Skip to content

Conversation

kikihakiem
Copy link

This PR

Adds Unleash Provider for OpenFeature Python SDK

Related Issues

Fixes #300
Fixes #3896 - Addresses community request for OpenFeature support in Unleash ecosystem

Notes

Follow-up Tasks

How to test

  1. Install dependencies: uv sync
  2. Run unit tests: uv run test -m "not integration"
  3. Run integration tests (requires Docker): uv run test -m integration
  4. Run type checking: uv run mypy-check
  5. Run linting: uv run ruff check

@kikihakiem kikihakiem requested review from a team as code owners September 2, 2025 08:14
Copy link

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Summary of Changes

Hello @kikihakiem, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request delivers a new OpenFeature provider specifically designed for Unleash, a popular feature management platform. Its primary purpose is to bridge the gap between OpenFeature's standardized API and Unleash's feature flagging capabilities for Python applications. This integration empowers developers to manage feature rollouts and experiments seamlessly, leveraging both OpenFeature's interoperability and Unleash's robust feature management features. The implementation is thorough, covering various flag types, event handling, and tracking, all backed by a comprehensive test suite.

Highlights

  • New Provider Added: Introduced a new OpenFeature provider for Unleash, enabling Python applications to integrate with Unleash for feature flagging and experimentation.
  • Community Request Fulfilled: This addition directly addresses community requests for OpenFeature support within the Unleash ecosystem.
  • Comprehensive Functionality: The provider supports boolean, string, integer, float, and object flag types, along with OpenFeature event handling and tracking capabilities.
  • Robust Testing: Includes extensive unit and integration tests, ensuring reliability and correctness of the Unleash provider.
  • Development Tooling: New pyproject.toml entries and utility scripts are added to streamline development, testing, type checking, and linting workflows.
Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.

Feature Command Description
Code Review /gemini review Performs a code review for the current pull request in its current state.
Pull Request Summary /gemini summary Provides a summary of the current pull request in its current state.
Comment @gemini-code-assist Responds in comments when explicitly tagged, both in issue comments and review comments.
Help /gemini help Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

  1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution.

Copy link

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This PR adds a new Unleash provider for the OpenFeature Python SDK. The implementation is well-structured and includes a comprehensive set of unit and integration tests, which is great to see. I've left a few comments with suggestions for improvement, mainly around security best practices in helper scripts and removing some potentially unreachable error handling code for better maintainability. Overall, this is a solid contribution.

Comment on lines 79 to 82
except requests.exceptions.HTTPError as e:
if e.response and e.response.status_code == 404:
raise FlagNotFoundError(f"Flag not found: {flag_key}") from e
raise GeneralError(f"HTTP error: {e}") from e

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

This except requests.exceptions.HTTPError block is likely unreachable. The UnleashClient.is_enabled() method evaluates flags based on its in-memory cache and does not perform network requests that would raise an HTTPError. Network errors during feature fetching are handled by the client's background poller. This exception handling can be removed to simplify the code.

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think this is an accurate comment. Unleash should not be making HTTP calls during is_enabled function. I'd remove this error handling.

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thank you for the assurance that a network error is unlikely. I'll remove this error handling.

Comment on lines 227 to 230
except requests.exceptions.HTTPError as e:
if e.response and e.response.status_code == 404:
raise FlagNotFoundError(f"Flag not found: {flag_key}") from e
raise GeneralError(f"HTTP error: {e}") from e

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

Similar to resolve_boolean_details, this except requests.exceptions.HTTPError block is likely unreachable because UnleashClient.get_variant() operates on the in-memory cache. This block can be removed for simplification.

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Also accurate

Signed-off-by: Kiki L Hakiem <[email protected]>
…eger_details, resolve_float_details and resolve_object_details

Signed-off-by: Kiki L Hakiem <[email protected]>
Signed-off-by: Kiki L Hakiem <[email protected]>
Signed-off-by: Kiki L Hakiem <[email protected]>
Signed-off-by: Kiki L Hakiem <[email protected]>
Signed-off-by: Kiki L Hakiem <[email protected]>
@kikihakiem kikihakiem force-pushed the feat/unleash-provider branch from d5f7e81 to 9a579d6 Compare September 2, 2025 08:19
Copy link

@gastonfournier gastonfournier left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for the contribution! I think it's heading in the right direction, but from Unleash design perspective, there are a few things that should be done differently. I believe it's a relatively simple change that will make the provider much better.

Comment on lines 79 to 82
except requests.exceptions.HTTPError as e:
if e.response and e.response.status_code == 404:
raise FlagNotFoundError(f"Flag not found: {flag_key}") from e
raise GeneralError(f"HTTP error: {e}") from e

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think this is an accurate comment. Unleash should not be making HTTP calls during is_enabled function. I'd remove this error handling.

Comment on lines 227 to 230
except requests.exceptions.HTTPError as e:
if e.response and e.response.status_code == 404:
raise FlagNotFoundError(f"Flag not found: {flag_key}") from e
raise GeneralError(f"HTTP error: {e}") from e

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Also accurate

self.app_name = app_name
self.api_token = api_token
self.environment = environment
self.client: Optional[UnleashClient] = None

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I don't think the client should be Optional. It's safe to initialize UnleashClient here, so you'd remove a lot of complexity of handling this being optional.

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I deliberately make UnleashClient optional and separate UnleashClient initialization from UnleashProvider constructor. It allows us to follow the OpenFeature provider lifecycle (NOT_READY -> READY -> ERROR). It also allows unit tests to create providers without immediately connecting to Unleash. Let me know if this is not the case with Unleash or if we don't really need distinction between NOT_READY and READY states.

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think the instantiation of UnleashClient does not make it ready. It will switch to ready when it receives the first update from upstream (or loads from bootstrap or backup). Not having it optional would simplify the design and should still be aligned with OF lifecycle. You can find Unleash documentation here: https://github.com/Unleash/unleash-python-sdk/?tab=readme-ov-file#lifecycle-events

Apparently, you can't disable fetching at startup with this SDK, but you shouldn't carry that design choice to this connector. I think you should use the bootstrapping mechanism to have a stable set of features and keep an invalid Unleash API url, which will fail with 404. Increasing the fetch interval could also help, but not with the initial fetch (if the docs are up to date)

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

thank you for your input. I've updated the Unleash lifecycle event handling. Please take a look

Comment on lines 186 to 187
if not self._provider.client:
raise GeneralError("Provider not initialized. Call initialize() first.")

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

is_enabled and resolve_variant paths should never throw exceptions. This is one of the design principles at Unleash. This would be resolved by removing the optional as in the other comment, and removing the exception handling from this method and is_enabled

@kikihakiem kikihakiem force-pushed the feat/unleash-provider branch from 01b3e09 to 048029b Compare September 3, 2025 07:57
@kikihakiem kikihakiem force-pushed the feat/unleash-provider branch 3 times, most recently from 1931868 to 7e02cb3 Compare September 10, 2025 12:20
@kikihakiem kikihakiem force-pushed the feat/unleash-provider branch from 7e02cb3 to 4133039 Compare September 10, 2025 12:28
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

[provider] Support Unleash provider
2 participants