Add Python 3.11

[Dreamsorcerer]

No description provided.

[codecov]

Codecov Report

Merging #618 (ca1e8dc) into master (a8b169e) will increase coverage by 0.02%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #618      +/-   ##
==========================================
+ Coverage   99.63%   99.66%   +0.02%     
==========================================
  Files          36       36              
  Lines        3560     3560              
==========================================
+ Hits         3547     3548       +1     
+ Misses         13       12       -1     

Flag	Coverage Δ
unit	99.66% <100.00%> (+0.02%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
tests/ut/backends/test_memcached.py	100.00% <ø> (ø)
tests/ut/backends/test_redis.py	100.00% <ø> (ø)
tests/ut/test_base.py	100.00% <ø> (ø)
aiocache/backends/redis.py	99.16% <100.00%> (ø)
aiocache/base.py	99.23% <100.00%> (ø)
tests/acceptance/conftest.py	100.00% <100.00%> (ø)
tests/acceptance/test_decorators.py	100.00% <100.00%> (ø)
tests/acceptance/test_lock.py	100.00% <100.00%> (ø)
tests/performance/test_footprint.py	100.00% <100.00%> (ø)
tests/ut/test_lock.py	100.00% <100.00%> (ø)
... and 2 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 84d9263...ca1e8dc. Read the comment docs.

[a68366]

Hello!
The tests are failing because of this change in 3.11:

Changed Enum.format() (the default for format(), str.format() and f-strings) of enums with mixed-in types (e.g. int, str) to also include the class name in the output, not just the member’s key. This matches the existing behavior of enum.Enum.str(), returning e.g. 'AnEnum.MEMBER' for an enum AnEnum(str, Enum) instead of just 'MEMBER'.

The _build_key in 3.11 returns Keys.KEY instead of using this enum values.

If you just need to fix the tests, then something like this works: a68366#2 .

I'm not sure if just fixing the tests is enough though, since this change in python's behavior makes aiocache unreliable in an environment where 3.11 and lower versions coexist and enums are used for cache keys.

[Dreamsorcerer]

@a68366 Thanks a lot for digging that out. The actual key values are not important, right? Only that they are consistent. So, what if we just do str(key) in _build_key()? That should produce the same result on all Python versions.

It'll invalidate existing caches when people upgrade to 0.12, but shouldn't cause any further issues after that.

[Dreamsorcerer]

@a68366 If you could double check the changes look sensible, that would be great.

[a68366]

I'm not so familiar with aiocache codebase, but LGTM.