properly deprecate the current behavior of the default_format option

[keewis]

As discussed in #1407, this rolls back the change in behavior of default_format (split the spec in mspec and uspec and decide for each part individually if the default should be used).

I also decided that this would be a good opportunity to rewrite the code such that mspec and uspec are separated as early as possible, such that fixing #1413 becomes a really small change.

• Closes pint 0.18 does not allow to explicitly override "~" modifier in default_format #1407
• Executed pre-commit run --all-files with no errors
• The change is fully covered by automated unit tests
• Documented in docs/ as appropriate
• Added an entry to the CHANGES file

[keewis]

apparently, there's a issue with installing uncertainties==3.0.1.

Also, I just noticed that UnitRegistry.default_format does not have a documentation entry, and that that's actually a property that sets class attributes on Quantity, Unit, and Measurement.

[jules-ch]

Yes I will make PR to resolve the error for uncertainties

[jules-ch]

Should be okay if you rebase @keewis

[keewis]

@hgrecco, this should be ready for reviewing and merging, after which we can issue the bugfix release we talked about in #1407 (maybe include #1422, too?)

[jules-ch]

Gonna look into this. but looks good so far !

[jules-ch]

@keewis can u rebase to fix conflicts

[keewis]

@hgrecco, @jules-ch, I think this should be ready for a final review. It would probably be good to include this, as well as #1422, in the bugfix release (which we should try to release soonish).

[jules-ch]

@keewis can you just change the condition that is duplicated to use if not mspec and default_mspec:
Idk if you received the notifications when I commented those lines.

[keewis]

@keewis can you just change the condition that is duplicated to use if not mspec and default_mspec:

sure

Idk if you received the notifications when I commented those lines.

I didn't (and I also can't see them anywhere), which is why I was confused when you posted #1450 (comment). Can you check if those comments are still pending and submit if so?

[jules-ch]

They are still pending, can't you see them? Weird
I can make those changes directly if you want.

[keewis]

if it says "pending" you have to submit the review

[jules-ch]

Oh I see, I thought those kind of review comments were automatically sent. My bad then.
Well I've just learned something it seems q_q

[hgrecco]

I like this PR and it is a great idea to split uspec and mspec early on