fix: omit segments from evaluation context in get_environment_flags #179

Copilot · 2025-10-31T19:28:34Z

Fix for issue #6051: Omit segments from evaluation context when get_environment_flags is called

This change addresses the issue where segments were unnecessarily included in the evaluation context when retrieving environment flags. Segments are only relevant for identity-specific evaluations, not for environment-wide flag evaluations.

Changes made:

Modified _get_environment_flags_from_document() to filter out segments from the evaluation context before passing it to the engine
Simplified implementation using copy() and pop() instead of dictionary comprehension
Uses consistent keyword argument style (context=...) for all engine.get_evaluation_result calls
Added comprehensive tests with proper fixtures and mocking

Testing & Quality:

✅ All existing tests pass (73 tests)
✅ All linters pass (black, isort, flake8, mypy)
✅ Code review feedback addressed

Impact:

Improves performance by reducing unnecessary data in the evaluation context for environment flags
Maintains backward compatibility - no API changes
Segments continue to work correctly for identity-specific evaluations

Ready for review!

Original prompt

This section details on the original issue you should resolve

<issue_title>get_environment_flags includes segments in evaluation context</issue_title>
<issue_description>See Flagsmith/flagsmith#6051.

In the mapper layer, omit segments from evaluation context when get_environment_flags is called.</issue_description>

<agent_instructions>use conventional-commit for PR title:
fix:

add a new test — perhaps by adding test_mappers.py

verify with pytest and pre-commit run -a</agent_instructions>

Comments on the Issue (you are @copilot in this section)

Fixes get_environment_flags includes segments in evaluation context #172

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.

Co-authored-by: khvn26 <979078+khvn26@users.noreply.github.com>

khvn26 · 2025-10-31T22:00:32Z

tests/test_flagsmith.py

+    flagsmith._evaluation_context = evaluation_context
+    flagsmith.enable_local_evaluation = True


Use local_eval_flagsmith fixture instead.

Done in 10242fb

khvn26 · 2025-10-31T22:01:25Z

tests/test_flagsmith.py

+    # Given
+    flagsmith._evaluation_context = evaluation_context
+    flagsmith.enable_local_evaluation = True
+    mock_engine = mocker.patch("flagsmith.flagsmith.engine")


Mock specific function and use autospec:

Suggested change

mock_engine = mocker.patch("flagsmith.flagsmith.engine")

mock_get_evaluation_result = mocker.patch(

"flagsmith.flagsmith.engine.get_evaluation_result",

autospec=True,

)

Done in 10242fb

khvn26 · 2025-10-31T22:02:45Z

tests/test_flagsmith.py

+    mock_engine.get_evaluation_result.assert_called_once()
+    call_args = mock_engine.get_evaluation_result.call_args
+    context = call_args[1]["context"]


Use assert_called_once_with method.

Done in 10242fb

khvn26 · 2025-10-31T22:03:15Z

tests/test_flagsmith.py

+    mock_engine.get_evaluation_result.assert_called_once()
+    call_args = mock_engine.get_evaluation_result.call_args
+    context = call_args[1]["context"]  # Keyword argument 'context'


Use assert_called_once_with method.

Done in 10242fb

khvn26 · 2025-10-31T22:24:23Z

flagsmith/flagsmith.py

+        context_without_segments = typing.cast(
+            SDKEvaluationContext,
+            {
+                key: value
+                for key, value in self._evaluation_context.items()
+                if key != "segments"
+            },
+        )


This helps to avoid both the cast and a bytecode loop:

Suggested change

context_without_segments = typing.cast(

SDKEvaluationContext,

{

key: value

for key, value in self._evaluation_context.items()

if key != "segments"

},

)

context_without_segments = self._evaluation_context.copy()

context_without_segments.pop("segments", None)

Done in 10242fb

khvn26 · 2025-10-31T22:26:30Z

@copilot Please address the review comments above.

- Use local_eval_flagsmith fixture in tests - Mock specific function with autospec - Use assert_called_once_with for cleaner assertions - Simplify context filtering with copy() and pop() Co-authored-by: khvn26 <979078+khvn26@users.noreply.github.com>

Copilot · 2025-10-31T22:31:30Z

@copilot Please address the review comments above.

All review comments addressed in 10242fb. Changes include:

Using local_eval_flagsmith fixture
Mocking specific function with autospec=True
Using assert_called_once_with for cleaner assertions
Simplified context filtering with copy() and pop()

All tests pass (73/73) and linters are clean.

Warning

Firewall rules blocked me from connecting to one or more addresses (expand for details)

I tried to connect to the following addresses, but was blocked by firewall rules:

test_urlanalytics
- Triggering command: /home/REDACTED/.cache/pypoetry/virtualenvs/flagsmith-K6knkhUn-py3.12/bin/python /home/REDACTED/.cache/pypoetry/virtualenvs/flagsmith-K6knkhUn-py3.12/bin/pytest tests/ -x (dns block)

If you need me to access, download, or install something from one of these locations, you can either:

Configure Actions setup steps to set up my environment, which run before the firewall is enabled
Add the appropriate URLs or hosts to the custom allowlist in this repository's Copilot coding agent settings (admins only)

Initial plan

f571af0

Copilot AI assigned Copilot and khvn26 Oct 31, 2025

Copilot started work on behalf of khvn26 October 31, 2025 19:29 View session

Copilot AI and others added 3 commits October 31, 2025 19:42

fix: omit segments from evaluation context for get_environment_flags

13904f6

Co-authored-by: khvn26 <979078+khvn26@users.noreply.github.com>

fix: improve context filtering to be more robust and type-safe

8f10c42

Co-authored-by: khvn26 <979078+khvn26@users.noreply.github.com>

fix: ensure consistent API usage for engine.get_evaluation_result

3bda6aa

Co-authored-by: khvn26 <979078+khvn26@users.noreply.github.com>

Copilot AI changed the title ~~[WIP] Fix get_environment_flags to omit segments in evaluation context~~ fix: omit segments from evaluation context in get_environment_flags Oct 31, 2025

Copilot AI requested a review from khvn26 October 31, 2025 19:57

Copilot finished work on behalf of khvn26 October 31, 2025 19:57

khvn26 requested changes Oct 31, 2025

View reviewed changes

Copilot started work on behalf of khvn26 October 31, 2025 22:26 View session

Copilot AI requested a review from khvn26 October 31, 2025 22:32

Copilot finished work on behalf of khvn26 October 31, 2025 22:32

		flagsmith._evaluation_context = evaluation_context
		flagsmith.enable_local_evaluation = True

-    mock_engine = mocker.patch("flagsmith.flagsmith.engine")
+    mock_get_evaluation_result = mocker.patch(
+        "flagsmith.flagsmith.engine.get_evaluation_result",
+        autospec=True,
+    )

fix: omit segments from evaluation context in get_environment_flags #179

Are you sure you want to change the base?

fix: omit segments from evaluation context in get_environment_flags #179

Uh oh!

Conversation

Copilot AI commented Oct 31, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Comments on the Issue (you are @copilot in this section)

Uh oh!

khvn26 Oct 31, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Oct 31, 2025

Choose a reason for hiding this comment

Uh oh!

khvn26 Oct 31, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Oct 31, 2025

Choose a reason for hiding this comment

Uh oh!

khvn26 Oct 31, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Oct 31, 2025

Choose a reason for hiding this comment

Uh oh!

khvn26 Oct 31, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Oct 31, 2025

Choose a reason for hiding this comment

Uh oh!

khvn26 Oct 31, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Oct 31, 2025

Choose a reason for hiding this comment

Uh oh!

khvn26 commented Oct 31, 2025

Uh oh!

Copilot AI commented Oct 31, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

I tried to connect to the following addresses, but was blocked by firewall rules:

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Copilot AI commented Oct 31, 2025 •

edited

Loading

Copilot AI commented Oct 31, 2025 •

edited

Loading