Repository navigation
The docstring of dict() should match the doc of dict() and it should be more understandable #137629
Closed as duplicate
Description
The docstring of dict() doesn't match the doc of dict() and it isn't so understandable as shown below:
print(help(dict))
# class dict(object)
# | dict() -> new empty dictionary
# | dict(mapping) -> new dictionary initialized from a mapping object's
# | (key, value) pairs
# | dict(iterable) -> new dictionary initialized as if via:
# | d = {}
# | for k, v in iterable:
# | d[k] = v
# | dict(**kwargs) -> new dictionary initialized with the name=value pairs
# | in the keyword argument list. For example: dict(one=1, two=2)So, it should be like below:
dict(**kwargs) -> E.g. dict(key1=value1, key2=value2)
dict(mapping, **kwargs) ->mappingis the dictionary made by mapping object
dict(iterable, **kwargs) ->iterableis e.g. [('key1', 'value1'), ('key2', 'value2')]) treated by {k:v for k, v in iterable}
Activity
picnixz commentedon Aug 11, 2025
I think the docstring is clear enough. It gives sufficient examples.
AA-Turner commentedon Aug 11, 2025
@hyperkai please avoid opening lots of new issues for related topics. In the last few months you've opened over 40 issues, the many of which maintainers have closed with no action needed.
Your suggestions and observations are valuable, but please combine these into fewer issues. For example the recent str/bytes/bytearray issues could have been in one meta-issue, or the recent few dict issues coudl have been in a single combined topic.
I would also suggest that you use a lower modality in your phrasing -- saying that "X should change" is instructive, and can be off-putting to the reader, especially when maintainers assess the issue and determine that no change is needed (as @picnixz suggests in this case).
A
serhiy-storchaka commentedon Aug 11, 2025
This will be done as a side effect of #73536.
picnixz commentedon Sep 2, 2025
Closing as per Serhiy's comment.