Merge branch 'major/3.0' into webb/openai-agents/limit-provider-name

Author: alexander-alderman-webb

.github/workflows/test-integrations-ai.yml

@@ -69,6 +69,10 @@ jobs:
         run: |
           set -x # print commands that are executed
           ./scripts/runtox.sh "py${{ matrix.python-version }}-openai-notiktoken"
+      - name: Test typesafe
+        run: |
+          set -x # print commands that are executed
+          ./scripts/runtox.sh "py${{ matrix.python-version }}-typesafe"
       - name: Upload coverage data
         if: ${{ !cancelled() }}
         continue-on-error: true

CHANGELOG.md

@@ -1,5 +1,52 @@
 # Changelog
 
+## 2.71.0
+
+### New Features ✨
+
+- **New integration:** TypeSafe (#4733) by @alexander-alderman-webb
+
+  Add the TypeSafe integration to your `sentry_sdk.init` call:
+
+```python
+import sentry_sdk
+from sentry_sdk.integrations.typesafe import TypeSafeIntegration
+
+sentry_sdk.init(
+    dsn="...",
+    traces_sample_rate=1.0,
+    integrations=[
+        TypeSafeIntegration(),
+    ]
+)
+```
+
+#### Typesafe
+
+- Record `gen_ai.output.messages` by @alexander-alderman-webb in [#7720](https://github.com/getsentry/sentry-python/pull/7720)
+- Set token usage attributes by @alexander-alderman-webb in [#7719](https://github.com/getsentry/sentry-python/pull/7719)
+- Record `gen_ai.input.messages` by @alexander-alderman-webb in [#7718](https://github.com/getsentry/sentry-python/pull/7718)
+- Record `gen_ai.response.model` by @alexander-alderman-webb in [#7717](https://github.com/getsentry/sentry-python/pull/7717)
+- Add integration with `system_one()` patches by @alexander-alderman-webb in [#7716](https://github.com/getsentry/sentry-python/pull/7716)
+
+### Bug Fixes 🐛
+
+- Keep event if processor raises by @sentrivana in [#7701](https://github.com/getsentry/sentry-python/pull/7701)
+- Isolate user callbacks, emit client reports by @sentrivana in [#7692](https://github.com/getsentry/sentry-python/pull/7692)
+
+### Documentation 📚
+
+- Promote `data_collection` in changelog by @sentrivana in [#7565](https://github.com/getsentry/sentry-python/pull/7565)
+
+### Internal Changes 🔧
+
+- (aws-lambda) Add `.gitignore` to lambda handlers by @pabloDeputter in [#7638](https://github.com/getsentry/sentry-python/pull/7638)
+- (litellm) Set `LITELLM_LOCAL_MODEL_COST_MAP=true` by @alexander-alderman-webb in [#7630](https://github.com/getsentry/sentry-python/pull/7630)
+- (otlp) Adjust to `OTLPSpanExporter._headers` removal by @alexander-alderman-webb in [#7722](https://github.com/getsentry/sentry-python/pull/7722)
+- Use top-level flush API in tests by @sentrivana in [#7725](https://github.com/getsentry/sentry-python/pull/7725)
+- Unsample on sampler exception by @sentrivana in [#7723](https://github.com/getsentry/sentry-python/pull/7723)
+- Use `sentry_init` fixture by @sentrivana in [#7577](https://github.com/getsentry/sentry-python/pull/7577)
+
 ## 2.70.0
 
 ### New Features ✨

MIGRATION_GUIDE.md

@@ -31,6 +31,7 @@ Looking to upgrade from Sentry SDK 2.x to 3.x? Here's a comprehensive list of wh
 - The option `attach_stacktrace` is now `True` by default, meaning the SDK will attach stack traces to messages.
 - The Django integration now creates spans for cache operations by default. Pass `DjangoIntegration(cache_spans=False)` to `sentry_sdk.init()` to turn them off.
 - The Django integration no longer force-enables cache spans when Spotlight is active and `settings.DEBUG` is `True`. The `cache_spans` option is now always respected as given.
+- Exception groups in exception chains are now properly unfurled.
 
 ### Logging
 
@@ -121,6 +122,7 @@ Looking to upgrade from Sentry SDK 2.x to 3.x? Here's a comprehensive list of wh
 - Dropped support for Pydantic AI below 1.76.
 - Dropped support for OpenAI Agents below 0.10.3.
 - Dropped support for MCP below 2.0.
+- Removed the `transaction_style` option from server integrations (DjangoIntegration, StarletteIntegration, FastApiIntegration, AioHttpIntegration, FlaskIntegration, PyramidIntegration, QuartIntegration, BottleIntegration, FalconIntegration, StarliteIntegration and LitestarIntegration).
 - Removed the RedisIntegration `max_data_size` option.
 - Removed the possibility to supply a specific client to the LaunchDarklyIntegration.
 - The `enable_tracing` option was removed. Use `traces_sample_rate=1.0` instead.

docs/apidocs.rst

@@ -35,9 +35,6 @@ API Docs
 .. autoclass:: sentry_sdk.attachments.Attachment
    :members:
 
-.. autoclass:: sentry_sdk.scrubber.EventScrubber
-   :members:
-
 .. autoclass:: sentry_sdk.monitor.Monitor
    :members:
 

docs/conf.py

@@ -31,7 +31,7 @@
 copyright = "2019-{}, Sentry Team and Contributors".format(datetime.now().year)
 author = "Sentry Team and Contributors"
 
-release = "2.70.0"
+release = "2.71.0"
 version = ".".join(release.split(".")[:2])  # The short X.Y version.
 
 

pyproject.toml

@@ -81,6 +81,7 @@ typing = [
     "aiohttp>=3.14.3",
     "openai-agents>=0.22.0",
     "mistralai>=2.10.1",
+    "typesafe-sdk>=0.7.2",
 ]
 test = [
     "dataclasses ; python_full_version < '3.7'",

scripts/populate_tox/config.py

@@ -235,6 +235,7 @@
         "package": "huggingface_hub",
         "deps": {
             "*": ["responses", "pytest-httpx"],
+            ">=2": ["httpx2-pytest"],
         },
     },
     "langchain-base": {
@@ -472,12 +473,13 @@
                 "pytest-asyncio",
                 "python-multipart",
                 "requests",
-                "anyio<4.0.0",
                 "jinja2",
                 "httpx",
             ],
             # See the comment on FastAPI's httpx bound for more info
             "<0.37.2": ["httpx<0.28.0"],
+            # Starlette <0.21 uses anyio.start_blocking_portal which was removed in anyio 4.
+            "<0.21": ["anyio<4"],
         },
     },
     "starlite": {
@@ -533,6 +535,13 @@
         "package": "typer",
         "num_versions": 2,
     },
+    "typesafe": {
+        "package": "typesafe-sdk",
+        "integration_name": "typesafe",
+        "deps": {
+            "*": ["pytest-asyncio"],
+        },
+    },
     "unleash": {
         "package": "UnleashClient",
         "num_versions": 2,