Change back to a streaming method for web server responses#730
Merged
Change back to a streaming method for web server responses#730
Conversation
PR #729 aimed to improve performance by removing the use of stream_with_context from the response handling. While this did improve performance it meant that we're now hitting the 32mb response size limit that cloud run imposes for non-streaming responses. This update reintroduces a streaming response, but using a different method that doesn't have the same small chunk size that was causing the performance issue previously. The docs on this are pretty minimal but this is pulling together some info from both the flask docs on streaming: https://flask.palletsprojects.com/en/stable/patterns/streaming/ and the fast api docs on streaming a file-like binary response: https://fastapi.tiangolo.com/advanced/custom-response/#using-streamingresponse-with-file-like-objects
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
dancoates
commented
Mar 26, 2025
|
|
||
| @app.route('/<dataset>/<path:filename>') | ||
| def handler( | ||
| def handler( # noqa: C901 |
Contributor
Author
There was a problem hiding this comment.
Adding the extra iterfile function took this over an arbitrary complexity limit, to fix would need to refactor out some of the other code in the handler function, not keen to do that in this PR as it would make the diff and intent of the change unclear.
violetbrina
approved these changes
Mar 26, 2025
Collaborator
violetbrina
left a comment
violetbrina
left a comment
There was a problem hiding this comment.
Thanks for the links in the description. Made this a super fast review. Conceptually the returning the generator makes sense. Hope this works!
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
PR #729 aimed to improve performance by removing the use of stream_with_context from the response handling. While this did improve performance it meant that we're now hitting the 32mb response size limit that cloud run imposes for non-streaming responses.
This update reintroduces a streaming response, but using a different method that doesn't have the same small chunk size that was causing the performance issue previously. The docs on this are pretty minimal but this is pulling together some info from both the flask docs on streaming:
https://flask.palletsprojects.com/en/stable/patterns/streaming/
and the fast api docs on streaming a file-like binary response: https://fastapi.tiangolo.com/advanced/custom-response/#using-streamingresponse-with-file-like-objects