I am trying to install numpy library on Python 2.7.15 in PyCharm but the
error message I get is:

ERROR: Could not find a version that satisfies the requirement numpy (from
> versions: none)
> ERROR: No matching distribution found for numpy
> c:\python27\lib\site-packages\pip\_vendor\urllib3\util\ssl_.py:164:
> InsecurePlatformWarning: A true SSLContext object is not available. This
> prevents urllib3 fro
> m configuring SSL appropriately and may cause certain SSL connections to
> fail. You can upgrade to a newer version of Python to solve this. For more
> information, see
> https://urllib3.readthedocs.io/en/latest/advanced-usage.html#ssl-warnings
> InsecurePlatformWarning,

Any clues?

On 12/06/2024 12:30, marc nicole wrote:
> I am trying to install numpy library on Python 2.7.15 in PyCharm but the
> error message I get is:
>
> ERROR: Could not find a version that satisfies the requirement numpy (from
>> versions: none)
>> ERROR: No matching distribution found for numpy
>> c:\python27\lib\site-packages\pip\_vendor\urllib3\util\ssl_.py:164:
>> InsecurePlatformWarning: A true SSLContext object is not available. This
>> prevents urllib3 fro
>> m configuring SSL appropriately and may cause certain SSL connections to
>> fail. You can upgrade to a newer version of Python to solve this. For more
>> information, see
>> https://urllib3.readthedocs.io/en/latest/advanced-usage.html#ssl-warnings
>> InsecurePlatformWarning,
>
>
> Any clues?

Why are you using Python 2? Come on, it's been 16 years. Ya gotta move
on at some point.

I am sure there is inertia to move from an older product and some people
need a reason like this where the old becomes untenable.

It seems Microsoft is having a problem where something lik 2/3 of Windows
users have not upgraded from Windows 10 after many years and have set a
deadline in a year or so for stopping updates. In that case, hardware was a
concern for some as Windows 11 did not work on their machines. With
upgrading python, the main concern is having to get someone to examine old
code and try to make it compatible.

But anyone doing new code in Python 2 in recent years should ...

-----Original Message-----
From: Python-list <python-list-bounces+avi.e.gross=gmail.com@python.org> On
Behalf Of Gordinator via Python-list
Sent: Wednesday, June 12, 2024 10:19 AM
To: python-list@python.org
Subject: Re: Couldn't install numpy on Python 2.7

On 12/06/2024 12:30, marc nicole wrote:
> I am trying to install numpy library on Python 2.7.15 in PyCharm but the
> error message I get is:
>
> ERROR: Could not find a version that satisfies the requirement numpy (from
>> versions: none)
>> ERROR: No matching distribution found for numpy
>> c:\python27\lib\site-packages\pip\_vendor\urllib3\util\ssl_.py:164:
>> InsecurePlatformWarning: A true SSLContext object is not available. This
>> prevents urllib3 fro
>> m configuring SSL appropriately and may cause certain SSL connections to
>> fail. You can upgrade to a newer version of Python to solve this. For
more
>> information, see
>> https://urllib3.readthedocs.io/en/latest/advanced-usage.html#ssl-warnings
>> InsecurePlatformWarning,
>
>
> Any clues?

Why are you using Python 2? Come on, it's been 16 years. Ya gotta move
on at some point.
--
https://mail.python.org/mailman/listinfo/python-list

On 2024-06-12 17:31, AVI GROSS via Python-list wrote:
> I am sure there is inertia to move from an older product and some people
> need a reason like this where the old becomes untenable.
>
> It seems Microsoft is having a problem where something lik 2/3 of Windows
> users have not upgraded from Windows 10 after many years and have set a
> deadline in a year or so for stopping updates. In that case, hardware was a
> concern for some as Windows 11 did not work on their machines. With
> upgrading python, the main concern is having to get someone to examine old
> code and try to make it compatible.
>
In the case of Windows, my PC is over 10 years old yet performs
perfectly well for my needs. It can't run Windows 11. Therefore, I'm in
the process of migrating to Linux, and I still have over a year to
achieve that before support ends.

> But anyone doing new code in Python 2 in recent years should ...
>
Indeed...

> -----Original Message-----
> From: Python-list <python-list-bounces+avi.e.gross=gmail.com@python.org> On
> Behalf Of Gordinator via Python-list
> Sent: Wednesday, June 12, 2024 10:19 AM
> To: python-list@python.org
> Subject: Re: Couldn't install numpy on Python 2.7
>
> On 12/06/2024 12:30, marc nicole wrote:
>> I am trying to install numpy library on Python 2.7.15 in PyCharm but the
>> error message I get is:
>>
>> ERROR: Could not find a version that satisfies the requirement numpy (from
>>> versions: none)
>>> ERROR: No matching distribution found for numpy
>>> c:\python27\lib\site-packages\pip\_vendor\urllib3\util\ssl_.py:164:
>>> InsecurePlatformWarning: A true SSLContext object is not available. This
>>> prevents urllib3 fro
>>> m configuring SSL appropriately and may cause certain SSL connections to
>>> fail. You can upgrade to a newer version of Python to solve this. For
> more
>>> information, see
>>> https://urllib3.readthedocs.io/en/latest/advanced-usage.html#ssl-warnings
>>> InsecurePlatformWarning,
>>
>>
>> Any clues?
>
> Why are you using Python 2? Come on, it's been 16 years. Ya gotta move
> on at some point.

Change is hard even when it may be necessary.

The argument often is about whether some things are necessary or not.

Python made a decision but clearly not a unanimous one.

My current PC was not upgradable because of the new hardware requirement
Microsoft decided was needed for Windows 11. I bought a new one a while back
and turned it on in another room and then set it aside because replacing the
current one in the current position will be a pain, especially with getting
all my wires and so on, and since I do not want to use a full copy of my
data including many obsolete things, that will be another pain to get what I
need, if I can remember. Complicating issues also include licenses for
things in fixed amounts and the likelihood of messing up the
hardware/software I have that records shows from cable to my hard disk,
possibly needing to buy a new one.

I mention this in the context of examples of why even people who are fairly
knowledgeable do not feel much need to fix what does not feel broken.

I have wondered if instead of doing what Microsoft wants, if maybe switching
to Linux of some kinds makes as much sense. I suspect some may simply
upgrade to an Apple product.

And think of all the PC's that may effectively be discarded as they may not
even be usable if donated.

We live in a rapidly developing age and hence one with regularly and
irregularly scheduled rounds of obsolescence.

When is Python 4 coming?

-----Original Message-----
From: Python-list <python-list-bounces+avi.e.gross=gmail.com@python.org> On
Behalf Of MRAB via Python-list
Sent: Wednesday, June 12, 2024 12:56 PM
To: python-list@python.org
Subject: Re: Couldn't install numpy on Python 2.7

On 2024-06-12 17:31, AVI GROSS via Python-list wrote:
> I am sure there is inertia to move from an older product and some people
> need a reason like this where the old becomes untenable.
>
> It seems Microsoft is having a problem where something lik 2/3 of Windows
> users have not upgraded from Windows 10 after many years and have set a
> deadline in a year or so for stopping updates. In that case, hardware was
a > concern for some as Windows 11 did not work on their machines. With
> upgrading python, the main concern is having to get someone to examine old
> code and try to make it compatible.
>
In the case of Windows, my PC is over 10 years old yet performs
perfectly well for my needs. It can't run Windows 11. Therefore, I'm in
the process of migrating to Linux, and I still have over a year to
achieve that before support ends.

> But anyone doing new code in Python 2 in recent years should ...
>
Indeed...

> -----Original Message-----
> From: Python-list <python-list-bounces+avi.e.gross=gmail.com@python.org>
On
> Behalf Of Gordinator via Python-list
> Sent: Wednesday, June 12, 2024 10:19 AM
> To: python-list@python.org
> Subject: Re: Couldn't install numpy on Python 2.7
>
> On 12/06/2024 12:30, marc nicole wrote:
>> I am trying to install numpy library on Python 2.7.15 in PyCharm but the
>> error message I get is:
>>
>> ERROR: Could not find a version that satisfies the requirement numpy
(from
>>> versions: none)
>>> ERROR: No matching distribution found for numpy
>>> c:\python27\lib\site-packages\pip\_vendor\urllib3\util\ssl_.py:164:
>>> InsecurePlatformWarning: A true SSLContext object is not available. This
>>> prevents urllib3 fro
>>> m configuring SSL appropriately and may cause certain SSL connections to
>>> fail. You can upgrade to a newer version of Python to solve this. For
> more
>>> information, see
>>>
https://urllib3.readthedocs.io/en/latest/advanced-usage.html#ssl-warnings
>>> InsecurePlatformWarning,
>>
>>
>> Any clues?
>
> Why are you using Python 2? Come on, it's been 16 years. Ya gotta move
> on at some point.

--
https://mail.python.org/mailman/listinfo/python-list

On Thu, 13 Jun 2024 at 03:41, AVI GROSS via Python-list
<python-list@python.org> wrote:
>
> Change is hard even when it may be necessary.
>
> The argument often is about whether some things are necessary or not.
>
> Python made a decision but clearly not a unanimous one.

What decision? To not release any new versions of Python 2? That isn't
actually the OP's problem here - the Python interpreter runs just
fine. But there's no numpy build for the OP's hardware and Python 2.7.

So if you want to complain about Python 2.7 being dead, all you have
to do is go through all of the popular packages and build binaries for
all modern computers. If that sounds easy, go ahead and do it; if it
sounds hard, realise that open source is not a democracy, and you
can't demand that other people do more and more and more unpaid work
just because you can't be bothered upgrading your code.

> My current PC was not upgradable because of the new hardware requirement
> Microsoft decided was needed for Windows 11.

Yes, and that's a good reason to switch to Linux for the older computer.

> I mention this in the context of examples of why even people who are fairly
> knowledgeable do not feel much need to fix what does not feel broken.

It doesn't feel broken, right up until it does. The OP has discovered
that it *IS* broken. Whining that it doesn't "feel broken" is nonsense
when it is, in fact, not working.

> When is Python 4 coming?

Is this just another content-free whine, or are you actually curious
about the planned future of Python? If the latter, there is **PLENTY**
of information out there and I don't need to repeat it here.

Please don't FUD.

ChrisA

On 6/12/2024 1:59 PM, Chris Angelico via Python-list wrote:
> On Thu, 13 Jun 2024 at 03:41, AVI GROSS via Python-list
> <python-list@python.org> wrote:
>>
>> Change is hard even when it may be necessary.
>>
>> The argument often is about whether some things are necessary or not.
>>
>> Python made a decision but clearly not a unanimous one.
>
> What decision? To not release any new versions of Python 2? That isn't
> actually the OP's problem here - the Python interpreter runs just
> fine. But there's no numpy build for the OP's hardware and Python 2.7.
>
> So if you want to complain about Python 2.7 being dead, all you have
> to do is go through all of the popular packages and build binaries for
> all modern computers. If that sounds easy, go ahead and do it; if it
> sounds hard, realise that open source is not a democracy, and you
> can't demand that other people do more and more and more unpaid work
> just because you can't be bothered upgrading your code.

I support a Tomcat project that has some java code and most of the code
is for Jython 2.7. Jython 2.7 is approximately on a par with Python
2.7. Any Python-only code from the standard library will probably run,
but of course any C extensions cannot. The nice thing about using
Jython in a java environment is that it can call any java object, and
java code can call Jython objects and their methods.

The project cannot move to a Python-3 compatible version because Jython
3.xx doesn't exist and may never exist. The saving grace is that my
project doesn't have to use packages like numpy, scipy, and so forth.
Also, the project is very mature and almost certainly won't need to
create functionality such packages would enable. It would be nice to be
able to use some newer parts of the standard library, but there it is.
Jython does support "from __future__ import" and I make use of that for
the print function and the like.

>> My current PC was not upgradable because of the new hardware requirement
>> Microsoft decided was needed for Windows 11.
>
> Yes, and that's a good reason to switch to Linux for the older computer.

I have a 2012-vintage laptop that in modern terms has a very small
supply of RAM and a very slow hard drive. When my newer Windows 10
computer was going to be out of service for a while, I put a Linux
distro on an external SSD and copied things I needed to work on to it,
including my Thunderbird email profile directory.

Thunderbird and everything else worked perfectly for me during that
week. True, there were a few Windows-only programs I missed, but I used
other similar programs even if I didn't like them as much. It's amazing
how little resources Linux installs need, even with a GUI. Of course,
4GB RAM is limiting whether you are on Linux or Windows - you can't
avoid shuffling all those GUI bits around - but with a little care it
worked great. And with the external SSD the laptop was a lot snappier
than it ever was when it was new.

>> I mention this in the context of examples of why even people who are fairly
>> knowledgeable do not feel much need to fix what does not feel broken.
>
> It doesn't feel broken, right up until it does. The OP has discovered
> that it *IS* broken. Whining that it doesn't "feel broken" is nonsense
> when it is, in fact, not working.
>
>> When is Python 4 coming?
>
> Is this just another content-free whine, or are you actually curious
> about the planned future of Python? If the latter, there is **PLENTY**
> of information out there and I don't need to repeat it here.
>
> Please don't FUD.
>
> ChrisA

On Thu, 13 Jun 2024 at 06:55, Thomas Passin via Python-list
<python-list@python.org> wrote:
> The project cannot move to a Python-3 compatible version because Jython
> 3.xx doesn't exist and may never exist. The saving grace is that my
> project doesn't have to use packages like numpy, scipy, and so forth.

Exactly. If you don't need to ALSO use something newer, there's
nothing stopping you from continuing with the old version. And that's
fine! As long as you're okay with not getting updates, you're welcome
to do whatever you like - including running Windows 98 on an ancient
PC and editing your documents on that. (Yes, I know someone who did
that, long after Win 98 was dead to most of us.)

> Thunderbird and everything else worked perfectly for me during that
> week. True, there were a few Windows-only programs I missed, but I used
> other similar programs even if I didn't like them as much.

It's true. And there ARE solutions to that, although it's a bit rough
trying to run them on low hardware (Wine works nicely for some
programs, less so for others; VirtualBox is really not gonna be happy
with a small fraction of your limited RAM). But if your needs are
simple, even a crazily low-end system is sufficient.

> It's amazing
> how little resources Linux installs need, even with a GUI. Of course,
> 4GB RAM is limiting whether you are on Linux or Windows - you can't
> avoid shuffling all those GUI bits around - but with a little care it
> worked great. And with the external SSD the laptop was a lot snappier
> than it ever was when it was new.

One of the big differences with Linux is that you have a choice of
desktop environments, from "none" (just boot straight into a terminal)
on up. Some of them are a bit of a compromise in terms of how well you
can get your work done, but let's say you had an even MORE ancient
system with maybe one gig of memory... I'd rather have a super-light
desktop environment even if it doesn't have everything I'm normally
accustomed to!

ChrisA

Chris,

Since you misunderstood, my statement was that making an incompatible set of
changes to create Python 3 in the first place was a decision made by some
and perhaps not one that thrilled others who already had an embedded base of
programs or ones in the pipeline that would need much work to become
comparable.

And, of course, users of a program who continued to use python 2, also have
to find a way to ...

But if the goal was to deprecate python 2 and in some sense phase it out, it
is perhaps not working well for some. Frankly, issuing so many updates like
2.7 and including backporting of new features has helped make it possible to
delay any upgrade.

And, yes, I was KIDDING about python 4. I am simply suggesting that there
may well be a time that another shift happens that may require another
effort to get people on board a new and perhaps incompatible setup. I have
seen things like that happen in multiple phases including phases where the
new tools are not an upgrade, but brand new. An example might be if
accompany decided to switch to another existing language because they want
better error detection and faster execution or new features that may take
forever to arrive in what they are using or that supply various services by
humans to help them.

The discussion though was about a specific OP asking if they can fix their
problem. One solution being suggested is to fix a deeper problem and simply
make their code work with a recent version of python 3. But another solution
could be to step backward to a version of python 2 that still has numpy
support, or as was suggested, find out what other modules they are using are
interfering with the program being satisfied with the last version of numpy
being used and perhaps find a way to get ...

In the long run, though, continuing with python 2 will likely cause ever
more such headaches if you want the latest and greatest of things like
numpy.

-----Original Message-----
From: Python-list <python-list-bounces+avi.e.gross=gmail.com@python.org> On
Behalf Of Chris Angelico via Python-list
Sent: Wednesday, June 12, 2024 2:00 PM
To: python-list@python.org
Subject: Re: Couldn't install numpy on Python 2.7

On Thu, 13 Jun 2024 at 03:41, AVI GROSS via Python-list
<python-list@python.org> wrote:
>
> Change is hard even when it may be necessary.
>
> The argument often is about whether some things are necessary or not.
>
> Python made a decision but clearly not a unanimous one.

What decision? To not release any new versions of Python 2? That isn't
actually the OP's problem here - the Python interpreter runs just
fine. But there's no numpy build for the OP's hardware and Python 2.7.

So if you want to complain about Python 2.7 being dead, all you have
to do is go through all of the popular packages and build binaries for
all modern computers. If that sounds easy, go ahead and do it; if it
sounds hard, realise that open source is not a democracy, and you
can't demand that other people do more and more and more unpaid work
just because you can't be bothered upgrading your code.

> My current PC was not upgradable because of the new hardware requirement
> Microsoft decided was needed for Windows 11.

Yes, and that's a good reason to switch to Linux for the older computer.

> I mention this in the context of examples of why even people who are
fairly
> knowledgeable do not feel much need to fix what does not feel broken.

It doesn't feel broken, right up until it does. The OP has discovered
that it *IS* broken. Whining that it doesn't "feel broken" is nonsense
when it is, in fact, not working.

> When is Python 4 coming?

Is this just another content-free whine, or are you actually curious
about the planned future of Python? If the latter, there is **PLENTY**
of information out there and I don't need to repeat it here.

Please don't FUD.

ChrisA
--
https://mail.python.org/mailman/listinfo/python-list

On Thu, 13 Jun 2024 at 07:36, <avi.e.gross@gmail.com> wrote:
> But if the goal was to deprecate python 2 and in some sense phase it out, it
> is perhaps not working well for some. Frankly, issuing so many updates like
> 2.7 and including backporting of new features has helped make it possible to
> delay any upgrade.

The goal was to improve Python. I don't think anyone's ever tried to
"kill off" Python 2 - not in the sense of stopping people from using
it - but there haven't been any changes, not even security fixes, in
over four years.

> And, yes, I was KIDDING about python 4. I am simply suggesting that there
> may well be a time that another shift happens that may require another
> effort to get people on board a new and perhaps incompatible setup.

Kidding, eh? It sure sounded like you were trying to imply that there
would inevitably be another major breaking change. It definitely
smelled like FUD.

Maybe your jokes just aren't funny.

ChrisA

On Wed, 12 Jun 2024 at 22:38, AVI GROSS via Python-list
<python-list@python.org> wrote:
>
> The discussion though was about a specific OP asking if they can fix their
> problem. One solution being suggested is to fix a deeper problem and simply
> make their code work with a recent version of python 3.

The OP has not replied with any explanation as to why they are using
Python 2.7 and has not said whether they have any existing code that
only works with Python 2.7. It is unclear at this point whether there
is any reason that they shouldn't just install a newer version of
Python.

They are seeing a warning that explicitly says "You can upgrade to a
newer version of Python to solve this". I don't know whether that SSL
warning is directly connected to pip not finding any versions of numpy
but with the available information so far that seems like the first
thing to consider.

It is entirely reasonable to start by suggesting to use a newer
version of Python until some reason is given for not doing that.

--
Oscar

On Thu, 13 Jun 2024 at 07:57, Oscar Benjamin via Python-list
<python-list@python.org> wrote:
> They are seeing a warning that explicitly says "You can upgrade to a
> newer version of Python to solve this". I don't know whether that SSL
> warning is directly connected to pip not finding any versions of numpy
> but with the available information so far that seems like the first
> thing to consider.

I think it is; AIUI, with an ancient SSL library, pip is unable to
download packages safely from the current pypi server. So if anyone
actually does need to use pip with Python 2.7, they probably need to
set up a local server, using older encryption protocols (which should
therefore NOT be made accessible to the internet). Since pip can't
contact the upstream pypi, there's no available numpy for it to
install.

ChrisA

On Wed, 12 Jun 2024 at 23:11, Chris Angelico via Python-list
<python-list@python.org> wrote:
>
> On Thu, 13 Jun 2024 at 07:57, Oscar Benjamin via Python-list
> <python-list@python.org> wrote:
> > They are seeing a warning that explicitly says "You can upgrade to a
> > newer version of Python to solve this". I don't know whether that SSL
> > warning is directly connected to pip not finding any versions of numpy
> > but with the available information so far that seems like the first
> > thing to consider.
>
> I think it is; AIUI, with an ancient SSL library, pip is unable to
> download packages safely from the current pypi server. So if anyone
> actually does need to use pip with Python 2.7, they probably need to
> set up a local server, using older encryption protocols (which should
> therefore NOT be made accessible to the internet). Since pip can't
> contact the upstream pypi, there's no available numpy for it to
> install.

I don't know much about SSL and related networking things especially
on Windows. I would be surprised if pip on old Python can't install
from current PyPI though. I imagine that something strange has
happened like a new version of pip running on an old version of Python
or old Python on new OS (or old PyCharm...).

There is no problem using Python 2.7 with pip and PyPI on this Linux
machine but I guess it has a newer SSL library provided by the OS:

$ pip install numpy
DEPRECATION: Python 2.7 reached the end of its life on January 1st,
2020. Please upgrade your Python as Python 2.7 is no longer
maintained. pip 21.0 will drop support for Python 2.7 in January 2021.
More details about Python 2 support in pip can be found at
https://pip.pypa.io/en/latest/development/release-process/#python-2-support
pip 21.0 will remove support for this functionality.
Collecting numpy
Downloading numpy-1.16.6-cp27-cp27mu-manylinux1_x86_64.whl (17.0 MB)
|████████████████████████████████| 17.0 MB 14.3 MB/s
Installing collected packages: numpy
Successfully installed numpy-1.16.6

If it is actually the case that pip on Python 2.7 (on Windows) cannot
download from PyPI then an easier option rather than creating a local
server would just be to download the numpy wheels from PyPI using a
browser:

https://pypi.org/project/numpy/1.15.4/#files

Then you can do

pip install .\numpy-1.15.4-cp27-none-win_amd64.whl

Using a newer version of Python is still my primary suggestion though.

--
Oscar

On 13/06/24 4:31 am, avi.e.gross@gmail.com wrote:
> It seems Microsoft is having a problem where something lik 2/3 of Windows
> users have not upgraded from Windows 10 after many years

At least Python 3 is a clear improvement over Python 2 in many ways.
Whereas the only thing Microsoft seems to have done with Windows in
recent times is change it in ways that nobody wants, so there is
understandable resistance to upgrading even if it's possible.

On 13/06/24 10:09 am, Chris Angelico wrote:
> So if anyone
> actually does need to use pip with Python 2.7, they probably need to
> set up a local server

You should also be able to download a .tar.gz from PyPI and use pip
to install that. Although you'll have to track down the dependencies
yourself in that case.

--
Greg

On Thu, 13 Jun 2024 at 08:46, Oscar Benjamin via Python-list
<python-list@python.org> wrote:
> I don't know much about SSL and related networking things especially
> on Windows. I would be surprised if pip on old Python can't install
> from current PyPI though. I imagine that something strange has
> happened like a new version of pip running on an old version of Python
> or old Python on new OS (or old PyCharm...).
>
> There is no problem using Python 2.7 with pip and PyPI on this Linux
> machine but I guess it has a newer SSL library provided by the OS:

Sadly, I would NOT be surprised if this is indeed a problem on
Windows. You're exactly right - on Linux, it can use a newer SSL
library from the OS. Of course, this does assume that you've updated
your OS, which is far from a guarantee, but since this has security
implications there's a good chance you can update it while retaining a
legacy system.

On Thu, 13 Jun 2024 at 08:51, Greg Ewing via Python-list
<python-list@python.org> wrote:
> On 13/06/24 10:09 am, Chris Angelico wrote:
> > So if anyone
> > actually does need to use pip with Python 2.7, they probably need to
> > set up a local server
>
> You should also be able to download a .tar.gz from PyPI and use pip
> to install that. Although you'll have to track down the dependencies
> yourself in that case.

Also a possibility; in my opinion, losing dependency management is too
big a cost, so I would be inclined to set up a local server. But then,
I would be using a newer SSL library and not have the problem in the
first place.

ChrisA

On Wed, 12 Jun 2024 at 23:52, Greg Ewing via Python-list
<python-list@python.org> wrote:
> On 13/06/24 10:09 am, Chris Angelico wrote:
> > So if anyone
> > actually does need to use pip with Python 2.7, they probably need to
> > set up a local server
>
> You should also be able to download a .tar.gz from PyPI and use pip
> to install that. Although you'll have to track down the dependencies
> yourself in that case.

It is almost certainly better to download the wheel (.whl) file rather
than the sdist (.tar.gz) file. Building NumPy from source needs not
just compilers etc but also you first need to build/install a BLAS
library.

--
Oscar

Chris,

I don't want to get off message and debate whether my "jokes" are jokes, let
alone funny. Obviously, they often aren't.

What I meant by joking here does seem relevant. As the years pass, there can
come a time when it is suggested that a language (any language including
python) is no longer useful to many in the community as it has not kept up
with changes in the industry or whatever. Suggestions are made for changes
and additions that that not be backward compatible. They can be somewhat
minor things like new keywords that have not been reserved and where
programs that exist might be scanned for use of that keyword, and you simply
replace those names with was_keyword or something and the programs will
generally run. But there can be major changes too and there can be a
choice to just create a new language that has some similarities to python 3
(or PERL version whatever) or just name it the same but a version higher
much like has happened.

My point was that version 4 COULD HAPPEN one day and I meant INCOMPATIBLE
version not 4. Obviously we can make a version 4 that is not incompatible
too.

I have experience in other languages where disconnects happen at various
levels. Some functions in a collection such as a package are removed perhaps
to replace them with a more abstract version that does much more. Do you
remove the old one immediately or do you make a new name for the new one and
perhaps in some way mark the old one for deprecation with a pointer to the
new one to be used as soon as reasonable? I have seen many approaches. I
have seen entire packages yanked. I have seen parts that used to be in the
distribution as if built-in and then taken out and vice versa.

The point is you do not need a 4.0 to be incompatible. The incompatibility,
or need to change, can happen anytime when you are importing things like
numpy which is released whenever they want to and is not part of the python
distribution. Also, as we have seen at times, other modules you may have
imported, in some languages, can mask names you are using in your program
that you may not even be aware are there. Much can go wrong with software
and keeping current can also give you problems when something released may
have inadvertent changes or bugs.

So much of our code is voluntary and as noted earlier, some python
variants/distributions simply may not have anyone interested in keeping them
up to date. You as a user, take your chances.

-----Original Message-----
From: Python-list <python-list-bounces+avi.e.gross=gmail.com@python.org> On
Behalf Of Chris Angelico via Python-list
Sent: Wednesday, June 12, 2024 5:52 PM
To: python-list@python.org
Subject: Re: Couldn't install numpy on Python 2.7

On Thu, 13 Jun 2024 at 07:36, <avi.e.gross@gmail.com> wrote:
> But if the goal was to deprecate python 2 and in some sense phase it out,
it
> is perhaps not working well for some. Frankly, issuing so many updates
like
> 2.7 and including backporting of new features has helped make it possible
to
> delay any upgrade.

The goal was to improve Python. I don't think anyone's ever tried to
"kill off" Python 2 - not in the sense of stopping people from using
it - but there haven't been any changes, not even security fixes, in
over four years.

> And, yes, I was KIDDING about python 4. I am simply suggesting that there
> may well be a time that another shift happens that may require another
> effort to get people on board a new and perhaps incompatible setup.

Kidding, eh? It sure sounded like you were trying to imply that there
would inevitably be another major breaking change. It definitely
smelled like FUD.

Maybe your jokes just aren't funny.

ChrisA
--
https://mail.python.org/mailman/listinfo/python-list

On Thu, 13 Jun 2024 at 09:20, <avi.e.gross@gmail.com> wrote:
> My point was that version 4 COULD HAPPEN one day and I meant INCOMPATIBLE
> version not 4. Obviously we can make a version 4 that is not incompatible
> too.

This is still FUD. Back your words with something, or stop trying to
imply that there's another incompatible change just around the corner.

Do you realise how insulting you are being to the developers of Python
by these implications?

ChrisA

Chris,

You seem to have perceived an insult that I remain unaware of.

I have no special knowledge, like you do, of plans made for changes to the
pthon language and implementation.

I was asking a hypothetical question about what some users would do if
python came out with a newer major version. I have seen people often wait
until some software that tries to get updated too frequently makes multiple
updates and then finally give in and skip the intermediates.

I wondered if something like a version 4.0 might get people still using
version 2 might finally come around and also if some version 3 users would
not be thrilled with something not stable enough.

I have no favorite ideas here and can see a balance between adding features
or fixing flaws and on the other side, not discomfiting many and especially
when in many cases, the original people who wrote software are no longer
there nor budgets to pay for changes.

I looked up FUD and sharply disagree with suggestions I am trying to somehow
cause Fear, Uncertainty or Doubt. I simply asked if another such update ...
as a hypothetical. Had I asked what impact Quantum Computers might have on
existing languages, would that also be FUD, or just a speculation in a
discussion.

Either way, I am taking any further discussion along these lines offline and
will not continue here.

-----Original Message-----
From: Python-list <python-list-bounces+avi.e.gross=gmail.com@python.org> On
Behalf Of Chris Angelico via Python-list
Sent: Wednesday, June 12, 2024 7:23 PM
To: python-list@python.org
Subject: Re: Couldn't install numpy on Python 2.7

On Thu, 13 Jun 2024 at 09:20, <avi.e.gross@gmail.com> wrote:
> My point was that version 4 COULD HAPPEN one day and I meant INCOMPATIBLE
> version not 4. Obviously we can make a version 4 that is not incompatible
> too.

This is still FUD. Back your words with something, or stop trying to
imply that there's another incompatible change just around the corner.

Do you realise how insulting you are being to the developers of Python
by these implications?

ChrisA
--
https://mail.python.org/mailman/listinfo/python-list

On Thu, 13 Jun 2024 at 10:58, <avi.e.gross@gmail.com> wrote:
>
> Chris,
>
> You seem to have perceived an insult that I remain unaware of.

If you're not aware that you're saying this, then don't say it.

> I looked up FUD and sharply disagree with suggestions I am trying to somehow
> cause Fear, Uncertainty or Doubt. I simply asked if another such update ...
> as a hypothetical. Had I asked what impact Quantum Computers might have on
> existing languages, would that also be FUD, or just a speculation in a
> discussion.

What DID you intend by your comments? Were you trying to imply that
work spent upgrading to Python 3 would have to be redone any day now
when this hypothetical massively-incompatible Python 4 is released? Or
what? What WERE you trying to say?

If you don't understand how damaging it can be to say that sort of
thing, **don't say it**. Otherwise, expect responses like this.

I *detest* the attitude that you can make vague disparaging comments
and then hide behind claims that you had no idea how damaging you were
being.

ChrisA

Chris Angelico <rosuav@gmail.com> wrote:
> On Thu, 13 Jun 2024 at 10:58, <avi.e.gross@gmail.com> wrote:
> >
> > Chris,
> >
> > You seem to have perceived an insult that I remain unaware of.
>
> If you're not aware that you're saying this, then don't say it.
>
Er, um, that really makes no sense! :-)

How can one not say something that one isn't aware of saying?

--
Chris Green
·