Thanks for the tip, pje. Your suggestion was very helpful. Perhaps I can ask
a follow-up.
In addition to calculating the tag at package-time (when setup.py is run for
sdist/bdist*), I am calculating the version at package-time.
The problem I have with this approach is with source distributions (s
On Sat, Jul 31, 2010 at 05:38:01PM +0100, Chris Withers wrote:
> Tres Seaver wrote:
> >By never, ever, ever distributing binary eggs, especially for
> >Linux.
>
> That's fine in a world without python packages with C and Fortran
> extensions that are a royal PITA to compile. (fragile much?!)
In m
At 12:39 PM 7/31/2010 -0700, Jason R. Coombs wrote:
I'm using setuptools 0.6c11.
I'd like to programmatically customize the way the egg_info command
is run. That is, in my setup.py, I would like to run some functions
to determine the tag-build, tag-date, and tag-svn-revision
parameters to egg
I'm using setuptools 0.6c11.
I'd like to programmatically customize the way the egg_info command is run.
That is, in my setup.py, I would like to run some functions to determine the
tag-build, tag-date, and tag-svn-revision parameters to egg_info. Are there
parameters
How is this task best
On Sun, Aug 1, 2010 at 2:28 AM, Éric Araujo wrote:
>> The static linking is almost always a bad idea, and mostly because of
>> distutils warts.
>
> Even if you’re happily working on another tool, it would be very
> appreciated if you could list those distutils issues in an email or
> better on bug
> The static linking is almost always a bad idea, and mostly because of
> distutils warts.
Even if you’re happily working on another tool, it would be very
appreciated if you could list those distutils issues in an email or
better on bugs.python.org.
Regards
_
On Sun, Aug 1, 2010 at 1:34 AM, Chris Withers wrote:
> David Cournapeau wrote:
>>
>> On Tue, Jul 27, 2010 at 5:36 PM, Chris Withers
>> wrote:
>>
>>> Indeed, but the other option requires a more complicated service to
>>> query.
>>> Being able to "serve" packages from a simple folder or from simpl
Barry Warsaw wrote:
Right, and IME, admins of those locked down systems aren't about to install
*anything* from the Cheeseshop anyway. :) They really really prefer
distribution packages (well, on platforms that have a packaging system ;).
The specific use case I have would involve an internal
Tres Seaver wrote:
GCC version shouldn't matter. It is barely possible that libc versions
might.
Practical experience on trying to migrate from Ubuntu 8.04 to 10.04
suggests both matter with the common scipy stack.
By never, ever, ever distributing binary eggs, especially for Linux.
Tha
David Cournapeau wrote:
On Tue, Jul 27, 2010 at 5:36 PM, Chris Withers wrote:
Indeed, but the other option requires a more complicated service to query.
Being able to "serve" packages from a simple folder or from simple folder
served via svn or Apache is a huge win.
I don't see how one precl
10 matches
Mail list logo