Sorry, there is a typo in my previous email...
It should be
$ python setup.py sdist bdist_wheel
for more information on this please check this example:
https://packaging.python.org/tutorials/packaging-projects/
On Fri, Jan 25, 2019 at 12:12 PM Goncalo Morgado <goncalo.ma...@gmail.com>
wrote:
> Hey William,
>
> Thanks, I will take a look.
>
> But please note that in order to have your dist package in PyPI you have
> to generate an archive:
> $ python sdist bdist_wheel # this generates both software dist and
> binary dist packages
> and push it with twine:
> $ twine upload --repository-url https://test.pypi.org/legacy/ dist/*
> # test.pypi is for testing purposes, but you can use the official when
> you're happy
>
> I haven't looked at your code, but I wonder if you're following these
> steps.
>
> (I hope we are not going off-topic)
>
> Cheers,
> Gonçalo
>
>
>
>
>
>
> On Fri, Jan 25, 2019 at 12:02 PM William Smith <willpjsm...@gmail.com>
> wrote:
>
>> Hi Gonçalo,
>>
>> Not at all, thanks for your interest!
>> Maybe I can provide that example - here's the link the CellModeller code
>> on GitHub. <https://github.com/HaseloffLab/CellModeller/>
>> Here you can find the setup.py script, which does indeed use the
>> setuptools package.
>> However, I couldn't find CellModeller listed on the PyPI repo
>> <https://pypi.org/>?
>>
>> What do you think?
>>
>> W
>>
>> Dr. William P J Smith
>> _____________________________________________
>> Postdoctoral researcher, Foster lab, University of Oxford
>>
>>
>> On Fri, Jan 25, 2019 at 12:46 PM Goncalo Morgado <goncalo.ma...@gmail.com>
>> wrote:
>>
>>>
>>> Apologies for jumping in, do you have an example of a project (link to
>>> repo) that depends on pyopencl and that is distributed through PyPI (uses
>>> setuptools/setup.py), that I could look at?
>>> Having to manually install numpy, pybind11 and mako is a bit of a
>>> problem. With pipenv Pipfile.lock I am able to install everything
>>> elegantly, but I have the problem of putting it in a setup.py...
>>> That would help me a lot.
>>>
>>> Many thanks,
>>> Gonçalo
>>>
>>>
>>>
>>> On Fri, Jan 25, 2019 at 9:27 AM Jerome Kieffer <jerome.kief...@esrf.fr>
>>> wrote:
>>>
>>>> On Fri, 25 Jan 2019 10:00:45 +0000
>>>> William Smith <willpjsm...@gmail.com> wrote:
>>>>
>>>> > My colleagues and I use PyOpenCL as part of an cell-based modelling
>>>> program
>>>> > (Tim Rudge's CellModeller, here <
>>>> http://haselofflab.github.io/CellModeller/>
>>>> > ).
>>>> >
>>>> > We would like to use CellModeller alongside a PDE Solver (FEniCS
>>>> > <https://fenicsproject.org/>), whose NumPy requirements (>=1.13)
>>>> conflict
>>>> > with those of PyOpenCL (1.9). Is it possible to make PyOpenCL
>>>> compatible
>>>> > with later NumPy versions?
>>>> >
>>>> > Here's the conflict in my conda environment:
>>>> >
>>>>
>>>> I guess the conflict is related to the conda packaging (poorly done?).
>>>>
>>>> We have no issue to deploy continuous integration of pyopencl 2018.2.2
>>>> with numpy 1.16:
>>>>
>>>> https://travis-ci.org/silx-kit/pyFAI/jobs/484293593
>>>>
>>>> As mentioned in a recent thread, numpy, pybind11 and mako have to be
>>>> installed beforehand.
>>>>
>>>> Cheers,
>>>> --
>>>> Jérôme Kieffer
>>>> _______________________________________________
>>>> PyOpenCL mailing list -- pyopencl@tiker.net
>>>> To unsubscribe send an email to pyopencl-le...@tiker.net
>>>>
>>> _______________________________________________
>>> PyOpenCL mailing list -- pyopencl@tiker.net
>>> To unsubscribe send an email to pyopencl-le...@tiker.net
>>>
>>
_______________________________________________
PyOpenCL mailing list -- pyopencl@tiker.net
To unsubscribe send an email to pyopencl-le...@tiker.net
