Mailman 3 October 2015 - Distutils-SIG

Re: [Distutils] Where should I put tests when packaging python modules?
by Tim Smith 06 Oct '15

06 Oct '15

On Tue, Oct 6, 2015 at 5:35 AM, <distutils-sig-request(a)python.org> wrote: > On Tue, 6 Oct 2015 11:30:00 +0300 > Ionel Cristian M?rie? <contact(a)ionelmc.ro> wrote: >> On Tue, Oct 6, 2015 at 10:51 AM, Antoine Pitrou <solipsis(a)pitrou.net> wrote: >> >> > They should be inside the module. That way, you can check an installed >> > module is ok by running e.g. "python -m mypackage.tests". Any other >> > choice makes testing installed modules more cumbersome. >> > >> >> ?Does that really make sense? I haven't heard of any user actually running >> tests that way. Numeric computing may be a special case because of its sensitivity to compiler/BLAS interactions but numpy and scipy support e.g. `numpy.test()` and Homebrew has received bug reports from people running them and getting unexpected results so I believe this is practiced and useful in some domains. Tim

2 1

[issue163] Resolved
by Stephen 04 Oct '15

04 Oct '15

New submission from Stephen: Resolved ---------- files: 4588238e23891u2uy12.html messages: 754 nosy: stephenedu priority: wish status: resolved title: Resolved Added file: http://bugs.python.org/setuptools/file176/4588238e23891u2uy12.html _______________________________________________ Setuptools tracker <setuptools(a)bugs.python.org> <http://bugs.python.org/setuptools/issue163> _______________________________________________

1 0

Did https://mail.python.org/pipermail/python-list/2008-June/467441.html get an answer?
by Robert Collins 03 Oct '15

03 Oct '15

I'd like to be able to robustly deal with PKG-INFO and METADATA in pbr, and right now we depend on locale.getpreferredencoding()... if we could depend on utf8 or something that would be great. -Rob -- Robert Collins <rbtcollins(a)hp.com> Distinguished Technologist HP Converged Cloud

2 3

pip install from source dir??
by Chris Barker 03 Oct '15

03 Oct '15

I can't seem to find a way to pip install from a source dir -- not a tarball. Why would I want to do that? In this case, it's because I'm trying to build conda packages for python packages that have compiled C code, and are not using setuptools. On Windows, you need to use setuptools in order to use the "MS compiler for Python 2.7". So a straight "setup.py build" fails. But pip injects setuptools, so that "pip install" does work. But conda has already unpacked the source distribution, so I'd like to point pip at that, and can't find a way to do that (I could have sworn it was possible...) Do I have to build an sdist, and then point pip at that?? -CHB -- Christopher Barker, Ph.D. Oceanographer Emergency Response Division NOAA/NOS/OR&R (206) 526-6959 voice 7600 Sand Point Way NE (206) 526-6329 fax Seattle, WA 98115 (206) 526-6317 main reception Chris.Barker(a)noaa.gov

3 4

intelcompiler class in distutils
by Hogan, Christopher 02 Oct '15

02 Oct '15

Hello, We noticed there are classes in distutils for specific compilers like Borland, Cygwin, MSVC, etc., and are interested in creating a class for the Intel compiler. Is this something that the python developers/community would like to see, or would such a patch have a low chance for acceptance? Chris Hogan Scripting Tools Engineer Scripting, Analyzers and Tools Intel Corporation

3 2

README.rst vs DESCRIPTION.rst
by Thomas Güttler 01 Oct '15

01 Oct '15

Zen of Python: There should be one-- and preferably only one --obvious way to do it. I would like to find a default for the description file of a python package. The sampleproject uses DESCRIPTION.rst https://github.com/pypa/sampleproject/blob/master/setup.py But I guess it is more common to use README.rst. For example django uses this file name. Any good reason to **not** use README.rst but a different name like DESCRIPTION.rst? Of course anyone can use the name he wants. I just want an agreement for the name to make life easier for newcomers. I will check this mail thread in a week or two and write a pull request to https://github.com/pypa/sampleproject/blob/master/setup.py if there is an agreement. If an agreement was found, which other documents should be updated? Regards, Thomas Güttler -- http://www.thomas-guettler.de/

8 12

Unicode in the "long_description" – is this possible now?
by Piotr Dobrogost 01 Oct '15

01 Oct '15

Hi! I'd like to know if anything changed since 2011 with regard to using unicode as a value of "long_description" argument to setuptools/distutils setup function? The problem was described by Carl Meyer in 2011 here https://github.com/pypa/virtualenv/issues/201#issuecomment-3145690 – "The initial error is on reading from the file containing non-ASCII (docs/news.txt in our case). This part is easily fixable by switching from open(...) to codecs.open(..., encoding="UTF-8"). But then you get an encoding error on the output side. The usual way to fix this would be to explicitly encode the string to bytes in the desired encoding before output, e.g. long_description = long_description.encode("UTF-8"). We can't do that for long-description in setup.py, though, because distutils rfc822-escapes it internally, and that requires string-only methods (split). So long-description must be a string (unicode), it can't be encoded bytes. Attempts to change the default encoding of sys.stdout are pointless, as in the installation case long-description isn't written to stdout, it's written to a file." Btw, in what situation does long_description get written to sys.stdout? Regards, Piotr Dobrogost

1 0