Hi Dan,
thank you very much for your kind hints - quite interesting idea to have
a more detailed look into this direction!
By the way, your response was the very first here, which I consider to
have a constructive notion; at least I did not felt very welcome here by
Chris until yet...
In the last project I developed the frontend on said C# - WinForms basis
(C# and .NET is quite awesome! Most of it meanwhile has even become
crossplatform! By the way, as Spyder also has this "just 3" notion, I
already switched to VSCodium and never regretted that - VSCodium is the
best free IDE for Python as well as C# yet, if you ask me:) and just
parts of the backend with IronPython (which also is nice, although it
just has access to a limited set of libraries). It worked fine, but I do
not like mixing languages if not necessary as I deem that to be a
software design weakness and it naturally comes with some overhead.
Imagine, if another person one day should continue this work, he must be
fluent in Python AND C#, not so easy to find someone free who is on the
market I guess...
Blythooon solves the current issues well, so at the moment, there is no
pressing reason for me to become frantic. But considering the long term,
those thoughts are naturally real. The obvious trend to force people to
switch to Python 3 might lead to people even eliminating the access to
the old packages Blythooon is using. This sword of Damocles is a heavy
burden.
If anybody thinks that is a little too much seeing on the black side,
then they should attentively follow what at the very moment is happening
with the current Qt version...
May I ask, do you have any knowledge or even experience about if resp.
how good Tauthon and Pypy2 works together with Qt 4.8?
From my experience the limitating factor during frontend development is
nearly always the GUI part. Kivy seems to be nice, but scientific
plotters alike PyQtGraph are Qt based and cannot easily be integrated in
Kivy yet.
Have a nice day,
Best Regards
Dominik
On 2021-03-27 07:01, Dan Stromberg wrote:
On Fri, Mar 26, 2021 at 10:37 PM Chris Angelico <ros...@gmail.com>
wrote:
On Sat, Mar 27, 2021 at 4:20 PM <pyt...@blackward.eu> wrote:
> By the way, some months ago I started trying to migrate to Python 3 and
> gave up in favor of creating said compilation. Compatibility of Python
> and its Packages decreased with V3 significantly. A whole lot of minor
> and major incompatibilities between your subversions and belonging
> packages. This was one reason, why Java took the route to its own death.
FUD. Lots and lots of FUD. More reasons to not promote your
distribution. Use it if you will, but it doesn't merit any further
visibility.
Chris, not everything you dislike is anti-Python FUD.
Dominik, if you want something like Python 2.7, you likely should try
Tauthon or Pypy2. Don't expect pip to work well on Tauthon; last I
heard
that was not happening. Also Pypy2 has some issues with C extension
modules, and I'm not confident it'll pip well either. It's very
worthwhile
to move to 3.x, but CPython has a rather sad compatibility story when
it
comes to C extension modules; hopefully CFFI is going to fix that in
the
long term. If you're avoiding porting pure Python code, then that feels
to
me a bit like foot dragging, as the pure Python changes are not that
big
and are pretty much limited to the 2.7 -> 3.0 transition.
I like to build versions of Python from 0.9 to 3.10alpha, for the sake
of
quickly ascertaining what features were introduced in what versions of
CPython. IOW, there are good reasons to keep around old Pythons.
Python
history is interesting.
--
https://mail.python.org/mailman/listinfo/python-list
