fix: Parsing endpoint content types with semicolon separator

[expobrain]

This PR allows to extract the content type safely removing any extra parameter like charset or boundary.

[codecov]

Codecov Report

Merging #727 (aaffff3) into main (f79bccb) will not change coverage.
The diff coverage is 100.00%.

@@            Coverage Diff            @@
##              main      #727   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           49        49           
  Lines         1971      1978    +7     
=========================================
+ Hits          1971      1978    +7     

Impacted Files	Coverage Δ
openapi_python_client/parser/openapi.py	100.00% <100.00%> (ø)
openapi_python_client/utils.py	100.00% <100.00%> (ø)

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[emann]

Great fix, but IMO implementation is a bit heavy handed. I could in theory see us wanting to use the parsing in email to get type/subtype/parameters separately in the future, but better to keep this as simple as possible for now IMO. Open to discussing though!

[expobrain]

Great fix, but IMO implementation is a bit heavy handed. I could in theory see us wanting to use the parsing in email to get type/subtype/parameters separately in the future, but better to keep this as simple as possible for now IMO. Open to discussing though!

The rationale is that there's already a function in the standard library that performs the operation we need and provides the confidence that it does it in the safest and most tested way than us.

[dbanty]

It definitely feels weird to use the email module to handle headers.. but the builtin http code is pretty unusable. The RFC space for header parsing is... dizzying, so it's unclear to me whether mail semantics are the same as HTTP in every scenario—however, I took a peek at how httpx deals with this and they also use the email.Message class.

So, given that we trust httpx as our interface to the internet, I think we follow their lead here and keep the implementation that @expobrain wrote.