

5. TODO: Add epub documentation format. It is available only for python32.

I'm planning to do this as series of micro-commits to not turn it into 
`task of my life' and to be sure I'm not break anything.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: [Fwd: dvdvideo-20120126_1 failed on amd64 8]

2012-04-16 Thread Ruslan Mahmatkhanov

wen heping wrote on 16.04.2012 17:16:

2012/4/16 Pav Lucistnik


On 2012/04/16 10:47, wen heping wrote:


Hi, all:

I attached a patch to resole it .
Any comments?



Are you going to pass over _all_ py* ports with that plist magic?



Not all, but those ports which could both build with python2 and python3.

Even that, there are many plist should be modified. But currently I have no
better resolve and I hope others
have a simple way to resolve it.

wen


I believe Pav means that this change should be actually added to 
bsd.python.mk, not the individual port's Makefiles.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: [Fwd: dvdvideo-20120126_1 failed on amd64 8]

2012-04-16 Thread Ruslan Mahmatkhanov

Pav Lucistnik wrote on 16.04.2012 17:31:

On 2012/04/16 15:26, Ruslan Mahmatkhanov wrote:

wen heping wrote on 16.04.2012 17:16:

2012/4/16 Pav Lucistnik


On 2012/04/16 10:47, wen heping wrote:


Hi, all:

I attached a patch to resole it .
Any comments?



Are you going to pass over _all_ py* ports with that plist magic?



Not all, but those ports which could both build with python2 and
python3.

Even that, there are many plist should be modified. But currently I
have no
better resolve and I hope others
have a simple way to resolve it.

wen


I believe Pav means that this change should be actually added to
bsd.python.mk, not the individual port's Makefiles.


Eh? Look at the patch, it is added to bsd.python.mk _and_ to individual
plist.


Yeah, it was me with inattentive look at this. Shutting up now, sorry.

--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: rhodecode port

2012-04-26 Thread Ruslan Mahmatkhanov

Hi Luiz!

Luiz Gustavo wrote on 26.04.2012 19:35:

Hi list !

I created a port from project rhodecode (www.rhodecode.org) to manage the hg and
git repositories.

The project is all done in python

I have already registered PR for them:
http://www.freebsd.org/cgi/query-pr.cgi?pr=167185 (dependency)
http://www.freebsd.org/cgi/query-pr.cgi?pr=167186

Anyway, I left in my repository available at:
https://bitbucket.org/gugabsd/freebsd_ports/src

Thanks !
---
Luiz Gustavo Costa (Powered by BSD)
*+*+*+*+*+*+*+*+*+*+*+*+*+*+*+*+*+*+*+*+*+*+*+*+*+
mundoUnix - Consultoria em Software Livre
http://www.mundounix.com.br
ICQ: 2890831 / MSN: cont...@mundounix.com.br
Tel: 55 (21) 4063-7110 / 8194-1905 / (11) 4063-0407
Blog: http://www.luizgustavo.pro.br


At glance this port should not use PKGNAMEPREFIX because it's a tool, 
not a library, so the port should be named just www/rhodecode. Also the 
port doesn't support python3, so it's USE_PYTHON should be adjusted to 
"-2.7". I also see that it build depends on PasteScript>=1.6.3 (per 
setup.py), but there is no such dependency in port's Makefile.

I will be able to deal with it at weekend if nobody picks it before.

--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: rhodecode port

2012-05-10 Thread Ruslan Mahmatkhanov

Hi Luiz,

Luiz Gustavo wrote on 10.05.2012 18:25:

Em Sáb, 2012-04-28 às 19:25 -0300, Luiz Gustavo escreveu:

On Fri, 27 Apr 2012 06:38:20 +0400
Ruslan Mahmatkhanov  wrote:


Hi Luiz!



At glance this port should not use PKGNAMEPREFIX because it's a tool,
not a library, so the port should be named just www/rhodecode. Also the
port doesn't support python3, so it's USE_PYTHON should be adjusted to
"-2.7". I also see that it build depends on PasteScript>=1.6.3 (per
setup.py), but there is no such dependency in port's Makefile.
I will be able to deal with it at weekend if nobody picks it before.



Hi Ruslan,

I made the changes you recommended, see:

https://bitbucket.org/gugabsd/freebsd_ports/changeset/aca830bcf349

I created a new PR: http://www.freebsd.org/cgi/query-pr.cgi?pr=167773

Very thanks for you support !


Just picked it. Will see tonight. Thanks.

--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: [PATCH] python2.4 removal zope and zope211 affected

2012-05-28 Thread Ruslan Mahmatkhanov

Hi, Martin,

Martin Wilke wrote on 28.05.2012 13:44:

Hi,

Finally I'm working on to remove python 24 from the ports tree, and the
reasons are clear. Support was dropped by 2008, and there are a lot of
unpatched security holes. We have warned about this since 2 years ago,
and now I will really take action to drop it from the ports tree.


I have no objections, but know people that do :). I ever know FreeBSD 
committer that using old zope.

We have only three ports, that wish exactly python 2.4 version:

net/honeyd (skreu...@freebsd.org)
www/zope   (esta...@augusta.de)
www/zope211 (esta...@augusta.de)

And only one port that depends on exact 2.5 version:

www/google-appengine (ni...@freebsd.org)

I prefer to remove both of python24 and python25 because they both EOL. 
But I'd still set them all to expiration period of month or two. We have 
such a tradition.



So here is the final patch [1] that has already been reviewed by lwshu@
and wen@, but I would like to discuss regarding the zope ports.

First of all I will clean up the zope version from the bsd.python.mk.
Now I would like to know what you guys think if I were to remove
www/zope and keep www/zope213. Another option is to merge www/zope213
to www/zope and just remove www/zope211 and www/zope213 completely.
What do you think about this?

Thanks Martin

[1]
http://miwifbsd.dyndns.org/patches/py24rm-3.diff



Our zope213 port didn't use ZOPESKELDIR, SZOPEBASEDIR so it may look 
like this:


+if ${ZOPE_VERSION} == "2.13"
+ZOPE_PORTSDIR= ${PORTSDIR}/www/zope213
 .else

I prefer to keep it as zope213 for some time.

--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: [PATCH] python2.4 removal zope and zope211 affected

2012-05-29 Thread Ruslan Mahmatkhanov

Hi Li-Wen,

Li-Wen Hsu wrote on 28.05.2012 22:51:

www/zope and www/zope211 can also be set deprecated, how about net/honeyd?
can it work with newer Python?  I looked at the project site, it seems moved to
somewhere else.


No, I can't recall the details, but net/honeyd isn't working with python 
> 2.4. I had investigate this some time ago. And no new versions is 
available.



And only one port that depends on exact 2.5 version:

www/google-appengine (ni...@freebsd.org)


I think we don't have a exact plan to remove 2.5 currently, do we?


I'd update www/google-appengine and drop 2.5 altogether with 2.4. It's 
simple plan :). But it more correct to let this port's expire first.



I prefer to keep it as zope213 for some time.


We do, too.  Do you have a plan for updating www/zope to 2.13?  BTW,
what's the status of
BlueBream?

Thanks,
Li-Wen


Not yet. Let the dust settle after www/zope removal. I think that people 
are generally aware where 2.13 is resides.
BB 1.0 is available since January 2011, but I'm not using it and don't 
have plans to port it atm. There will be Plone 4.2 available soon, and

it will go right ahead to www/plone, since it backward-compatible with 4.1.

--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: [PATCH] python2.4 removal zope and zope211 affected

2012-05-29 Thread Ruslan Mahmatkhanov

Hi Martin,

Martin Wilke wrote on 29.05.2012 03:02:

Nope, we have already mention that ever and ever and ever again, this
time i dont do that where a problem with that is this time not mine,
the python@ histroy can prove that. I wont postpone it again


As I said already, I have no objections and actually all I did in past 
it's a dropping python24/25 support in ports :). Expiring before 
deleting it's a common practice. And to me it's not a technical aspect, 
but our users awareness and loyalty aspect.



+1 for py25 removal.. since is the same situation..


So let's get them out both in a row. I think that holding 25 is more 
pointless, than ever holding 24.



Will fix the patch


Ok, thank you.

--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Fwd: cvs commit: ports/devel/mercurial Makefile distinfo pkg-descr pkg-plist

2012-06-03 Thread Ruslan Mahmatkhanov

Hi!

Sorry, I know I didn't discussed maintainership transferring step with 
anybody from python@. But I think that mercurial port is important 
enough to keep it's maintenance under collective effort.


 Original Message 
Subject: cvs commit: ports/devel/mercurial Makefile distinfo pkg-descr 
   pkg-plist

Date: Sun, 3 Jun 2012 19:38:59 + (UTC)
From: Ruslan Mahmatkhanov 
To: ports-committ...@freebsd.org, cvs-po...@freebsd.org, cvs-...@freebsd.org

rm  2012-06-03 19:38:59 UTC

  FreeBSD ports repository

  Modified files:
devel/mercurial  Makefile distinfo pkg-descr pkg-plist
  Log:
  - update to 2.2.2
  - add LICENSE (GPLv2)
  - pass maintainership to python@

  PR: 168552
  Submitted by:   koobs 
  Approved by:roberto (maintainer, via email)

  Revision  ChangesPath
  1.72  +4 -2  ports/devel/mercurial/Makefile
  1.47  +2 -2  ports/devel/mercurial/distinfo
  1.2   +1 -1  ports/devel/mercurial/pkg-descr
  1.46  +10 -1 ports/devel/mercurial/pkg-plist


___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: cvs commit: ports/devel/mercurial Makefile distinfo pkg-descr pkg-plist

2012-06-03 Thread Ruslan Mahmatkhanov

Eitan Adler wrote on 04.06.2012 01:24:

On 3 June 2012 12:52, Ruslan Mahmatkhanov  wrote:

Hi!

Sorry, I know I didn't discussed maintainership transferring step with
anybody from python@. But I think that mercurial port is important enough to
keep it's maintenance under collective effort.


Teams are treated exactly like individuals. Unless you ask first
please don't transfer ports around like this.
(note, I'm not speaking for python@)


Ok, will do next time. I didn't actually bother to ask in this case 
because I'm interested in this port and I'm in python@ :).


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Python conflicting with other python when installed from packages

2012-06-05 Thread Ruslan Mahmatkhanov

Baptiste Daroussin wrote on 05.06.2012 16:07:

Hi,

With pkgng build on pointyhat I discovered that all the python packages when
built on pointyhat, tinderbox or even poudriere all conflicts which each other
meaning you can't install multiple python version in parallel without
overwritting files with old package system and pkgng just forbid you to install
the second package to avoid overwritting the file.

The thing is that they are conflicting only for the bin/2to3 file. It would be
nice if this file could be for example prefixed with the release of python like:

bin/py27-2to3 or something like that. I'm aware that when just installing from
ports only the first version of python installed got it, but if you are thinking
in package only world it can't works that way.

regards,
Bapt


Packages that using 2to3 to convert their code on install to be python3 
compatible looking exactly for `2to3' executable. So such a modification 
would require modifying all the setup.py's according the python version 
user has installed as default. It's just a though. I didn't see how to 
resolve this correctly.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Python conflicting with other python when installed from packages

2012-06-05 Thread Ruslan Mahmatkhanov

John Hein wrote on 05.06.2012 21:09:

Ruslan Mahmatkhanov wrote at 16:14 +0400 on Jun  5, 2012:
  >  Baptiste Daroussin wrote on 05.06.2012 16:07:
  >  >  Hi,
  >  >
  >  >  With pkgng build on pointyhat I discovered that all the python packages 
when
  >  >  built on pointyhat, tinderbox or even poudriere all conflicts which 
each other
  >  >  meaning you can't install multiple python version in parallel without
  >  >  overwritting files with old package system and pkgng just forbid you to 
install
  >  >  the second package to avoid overwritting the file.
  >  >
  >  >  The thing is that they are conflicting only for the bin/2to3 file. It 
would be
  >  >  nice if this file could be for example prefixed with the release of 
python like:
  >  >
  >  >  bin/py27-2to3 or something like that. I'm aware that when just 
installing from
  >  >  ports only the first version of python installed got it, but if you are 
thinking
  >  >  in package only world it can't works that way.
  >  >
  >  >  regards,
  >  >  Bapt
  >
  >  Packages that using 2to3 to convert their code on install to be python3
  >  compatible looking exactly for `2to3' executable. So such a modification
  >  would require modifying all the setup.py's according the python version
  >  user has installed as default. It's just a though. I didn't see how to
  >  resolve this correctly.

One way: pull out 2to3 to its own separate port and have those
ports that need it specify the dependency.  And, no, I don't
have patches.


Well, then we'll get separate 2to3 ports that is conflicting :). I'm not 
sure that 2to3 is the same in all python distributions.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/155526: [PATCH] devel/py-elementtree: ignore if python >= 2.5

2012-06-07 Thread Ruslan Mahmatkhanov
The following reply was made to PR ports/155526; it has been noted by GNATS.

From: Ruslan Mahmatkhanov 
To: Marcus von Appen 
Cc: bug-follo...@freebsd.org
Subject: Re: ports/155526: [PATCH] devel/py-elementtree: ignore if python
 >= 2.5
Date: Thu, 07 Jun 2012 20:13:42 +0400

 Marcus von Appen wrote on 07.06.2012 19:39:
 > As long as www/py-Products.TinyMCE depends on it, there is no way to
 > delete it (without fixing www/py-Products.TinyMCE beforehand for python
 >> = 2.5).
 >
 > Also, as it does not do any harm (no package name confusion with
 > xml.etree.elementTree and _elementtree), it is not really necessary to
 > force the deletion of it.
 >
 > Cheers
 > Marcus
 
 Yes, but why do we need to hold the ports, that nobody using and will 
 not be anyway? All the ports I saw, that uses elementTree, first 
 checking if they can import xml.etree and if they not, then they will 
 try to fallback to standalone _elementtree. And since python24/25 will 
 be removed soon, there is no sense to hold them.
 
 The only difference between them is HTMLTreeBuilder class that is 
 missing in xml.etree (and that py-Products.TinyMCE is using), so 
 py-elementtree should stay until Products.TinyMCE will be ported to 
 py-lxml, and py-celementtree is safe to remove (because nobody using 
 it). This ports/155526 should be closed too, because it's now impossible 
 to remove this port.
 
 PS. The same belongs to simplejson/json case. Most ports try to use 
 bundled json module first, and try to import simplejson if they fail 
 (python<2.6). But there is couple of ports that need exactly simplejson 
 implementation.
 
 -- 
 Regards,
 Ruslan
 
 Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


python32 build broken in tinderbox

2012-06-21 Thread Ruslan Mahmatkhanov

Hi Marcus,

I'm trying to build net/py-smb with python32 in tinderbox and it always 
stuck on python32 with this messages in builds\make.0:


Graph cycles through `gettext-0.18.1.1.tbz'
Graph cycles through `libiconv-1.14.tbz'
Graph cycles through `python32-3.2.3_1.tbz'
Graph cycles through `python32-3.2.3_1.tbz'
Graph cycles through `libiconv-1.14.tbz'
Graph cycles through `python32-3.2.3_1.tbz'
Graph cycles through `libtool-2.4.2.tbz'
Graph cycles through `python32-3.2.3_1.tbz'

I have nothing in logs/ and errors/.

Looks like it was introduced in rev 1.187/1.188, because I was able to 
test ports in python32 tinderbox successfully about a two days ago. Any 
ideas?


--
Regards,
Ruslan

Tinderboxing kills... the drives.

___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: python32 build broken in tinderbox

2012-06-21 Thread Ruslan Mahmatkhanov

Ok, revision 1.187 is the culprit. Reverting to 1.186 fixed things for me.


Ruslan Mahmatkhanov wrote on 21.06.2012 22:35:

Hi Marcus,

I'm trying to build net/py-smb with python32 in tinderbox and it always
stuck on python32 with this messages in builds\make.0:

Graph cycles through `gettext-0.18.1.1.tbz'
Graph cycles through `libiconv-1.14.tbz'
Graph cycles through `python32-3.2.3_1.tbz'
Graph cycles through `python32-3.2.3_1.tbz'
Graph cycles through `libiconv-1.14.tbz'
Graph cycles through `python32-3.2.3_1.tbz'
Graph cycles through `libtool-2.4.2.tbz'
Graph cycles through `python32-3.2.3_1.tbz'

I have nothing in logs/ and errors/.

Looks like it was introduced in rev 1.187/1.188, because I was able to
test ports in python32 tinderbox successfully about a two days ago. Any
ideas?





--
Regards,
Ruslan

Tinderboxing kills... the drives.

___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: [SPAM] Re: python32 build broken in tinderbox

2012-06-21 Thread Ruslan Mahmatkhanov

Marcus von Appen wrote on 21.06.2012 23:47:

Hi Ruslan,

On, Thu Jun 21, 2012, Ruslan Mahmatkhanov wrote:


Ok, revision 1.187 is the culprit. Reverting to 1.186 fixed things for me.


[...]

Does the latest one (1.188) work for you? I can't reproduce the issue
with my tinderboxes. How are the ports and dependencies configured on
yours?

Cheers
Marcus


No, it is not. I consequently revert 1.188 (with no success), then 1.187 
(the same) and now it builds fine when I'm on 1.186.
All ports options are default. Tinderbox is i386. The only difference - 
python32 is system-wide default:


nx7400> cat scripts/etc/env/build.9.0-FreeBSD-py3
USE_PYTHON=3.2
PYTHON_VERSION=python3.2
PYTHON_DEFAULT_VERSION=3.2

--
Regards,
Ruslan

Tinderboxing kills... the drives.


___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: [SPAM] Re: [SPAM] Re: python32 build broken in tinderbox

2012-06-23 Thread Ruslan Mahmatkhanov

Marcus von Appen wrote on 22.06.2012 00:45:

On, Thu Jun 21, 2012, Ruslan Mahmatkhanov wrote:


Marcus von Appen wrote on 21.06.2012 23:47:

Hi Ruslan,

On, Thu Jun 21, 2012, Ruslan Mahmatkhanov wrote:


Ok, revision 1.187 is the culprit. Reverting to 1.186 fixed things for me.


[...]

Does the latest one (1.188) work for you? I can't reproduce the issue
with my tinderboxes. How are the ports and dependencies configured on
yours?

Cheers
Marcus


No, it is not. I consequently revert 1.188 (with no success), then 1.187
(the same) and now it builds fine when I'm on 1.186.
All ports options are default. Tinderbox is i386. The only difference -
python32 is system-wide default:

nx7400> cat scripts/etc/env/build.9.0-FreeBSD-py3
USE_PYTHON=3.2
PYTHON_VERSION=python3.2
PYTHON_DEFAULT_VERSION=3.2


The issue is that USE_PYTHON=xx will introduce python as dependency for
_every_ port. Do not set it, since it is for usage in port Makefiles,
not for the environmental flags.
The reason for your problem is the circular dependency introduced by
that.

Cheers
Marcus


Sorry for delay. Yes, you are right. Removing two first lines fixed it. 
Strange that it worked fine until rev 1.187 though.


--
Regards,
Ruslan

Tinderboxing kills... the drives.


___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: [SPAM] Re: [SPAM] Re: python32 build broken in tinderbox

2012-06-23 Thread Ruslan Mahmatkhanov

Sunpoet Hsieh wrote on 22.06.2012 05:20:

On Fri, Jun 22, 2012 at 4:45 AM, Marcus von Appen  wrote:

On, Thu Jun 21, 2012, Ruslan Mahmatkhanov wrote:


Marcus von Appen wrote on 21.06.2012 23:47:

Hi Ruslan,

On, Thu Jun 21, 2012, Ruslan Mahmatkhanov wrote:


Ok, revision 1.187 is the culprit. Reverting to 1.186 fixed things for me.


[...]

Does the latest one (1.188) work for you? I can't reproduce the issue
with my tinderboxes. How are the ports and dependencies configured on
yours?

Cheers
Marcus


No, it is not. I consequently revert 1.188 (with no success), then 1.187
(the same) and now it builds fine when I'm on 1.186.
All ports options are default. Tinderbox is i386. The only difference -
python32 is system-wide default:

nx7400> cat scripts/etc/env/build.9.0-FreeBSD-py3
USE_PYTHON=3.2
PYTHON_VERSION=python3.2
PYTHON_DEFAULT_VERSION=3.2


The issue is that USE_PYTHON=xx will introduce python as dependency for
_every_ port. Do not set it, since it is for usage in port Makefiles,
not for the environmental flags.
The reason for your problem is the circular dependency introduced by
that.

Cheers
Marcus


Hi,

In tinderbox env, I'm sure that PYTHON_DEFAULT_VERSION=python3.2 is
enough for specify the python port.
HTH

Regards,


Hi,

Yes, it's true. By the way, missing PYTHON_DEFAULT_VERSION was the 
reason of py-distribute packaging break, that I asked you about a week ago.


--
Regards,
Ruslan

Tinderboxing kills... the drives.


___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: twisted/internet gtk2reactor.py issue. (breaks for instance deluge);

2012-07-03 Thread Ruslan Mahmatkhanov

Hi,

Jesper Schmitz Mouridsen wrote on 23.06.2012 11:50:

Hello maintainers of python.

I'm using python27 from ports (upgraded)

This is an issue with gtk2reactor.py

running
#
import gobject
from twisted.internet import gtk2reactor
reactor = gtk2reactor.install()
#

gives :
/usr/local/lib/python2.7/site-packages/gtk-2.0/gtk/__init__.py:40:
RuntimeWarning: tp_compare didn't return -1 or -2 for exception
   from gtk import _gtk
ImportError: No module named gi


Jesper, this issue was just fixed in ports with your solution, thank 
you. It seems it was triggered by this commit:

http://twistedmatrix.com/trac/ticket/5494

I'll contact upstream about this problem.

--
Regards,
Ruslan

Tinderboxing kills... the drives.

___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/168977: [PATCH] devel/mercurial: Fix WITHOUT_NLS option handling.

2012-07-09 Thread Ruslan Mahmatkhanov

arrowdodger wrote on 19.06.2012 23:07:

On Tue, Jun 19, 2012 at 10:42 PM,  wrote:


Synopsis: [PATCH] devel/mercurial: Fix WITHOUT_NLS option handling.

State-Changed-From-To: open->feedback
State-Changed-By: rm
State-Changed-When: Tue Jun 19 18:42:23 UTC 2012
State-Changed-Why:
Hi arrowdodger, thank you for submission. Would you please clarify what
the point of this change? I checked the port in it's current state and
if works properly (only installing locale dirs if WITHOUT_NLS is not
defined).

http://www.freebsd.org/cgi/query-pr.cgi?pr=168977



Uhm. Without this patch i get "hg" in russian despite WITHOUT_NLS=YES in
/etc/make.conf. And it installs po dirs for me.


Hi,

I'm sorry for delay. Just looked into this and Marcus was right. The 
current port fails to follow WITHOUT_NLS if msgfmt(1) (devel/gettext) is 
installed. So I filed upstream bug-report:

http://bz.selenic.com/show_bug.cgi?id=3536

Right now I'm testing proposed patch and will commit it later today 
(along with 2.2.3 update). Thanks.


--
Regards,
Ruslan

Tinderboxing kills... the drives.

___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Mercurial CA Certificates

2012-08-29 Thread Ruslan Mahmatkhanov

Hi,

Alexander Kapshuk wrote on 19.08.2012 21:45:

How do I add an CA Certificate for Mercurial on FreeBSD please?

:; uname -a
FreeBSD box2 9.0-RELEASE-p4 FreeBSD 9.0-RELEASE-p4 #0: Fri Aug 17
21:53:39 EEST 2012 root@box2:/usr/obj/usr/src/sys/GENERIC  i386

:; pkg_info -xc mercurial | sed 1q
Information for mercurial-2.3:

Thanks.


As far I understand, FreeBSD doesn't come prepackaged with root CA 
certificates like other systems do (it's not 146%, but seems so). So, 
I'd follow recommendation from [1] in part "2.7. Other platforms".


- download CA list from [2]
- put this lines into your ~/.hgrc:

[web]
cacerts = /place/where/you/put/cacert.pem

Hope this helps.

[1] http://mercurial.selenic.com/wiki/CACertificates
[2] http://curl.haxx.se/docs/caextract.html

--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Mercurial CA Certificates

2012-08-30 Thread Ruslan Mahmatkhanov

Alexander Kapshuk wrote on 30.08.2012 00:34:

On 08/29/12 22:00, Ruslan Mahmatkhanov wrote:

Hi,

Alexander Kapshuk wrote on 19.08.2012 21:45:

How do I add an CA Certificate for Mercurial on FreeBSD please?

:; uname -a
FreeBSD box2 9.0-RELEASE-p4 FreeBSD 9.0-RELEASE-p4 #0: Fri Aug 17
21:53:39 EEST 2012 root@box2:/usr/obj/usr/src/sys/GENERIC i386

:; pkg_info -xc mercurial | sed 1q
Information for mercurial-2.3:

Thanks.


As far I understand, FreeBSD doesn't come prepackaged with root CA
certificates like other systems do (it's not 146%, but seems so). So,
I'd follow recommendation from [1] in part "2.7. Other platforms".

- download CA list from [2]
- put this lines into your ~/.hgrc:

[web]
cacerts = /place/where/you/put/cacert.pem

Hope this helps.

[1] http://mercurial.selenic.com/wiki/CACertificates
[2] http://curl.haxx.se/docs/caextract.html


OK. Thanks. I'll give that a try.


Just realized that there is security/ca_root_nss that installs the certs 
into /usr/local/share/certs and the port itself asks if you want to 
create symlink to it in /etc/ssl/cert.pem. You may add any of them into 
your ~/.hgrc. Please let me know if it works for you, and if it is, it 
should be added into mercurial's wiki, and maybe to our mercurial port 
as pkg-message or so.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Simple change into Mk/bsd.python.mk

2012-09-02 Thread Ruslan Mahmatkhanov

Hi,

I'm preparing an update of www/plone and discovered thing that's need to 
be changed to make USE_ZOPE knob actually work. We do not define 
ZOPESKELDIR anymore, so the RUN_DEPENDS line should be fixed.
While here, trim some leftover from py24/25 removal, that's now always 
true in current ports tree. Would you please approve this change?


PS. It doesn't needed an exp-run because at the time there is no ports, 
that use USE_ZOPE.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
Index: bsd.python.mk
===
--- bsd.python.mk   (revision 303535)
+++ bsd.python.mk   (working copy)
@@ -501,7 +501,7 @@
(defined(INSTALLS_EGGINFO) ||   \
(defined(USE_PYDISTUTILS) && \
 ${USE_PYDISTUTILS} != "easy_install")) && \
-defined(PYTHON_REL) && ${PYTHON_REL} >= 250
+defined(PYTHON_REL)
 . for egg in ${PYDISTUTILS_EGGINFO}
 PLIST_FILES+=  ${PYDISTUTILS_EGGINFODIR:S;${PREFIX}/;;}/${egg}
 . endfor
@@ -543,7 +543,7 @@
 .endif # ${PYTHON_NO_DEPENDS} == "NO"
 
 .if defined(USE_ZOPE)
-RUN_DEPENDS+=  ${ZOPESKELDIR}/bin/zopectl.in:${ZOPE_PORTSDIR}
+RUN_DEPENDS+=  ${LOCALBASE}/bin/zopectl:${ZOPE_PORTSDIR}
 .endif
 
 # set $PREFIX as Python's one
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Simple change into Mk/bsd.python.mk

2012-09-02 Thread Ruslan Mahmatkhanov

Hi Li-Wen,

Li-Wen Hsu wrote on 02.09.2012 14:03:

On Sun, Sep 2, 2012 at 3:31 PM, Ruslan Mahmatkhanov  wrote:

Hi,

I'm preparing an update of www/plone and discovered thing that's need to be
changed to make USE_ZOPE knob actually work. We do not define ZOPESKELDIR
anymore, so the RUN_DEPENDS line should be fixed.
While here, trim some leftover from py24/25 removal, that's now always true
in current ports tree. Would you please approve this change?

PS. It doesn't needed an exp-run because at the time there is no ports, that
use USE_ZOPE.


Looks good to me, and if you don't mind, how about doing this:

 RUN_DEPENDS+=  ${LOCALBASE}/bin/zopectl:${ZOPE_PORTSDIR}

in this way:

 RUN_DEPENDS+=  zope213>0:${ZOPE_PORTSDIR}

or even replace this:

 ZOPE_PORTSDIR= ${PORTSDIR}/www/zope213

with:

 ZOPE_DEPENDS=  zope213>0:${PORTSDIR}/www/zope213

and then, .if defined(USE_ZOPE) section can be written as:

 RUN_DEPENDS+=  ${ZOPE_DEPENDS}

How do you think?



I believe that original intention was not to hardcode the zope port 
version. Yes, there is only one zope version at the moment, but there 
may be more in future. Just though about ports that needs different zope 
versions. Please tell me if I understand it wrong.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Simple change into Mk/bsd.python.mk

2012-09-02 Thread Ruslan Mahmatkhanov

Li-Wen Hsu wrote on 02.09.2012 20:37:

On Sun, Sep 02, 2012 at 16:01:51 +0400, Ruslan Mahmatkhanov wrote:


I believe that original intention was not to hardcode the zope port
version. Yes, there is only one zope version at the moment, but there
may be more in future. Just though about ports that needs different zope
versions. Please tell me if I understand it wrong.



We have the same thought.  Sorry that my expression was not clear.
My proposed patch is attached, and what I just wanted was reducing
hardcoded dependency of ${LOCALBASE}/bin/zopectl .  I am not sure which
one is better, so please just commit one you like.


Ah, understood. I like the idea. Thank you! If there will not be strong 
objections, this will be committed tomorrow.




BTW, should we bother portmgr at this point?  I believe they have much
more important things need to deal with.  But I am still happy to hear
any suggestion from them if they want to step in.

Regards,
Li-Wen


To be honest, I was under impression that anything under Mk/ should be 
portmgr approved, but Chris told me that this is not the case.



--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/171287: [UPDATE] devel/mercurial to 2.3.1

2012-09-04 Thread Ruslan Mahmatkhanov
The following reply was made to PR ports/171287; it has been noted by GNATS.

From: Ruslan Mahmatkhanov 
To: Olivier Duchateau 
Cc: bug-follo...@freebsd.org
Subject: Re: ports/171287: [UPDATE] devel/mercurial to 2.3.1
Date: Tue, 04 Sep 2012 11:17:46 +0400

 Hi Olivier,
 
 there is upstream feature request for providing a way to disable 
 localizations installation [1], that is work in progress now, as I 
 understand. So I prefer to not hack/complicate our port more than it is 
 now and wait for upstream solution, furthermore we already have a 
 working temporary solution anyway.
 So, at this time I'd suggest to commit only the update itself and 
 optionsng change, but not NLS hacking. What do you think?
 
 [1] http://bz.selenic.com/show_bug.cgi?id=3536
 
 PS. By the way, if you have a vision/patch on how to implement this 
 nicely in mercurial's setup.py, you may share this in aforementioned 
 bug-report to ease upstream's job.
 
 Olivier Duchateau wrote on 03.09.2012 21:59:
 >> Number: 171287
 >> Category:   ports
 >> Synopsis:   [UPDATE] devel/mercurial to 2.3.1
 >> Confidential:   no
 >> Severity:   non-critical
 >> Priority:   low
 >> Responsible:freebsd-ports-bugs
 >> State:  open
 >> Quarter:
 >> Keywords:
 >> Date-Required:
 >> Class:  sw-bug
 >> Submitter-Id:   current-users
 >> Arrival-Date:   Mon Sep 03 18:00:20 UTC 2012
 >> Closed-Date:
 >> Last-Modified:
 >> Originator: Olivier Duchateau
 >> Release:FreeBSD 8.3-RELEASE-p3 i386
 >> Organization:
 >> Environment:
 >> Description:
 > - Update to 2.3.1, 
 > http://mercurial.selenic.com/wiki/WhatsNew#Mercurial_2.3.1_.282012-09-01.29
 > - Convert to new framework OptionsNG
 >> How-To-Repeat:
 >
 >> Fix:
 >
 >
 > Patch attached with submission follows:
 >
 > Index: distinfo
 > ===
 > --- distinfo (r�vision 303607)
 > +++ distinfo (copie de travail)
 > @@ -1,2 +1,2 @@
 > -SHA256 (mercurial-2.3.tar.gz) = 
 > 456cb8a9fb15606a04d81589a3459d258149170008ca64c62e5a6cbaf5f0fe5f
 > -SIZE (mercurial-2.3.tar.gz) = 3549002
 > +SHA256 (mercurial-2.3.1.tar.gz) = 
 > ef587fdd3b062e08eb4012e7870f4d0a00fd834f11578bfba80313b10f808541
 > +SIZE (mercurial-2.3.1.tar.gz) = 3558058
 > Index: files/patch-setup.py
 > ===
 > --- files/patch-setup.py (r�vision 303607)
 > +++ files/patch-setup.py (copie de travail)
 > @@ -1,10 +0,0 @@
 >  setup.py.orig   2012-08-02 10:49:31.0 +0400
 > -+++ setup.py2012-08-08 18:34:01.0 +0400
 > -@@ -227,6 +227,7 @@
 > - description = "build translations (.mo files)"
 > -
 > - def run(self):
 > -+%%NLS%%return
 > - if not find_executable('msgfmt'):
 > - self.warn("could not find msgfmt executable, no translations "
 > -  "will be built")
 > Index: files/extra-patch-setup.py
 > ===
 > --- files/extra-patch-setup.py   (r�vision 0)
 > +++ files/extra-patch-setup.py   (copie de travail)
 > @@ -0,0 +1,29 @@
 > +--- setup.py.orig   2012-08-02 06:49:31.0 +
 >  setup.py2012-08-05 08:18:26.0 +
 > +@@ -212,7 +212,7 @@
 > + class hgbuild(build):
 > + # Insert hgbuildmo first so that files in mercurial/locale/ are found
 > + # when build_py is run next.
 > +-sub_commands = [('build_mo', None),
 > ++sub_commands = [
 > +
 > + # We also need build_ext before build_py. Otherwise, when 2to3 is
 > + # called (in build_py), it will not find osutil & friends,
 > +@@ -402,7 +402,6 @@
 > + fp.close()
 > +
 > + cmdclass = {'build': hgbuild,
 > +-'build_mo': hgbuildmo,
 > + 'build_ext': hgbuildext,
 > + 'build_py': hgbuildpy,
 > + 'build_hgextindex': buildhgextindex,
 > +@@ -463,8 +462,7 @@
 > + extmodules.append(inotify)
 > + packages.extend(['hgext.inotify', 'hgext.inotify.linux'])
 > +
 > +-packagedata = {'mercurial': ['locale/*/LC_MESSAGES/hg.mo',
 > +- 'help/*.txt']}
 > ++packagedata = {'mercurial': ['help/*.txt']}
 > +
 > + def ordinarypath(p):
 > + return p and p[0] != '.' and p[-1] != '~'
 > Index: Makefile
 > ===
 > --- Makefile (rï

Re: Checksum mismatch for lang/python27

2012-09-20 Thread Ruslan Mahmatkhanov

David Ehrmann wrote on 21.09.2012 09:59:

I've been getting a checksum mismatch when installing/upgrading
lang/python27. The file that seems to have the issue is Python-2.7.3.tgz.


Can't confirm. Please try to do `make distclean && make fetch` in 
lang/python27 directory.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: bin/python3 symlink for python3X ports

2012-09-26 Thread Ruslan Mahmatkhanov

Chris Rees wrote on 26.09.2012 13:03:

On 26 Sep 2012 08:13, "Dmitry Sivachenko"  wrote:


Hello!

It is common to have both python-2 and python-3 packages installed.
To help transition it is convenient to have bin/python to point to

python-2 binary.


Now for every X in python3X we have a bin/python3.X binary.

So for every new python3.X release one should change shebang line in

scripts.


I propose to tweak python3X ports so they create bin/python3 symlink

pointing to the installed python3.X binary so that people can always use
'python3' name.


What do you think?


I think it's a great idea, at least minimising weird breakage when python
is updated.

Have you a patch?  Also, what do other OSes do?

Chris


AFAIK, linuces has python2 and python3 symlinks. So this is generally 
common idea. And I +1 for it.



--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: bin/python3 symlink for python3X ports

2012-09-26 Thread Ruslan Mahmatkhanov

Dmitry Sivachenko wrote on 26.09.2012 13:14:

On 26.09.2012 13:07, Ruslan Mahmatkhanov wrote:

Chris Rees wrote on 26.09.2012 13:03:

On 26 Sep 2012 08:13, "Dmitry Sivachenko"  wrote:


Hello!

It is common to have both python-2 and python-3 packages installed.
To help transition it is convenient to have bin/python to point to

python-2 binary.


Now for every X in python3X we have a bin/python3.X binary.

So for every new python3.X release one should change shebang line in

scripts.


I propose to tweak python3X ports so they create bin/python3 symlink

pointing to the installed python3.X binary so that people can always use
'python3' name.


What do you think?


I think it's a great idea, at least minimising weird breakage when
python
is updated.

Have you a patch?  Also, what do other OSes do?




I don't have a patch but I can make one if you like.


Please do.


I don't use other OSes so I have no idea what do they do with that
problem and I don't see any reason why should we bother...


To no reinvent the wheel, for example. And because many software 
installers (setup.py, f.e.) are somewhat linux-centric: they looking for 
python2 or python3 binary (that we change to ${PYTHON_CMD} in our 
Makefile's).


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: bin/python3 symlink for python3X ports

2012-09-26 Thread Ruslan Mahmatkhanov

Dmitry Sivachenko wrote on 26.09.2012 15:18:

On 26.09.2012 13:20, Ruslan Mahmatkhanov wrote:

Dmitry Sivachenko wrote on 26.09.2012 13:14:

I don't have a patch but I can make one if you like.


Please do.



Please consider the attached patch.


The patch seems good to me. The only thing: I'd rather define 
${PYTHON_MAJOR} (or ${PYTHON_MAJOR_VER}) in bsd.python.mk to make it 
reusable when USE_PYTHON is used (sorry for the pun).


PYTHON_MAJOR_VER=${PYTHON_VER:R}

because there will be other ports that need it. And as Li-Wen said, 
there should be the same change for python2. And it needs an exp-run, I 
believe.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Python distribution: switching to tar.xz

2012-10-01 Thread Ruslan Mahmatkhanov


Hello,

The python.org ftp provides lang/pythonXX source tarballs in a many formats:
- tar.bz2
- tar.xz
- tgz

Using .tgz is not in sports-way this days, so I'd prefer to use tar.xz 
for new versions. But I can't actually choose there because tarball 
format is hardcoded in Mk/bsd.python.mk as PYTHON_DISTFILE variable:


PYTHON_DISTFILE=
Python-${PYTHON_PORTVERSION:S/.rc/rc/}.tgz

And in particular python port we have this:

DISTFILES=  ${PYTHON_DISTFILE}

Since I can't just choose tar.xz for some new python33 port, I suggest 
to change PYTHON_DISTFILE to use tar.xz and to change all the distinfos 
for lang/pythonXX (with changing SHA256/SIZE fields). What do you think?


Surely, I will compare the contents of current tgz with new tar.xz for 
the case, but does the whole idea is correct? May be it's more correct 
to not use ${PYTHON_DISTFILE} in first place? I'd prefer to change it in 
bsd.python.mk, but it will require to redownload the new tarballs.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Python distribution: switching to tar.xz

2012-10-01 Thread Ruslan Mahmatkhanov

Li-Wen Hsu wrote on 01.10.2012 21:51:

On Mon, Oct 1, 2012 at 10:10 PM, Marcus von Appen  wrote:


Ruslan Mahmatkhanov :


Hello,

The python.org ftp provides lang/pythonXX source tarballs in a many
formats:
- tar.bz2
- tar.xz
- tgz

Using .tgz is not in sports-way this days, so I'd prefer to use tar.xz for
new versions. But I can't actually choose there because tarball format is
hardcoded in Mk/bsd.python.mk as PYTHON_DISTFILE variable:

PYTHON_DISTFILE=
Python-${PYTHON_PORTVERSION:S/.rc/rc/}.tgz

And in particular python port we have this:

DISTFILES=  ${PYTHON_DISTFILE}

Since I can't just choose tar.xz for some new python33 port, I suggest to
change PYTHON_DISTFILE to use tar.xz and to change all the distinfos for
lang/pythonXX (with changing SHA256/SIZE fields). What do you think?



I would rather see PYTHON_DISTFILE removed from bsd.python.mk and being
moved back
to the individual ports instead. Or, to stay compatible for whatever reason,
change
it to PYTHON_DISTFILE?= ..., so ports can override it.

Or, even better in making things complicated, change it to

PYTHON_DISTFILE = ${PYTHON_IMPLTYPE}_${PORTVERSION}${EXTRACT_SUFX}



I'm not sure introducing ${PYTHON_IMPLTYPE} is a good idea... but it
might be good to
change hard-coded .tgz:

PYTHON_DISTFILE= Python-${PYTHON_PORTVERSION:S/.rc/rc/}.tgz

to:

EXTRACT_SUFX?= .tgz
PYTHON_DISTFILE= Python-${PYTHON_PORTVERSION:S/.rc/rc/}}${EXTRACT_SUFX}


Yes, I also bow to this option.


then we can define USE_XZ= yes in the ports we want to use xz
distfile, which handles
${EXTRACT_SUFX} and ${EXTRACT_DEPENDS} correctly.  After python26 decommission,
we can even move this into  bsd.python.mk


python26 distributes in tar.xz too [1]. Or you meant something other?

Ok, so I get the point. I'll prepare an patch that make ${EXTRACT_SUFX} 
redefinable for python ports, so we can gradually switch all this ports 
to xz. Thanks Marcus and Li-Wen.


[1] http://www.python.org/ftp/python/2.6.8/Python-2.6.8.tar.xz

--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Python distribution: switching to tar.xz

2012-10-01 Thread Ruslan Mahmatkhanov

Li-Wen Hsu wrote on 01.10.2012 22:11:

On Mon, Oct 01, 2012 at 22:03:48 +0400, Ruslan Mahmatkhanov wrote:

Li-Wen Hsu wrote on 01.10.2012 21:51:


EXTRACT_SUFX?= .tgz
PYTHON_DISTFILE= Python-${PYTHON_PORTVERSION:S/.rc/rc/}}${EXTRACT_SUFX}


Yes, I also bow to this option.


then we can define USE_XZ= yes in the ports we want to use xz
distfile, which handles
${EXTRACT_SUFX} and ${EXTRACT_DEPENDS} correctly.  After python26 decommission,
we can even move this into  bsd.python.mk


python26 distributes in tar.xz too [1]. Or you meant something other?

Ok, so I get the point. I'll prepare an patch that make ${EXTRACT_SUFX}
redefinable for python ports, so we can gradually switch all this ports
to xz. Thanks Marcus and Li-Wen.

[1] http://www.python.org/ftp/python/2.6.8/Python-2.6.8.tar.xz



Oh I was not aware that, I just did not see XZ information at:

http://python.org/download/releases/2.6.8/

So maybe we can simply put USE_XZ= yes in bsd.python.mk now.


So we returned to initial question :). Is this acceptable to do this, 
because it will require all the distinfos change "w/o a reason"? I 
didn't see anything criminal there, but other people may know better.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: bin/python3 symlink for python3X ports

2012-10-02 Thread Ruslan Mahmatkhanov

Dmitry Sivachenko wrote on 26.09.2012 13:14:

On 26.09.2012 13:07, Ruslan Mahmatkhanov wrote:

Chris Rees wrote on 26.09.2012 13:03:

On 26 Sep 2012 08:13, "Dmitry Sivachenko"  wrote:


Hello!

It is common to have both python-2 and python-3 packages installed.
To help transition it is convenient to have bin/python to point to

python-2 binary.


Now for every X in python3X we have a bin/python3.X binary.

So for every new python3.X release one should change shebang line in

scripts.


I propose to tweak python3X ports so they create bin/python3 symlink

pointing to the installed python3.X binary so that people can always use
'python3' name.


What do you think?


I think it's a great idea, at least minimising weird breakage when
python
is updated.

Have you a patch?  Also, what do other OSes do?




I don't have a patch but I can make one if you like.
I don't use other OSes so I have no idea what do they do with that
problem and I don't see any reason why should we bother...


I just realized there is such a document:
http://www.python.org/dev/peps/pep-0394/

According to it:

python2 will refer to some version of Python 2.x
python3 will refer to some version of Python 3.x
python should refer to the same target as python2 but may refer to 
python3 on some bleeding edge distributions


And it has full details about all the symlinks etc. So it's not just 
linuces. I'd read it before preparing the patch.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Python distribution: switching to tar.xz

2012-10-02 Thread Ruslan Mahmatkhanov

Li-Wen Hsu wrote on 01.10.2012 22:11:

On Mon, Oct 01, 2012 at 22:03:48 +0400, Ruslan Mahmatkhanov wrote:

Li-Wen Hsu wrote on 01.10.2012 21:51:


EXTRACT_SUFX?= .tgz
PYTHON_DISTFILE= Python-${PYTHON_PORTVERSION:S/.rc/rc/}}${EXTRACT_SUFX}


Yes, I also bow to this option.


then we can define USE_XZ= yes in the ports we want to use xz
distfile, which handles
${EXTRACT_SUFX} and ${EXTRACT_DEPENDS} correctly.  After python26 decommission,
we can even move this into  bsd.python.mk


python26 distributes in tar.xz too [1]. Or you meant something other?

Ok, so I get the point. I'll prepare an patch that make ${EXTRACT_SUFX}
redefinable for python ports, so we can gradually switch all this ports
to xz. Thanks Marcus and Li-Wen.

[1] http://www.python.org/ftp/python/2.6.8/Python-2.6.8.tar.xz



Oh I was not aware that, I just did not see XZ information at:

http://python.org/download/releases/2.6.8/

So maybe we can simply put USE_XZ= yes in bsd.python.mk now.


Ok, here is the patch [1]. Unfortunately, we can't just include USE_XZ 
into bsd.python.mk because this will affect *all* the ports, that define 
USE_PYTHON in their Makefile's. I can't think a way how to (nicely) 
apply USE_XZ to limited set of ports inside of bsd.python.mk. Using 
tar.xz instead of .tgz in ${PYTHON_DISTFILE} will not work on systems, 
that lacks xz in a base system and I dislike the duplication of USE_XZ 
internals in bsd.python.mk for them.


So this patch just replaces .tgz with ${EXTRACT_SUFX} and USE_XZ is used 
in individual python ports. I did some test builds of python and it's 
slaves (py-sqlite3, py-bdsdb, py-gdbm, py-tkinter) and everything seems 
fine. I also diffed all the four pairs tgz/tar.xz of each python version 
- and, as it was expected, their content is equal.


Any objections, suggestions etc?

[1] http://people.freebsd.org/~rm/tgz-to-tx-conversion.diff.txt

--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Python distribution: switching to tar.xz

2012-10-02 Thread Ruslan Mahmatkhanov

Chris Rees wrote on 02.10.2012 22:12:

On 2 Oct 2012 18:55, "Ruslan Mahmatkhanov"  wrote:


[...]


Any objections, suggestions etc?

[1] http://people.freebsd.org/~rm/tgz-to-tx-conversion.diff.txt


My suggestion is to do it :)

Thanks for doing the work.

Chris


Just committed.

--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: py-distribute update leftover

2012-10-03 Thread Ruslan Mahmatkhanov

Hello,

Volodymyr Kostyrko wrote on 03.10.2012 15:20:

Hello.

Recent py-ditribute update doesn't update easy_install.pth on python2.7:

===>>> The following actions were performed:
 Re-installation of py27-distribute-0.6.28

# cd /usr/local/lib/python2.7/site-packages/
# cat easy-install.pth
import sys; sys.__plen = len(sys.path)
./distribute-0.6.27-py2.7.egg
./Pyrseas-0.5.0-py2.7.egg
./Babel-0.9.6-py2.7.egg
./Genshi-0.6-py2.7.egg
./Pygments-1.5-py2.7.egg
./Trac-0.12.3-py2.7.egg
./hg_git-0.3.3-py2.7.egg
./graphviz-0.7.4-py2.7.egg
./TracTocMacro-11.0.0.3-py2.7.egg
./pytz-2012f-py2.7.egg
import sys; new=sys.path[sys.__plen:]; del sys.path[sys.__plen:];
p=getattr(sys,'__egginsert',0); sys.path[p:p]=new; sys.__egginsert =
p+len(new)

And a new file was created:

# cat easy-install.pth.dist
import sys; sys.__plen = len(sys.path)
./distribute-0.6.28-py2.7.egg
import sys; new=sys.path[sys.__plen:]; del sys.path[sys.__plen:];
p=getattr(sys,'__egginsert',0); sys.path[p:p]=new; sys.__egginsert =
p+len(new)


It's in pkg-plist logic:
"""
%%PYTHON_SITELIBDIR%%/easy-install.pth.dist
@exec if [ ! -f %B/easy-install.pth ]; then cp %B/%f %B/easy-install.pth; fi
"""

easy-install.pth.dist is always installed, and it will be only copied to 
easy-install.pth if there is no one already.


What about easy-install.pth contents is not updated - I need an testing 
environment. Would you describe the steps you did to stick with 
distribute instead of setuptools and I'll try to reproduce?


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: py-distribute update leftover

2012-10-03 Thread Ruslan Mahmatkhanov

Volodymyr Kostyrko wrote on 03.10.2012 17:01:

03.10.2012 15:11, Ruslan Mahmatkhanov wrote:

Recent py-ditribute update doesn't update easy_install.pth on python2.7:



[...]



PS: I'm also using unegg patch.


Ok, thanks for detailed report. I'll try it on my system at weekend.


PPS: This is the only bad thing in switching to py-distribute -
everything else works: trac, viewvc, others...


The main problem as for me is that this two systems conflicting. And we 
have some amount of ports that need to be fixed before trying to ever do 
distribute exp-runs - the ones that depend on setuptools directly, not 
via USE_PYDISTUTILS (most of trac ports for example).


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: py-distribute update leftover

2012-10-03 Thread Ruslan Mahmatkhanov

Volodymyr Kostyrko wrote on 03.10.2012 17:30:

03.10.2012 16:19, Ruslan Mahmatkhanov wrote:

PPS: This is the only bad thing in switching to py-distribute -
everything else works: trac, viewvc, others...


The main problem as for me is that this two systems conflicting. And we
have some amount of ports that need to be fixed before trying to ever do
distribute exp-runs - the ones that depend on setuptools directly, not
via USE_PYDISTUTILS (most of trac ports for example).


Hmm, if I will come up with a patch would that count?

I'm already running patched trac locally and everything I had done was
just unhardcoding setuptools dependency.


Sure, please do, I started it couple of months ago, but never finished.

--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/172657

2012-10-16 Thread Ruslan Mahmatkhanov
The following reply was made to PR ports/172657; it has been noted by GNATS.

From: Ruslan Mahmatkhanov 
To: bug-follo...@freebsd.org
Cc:  
Subject: Re: ports/172657
Date: Tue, 16 Oct 2012 16:43:26 +0400

 Thanks, but it should be USE_TWISTED_RUN=yes however.
 Chris, would you approve?
 
 -- 
 Regards,
 Ruslan
 
 Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Building math/py-pandas against py3k (WAS: ports/172528: [maintainer update] [patch] Update math/py-pandas to 0.9.0)

2012-10-17 Thread Ruslan Mahmatkhanov

Hi John,

John W. O'Brien wrote on 17.10.2012 18:17:

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 10/12/2012 08:17 AM, John W. O'Brien wrote:

On 10/11/2012 03:55 PM, Ruslan Mahmatkhanov wrote:

I delayed the response becaused I decide to check the build with
  python32. It's fore sure that py-pandas pkg-plist should be
patched to make packaging correct, but unfortunately the build is
stuck on some of it's dependencies (py-numpy, maybe) so I wasn't
able to test it.


I appreciate your attempt. I'll see what I can do to probe that
issue this weekend. It will be the first time I try building
everything against python 3, but I've got a fast machine and a load
of disk space.


Ruslan (and freebsd-python@),

It looks to me like there are three, maybe four other ports that need
attention before we can successfully build and test math/py-pandas
against py3k.

devel/py-dateutil is at 1.5 which is needed for Python < 3, but Python

= 3 requires dateutil 2.0.


devel/py-pytz will build, install, and package cleanly, but...


[...]


devel/py-nose doesn't package cleanly because of plist problems (looks
like mainly lingering *.pyo and *.pyc entries). I also tried 1.2.1
using the patch from ports/172161, but the plist is still not
py3k-friendly.

math/py-numpy also has plist problems, and fails 14 of 3556 unit tests
(results available on request).

In summary, devel/py-dateutil is the big blocker. math/py-pandas,
devel/py-nose, and math/py-numpy need plist patches. devel/py-pytz and
math/py-numpy have runtime bugs that need further analysis.

As time permits, I will learn the plist magic that makes ports
py3k-friendly, and secondarily poke at the dateutils upgrade. I'm
probably out of my depth, but I'm happy to take direction and correction.

Cheers,
John


Thank you for the investigation. Yes, sadly dateutil upstream developing 
two separate branches for 2.x/3.x, so it will require some bsd.python.mk 
tweaks (like setuptools vs distribute) so the correct dependency picked 
according to python version used.


pytz needs an bug report on launchpad.

nose and numpy issues are solvable, and I will look at it in near time.

--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/172851: [PATCH] sysutils/py-supervisor: update to 3.0b1 + OPTIONS + more

2012-10-18 Thread Ruslan Mahmatkhanov
The following reply was made to PR ports/172851; it has been noted by GNATS.

From: Ruslan Mahmatkhanov 
To: Kubilay Kocak 
Cc: freebsd-gnats-sub...@freebsd.org, hi...@vyborg.ru
Subject: Re: ports/172851: [PATCH] sysutils/py-supervisor: update to 3.0b1
 + OPTIONS + more
Date: Thu, 18 Oct 2012 16:12:24 +0400

 Hi koobs!
 
 Kubilay Kocak wrote on 18.10.2012 16:05:
 >
 >> Number: 172851
 >> Category:   ports
 >> Synopsis:   [PATCH] sysutils/py-supervisor: update to 3.0b1 + OPTIONS + 
 >> more
 
 Please note that most of this changes is already committed today, sorry 
 :). Please make the patch against current version in the tree.
 
 -- 
 Regards,
 Ruslan
 
 Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: FreeBSD Port: python32-3.2.3_1

2012-10-29 Thread Ruslan Mahmatkhanov

Hi,

Finn Espen Gundersen wrote on 29.10.2012 13:36:

Hi,



Where can i subscribe or look to get notified when Python 3.3 is ported to
FreeBSD?


Please see there:
http://lists.freebsd.org/pipermail/freebsd-python/2012-October/004688.html

I hope this will be added after 9.1-R.

--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/173501: irc/py-irclib build failure on amd64

2012-11-09 Thread Ruslan Mahmatkhanov

Bryan Drewery wrote on 09.11.2012 21:40:

The following reply was made to PR ports/173501; it has been noted by GNATS.

From: Bryan Drewery 
To: bug-follo...@freebsd.org, willia...@gmail.com
Cc:
Subject: Re: ports/173501: irc/py-irclib build failure on amd64
Date: Fri, 09 Nov 2012 11:30:58 -0600

  I think the problem is that it is trying to use the network during the
  "configure" phase. It should only be using the network during the
  "fetch" phase.

  Bryan


Sure, but according to this [1] change, it shouldn't ever try to fetch 
anything by itself.


[1] 
http://www.freebsd.org/cgi/cvsweb.cgi/ports/Mk/bsd.python.mk.diff?r1=1.141;r2=1.142


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/173501: irc/py-irclib build failure on amd64

2012-11-13 Thread Ruslan Mahmatkhanov

William Grzybowski wrote on 13.11.2012 16:07:

FYI the PRs I opened are ports/173506 and ports/173507.

Thanks rm.


Yes, I taken them both along with original 173501 pointyhat build 
failure report. Will check them and commit shortly.



On Tue, Nov 13, 2012 at 10:03 AM,  wrote:


Synopsis: irc/py-irclib build failure on amd64

Responsible-Changed-From-To: freebsd-python->rm
Responsible-Changed-By: rm
Responsible-Changed-When: Tue Nov 13 12:03:28 UTC 2012
Responsible-Changed-Why:
I will take it.

http://www.freebsd.org/cgi/query-pr.cgi?pr=173501


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/173501: irc/py-irclib build failure on amd64

2012-11-13 Thread Ruslan Mahmatkhanov

William Grzybowski wrote on 13.11.2012 16:07:

FYI the PRs I opened are ports/173506 and ports/173507.

Thanks rm.


Ok, done. To be honest I think it will be better to update irclib to 
latest version (3.6):

http://pypi.python.org/pypi/irc

Author changed his mind and proposing regular setuptools install along 
with paver install:


"""
 You have several options to install the IRC project.

  * Use "easy_install irc" or "pip install irc" to grab the latest
version from the cheeseshop (recommended).
  * Run "python setup.py install" (from the source distribution) or
  * Run "paver install" (from repo checkout, requires paver) or
  * Copy irc directory to appropriate site-packages directory.
"""

After update this port can be converted to old good 
USE_DISTUTILS=easy_install



On Tue, Nov 13, 2012 at 10:03 AM,  wrote:


Synopsis: irc/py-irclib build failure on amd64

Responsible-Changed-From-To: freebsd-python->rm
Responsible-Changed-By: rm
Responsible-Changed-When: Tue Nov 13 12:03:28 UTC 2012
Responsible-Changed-Why:
I will take it.

http://www.freebsd.org/cgi/query-pr.cgi?pr=173501

--
Regards,
Ruslan

Tinderboxing kills... the drives.

___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/173828: irc/py-irclib: update to 5.0.1

2012-11-22 Thread Ruslan Mahmatkhanov

William Grzybowski wrote on 22.11.2012 14:00:


  Hi,

  I didn't know there was a new upstream, thats nice. Patch approved on my
  side, thanks!


Just done, thanks.

Yes, please see the reference on old offsite: 
http://python-irclib.sourceforge.net/


I also pointed that out after committing your build fix (by introducing 
py-hgtools dependency). This port now may be installed both with paver 
and setuptools (that we prefer) and the port no more need to depend on 
devel/py-setuptools directly.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Postgresql + Python

2012-11-27 Thread Ruslan Mahmatkhanov

Rafael Espíndola wrote on 27.11.2012 16:17:

Hi,

I just solve as follows, first of all I installed Python32, then changed the 
Makefile Postgresql91 follows:

CONFIGURE_ARGS + = - with-libraries = $ {PREFIX} / lib \
  - with-includes = $ {PREFIX} / include \
  - enable-thread-safety \
  - with-python

Added the - with-python, worked perfectly, installed the library "plpython3u", 
I appreciate your interest in helping, this is very good. A hug!


Strange enough, slave's Makefile has ``CONFIGURE_ARGS=	--with-python'' 
and it worked fine earlier in my tests.


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: Postgresql + Python

2012-11-27 Thread Ruslan Mahmatkhanov

Rafael Espíndola wrote on 27.11.2012 16:29:

Well, if I'm not mistaken had tested "CONFIGURE_ARGS = - with-python'', but had 
no success, but I could be mistaken. Anyway I did what I told you earlier and it 
worked perfectly, and this I care ehheheh, I was afflicted with it.

I come once again thank the interest in helping, it is important for the 
FreeBSD project. I'm Brazilian, sorry for my English, or maybe I should say 
sorry for the English Google :)


Just checked the build in my test environment (FreeBSD 9.0 i386, python 
3.2.3, PYTHON_DEFAULT_VERSION=python3.2 in /etc/make.conf) and the build 
is indeed failed, but this way:


cc -O2 -pipe -fno-strict-aliasing -Wall -Wmissing-prototypes 
-Wpointer-arith -Wdeclaration-after-statement -Wendif-labels 
-fno-strict-aliasing -fwrapv -fPIC -DPIC -I. 
-I/usr/local/include/python3.2mu -I../../../src/include 
-I/usr/local/include -I/usr/local/include  -c -o plpython.o plpython.c

plpython.c: In function 'PLy_procedure_call':
plpython.c:1170: warning: passing argument 1 of 'PyEval_EvalCode' from 
incompatible pointer type

plpython.c:1187: warning: null format string
plpython.c: In function 'PLy_procedure_get':
plpython.c:1321: warning: implicit declaration of function 'PyCObject_Check'
plpython.c:1324: warning: implicit declaration of function 
'PyCObject_AsVoidPtr'
plpython.c:1324: warning: assignment makes pointer from integer without 
a cast

plpython.c: In function 'PLy_procedure_create':
plpython.c:1553: warning: implicit declaration of function 
'PyCObject_FromVoidPtr'
plpython.c:1553: warning: assignment makes pointer from integer without 
a cast

plpython.c: In function 'PLy_spi_prepare':
plpython.c:2925: warning: null format string
plpython.c: In function 'PLy_spi_execute_plan':
plpython.c:3076: warning: null format string
plpython.c: In function 'PLy_spi_execute_query':
plpython.c:3122: warning: null format string
cc -O2 -pipe -fno-strict-aliasing -Wall -Wmissing-prototypes 
-Wpointer-arith -Wdeclaration-after-statement -Wendif-labels 
-fno-strict-aliasing -fwrapv -fPIC -DPIC -shared -o plpython3.so 
plpython.o -L../../../src/port -L/usr/local/lib  -L/usr/local/lib 
-Wl,--as-needed -Wl,-R'/usr/local/lib/python3.2/config' 
-L/usr/local/lib/python3.2/config -lpython3.2 -lutil -lm
/usr/bin/ld: --just-symbols may not be used on DSO: 
/usr/local/lib/python3.2/config

gmake: *** [plpython3.so] Error 1
*** Error code 2

It worked before, and I'm not sure how to fix this one. Python2 build 
works as expected. Any ideas? (Mostly asking Palle and pgsql@)


--
Regards,
Ruslan

Tinderboxing kills... the drives.
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


Re: ports/141133

2010-03-10 Thread Ruslan Mahmatkhanov
The following reply was made to PR ports/141133; it has been noted by GNATS.

From: Ruslan Mahmatkhanov 
To: bug-follo...@freebsd.org
Cc:  
Subject: Re: ports/141133
Date: Wed, 10 Mar 2010 18:42:58 +0300

 Hello.
 I want to maintain this port.
 
 -- 
 Regards,
 Ruslan
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"


How to depend on package in EGG?

2010-04-21 Thread Ruslan Mahmatkhanov

Good day!

I want add option in port's Makefile to depend on textproc/py-enchant.
But how i can do it? It installs as python egg.

.if defined(WITH_ENCHANT)
RUN_DEPENDS+=${LOCALBASE}//enchant/__init__.py:${PORTSDIR}/textproc/py-enchant
.endif

What should i wrote instead  in left part of this RUN_DEPENDS string?

Thanks in advance for answer.

--
Regards,
Ruslan
___
freebsd-python@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-python
To unsubscribe, send any mail to "freebsd-python-unsubscr...@freebsd.org"