|
1 | | -# 6.8.0 - 2025-12-01 |
| 1 | +# 7.2.0 - 2025-12-01 |
2 | 2 |
|
3 | | -- feat: add `$feature_flag_evaluated_at` properties to `$feature_flag_called` events |
| 3 | +feat: add $feature_flag_evaluated_at properties to $feature_flag_called events |
| 4 | + |
| 5 | + |
| 6 | +# 7.1.0 - 2025-11-26 |
| 7 | + |
| 8 | +Add support for the async version of Gemini. |
| 9 | + |
| 10 | +# 7.0.2 - 2025-11-18 |
| 11 | + |
| 12 | +Add support for Python 3.14. |
| 13 | +Projects upgrading to Python 3.14 should ensure any Pydantic models passed into the SDK use Pydantic v2, as Pydantic v1 is not compatible with Python 3.14. |
| 14 | + |
| 15 | +# 7.0.1 - 2025-11-15 |
| 16 | + |
| 17 | +Try to use repr() when formatting code variables |
| 18 | + |
| 19 | +# 7.0.0 - 2025-11-11 |
| 20 | + |
| 21 | +NB Python 3.9 is no longer supported |
| 22 | + |
| 23 | +- chore(llma): update LLM provider SDKs to latest major versions |
| 24 | + - openai: 1.102.0 → 2.7.1 |
| 25 | + - anthropic: 0.64.0 → 0.72.0 |
| 26 | + - google-genai: 1.32.0 → 1.49.0 |
| 27 | + - langchain-core: 0.3.75 → 1.0.3 |
| 28 | + - langchain-openai: 0.3.32 → 1.0.2 |
| 29 | + - langchain-anthropic: 0.3.19 → 1.0.1 |
| 30 | + - langchain-community: 0.3.29 → 0.4.1 |
| 31 | + - langgraph: 0.6.6 → 1.0.2 |
| 32 | + |
| 33 | +# 6.9.3 - 2025-11-10 |
| 34 | + |
| 35 | +- feat(ph-ai): PostHog properties dict in GenerationMetadata |
| 36 | + |
| 37 | +# 6.9.2 - 2025-11-10 |
| 38 | + |
| 39 | +- fix(llma): fix cache token double subtraction in Langchain for non-Anthropic providers causing negative costs |
| 40 | + |
| 41 | +# 6.9.1 - 2025-11-07 |
| 42 | + |
| 43 | +- fix(error-tracking): pass code variables config from init to client |
| 44 | + |
| 45 | +# 6.9.0 - 2025-11-06 |
| 46 | + |
| 47 | +- feat(error-tracking): add local variables capture |
| 48 | + |
| 49 | +# 6.8.0 - 2025-11-03 |
| 50 | + |
| 51 | +- feat(llma): send web search calls to be used for LLM cost calculations |
| 52 | + |
| 53 | +# 6.7.14 - 2025-11-03 |
| 54 | + |
| 55 | +- fix(django): Handle request.user access in async middleware context to prevent SynchronousOnlyOperation errors in Django 5+ (fixes #355) |
| 56 | +- test(django): Add Django 5 integration test suite with real ASGI application testing async middleware behavior |
| 57 | + |
| 58 | +# 6.7.13 - 2025-11-02 |
| 59 | + |
| 60 | +- fix(llma): cache cost calculation in the LangChain callback |
| 61 | + |
| 62 | +# 6.7.12 - 2025-11-02 |
| 63 | + |
| 64 | +- fix(django): Restore process_exception method to capture view and downstream middleware exceptions (fixes #329) |
| 65 | +- fix(ai/langchain): Add LangChain 1.0+ compatibility for CallbackHandler imports (fixes #362) |
| 66 | + |
| 67 | +# 6.7.11 - 2025-10-28 |
| 68 | + |
| 69 | +- feat(ai): Add `$ai_framework` property for framework integrations (e.g. LangChain) |
| 70 | + |
| 71 | +# 6.7.10 - 2025-10-24 |
| 72 | + |
| 73 | +- fix(django): Make middleware truly hybrid - compatible with both sync (WSGI) and async (ASGI) Django stacks without breaking sync-only deployments |
| 74 | + |
| 75 | +# 6.7.9 - 2025-10-22 |
| 76 | + |
| 77 | +- fix(flags): multi-condition flags with static cohorts returning wrong variants |
| 78 | + |
| 79 | +# 6.7.8 - 2025-10-16 |
| 80 | + |
| 81 | +- fix(llma): missing async for OpenAI's streaming implementation |
| 82 | + |
| 83 | +# 6.7.7 - 2025-10-14 |
| 84 | + |
| 85 | +- fix: remove deprecated attribute $exception_personURL from exception events |
| 86 | + |
| 87 | +# 6.7.6 - 2025-09-16 |
| 88 | + |
| 89 | +- fix: don't sort condition sets with variant overrides to the top |
| 90 | +- fix: Prevent core Client methods from raising exceptions |
4 | 91 |
|
5 | 92 | # 6.7.5 - 2025-09-16 |
6 | 93 |
|
|
0 commit comments