gh-108191: Add support of positional argument in SimpleNamespace constructor

[serhiy-storchaka]

SimpleNamespace({'a': 1, 'b': 2}) and SimpleNamespace([('a', 1), ('b', 2)]) are now the same as SimpleNamespace(a=1, b=2).

• Issue: Allow specifying SimpleNamespace attributes as a mapping/iterable in a positional argument #108191

---

📚 Documentation preview 📚: https://cpython-previews--108195.org.readthedocs.build/

[carljm]

This looks like a convenient option to provide in the constructor, with no downside, and established precedent in the dict constructor.

[erlend-aasland]

Thanks! The "polish" commit was a nice clean-up; it made the code more readable, IMO.

[ericsnowcurrently]

I have no real objections. I had considered doing this originally but figured it could wait. Here we are. 😄

[serhiy-storchaka]

I tried to implement SimpleNamespace.__replace__, and the code may be simpler with this feature:

args = (self.__dict__,)
return type(self)(*args, **kwargs)

instead of

d = {}
d.update(self.__dict__)
d.update(kwargs)
return type(self)(**d)

[serhiy-storchaka]

On other hand, I decided to use other way (#109175), which is more consistent with pickling and copying:

result = type(self)()
result.__dict__.update(self.__dict__)
result.__dict__.update(kwargs)
return result

So this is no longer a blocker.

[bedevere-app]

When you're done making the requested changes, leave the comment: I have made the requested changes; please review again.

[serhiy-storchaka]

I have made the requested changes; please review again.

[bedevere-app]

Thanks for making the requested changes!

@erlend-aasland, @ericsnowcurrently, @AA-Turner, @carljm: please review the changes made to this pull request.

[ericsnowcurrently]

LGTM

I left one comment about something mostly unrelated to this PR, so I don't consider it a blocker.

[serhiy-storchaka]

It would be good to get @rhettinger's feedback, but this PR already received many positive feedbacks, and I already invested in this issue so much, that I'm going to merge it anyway, even if initially I was not so interested in this.

[carljm]

Seems to be a reasonable amount of interest in this, and plenty of review approvals. Can we go ahead and merge for 3.13?

[carljm]

Looks like a merge conflict needs to be resolved.