New submission from Glenn Gribble <[email protected]>:
At present, it is not possible to use the shorthand notation to define a
NamedTuple with typename or fields. I.e., NamedTuple('MyTuple', typename=str,
fields=int) does not work. Changing the parameter names to _typename and
_fields would allow any non-private, legal identifier to be used in the
shorthand notation.
>>> import typing
>>> typing.NamedTuple('Example', fieldz=int)
<class '__main__.Example'>
>>> typing.NamedTuple('Example2', fields=int)
Traceback (most recent call last):
File "<input>", line 1, in <module>
File "C:\Program Files\Python37\lib\typing.py", line 1411, in __new__
return _make_nmtuple(typename, fields)
File "C:\Program Files\Python37\lib\typing.py", line 1326, in _make_nmtuple
types = [(n, _type_check(t, msg)) for n, t in types]
TypeError: 'type' object is not iterable
Of course, it is fairly easy to work around the issue by using fields parameter:
>>> typing.NamedTuple('Example3', [('fields', int)])
<class '__main__.Example3'>
There would be backwards compatibility issues with any code using named
arguments for fields or typename.
----------
components: Library (Lib)
messages: 352579
nosy: gribbg
priority: normal
severity: normal
status: open
title: typing.NamedTuple() should prefix parameters with '_'
type: enhancement
versions: Python 3.7, Python 3.8, Python 3.9
_______________________________________
Python tracker <[email protected]>
<https://bugs.python.org/issue38191>
_______________________________________
_______________________________________________
Python-bugs-list mailing list
Unsubscribe:
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
