Add support for batching

[patrick91]

This PR adds support for batching. I want to make sure all the views get this :)

TODO:

• Consolidate all the views (this can be done in another PR if I don't manage to have enough time)
• Allow to disable batching
• Support of async (with gather)

Closes #1383

[codecov]

Codecov Report

Merging #1798 (13949fd) into main (c64a9f9) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##             main    #1798   +/-   ##
=======================================
  Coverage   98.15%   98.15%           
=======================================
  Files         139      139           
  Lines        5252     5262   +10     
  Branches      955      960    +5     
=======================================
+ Hits         5155     5165   +10     
  Misses         52       52           
  Partials       45       45           

[justuswilhelm]

This looks good. Looking forward to seeing this get merged.

[patrick91]

hi @justuswilhelm! that's good to know, do you need this for django or other frameworks? 😊

[justuswilhelm]

hi @justuswilhelm! that's good to know, do you need this for django or other frameworks? 😊

Thanks for your fast response. Yes, I am using strawberry in django together with apollo in the frontend

[camflan]

@patrick91 I'd love to get this merged. Is there anything I can help with?

[patrick91]

@camflan I'll try to work on this in the next few weeks, I'm a bit busy and don't have a lot of free time :D

Good to know there's people that need this 😊

[camflan]

@camflan I'll try to work on this in the next few weeks, I'm a bit busy and don't have a lot of free time :D

Good to know there's people that need this 😊

Understood! We've started an initiative to get off graphene and this will be one of our blockers for rolling out to clients. If there's anything I can do to help please let me know

[patrick91]

@camflan thanks for offering! Please don't hesitate to ping me in discord if I forget :)

What integration are you using? :)

[patrick91]

@camflan I'll think about how we can enable this feature using schema extensions, if I don't find a good solution I'll keep working on the solution from this PR

[camflan]

@camflan thanks for offering! Please don't hesitate to ping me in discord if I forget :)

What integration are you using? :)

We'd be mostly using a django integration

[patrick91]

@camflan I'm working on refactoring the HTTP view layer before adding this. Shouldn't take too long 😊

[camflan]

@camflan I'm working on refactoring the HTTP view layer before adding this. Shouldn't take too long 😊

Famous last words 🙈 😅

thanks for the update!

[patrick91]

@camflan that's been merged! I'll work on batching soon :)

Do you have any requirements? I need to think about security requirements for this, for now I think I'll add the following options:

1. Enable disable batching
2. Batching limit (max x operations)

but I wonder if we should put more :)

[camflan]

@camflan that's been merged! I'll work on batching soon :)

Yay! 🥳

Do you have any requirements? I need to think about security requirements for this, for now I think I'll add the following options:

1. Enable disable batching

2. Batching limit (max x operations)

I think both of those would cover what we need for now. Batching limit, rate limiting, etc could even be tackled later as a middleware instead of the core implementation, right?

[patrick91]

I think both of those would cover what we need for now. Batching limit, rate limiting, etc could even be tackled later as a middleware instead of the core implementation, right?

depends on what you mean by middlewares, Strawberry extensions at the moment only work on a single operation.

But you'll be able to extend the GraphQL view/router to do additional validation. I think there's going to be a method called run_batch that will validate and execute the batching (and maybe a validate_batch method that can be used to change the validation if needed)

[patrick91]

closed in favour of #2698