Draft: Use StrEnum and unique for chroottype

• chroottype: enforce unique enum and strenum

added 1 commit

• 5370a4c4 - chroottype: enforce unique enum and strenum

Compare with previous version

changed title from anjan/strenum to Use StrEnum and unique for chroottype

changed the description

added 1 commit

• 2ad58aad - chroottype: enforce unique enum and strenum

Compare with previous version

added 1 commit

• 1df4dce5 - chroottype: enforce unique enum and strenum

Compare with previous version

Could you write some brief description justifying this change? Is there some specific benefit or what motivated it?

seems totally sensible just trying to understand the context a bit

In python 3.11, StrEnums were introduced. I want to use StrEnums as they are just more concise and you dont need to call .value everywhere or define str. Since 3.10 doesnt support StrEnum, you can make the equivalent of StrEnum by inheriting from str and Enum. However, there is this issue with the str and Enum inherit:

https://github.com/python/cpython/issues/100458

Your call, I wanted to open this MR to see what you thought would be the best approach. All that said, I do think we should always use the @unique decorator on this and other enums which are unique. It makes scanning the enum and understanding the == operator for that enum easier.

I had to add a __str__ overload in mrhlpr because it wasn't using the actual value of the enum but rather the name of it for the string representation even though it inherited from str: mrhlpr@d49a1ccf

Is this different somehow?

Ya, you will need to add __str__ overload to this anyway on python 3.10. Newer python interpreters dont have this issue. Closing this issue....Lets wait till we use 3.11 and we can just use strenums

added 69 commits

• 1df4dce5...a8283033 - 68 commits from branch postmarketOS:master
• 883d3e3c - chroottype: enforce unique enum and strenum

Compare with previous version

marked this merge request as draft

Marking as draft to get this out of the review queue as !2486 (comment 457065) hasn't been answered.

closed