Complete removal of SplitResult as an internal

[bdraco]

partially implements #172

It was suggested in #172 to only upgrade to SplitResult when needed. This will give us more flexibility to change how the internal structure works in the future and we won't be forever bound to using SplitResult.

The only place we can't fully get rid of it is for pickle without a breaking change. #172 suggested leaving it for compat, so thats what I did.

In the future we may want to change the split functions to unpack the values into named variables in future. ie self._scheme, self._netloc, self._path, self._query, self._fragment = val or store everything in self._cache as the single source of truth; However that was too large of a redesign for this PR since it would also require changing all the tuple unpacking and pre cache in __new__.

One consideration about getting rid of the _val tuple is that SplitResult is exposed as an allowed incoming value in new we would have to have code to convert it. With _val being a normal tuple, SplitResult can be passed in and its still compatible.

It would be a lot nicer if we stored everything in self._cache though since it means the encode_url and pre_encoded_url calls would return a simple dict that was pre-populated, and it also means propcache would have more keys pre-populated.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 96.01%. Comparing base (24f814d) to head (facbe5b).
Report is 1 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #1396   +/-   ##
=======================================
  Coverage   96.00%   96.01%           
=======================================
  Files          31       31           
  Lines        5683     5692    +9     
  Branches      353      353           
=======================================
+ Hits         5456     5465    +9     
  Misses        201      201           
  Partials       26       26           

Flag	Coverage Δ
CI-GHA	96.01% <100.00%> (+<0.01%)	⬆️
MyPy	48.83% <63.75%> (+0.06%)	⬆️
OS-Linux	99.55% <100.00%> (+<0.01%)	⬆️
OS-Windows	99.61% <100.00%> (+<0.01%)	⬆️
OS-macOS	99.29% <100.00%> (+<0.01%)	⬆️
Py-3.10.11	99.27% <100.00%> (+<0.01%)	⬆️
Py-3.10.15	99.50% <100.00%> (+<0.01%)	⬆️
Py-3.11.10	99.50% <100.00%> (+<0.01%)	⬆️
Py-3.11.9	99.27% <100.00%> (+<0.01%)	⬆️
Py-3.12.7	99.50% <100.00%> (+<0.01%)	⬆️
Py-3.13.0	99.50% <100.00%> (+<0.01%)	⬆️
Py-3.9.13	99.22% <100.00%> (+<0.01%)	⬆️
Py-3.9.20	99.46% <100.00%> (+<0.01%)	⬆️
Py-pypy7.3.16	99.52% <100.00%> (+<0.01%)	⬆️
Py-pypy7.3.17	99.54% <100.00%> (+<0.01%)	⬆️
VM-macos-latest	99.29% <100.00%> (+<0.01%)	⬆️
VM-ubuntu-latest	99.55% <100.00%> (+<0.01%)	⬆️
VM-windows-latest	99.61% <100.00%> (+<0.01%)	⬆️
pytest	99.55% <100.00%> (+<0.01%)	⬆️

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.

[codspeed-hq]

CodSpeed Performance Report

Merging #1396 will improve performances by ×2.2

_{Comparing get_rid_splitresult (facbe5b) with master (24f814d)}

Summary

⚡ 34 improvements
✅ 48 untouched benchmarks

Benchmarks breakdown

	Benchmark	master	get_rid_splitresult	Change
⚡	test_update_query_empty	125.9 µs	94.3 µs	+33.57%
⚡	test_update_query_none	122.7 µs	90.9 µs	+35.03%
⚡	test_url_build_access_fragment	607.8 µs	507.1 µs	+19.86%
⚡	test_url_build_access_raw_host	717.6 µs	616.6 µs	+16.37%
⚡	test_url_build_access_raw_path	666.5 µs	564.4 µs	+18.1%
⚡	test_url_build_access_username_password	799.8 µs	697.8 µs	+14.61%
⚡	test_url_build_encoded_with_host_and_port	500.6 µs	405.3 µs	+23.52%
⚡	test_url_build_no_netloc	448.4 µs	356.9 µs	+25.63%
⚡	test_url_build_no_netloc_relative	447.3 µs	354.3 µs	+26.27%
⚡	test_url_build_with_different_hosts	1.3 ms	1 ms	+22.87%
⚡	test_url_build_with_host	504.1 µs	410 µs	+22.96%
⚡	test_url_build_with_host_and_port	717.7 µs	617.6 µs	+16.21%
⚡	test_url_build_with_host_path_and_port	636.6 µs	537.9 µs	+18.35%
⚡	test_url_equality	780.1 µs	358 µs	×2.2
⚡	test_url_extend_query_existing_query_string	224.3 µs	194 µs	+15.65%
⚡	test_url_hash	323.3 µs	195 µs	+65.77%
⚡	test_url_join	469.2 µs	320.4 µs	+46.43%
⚡	test_url_joinpath_encoded	871.8 µs	751.4 µs	+16.01%
⚡	test_url_joinpath_with_truediv	1,012.9 µs	883.1 µs	+14.7%
⚡	test_url_to_string	308 µs	239.8 µs	+28.45%
...	...	...	...	...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

[bdraco]

For the future we could store everything in self._cache as the single source of truth.

That would get rid of the tuple.. but thats a major change, and since SplitResult is
exposed as an allowed incoming value in new we would have to have code to convert it

It would be a lot nicer though since it means the encode_url and unencoded_url calls
would return a simple dict that was pre-populated, and it also means propcache
would have more keys pre-populated

[bdraco]

#1397 Aims to fully eliminate Val and only construct it lazily