Allow propcache objects to be utilized in cython

[Vizonex]

What do these changes do?

After doing a bit of research and digging I've discovered that it is possible to use propcache's member descriptors in Cython and I even wrote a little code snippet to prove it works. I think some low level programmers may take interest in this feature and I am looking to use it myself in my own projects but this might be the last thing needed if we were looking to embed yarl's URL object more deeply in Cython (Which is one of my goals which was to cimport the URL object to Cython to improve performance of aiohttp's http_parser, not to mention the new Multidict CAPI that I am really excited to start using in Cython).

# test_snippet.pyx
from propcache cimport under_cached_property

cdef class CachedItem:
    cdef int i
    # as long as _cache is public, under_cached_property works every time...
    cdef public dict _cache

    def __init__(self, int i) -> None:
        self.i = i
        self._cache = {}
    
    @under_cached_property
    def real_i(self):
        return f"<{self.i}>"

from test_snippet import CachedItem

i = CachedItem(43)
assert i.real_i == "<43>"
print("it works!!!")

Are there changes in behavior for the user?

Cython Programmers can now utilize propcache. This would benefit projects like yarl or other low level libraries that may use cached properties. I'll see about documenting these features here in a bit if we decide this is something new to explore. I don't want to flush my work down the toilet after writing documentation if this idea is rejected.

Related issue number

Checklist

• I think the code is well written
• Unit tests for the changes exist
• Documentation reflects the changes

[codspeed-hq]

CodSpeed Performance Report

Merging #130 will not alter performance

_{Comparing Vizonex:cython-support (e324901) with master (3a8c2d0)}

Summary

✅ 4 untouched

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 97.67%. Comparing base (3a8c2d0) to head (e324901).
⚠️ Report is 22 commits behind head on master.

❌ Your project status has failed because the head coverage (97.44%) is below the target coverage (98.20%). You can increase the head coverage or adjust the target coverage.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #130      +/-   ##
==========================================
+ Coverage   97.63%   97.67%   +0.03%     
==========================================
  Files          17       18       +1     
  Lines         847      861      +14     
  Branches       44       44              
==========================================
+ Hits          827      841      +14     
  Misses         12       12              
  Partials        8        8              

Flag	Coverage Δ
CI-GHA	97.67% <100.00%> (+0.03%)	⬆️
MyPy	90.91% <35.71%> (-0.93%)	⬇️
OS-Linux	97.56% <100.00%> (+0.07%)	⬆️
OS-Windows	94.69% <100.00%> (+0.16%)	⬆️
OS-macOS	94.69% <100.00%> (+0.16%)	⬆️
Py-3.10.11	92.92% <100.00%> (+0.22%)	⬆️
Py-3.10.18	94.69% <100.00%> (+0.16%)	⬆️
Py-3.10.19	?
Py-3.11.13	96.23% <100.00%> (+0.12%)	⬆️
Py-3.11.14	?
Py-3.11.9	94.46% <100.00%> (+0.17%)	⬆️
Py-3.12.10	94.46% <100.00%> (+0.17%)	⬆️
Py-3.12.11	96.23% <100.00%> (+0.12%)	⬆️
Py-3.12.12	?
Py-3.13.5	94.46% <100.00%> (+0.17%)	⬆️
Py-3.13.7	96.23% <100.00%> (+0.12%)	⬆️
Py-3.13.8	?
Py-3.13.9	94.46% <100.00%> (-1.65%)	⬇️
Py-3.13.9t	96.23% <100.00%> (+0.12%)	⬆️
Py-3.14.0	96.20% <100.00%> (+0.12%)	⬆️
Py-3.14.0t	96.20% <100.00%> (+0.12%)	⬆️
Py-3.9.13	92.90% <100.00%> (+0.22%)	⬆️
Py-3.9.23	94.67% <100.00%> (+0.17%)	⬆️
Py-3.9.24	?
Py-pypy3.10.16-7.3.19	76.10% <14.28%> (-1.98%)	⬇️
Py-pypy3.9.19-7.3.16	76.05% <14.28%> (-1.98%)	⬇️
VM-macos-latest	94.69% <100.00%> (+0.16%)	⬆️
VM-ubuntu-latest	97.56% <100.00%> (+0.07%)	⬆️
VM-windows-11-arm	94.46% <100.00%> (+0.17%)	⬆️
VM-windows-latest	94.69% <100.00%> (+0.16%)	⬆️
pytest	97.56% <100.00%> (+0.07%)	⬆️

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

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Vizonex]

I'm going to close this for now I might rethink about this in the future but for now my desire for cython to have this is kinda dead. However if someone needs this in cython I might just let people know that if you want it copying it off should be doable and to give citations as to where it came from.