gh-136282: Configparser: create unnamed sections via mapping protocol access

[Rogdham]

Using the mapping protocol access to create unnamed section was not supported.

With this PR you can use it:

>>> cfg = configparser.ConfigParser(allow_unnamed_section=True)
>>> cfg[configparser.UNNAMED_SECTION] = {"foo": "bar"}
>>> cfg.write(fobj)

• Issue: configparser: create unnamed section with mapping protocol access #136282

[Rogdham]

I have opened this PR as a feature request, but if you think @jaraco that this should be a bugfix (to be included in 3.14) for gh-123049, tell me and I will open an alternative PR accordingly!

[ZeroIntensity]

If we do decide to backport (which I don't think we should), we can automatically backport to 3.14 with the bot. No need for an extra PR.

[Rogdham]

we can automatically backport to 3.14 with the bot. No need for an extra PR

Yes you are right! I meant to reference gh-123049 and not gh-136282 in that case, but I guess it does not matter much at the end of the day

[merwok]

I would see this as a bugfix, I can’t imagine someone using mapping protocol + unnamed section and wanting the output [<UNNAMED_SECTION>].

PR LGTM

[merwok]

@encukou what do you think of fix vs feature classification? (with the consequence of backporting or not)

[ZeroIntensity]

I don't remember my reasons for objecting to this as a bugfix -- if both of you are happy with backporting, let's do that.

[encukou]

@hugovk is the authority here.

I'm leaning towards bugfix.

[hugovk]

Bugfix is fine by me.

[miss-islington-app]

Thanks @Rogdham for the PR, and @encukou for merging it 🌮🎉.. I'm working now to backport this PR to: 3.14.
🐍🍒⛏🤖

[bedevere-app]

GH-142978 is a backport of this pull request to the 3.14 branch.