[issue27721] distutils strtobool returns 0 and 1 rather than False and True

[Nathaniel Manista]

Change by Nathaniel Manista :


--
nosy: +Nathaniel Manista

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue27721] distutils strtobool returns 0 and 1 rather than False and True

[SilentGhost]

Changes by SilentGhost :


--
resolution: not a bug -> wont fix
stage: patch review -> resolved

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue27721] distutils strtobool returns 0 and 1 rather than False and True

[Raymond Hettinger]

Raymond Hettinger added the comment:

Sorry Joseph, I concur that with Ned that we're best off leaving this alone 
(risk of breaking doctests with the changing repr, risk of fooling an overly 
specific type test, etc).  The code is not incorrect and does correspond with 
its docstring which claims 1 for true and 2 for false.

The sentiment is appreciated, but in general, the time to get APIs perfect is 
when it is initially released.  Afterwards, we really only want to make 
implementation improvements or API extensions until the API is flat-out broken 
or has major usability problems (even then, it always causes consternation to 
change the API).

--
nosy: +rhettinger
resolution:  -> not a bug
status: open -> closed

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue27721] distutils strtobool returns 0 and 1 rather than False and True

[Joseph Bane]

Joseph Bane added the comment:

Thank you for your feedback. I definitely understand your reasoning and 
commitment to not introducing breaking changes in externally facing codes. I 
also agree that this does fall slightly more on the side of aesthetic rather 
than functional changes. However, this change doesn't seem like it is 
significantly altering the interface of strtobool; if anything it is making it 
clearer and less surprising, especially for Python 2/3+ users.

I know, as someone in that category, having learned Python at version 2.5, that 
it was a bit surprising when I got back an integer from this function rather 
than the native True and False values. I also want to emphasize that this 
change is specifically targeting Python 3.6, not 2.7, 3.4, or 3.5. Making this 
small change in a not-yet-released version should be safe, but, of course, that 
is my outsider opinion. You absolutely have a valid point if I were asking that 
we change a released version and it is not something I take lightly.

Lastly, this is the kind of change I would hope I could make easily as a first 
time contributor to the project because of it being more of an aesthetic one. 
After all, isn't the whole point of making changes to a project about making it 
cleaner, modern, and clearer?

Thanks again. I really appreciate the feedback and discussion. P.S. If there is 
support to introduce this change, I will certainly update the tests and docs as 
pointed out.

--

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue27721] distutils strtobool returns 0 and 1 rather than False and True

[Ned Deily]

Ned Deily added the comment:

Joseph, thanks for taking an interest in improving Python and submitting this 
patch.  Alas, I don't think we should apply it.  There are always tradeoffs 
that have to be considered when changing existing code.  Distutils is one of 
the oldest and most external facing of Python's standard library modules.  
Distutils is a key dependency of many other critical third-party projects, like 
setuptools and pip. Also, Numpy subclasses and extends parts of Distutils.  
Many third-party projects are set up to work with multiple versions of Python 
so we always need to be careful about introducing changes in one version that 
might break that version independence.  On the one hand, this change is small 
and seems to be likely to be safe, although it is very difficult to know for 
sure how third-parties may currently be using strtobool() in their own code.  
On the other hand, what does making this change buy us and users?  The only 
thing I can see is a more modern return type and slightly cleaner code.
   In cases like that, we usually say that the status quo wins: we choose to 
not make changes that have no real benefit (other than a minor aesthetic one) 
but could possibly introduce an incompatibility.  Yes, if we were writing 
Distutils today, it would definitely return a boolean but that's not a good 
enough reason to change it now.  Other opinions?

--
nosy: +ned.deily

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue27721] distutils strtobool returns 0 and 1 rather than False and True

[Berker Peksag]

Berker Peksag added the comment:

Could you also update documentation in Doc/distutils/apiref.rst?

Also, test_strtobool in Lib/distutils/tests/test_util.py doesn't test the 
following case:

>>> from distutils.util import strtobool
>>> strtobool('x')
Traceback (most recent call last):
  File "", line 1, in 
  File "/home/berker/projects/cpython/default/Lib/distutils/util.py", line 
317, in strtobool
raise ValueError("invalid truth value %r" % (val,)) 
ValueError: invalid truth value 'x'

Bonus points if you add a test for that :)

Thanks!

--
nosy: +berker.peksag
stage: commit review -> patch review

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue27721] distutils strtobool returns 0 and 1 rather than False and True

[SilentGhost]

SilentGhost added the comment:

LGTM

--
stage:  -> commit review

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue27721] distutils strtobool returns 0 and 1 rather than False and True

[Joseph Bane]

Joseph Bane added the comment:

Thank you for the feedback. Please find an updated patch attached.

--
Added file: http://bugs.python.org/file44061/patch.diff

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue27721] distutils strtobool returns 0 and 1 rather than False and True

[SilentGhost]

SilentGhost added the comment:

Couple of things: your patch doesn't seem to apply cleanly for whatever reason 
(most likely because it's a patch against python2); the change from lowercasing 
to casefolding doesn't seem justified to me; the formatted value needs to be 
repr-formatted - though again - a change that's beyond the scope of this issue.

The tests are unaffected by this change, however.

--
nosy: +SilentGhost
type: enhancement -> behavior

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue27721] distutils strtobool returns 0 and 1 rather than False and True

[R. David Murray]

Changes by R. David Murray :


--
title: strtobool returns 0 and 1 rather than False and True -> distutils 
strtobool returns 0 and 1 rather than False and True

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue27721] distutils strtobool returns 0 and 1 rather than False and True

[R. David Murray]

Changes by R. David Murray :


--
versions: +Python 3.6 -Python 2.7, Python 3.5

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com