[Python-Dev] Exact date of Python 2 EOL?

2017-03-23 Thread Victor Stinner
Hi,

I looked at the "Status of Python branches" to check if it was up to
date. It's the case, thanks :-) But it recalled me that no exact date
was decided for the official end of line of the the Python 2 branch
(2.7 EOL).

https://docs.python.org/devguide/#status-of-python-branches
says January 1st, 2020

https://pythonclock.org/ picked April 12th, 2020 (Pycon US)

"Python 2.7 will not be maintained past 2020. No official date has
been given, so this clock counts down until April 12th, 2020, which
will be roughly the time of the 2020 PyCon. I am hereby suggesting we
make PyCon 2020 the official end-of-life date, and we throw a massive
party to celebrate all that Python 2 has done for us. (If this sounds
interesting to you, email [email protected])."

Can we pick an official date?


By the way, maybe we can also start to list vendors (Linux vendors?)
who plan to offer commercial extended support? For example, you can
expect longer support than 2020 from RHEL:
https://access.redhat.com/support/policy/updates/errata

Ubuntu 12.04 reached its end of life after 5 years, but it seems like
Canonical also starts to offer extended support to customers:
http://www.omgubuntu.co.uk/2017/03/ubuntu-12-04-esm-support

You can expect longer Python 2 support indirectely ;-)

Victor
___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


[Python-Dev] 3.5 unittest does not support namespace packages for discovering

2017-03-23 Thread Ilya Kazakevich
Hello.
I have following layout:

\---tests
|   test_module.py
|   __init__.py


When I launch "python.exe" -m unittest discover -t . -s tests" it works
perfectly.
But when I remove " __init__.py" it says

 Start directory is not importable: "tests'"

``loader.py``:
if start_dir != top_level_dir:
is_not_importable = not os.path.isfile(os.path.join(start_dir,
'__init__.py'))


I believe ``__init__.py`` does not play any role since Python 3.3, am I
right?
If so, it seems to be a bug. Should I create an issue?

Ilya.
___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] 3.5 unittest does not support namespace packages for discovering

2017-03-23 Thread INADA Naoki
There is already http://bugs.python.org/issue29642

On Thu, Mar 23, 2017 at 11:04 PM, Ilya Kazakevich
 wrote:
> Hello.
> I have following layout:
>
> \---tests
> |   test_module.py
> |   __init__.py
>
>
> When I launch "python.exe" -m unittest discover -t . -s tests" it works
> perfectly.
> But when I remove " __init__.py" it says
>
>  Start directory is not importable: "tests'"
>
> ``loader.py``:
> if start_dir != top_level_dir:
> is_not_importable = not os.path.isfile(os.path.join(start_dir,
> '__init__.py'))
>
>
> I believe ``__init__.py`` does not play any role since Python 3.3, am I
> right?
> If so, it seems to be a bug. Should I create an issue?
>
> Ilya.
>
> ___
> Python-Dev mailing list
> [email protected]
> https://mail.python.org/mailman/listinfo/python-dev
> Unsubscribe:
> https://mail.python.org/mailman/options/python-dev/songofacandy%40gmail.com
>
___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] 3.5 unittest does not support namespace packages for discovering

2017-03-23 Thread INADA Naoki
And this issue is relating to it too: http://bugs.python.org/issue29716

In short, "namespace package" is for make it possible to `pip install
foo_bar foo_baz`,
when foo_bar provides `foo.bar` and foo_baz provides `foo.baz`
package.  (foo is namespace package).

If unittests searches normal directly, it may walk deep into very
large tree containing
millions of directories.  I don't like it.


On Fri, Mar 24, 2017 at 12:53 AM, INADA Naoki  wrote:
> There is already http://bugs.python.org/issue29642
>
> On Thu, Mar 23, 2017 at 11:04 PM, Ilya Kazakevich
>  wrote:
>> Hello.
>> I have following layout:
>>
>> \---tests
>> |   test_module.py
>> |   __init__.py
>>
>>
>> When I launch "python.exe" -m unittest discover -t . -s tests" it works
>> perfectly.
>> But when I remove " __init__.py" it says
>>
>>  Start directory is not importable: "tests'"
>>
>> ``loader.py``:
>> if start_dir != top_level_dir:
>> is_not_importable = not os.path.isfile(os.path.join(start_dir,
>> '__init__.py'))
>>
>>
>> I believe ``__init__.py`` does not play any role since Python 3.3, am I
>> right?
>> If so, it seems to be a bug. Should I create an issue?
>>
>> Ilya.
>>
>> ___
>> Python-Dev mailing list
>> [email protected]
>> https://mail.python.org/mailman/listinfo/python-dev
>> Unsubscribe:
>> https://mail.python.org/mailman/options/python-dev/songofacandy%40gmail.com
>>
___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] Exact date of Python 2 EOL?

2017-03-23 Thread Skip Montanaro
By the way, maybe we can also start to list vendors (Linux vendors?)
who plan to offer commercial extended  ...


Delurking ever so briefly...

Might be worthwhile to list published vendor EOL dates no matter if they
are before or after the 2020 EOL date. Different Linux distros have
different focuses. This could very well be a useful gauge for potential
users.

Skip
___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] Exact date of Python 2 EOL?

2017-03-23 Thread Barry Warsaw
On Mar 23, 2017, at 09:41 AM, Victor Stinner wrote:

>Can we pick an official date?

Benjamin should pick the date and update PEP 373.

>Ubuntu 12.04 reached its end of life after 5 years, but it seems like
>Canonical also starts to offer extended support to customers:
>http://www.omgubuntu.co.uk/2017/03/ubuntu-12-04-esm-support

Ubuntu 14.04 and 16.04 both have Python 2.7 so those will be supported until
2019 and 2021 respectively.

I'm still hoping to relegate 2.7 to universe for 18.04 LTS.  I don't think
we'll be able to drop it entirely (nor should we).  If it stays in main, it
will get official support until 2023.

https://www.ubuntu.com/info/release-end-of-life

Cheers,
-Barry
___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] Exact date of Python 2 EOL?

2017-03-23 Thread Brian Curtin
On Thu, Mar 23, 2017 at 12:19 PM, Barry Warsaw  wrote:
> On Mar 23, 2017, at 09:41 AM, Victor Stinner wrote:
>
>>Can we pick an official date?
>
> Benjamin should pick the date and update PEP 373.

Not to start a bikeshed (calendarshed?), but how about 8 February
2020, or 2/8 as some in the US would write it?
___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] Exact date of Python 2 EOL?

2017-03-23 Thread Steve Dower

On 23Mar2017 0937, Brian Curtin wrote:

On Thu, Mar 23, 2017 at 12:19 PM, Barry Warsaw  wrote:

On Mar 23, 2017, at 09:41 AM, Victor Stinner wrote:


Can we pick an official date?


Benjamin should pick the date and update PEP 373.


Not to start a bikeshed (calendarshed?), but how about 8 February
2020, or 2/8 as some in the US would write it?


Whatever date is it will have to have a release on it. There's no point 
in taking fixes after the final release, and I'd suggest that given it 
is the last one that we'll want a longer RC period with RM approval 
required for any changes after that. (I predict a slight chance of a 
last minute rush to get contribution made - possibly it's worth locking 
things down from the second-to-last release?)


All of which to say, there are more important concerns than choosing a 
cute date :) I've got no doubt that Benjamin will come up with a plan 
that leads to a stable and reliable release.


Cheers,
Steve
___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] Exact date of Python 2 EOL?

2017-03-23 Thread Ned Deily
On Mar 23, 2017, at 04:41, Victor Stinner  wrote:
> By the way, maybe we can also start to list vendors (Linux vendors?)
> who plan to offer commercial extended support?

IMO, we should definitely not list 2.7 vendor and their plans on a python.org 
web page.  One, it would be a maintenance headache for the web site / devguide. 
 Two, it puts us and the PSF in the position of curating such a list and 
deciding who get to be on it and in what order and with keeping up with 
vendors' changing plans etc etc.  Third, most likely they are already using a 
particular vendor so they are not likely to change just because of Python 2 
support.  And, fourth and most important, we don't really want to be 
encouraging people to stay on 2.7 anyway!  If it is important enough for them, 
they can do their research and find out directly from the vendors.

--
  Ned Deily
  [email protected] -- []

___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] Exact date of Python 2 EOL?

2017-03-23 Thread Terry Reedy

On 3/23/2017 4:41 AM, Victor Stinner wrote:

Hi,

I looked at the "Status of Python branches" to check if it was up to
date. It's the case, thanks :-) But it recalled me that no exact date
was decided for the official end of line of the the Python 2 branch
(2.7 EOL).


> https://docs.python.org/devguide/#status-of-python-branches
> says January 1st, 2020


Can we pick an official date?


The devguide list is effectively the official list.  No one should plan 
on getting anything from pydev after 1/1/20.


As far as ''we' are concerned, 2.7 EOL will be when the final 2.7.n 
release is made.  If Benjamin continues twice-a-year releases, n could 
be 20 in mid 2020, about 10 years after 2.7.0 in 2010-7-3.  Or perhaps 
he will push a possible 2.7.19 'Dec 2019' release to 2020-01-01. 
Especially if the rate of commits decreases, he might lengthen the 
interval, thereby reducing the final ''n'.  In the meanwhile, I think we 
should let him continue planning one release ahead.


https://www.python.org/dev/peps/pep-0373/
  Planned future release dates: 2.7.14 in mid 2017

--
Terry Jan Reedy

___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] 3.5 unittest does not support namespace packages for discovering

2017-03-23 Thread Robert Collins
On 24 March 2017 at 04:59, INADA Naoki  wrote:
> And this issue is relating to it too: http://bugs.python.org/issue29716
>
> In short, "namespace package" is for make it possible to `pip install
> foo_bar foo_baz`,
> when foo_bar provides `foo.bar` and foo_baz provides `foo.baz`
> package.  (foo is namespace package).
>
> If unittests searches normal directly, it may walk deep into very
> large tree containing
> millions of directories.  I don't like it.

That is a risk, OTOH I think the failure to do what folk expect is a
bigger risk.

-Rob
___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] Exact date of Python 2 EOL?

2017-03-23 Thread MRAB

On 2017-03-23 16:37, Brian Curtin wrote:

On Thu, Mar 23, 2017 at 12:19 PM, Barry Warsaw  wrote:

On Mar 23, 2017, at 09:41 AM, Victor Stinner wrote:


Can we pick an official date?


Benjamin should pick the date and update PEP 373.


Not to start a bikeshed (calendarshed?), but how about 8 February
2020, or 2/8 as some in the US would write it?


If you see 2/8, is that 2 August or February 8?

We could avoid ambiguity if day_of_month == month or day_of_month > 12.

___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] 3.5 unittest does not support namespace packages for discovering

2017-03-23 Thread Ned Batchelder
On 3/23/17 3:14 PM, Robert Collins wrote:
> On 24 March 2017 at 04:59, INADA Naoki  wrote:
>> And this issue is relating to it too: http://bugs.python.org/issue29716
>>
>> In short, "namespace package" is for make it possible to `pip install
>> foo_bar foo_baz`,
>> when foo_bar provides `foo.bar` and foo_baz provides `foo.baz`
>> package.  (foo is namespace package).
>>
>> If unittests searches normal directly, it may walk deep into very
>> large tree containing
>> millions of directories.  I don't like it.
> That is a risk, OTOH I think the failure to do what folk expect is a
> bigger risk.

The issue here is, what do folks expect?  PEP 420 is pretty clear on its
purpose. The first sentence of the abstract:

> Namespace packages are a mechanism for splitting a single Python
package across multiple directories on disk.

And the first sentence of the specification:

> Regular packages will continue to have an __init__.py and will reside
in a single directory.

PEP 420 is not meant to make all __init__.py files optional. It has a
specific purpose.  These proposed changes are not in support of that
purpose.  We should not bend over backwards to support getting rid of
__init__.py files just because people don't like empty __init__.py
files. That's not what PEP 420 is for.

--Ned.

___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] Exact date of Python 2 EOL?

2017-03-23 Thread Antoine Pitrou
On Thu, 23 Mar 2017 09:41:11 +0100
Victor Stinner  wrote:
> Hi,
> 
> I looked at the "Status of Python branches" to check if it was up to
> date. It's the case, thanks :-) But it recalled me that no exact date
> was decided for the official end of line of the the Python 2 branch
> (2.7 EOL).
> 
> https://docs.python.org/devguide/#status-of-python-branches
> says January 1st, 2020
> 
> https://pythonclock.org/ picked April 12th, 2020 (Pycon US)

I suggest we pick April 1st, 2020, so as to extra confuse users who are
still on Python 2 ;-)

Regards

Antoine.


___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] 3.5 unittest does not support namespace packages for discovering

2017-03-23 Thread Paul Moore
On 23 March 2017 at 20:41, Ned Batchelder  wrote:
>>> If unittests searches normal directly, it may walk deep into very
>>> large tree containing
>>> millions of directories.  I don't like it.
>
>> That is a risk, OTOH I think the failure to do what folk expect is a
>> bigger risk.
>
> The issue here is, what do folks expect?  PEP 420 is pretty clear on its
> purpose.

The unittest docs say
(https://docs.python.org/3.6/library/unittest.html#test-discovery)

"""
Unittest supports simple test discovery. In order to be compatible
with test discovery, all of the test files must be modules or packages
(including namespace packages) importable from the top-level directory
of the project (this means that their filenames must be valid
identifiers).
"""

Personally, that to me pretty clearly implies that namespace packages
*should* work. They are explicitly mentioned, after all. I agree that
the implication that any random directory will get scanned is
unexpected - after all *I* know which directories I intended to be
namespace packages, it's just the computer that doesn't. Either way,
someone is likely to get surprised. But I'd argue that if we retain
the current __init__.py check, we need to amend the docs - and that's
then a change in the documented behaviour.

I do think it's a relatively minor point either way, and practicality
may well imply that we're better avoiding the risk of discovery being
impractical in projects with large directories full of resource files
or similar. But on the expectation side of things, I'd consider the
current behaviour as unexpected.

Paul
___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] Exact date of Python 2 EOL?

2017-03-23 Thread Barry Warsaw
On Mar 23, 2017, at 08:02 PM, MRAB wrote:

>If you see 2/8, is that 2 August or February 8?

I think that's 0.25 which doesn't look like a date to me .  ISO 8601
dates please: 2020-02-08 is unambiguous.

-Barry
___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] Exact date of Python 2 EOL?

2017-03-23 Thread Joao S. O. Bueno
On 23 March 2017 at 19:47, Barry Warsaw  wrote:
> On Mar 23, 2017, at 08:02 PM, MRAB wrote:
>
>>If you see 2/8, is that 2 August or February 8?
>
> I think that's 0.25 which doesn't look like a date to me .  ISO 8601
> dates please: 2020-02-08 is unambiguous.

In Python 2, 2/8 is just 0.
>
> -Barry
> ___
> Python-Dev mailing list
> [email protected]
> https://mail.python.org/mailman/listinfo/python-dev
> Unsubscribe: 
> https://mail.python.org/mailman/options/python-dev/jsbueno%40python.org.br
___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] Exact date of Python 2 EOL?

2017-03-23 Thread Victor Stinner
Le 23 mars 2017 19:56, "Terry Reedy"  a écrit :


> https://docs.python.org/devguide/#status-of-python-branches
> says January 1st, 2020

Can we pick an official date?
>

The devguide list is effectively the official list.  No one should plan on
getting anything from pydev after 1/1/20.


Oh sorry, I forgot to mention that *I* wrote this table. I chose this date
just to fill quickly the table. But I now would like a precise date to
prevent future confusion.

Victor
___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] Exact date of Python 2 EOL?

2017-03-23 Thread Zero Piraeus
:

On Thu, 2017-03-23 at 21:08 -0300, Joao S. O. Bueno wrote:
> In Python 2, 2/8 is just 0.

27/7 is 3 in Python 2, and between 3.8 and 3.9 in Python 3 (which is
probably about where 3.x will be).

 -[]z.
___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] Exact date of Python 2 EOL?

2017-03-23 Thread Nick Coghlan
On 24 March 2017 at 02:19, Barry Warsaw  wrote:

> On Mar 23, 2017, at 09:41 AM, Victor Stinner wrote:
>
> >Can we pick an official date?
>
> Benjamin should pick the date and update PEP 373.
>

+1.

The other detail to be clarified is that we know 2020-MM (whatever MM ends
up being) will be the last upstream 2.7.x binary release (so the
security-updates-for-bundled-OpenSSL problem will go away), but we haven't
decided yet how other security updates after that point will be handled.

If It's just RHEL/CentOS (and maybe Software Collections) keeping the
branch alive past the upstream end-of-community-support date it's not a big
deal, but if there are other redistributors or folks running self-supported
binaries still monitoring it for security updates, then it may make sense
to continue putting any required security patches in a common location,
even if we stop doing any new upstream releases.


> I'm still hoping to relegate 2.7 to universe for 18.04 LTS.  I don't think
> we'll be able to drop it entirely (nor should we).  If it stays in main, it
> will get official support until 2023.
>
> https://www.ubuntu.com/info/release-end-of-life
>

As far as I'm aware, Samba is the main remaining challenge for Fedora
Server on that front, but at least all of the libraries it depends on have
received the necessary updates to make them Python 2/3 compatible:
http://fedora.portingdb.xyz/pkg/samba/

Cheers,
Nick.

-- 
Nick Coghlan   |   [email protected]   |   Brisbane, Australia
___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] Exact date of Python 2 EOL?

2017-03-23 Thread Nick Coghlan
On 24 March 2017 at 02:59, Ned Deily  wrote:

> On Mar 23, 2017, at 04:41, Victor Stinner 
> wrote:
> > By the way, maybe we can also start to list vendors (Linux vendors?)
> > who plan to offer commercial extended support?
>
> IMO, we should definitely not list 2.7 vendor and their plans on a
> python.org web page.  One, it would be a maintenance headache for the web
> site / devguide.  Two, it puts us and the PSF in the position of curating
> such a list and deciding who get to be on it and in what order and with
> keeping up with vendors' changing plans etc etc.  Third, most likely they
> are already using a particular vendor so they are not likely to change just
> because of Python 2 support.  And, fourth and most important, we don't
> really want to be encouraging people to stay on 2.7 anyway!  If it is
> important enough for them, they can do their research and find out directly
> from the vendors.
>

Agreed, although I'd be open to maintaining such a list in my Python 3 Q&A
at
http://python-notes.curiousefficiency.org/en/latest/python3/questions_and_answers.html

That's explicitly caveated as being both from my personal perspective and
only intermittently updated. It also already mentions the 2024 date, since
that comes from RHEL 7's 2014 release date and 10 year support cycle. All
the other commercial support end dates we know for certain at this point
finish before 2020, with Canonical being a "maybe" on 2023 (depending on
what happens in Ubuntu 18.04).

Cheers,
Nick.

-- 
Nick Coghlan   |   [email protected]   |   Brisbane, Australia
___
Python-Dev mailing list
[email protected]
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com