File Cache Control Headers Support

[ChihweiLHBird]

Related issue: #2441

Would like to get some feedback about the implementation.

The behavior was made to be similar to Werkzeug. But I think we can only keep the auto headers option (and remove last_modified) since it is a fact from the file stat and user can still override the it with a headers dictionary passed in.

[codecov]

Codecov Report

Merging #2447 (035bbd2) into main (2f90a85) will increase coverage by 0.000%.
The diff coverage is 88.888%.

@@            Coverage Diff            @@
##              main     #2447   +/-   ##
=========================================
  Coverage   87.246%   87.247%           
=========================================
  Files           60        60           
  Lines         5073      5089   +16     
  Branches       907       912    +5     
=========================================
+ Hits          4426      4440   +14     
  Misses         476       476           
- Partials       171       173    +2     

Impacted Files	Coverage Δ
sanic/response.py	89.230% <88.888%> (-0.155%)	⬇️

---

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 2f90a85...035bbd2. Read the comment docs.

[ChihweiLHBird]

Currently, the headers look like:

[ChihweiLHBird]

@sanic-org/framework @ahopkins May I have some feedbacks about the draft? Thanks!

[Tronic]

@ChihweiLHBird Please avoid force pushing on PRs. Use normal push so that the notification don't give an error about commits not being found.

[ChihweiLHBird]

@ChihweiLHBird Please avoid force pushing on PRs. Use normal push so that the notification don't give an error about commits not being found.

Thanks. Sorry about that, and I didn't know that before. I wanted to undo the commit because the change was messed up. Won't do it in the future PRs.

[ahopkins]

Marking it as a draft. Change it back when it is ready for final review 😎

[ahopkins]

This is a good first step, but I wonder if we should raise a warning or something if cache-control was set in the headers? Will merge as is, but we can think about this for the next round.

[ChihweiLHBird]

This is a good first step, but I wonder if we should raise a warning or something if cache-control was set in the headers? Will merge as is, but we can think about this for the next round.

How to? Maybe mention it in the docs? Or have a parameter called something like enable_auto_cache_control?