Re: pyxs review

2012-12-10 Thread Jakub Wilk 

* Maykel Moya , 2012-11-09, 13:44:
I'll take care of the issues noted here, including contacting upstream 
regarding the blob and the inclusion of GPL-3 text. Thanks for the 
review.


It's been over a month. Any news?

--
Jakub Wilk


--
To UNSUBSCRIBE, email to debian-python-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20121210155754.ga1...@jwilk.net

Re: Problem with distutils and the experimental multi-arch Python2.7 packages

2012-12-10 Thread Jakub Wilk 

* Yannick Roehlly , 2012-12-09, 17:08:
While trying the experimental Python 2.7 package with multiarch 
support, I found that distutils is not working.


When I try to build some package, I have this error:

error: invalid Python installation: unable to open
/usr/lib/python2.7/config/Makefile (No such file of directory)


This looks like bug #695596.

--
Jakub Wilk


--
To UNSUBSCRIBE, email to debian-python-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20121210172342.gb2...@jwilk.net

Re: Third round of advise on packaging python-csb

2012-12-10 Thread Jakub Wilk 

* Tomás Di Domenico , 2012-12-09, 22:27:
I have also tried my hand at generating a Python3 binary package, which 
seems to be working. I would appreciate if someone could take a look at 
the debian/rules file and tell me how to improve it. I'm sure my very 
simple version is not doing all that can/should be done.


Two comments:


PYVERS :=  $(shell pyversions -r)
PY3VERS := $(shell py3versions -r)


Using simple "=" instead of ":=" should be cheaper.


.PHONY: override_dh_auto_clean
override_dh_auto_clean:
  set -e; \
  for py in $(PYVERS) $(PY3VERS); do \
  $$py -B setup.py clean; \
  rm -rf build; \
  done


"rm -rf build" inside the loop looks weird... Shouldn't it be outside 
it? Anyway, if you called "setup.py clean" with the -a option, then 
removing build shouldn't be needed.


--
Jakub Wilk


--
To UNSUBSCRIBE, email to debian-python-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20121210221058.ga3...@jwilk.net

Re: Problem with distutils and the experimental multi-arch Python2.7 packages

2012-12-10 Thread Yannick Roehlly 
Jakub Wilk wrote:

> This looks like bug #695596.

Yes. Thanks for correcting this.

Yannick

-- 
Between infinite and short there is a big difference.
-- G. H. Gonnet



-- 
To UNSUBSCRIBE, email to debian-python-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/ka5nc9$v3p$1...@ger.gmane.org

Re: Third round of advise on packaging python-csb

2012-12-10 Thread Bradley M. Froehle 
Interesting, the LibraryStyleGuide [1] suggests a plain `=` and the
AppStyleGuide [2] suggests `:=`.  (The difference of course is that `=`
does delayed evaluation meaning the command is run once for every time the
variable is needed, and `:=` does immediate evaluation meaning the command
is run only once when the variable is set.

Naively I would have always reached for `:=`, but I think you are correct
that `=` would be slightly cheaper in that the variable is not used every
time it is processed by make.

[1]: http://wiki.debian.org/Python/LibraryStyleGuide
[2]: http://wiki.debian.org/Python/AppStyleGuide

Two comments:
>
>  PYVERS :=  $(shell pyversions -r)
>> PY3VERS := $(shell py3versions -r)
>>
>
> Using simple "=" instead of ":=" should be cheaper.
>

Re: Third round of advise on packaging python-csb

2012-12-10 Thread Barry Warsaw 
On Dec 10, 2012, at 02:33 PM, Bradley M. Froehle wrote:

>Interesting, the LibraryStyleGuide [1] suggests a plain `=` and the
>AppStyleGuide [2] suggests `:=`.  (The difference of course is that `=`
>does delayed evaluation meaning the command is run once for every time the
>variable is needed, and `:=` does immediate evaluation meaning the command
>is run only once when the variable is set.

I'm nearly positive not much thought went into those choices - they were
probably cargo-culted along the way.  I don't have a strong opinion about what
should be used since I doubt the optimization would be noticed, but let's do
at least make them consistent.

Cheers,
-Barry


-- 
To UNSUBSCRIBE, email to debian-python-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20121210180605.31073...@limelight.wooz.org