Use json rather than flask.json to generate ETag

[lafrech]

Historically, we used flask.json because it serializes lazy strings but this should not be the responsibility of this framework.

Then, we decided not to force sort_keys for consistency with the way flask dumps the payload (#305). But there is no reason to do that for the "explicit ETag" (arbitrary data) case. And even if the app doesn't sort keys, doing it for ETag computation makes sense.

It is not easy to provide a one-size-fits-all solution. User is free to override _generate_etag for specific cases (specific json serializer, sort_keys=False,...).

Things can probably be improved in a wider ETag rework. Meanwhile, this change simplifies the code and fixes compatibility with Flask 2.2+ without breaking compatibility with Flask <2.2.

[codecov]

Codecov Report

Merging #393 (a136e30) into master (d508476) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master     #393   +/-   ##
=======================================
  Coverage   99.87%   99.87%           
=======================================
  Files          13       13           
  Lines         816      817    +1     
  Branches      178      178           
=======================================
+ Hits          815      816    +1     
  Partials        1        1           

Impacted Files	Coverage Δ
flask_smorest/etag.py	100.00% <100.00%> (ø)

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.