Re: [Python-Dev] git history conundrum

2019-04-28 Thread Robert Collins
Share your own username with Michael or I and we'll add you there.

Rob

On Mon, 29 Apr 2019, 09:55 Chris Withers,  wrote:

> On 28/04/2019 22:21, Robert Collins wrote:
> > Thank you!
>
> Thank me when we get there ;-) Currently in Dec 2018 with a wonderful
> Py2 failure:
>
> ==
> ERROR: test_autospec_getattr_partial_function
> (mock.tests.testhelpers.SpecSignatureTest)
> --
> Traceback (most recent call last):
>File "mock/tests/testhelpers.py", line 973, in
> test_autospec_getattr_partial_function
>  autospec = create_autospec(proxy)
>File "mock/mock.py", line 2392, in create_autospec
>  for entry in dir(spec):
> TypeError: __dir__() must return a list, not str
>
> Once we're done, I'll need a username/password that can write to
> https://pypi.org/project/mock/ ...
>
> > If I understand correctly this is just the hg style branch backport
> > consequence, multiple copies of a change. Should be safe to skip those.
>
> Yep, current script I've been using is here, high level highlighted:
>
> https://github.com/cjw296/mock/blob/backporting/backport.py#L102-L125
>
> cheers,
>
> Chris
>
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] git history conundrum

2019-04-28 Thread Michael Foord



> On 28 Apr 2019, at 22:55, Chris Withers  wrote:
> 
>> On 28/04/2019 22:21, Robert Collins wrote:
>> Thank you!
> 
> Thank me when we get there ;-) Currently in Dec 2018 with a wonderful Py2 
> failure:
> 
> ==
> ERROR: test_autospec_getattr_partial_function 
> (mock.tests.testhelpers.SpecSignatureTest)
> --
> Traceback (most recent call last):
>  File "mock/tests/testhelpers.py", line 973, in 
> test_autospec_getattr_partial_function
>autospec = create_autospec(proxy)
>  File "mock/mock.py", line 2392, in create_autospec
>for entry in dir(spec):
> TypeError: __dir__() must return a list, not str
> 
> Once we're done, I'll need a username/password that can write to 
> https://pypi.org/project/mock/ ...

I can add you as a maintainer. Ping me off-list. 

Michael



> 
>> If I understand correctly this is just the hg style branch backport 
>> consequence, multiple copies of a change. Should be safe to skip those.
> 
> Yep, current script I've been using is here, high level highlighted:
> 
> https://github.com/cjw296/mock/blob/backporting/backport.py#L102-L125
> 
> cheers,
> 
> Chris
> ___
> Python-Dev mailing list
> Python-Dev@python.org
> https://mail.python.org/mailman/listinfo/python-dev
> Unsubscribe: 
> https://mail.python.org/mailman/options/python-dev/fuzzyman%40voidspace.org.uk

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


Re: [Python-Dev] git history conundrum

2019-04-28 Thread Chris Withers

On 28/04/2019 22:21, Robert Collins wrote:

Thank you!


Thank me when we get there ;-) Currently in Dec 2018 with a wonderful 
Py2 failure:


==
ERROR: test_autospec_getattr_partial_function 
(mock.tests.testhelpers.SpecSignatureTest)

--
Traceback (most recent call last):
  File "mock/tests/testhelpers.py", line 973, in 
test_autospec_getattr_partial_function

autospec = create_autospec(proxy)
  File "mock/mock.py", line 2392, in create_autospec
for entry in dir(spec):
TypeError: __dir__() must return a list, not str

Once we're done, I'll need a username/password that can write to 
https://pypi.org/project/mock/ ...


If I understand correctly this is just the hg style branch backport 
consequence, multiple copies of a change. Should be safe to skip those.


Yep, current script I've been using is here, high level highlighted:

https://github.com/cjw296/mock/blob/backporting/backport.py#L102-L125

cheers,

Chris
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] git history conundrum

2019-04-28 Thread Robert Collins
Thank you!

If I understand correctly this is just the hg style branch backport
consequence, multiple copies of a change. Should be safe to skip those.

Rob

On Sun, 28 Apr 2019, 07:11 Chris Withers,  wrote:

> Hi All,
>
> I'm in the process of bringing the mock backport up to date, but this
> has got me stumped:
>
> $ git log --oneline  --no-merges
> 5943ea76d529f9ea18c73a61e10c6f53bdcc864f.. -- Lib/unittest/mock.py
> Lib/unittest/test/testmock/ | tail
> 362f058a89 Issue #28735: Fixed the comparison of mock.MagickMock with
> mock.ANY.
> d9c956fb23 Issue #20804: The unittest.mock.sentinel attributes now
> preserve their identity when they are copied or pickled.
> 84b6fb0eea Fix unittest.mock._Call: don't ignore name
> 161a4dd495 Issue #28919: Simplify _copy_func_details() in unittest.mock
> ac5084b6c7 Fixes issue28380: unittest.mock Mock autospec functions now
> properly support assert_called, assert_not_called, and assert_called_once.
> 0be894b2f6 Issue #27895:  Spelling fixes (Contributed by Ville Skyttä).
> 15f44ab043 Issue #27895:  Spelling fixes (Contributed by Ville Skyttä).
> d4583d7fea Issue #26750: use inspect.isdatadescriptor instead of our own
> _is_data_descriptor().
> 9854789efe Issue #26750: unittest.mock.create_autospec() now works
> properly for subclasses of property() and other data descriptors.
> 204bf0b9ae English spelling and grammar fixes
>
> Right, so I've merged up to 15f44ab043, what comes next?
>
> $ git log --oneline  --no-merges 15f44ab043.. -- Lib/unittest/mock.py
> Lib/unittest/test/testmock/ | tail -n 3
> 161a4dd495 Issue #28919: Simplify _copy_func_details() in unittest.mock
> ac5084b6c7 Fixes issue28380: unittest.mock Mock autospec functions now
> properly support assert_called, assert_not_called, and assert_called_once.
> 0be894b2f6 Issue #27895:  Spelling fixes (Contributed by Ville Skyttä).
>
> Okay, no idea why 0be894b2f6 is there, appears to be a totally identical
> commit to 15f44ab043, so let's skip it:
>
> $ git log --oneline  --no-merges 0be894b2f6.. -- Lib/unittest/mock.py
> Lib/unittest/test/testmock/ | tail -n 3
> 161a4dd495 Issue #28919: Simplify _copy_func_details() in unittest.mock
> ac5084b6c7 Fixes issue28380: unittest.mock Mock autospec functions now
> properly support assert_called, assert_not_called, and assert_called_once.
> 15f44ab043 Issue #27895:  Spelling fixes (Contributed by Ville Skyttä).
>
> Wat?! Why is 15f44ab043 showing up again?!
>
> What's the git subtlety I'm missing here?
>
> Chris
> ___
> Python-Dev mailing list
> Python-Dev@python.org
> https://mail.python.org/mailman/listinfo/python-dev
> Unsubscribe:
> https://mail.python.org/mailman/options/python-dev/robertc%40robertcollins.net
>
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] git history conundrum

2019-04-28 Thread Antoine Pitrou
On Sun, 28 Apr 2019 08:25:30 +0100
Chris Withers  wrote:
> 
> What's the best way to spell "show me all the revisions on master that 
> affect {mock files} from commit x to HEAD, not including x"?

Something like:

$ git log x...HEAD -- {mock files}

perhaps?

Regards

Antoine.


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


Re: [Python-Dev] git history conundrum

2019-04-28 Thread Chris Withers

On 28/04/2019 03:51, Martin Panter wrote:

On Sat, 27 Apr 2019 at 19:07, Chris Withers  wrote:

Right, so I've merged up to 15f44ab043, what comes next?

$ git log --oneline  --no-merges 15f44ab043.. -- Lib/unittest/mock.py
Lib/unittest/test/testmock/ | tail -n 3


This Git command line means list all the revisions except 15f44ab043
and those leading up to it.


That seems at odds with what I've found searching online and with the 
backporting instructions left in the mock backport docs.


My understanding is that 15f44ab043.. expands out to 15f44ab043..HEAD 
and means "all revs between 15f44ab043 and master":


https://stackoverflow.com/a/7693298/216229

Can you explain what leads you to expect that to behave differently?


The convention at the time was to keep the 3.5 branch merged into
Default (Master). That is why my 3.5 backport appears in your history
of Master.


Ah, okay.


Because you are asked for all the revisions except my backport and its
ancestors. As far as Git is concerned, the original spelling fixes are
not an ancestor of my backport.

I don’t have a copy of the Git repository to try, but I suggest the
following command is what you want:

git log --oneline  --no-merges HEAD ^15f44ab043 ^0be894b2f6 --
Lib/unittest/mock.py Lib/unittest/test/testmock/


What's the best way to spell "show me all the revisions on master that 
affect {mock files} from commit x to HEAD, not including x"?


cheers,

Chris
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] git history conundrum

2019-04-28 Thread Chris Withers

On 28/04/2019 03:51, Martin Panter wrote:

On Sat, 27 Apr 2019 at 19:07, Chris Withers  wrote:

Right, so I've merged up to 15f44ab043, what comes next?

$ git log --oneline --no-merges 15f44ab043.. -- Lib/unittest/mock.py
Lib/unittest/test/testmock/ | tail -n 3


This Git command line means list all the revisions except 15f44ab043
and those leading up to it.


That seems at odds with what I've found searching online and with the 
backporting instructions left in the mock backport docs.


My understanding is that 15f44ab043.. expands out to 15f44ab043..HEAD 
and means "all revs between 15f44ab043 and master":


https://stackoverflow.com/a/7693298/216229

Can you explain what leads you to expect that to behave differently?


The convention at the time was to keep the 3.5 branch merged into
Default (Master). That is why my 3.5 backport appears in your history
of Master.


Ah, okay.


Because you are asked for all the revisions except my backport and its
ancestors. As far as Git is concerned, the original spelling fixes are
not an ancestor of my backport.

I don’t have a copy of the Git repository to try, but I suggest the
following command is what you want:

git log --oneline --no-merges HEAD ^15f44ab043 ^0be894b2f6 --
Lib/unittest/mock.py Lib/unittest/test/testmock/


What's the best way to spell "show me all the revisions on master that 
affect {mock files} from commit x to HEAD, not including x"?


cheers,

Chris

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


Re: [Python-Dev] git history conundrum

2019-04-27 Thread Martin Panter
On Sat, 27 Apr 2019 at 19:07, Chris Withers  wrote:
> Right, so I've merged up to 15f44ab043, what comes next?
>
> $ git log --oneline  --no-merges 15f44ab043.. -- Lib/unittest/mock.py
> Lib/unittest/test/testmock/ | tail -n 3

This Git command line means list all the revisions except 15f44ab043
and those leading up to it.

> 161a4dd495 Issue #28919: Simplify _copy_func_details() in unittest.mock
> ac5084b6c7 Fixes issue28380: unittest.mock Mock autospec functions now
> properly support assert_called, assert_not_called, and assert_called_once.
> 0be894b2f6 Issue #27895:  Spelling fixes (Contributed by Ville Skyttä).
>
> Okay, no idea why 0be894b2f6 is there, appears to be a totally identical
> commit to 15f44ab043

Git revision 15f44ab043 is the original spelling fixes, which were
pushed to the Mercurial “default” branch (= Git master) for Python 3.6
by Raymond. Revision 0be894b2f6 is my backport to the 3.5 branch, done
about a week later. The backport is probably a subset of the original,
rather than identical (e.g. the datetime.rst change was not applicable
to 3.5).

The convention at the time was to keep the 3.5 branch merged into
Default (Master). That is why my 3.5 backport appears in your history
of Master.

> so let's skip it:
>
> $ git log --oneline  --no-merges 0be894b2f6.. -- Lib/unittest/mock.py
> Lib/unittest/test/testmock/ | tail -n 3
> 161a4dd495 Issue #28919: Simplify _copy_func_details() in unittest.mock
> ac5084b6c7 Fixes issue28380: unittest.mock Mock autospec functions now
> properly support assert_called, assert_not_called, and assert_called_once.
> 15f44ab043 Issue #27895:  Spelling fixes (Contributed by Ville Skyttä).
>
> Wat?! Why is 15f44ab043 showing up again?!

Because you are asked for all the revisions except my backport and its
ancestors. As far as Git is concerned, the original spelling fixes are
not an ancestor of my backport.

I don’t have a copy of the Git repository to try, but I suggest the
following command is what you want:

git log --oneline  --no-merges HEAD ^15f44ab043 ^0be894b2f6 --
Lib/unittest/mock.py Lib/unittest/test/testmock/
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


[Python-Dev] git history conundrum

2019-04-27 Thread Chris Withers

Hi All,

I'm in the process of bringing the mock backport up to date, but this 
has got me stumped:


$ git log --oneline  --no-merges 
5943ea76d529f9ea18c73a61e10c6f53bdcc864f.. -- Lib/unittest/mock.py 
Lib/unittest/test/testmock/ | tail
362f058a89 Issue #28735: Fixed the comparison of mock.MagickMock with 
mock.ANY.
d9c956fb23 Issue #20804: The unittest.mock.sentinel attributes now 
preserve their identity when they are copied or pickled.

84b6fb0eea Fix unittest.mock._Call: don't ignore name
161a4dd495 Issue #28919: Simplify _copy_func_details() in unittest.mock
ac5084b6c7 Fixes issue28380: unittest.mock Mock autospec functions now 
properly support assert_called, assert_not_called, and assert_called_once.

0be894b2f6 Issue #27895:  Spelling fixes (Contributed by Ville Skyttä).
15f44ab043 Issue #27895:  Spelling fixes (Contributed by Ville Skyttä).
d4583d7fea Issue #26750: use inspect.isdatadescriptor instead of our own 
_is_data_descriptor().
9854789efe Issue #26750: unittest.mock.create_autospec() now works 
properly for subclasses of property() and other data descriptors.

204bf0b9ae English spelling and grammar fixes

Right, so I've merged up to 15f44ab043, what comes next?

$ git log --oneline  --no-merges 15f44ab043.. -- Lib/unittest/mock.py 
Lib/unittest/test/testmock/ | tail -n 3

161a4dd495 Issue #28919: Simplify _copy_func_details() in unittest.mock
ac5084b6c7 Fixes issue28380: unittest.mock Mock autospec functions now 
properly support assert_called, assert_not_called, and assert_called_once.

0be894b2f6 Issue #27895:  Spelling fixes (Contributed by Ville Skyttä).

Okay, no idea why 0be894b2f6 is there, appears to be a totally identical 
commit to 15f44ab043, so let's skip it:


$ git log --oneline  --no-merges 0be894b2f6.. -- Lib/unittest/mock.py 
Lib/unittest/test/testmock/ | tail -n 3

161a4dd495 Issue #28919: Simplify _copy_func_details() in unittest.mock
ac5084b6c7 Fixes issue28380: unittest.mock Mock autospec functions now 
properly support assert_called, assert_not_called, and assert_called_once.

15f44ab043 Issue #27895:  Spelling fixes (Contributed by Ville Skyttä).

Wat?! Why is 15f44ab043 showing up again?!

What's the git subtlety I'm missing here?

Chris
___
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com