[Debian-science-sagemath] [RFP]: primecount -- count the primes below an integer

[Tobias Hansen]

On 1/31/22 12:35, Vincent Delecroix wrote:
> Le 31/01/2022 à 13:31, Dima Pasechnik a écrit :
>> On Mon, Jan 31, 2022 at 10:17 AM Tobias Hansen <thansen at debian.org> wrote:
>>>
>>> On 1/28/22 10:32, Dima Pasechnik wrote:
>>>> I'm the main author of primecountpy, let me know if you have problems with it.
>>>>
>>>> We're able to build binary wheels for it, see
>>>> https://pypi.org/project/primecountpy/
>>>> (this is done using PyPI's cibuildwheel using GitHub Actions:
>>>> https://github.com/dimpase/primecountpy/blob/master/.github/workflows/wheels.yml)
>>>>
>>>> Cheers
>>>> Dima
>>>>
>>>
>>> Thank you. Jerome uploaded primecount (it's in new) and I already did a successful test build of primecountpy and sagemath 9.5.
>>>
>>> Is there a reason why you didn't call the python module primecount? (Of course I understand why the project is called primecountpy.)
>>> The "import primecountpy as primecount" in the readme suggests that's the module name you really wanted.
>>
>> "primecount" was taken on PyPI - it's an interface to primecount
>> library, lacking features etc. in many ways, so I instead created
>> primecountpy (well, re-using the cython code from Sage).
>
> I second Tobias. It is much nicer to write in python
>
>     sage: import primecount
>
> rather than
>
>     sage: import primecountpy
>
> This is what is done for example with pplpy (name of the package)
> which is imported via
>
>     sage: import ppl
>
> This has nothing to do with the package name on PyPI.
>
> Vincent

But if another python module already uses the name, it's better not to hijack the name to make it clear which module is meant with an import.

Another small comment: The license is stated as GPL-2+ in primecount.pyx but the LICENSE file contains the GPL-3. It would be nice to make this more clear/consistent in the future.

Best,

Tobias