Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 27 Mar 2018 15:16:47 -0700
From:      Pete Wright <pete@nomadlogic.org>
To:        Guido Falsi <madpilot@FreeBSD.org>, "D.-C. M." <my-roaming-data@outlook.com>, "kde@FreeBSD.org" <kde@FreeBSD.org>
Cc:        "ports@FreeBSD.org" <ports@FreeBSD.org>
Subject:   Re: FreeBSD Port: py27-qt5-core / Py36-qt5-core
Message-ID:  <67cf2069-85d7-7531-6177-e4d258009df9@nomadlogic.org>
In-Reply-To: <71bf65f9-20ad-a30c-0fdd-bc78b31e666c@FreeBSD.org>
References:  <AM5PR0901MB1139637F840990FAB019C890A9AC0@AM5PR0901MB1139.eurprd09.prod.outlook.com> <c600a76b-b01c-6a65-c0c4-ecb2bd7ff105@FreeBSD.org> <8b5a9d2d-3373-f164-9a1d-e3acf19e1ec9@nomadlogic.org> <71bf65f9-20ad-a30c-0fdd-bc78b31e666c@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help


On 03/27/2018 15:06, Guido Falsi wrote:
> On 03/28/18 00:00, Pete Wright wrote:
>>> I'm not a python expert, but I understand that python 2.7 and python 3
>>> are two slightly different languages not fully compatible with each
>>> other.
>>>
>>> I also understand(but have not gone into depth about this) that there is
>>> some resistance to python 3, with many developers being reluctant to
>>> move to version 3, for whatever reason(I imagine it's language design
>>> choices, but I really don't know)
>>>
>>> I'm stating this because it means such incompatibilities are not going
>>> away easily. It's not just a ports system problem, but an actual python
>>> ecosystem problem.
>>>
>>> Too say it in other words, python 2.7 isn't really just "the old
>>> version" and python 3 is not just "the new version". They have parallel
>>> lifes.
>> I'm not %100 sure that's really an accurate assessment of the slow
>> uptake in Python3.
> I'd like to make it clear I don't know the details, I just stated what I
> heard. I know this could not be accurate.
sorry - that came out wrong - i wasn't trying to be combative!  i'm in 
the same boat as you here :)

>
>> Regardless, the clock is ticking on the 2.x codebase
>> as it is reaching EOL status in 2020:
>>
>> https://www.python.org/dev/peps/pep-0373/
>>
>> Hopefully a solid deadline (which has already been pushed back) will
>> motivate developers to accelerate the task of migrating to py3 sooner
>> rather than later.
> Speaking strictly as the maintainer of the calibre port and having
> discovered just now about this deadline:
>
> I don't know what the calibre developer plans to do about this, I'm
> certainly unable to port calibre to python 3, so I will do the best to
> keep it working for as long as python 2.7 is available in the ports, or
> update the port to use python 3 once the upstream does port it to that
> version.
>
this is a really tricky situation to be in no doubt, i wonder if 
surfacing concerns about the impending 2.x EOL with upstream maintainers 
would be a good way to nudge them into supporting py3? it's certainly 
possible that the deadline in pep-373 hasn't been widely disseminated to 
the developer community?

i'm not super active in the python community to be honest - but in my 
role as a systems engineer this is something i've highlighted with teams 
whose code i help support and have had mixed success with.  usually 
along the lines of "hey, so py2.7 is EOL'ing in 2020 do we have a 
document with our migration strategy?"

cheers,
-pete


-- 
Pete Wright
pete@nomadlogic.org
@nomadlogicLA




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?67cf2069-85d7-7531-6177-e4d258009df9>