Fix a bug in HyperFeatureRequester::get_all()

[aalexandrov]

The FeatureRequester::get_all() implementation for the HyperFeatureRequester is currently reading the entire response (as a test try a payload that is more than 8000 bytes long).

Requirements

• I have added test coverage for new or changed functionality
• I have followed the repository's pull request submission guidelines
• I have validated my changes against all supported platform versions

Related issues

Provide links to any issues in this repository or elsewhere relating to this pull request.

Describe the solution you've provided

Provide a clear and concise description of what you expect to happen.

Describe alternatives you've considered

Provide a clear and concise description of any alternative solutions or features you've considered.

Additional context

Add any other context about the pull request here.

[keelerm84]

My apologies for the delay on this PR. There are a few other matters I still have to attend to this week, but I will try to get some resolution on this for you by the end of the week.

Thank you for your patience.

[aalexandrov]

@keelerm84 It's not urgent on our side, it's just something I noticed when I was trying something out.

[keelerm84]

@aalexandrov Thanks for letting me know this wasn't urgent as it gave me some breathing room. I appreciate that!

I was looking into this change this morning, and I was curious what issue you are trying to solve. You mention in the original write-up that I should:

as a test try a payload that is more than 8000 bytes long

with the default rust SDK version, this seems to work fine for my payload (~12k bytes). Were you experiencing a memory or CPU spike?

[aalexandrov]

I think I was trying to create a mock API endpoint that only implements the regular HTTP protocol (no streaming API) and returns a sizeable JSON response. I wasn't able to make my client SDK consume the entire response without this change.

[keelerm84]

Hmmm. I can't seem to reproduce this problem against our systems. I am using a payload that is around 17k bytes and it seems to work without issue.

Are you able to provide a reproduction case?

[aalexandrov]

I'm sorry, I think my mock Python server was on a WIP branch that has been lost.

[keelerm84]

No worries at all. However, without a clear reproduction case, I'm going to close this PR. Please do not hesitate to re-open this issue if you are able to reproduce this error.

Thank you for all of the time and effort you have invested!