Merge remote-tracking branch 'origin/master' into ep/django-tests-asgi-params

Author: ericapisani

.github/workflows/test.yml

@@ -115,7 +115,7 @@ jobs:
           uv run coverage combine .coverage-sentry-*
           uv run coverage xml
       - name: Report coverage and test results
-        uses: getsentry/codecov-action@66554aa51d8656c657e233a9e98bcb18e95f2c9a # main
+        uses: getsentry/codecov-action@d90e69cdf071dfbb0430159125321dc09c424d4c # main
         with:
           token: ${{ secrets.GITHUB_TOKEN }}
           files: coverage.xml

CHANGELOG.md

@@ -1,5 +1,40 @@
 # Changelog
 
+## 2.68.1
+
+### `enable_logs`
+
+- (logs) Don't stop sending auto-collected logs when `enable_logs=True` by @sentrivana in [#7237](https://github.com/getsentry/sentry-python/pull/7237)
+
+  If you have `enable_logs` set to `True`, our logging integrations for the standard library logging module as well as Loguru will auto-collect logs and send them to Sentry as Sentry logs by default, preserving old behavior. Turning automatic collection off for a specific integration can still be achieved using the `capture_sentry_logs` integration option.
+
+    ```python
+    import sentry_sdk
+    from sentry_sdk.integrations.logging import LoggingIntegration
+    from sentry_sdk.integrations.loguru import LoguruIntegration
+
+    sentry_sdk.init(
+        enable_logs=True,
+        integrations=[
+            LoggingIntegration(capture_sentry_logs=True),
+            LoguruIntegration(capture_sentry_logs=False),
+        ],
+    )
+    ```
+
+  Please note that the `enable_logs` option is deprecated and will be removed in the next major release. The `sentry_sdk.logger.X` API now works regardless of it, and auto-collection can be opted into via the `capture_sentry_logs` integration-level options, which are `False` by default, unless you have `enable_logs=True`. We've added this compatibility layer to make the transition to a `enable_logs`-free world easier.
+
+### Bug Fixes 🐛
+
+- (django) Add `failed_request_status_codes` by @mgaligniana in [#7140](https://github.com/getsentry/sentry-python/pull/7140)
+- (anthropic) Gate `gen_ai.response.tool_calls` on outputs, not inputs by @ericapisani in [#7207](https://github.com/getsentry/sentry-python/pull/7207)
+- (google_genai) Gate streaming `gen_ai.response.tool_calls` on outputs, not inputs by @ericapisani in [#7210](https://github.com/getsentry/sentry-python/pull/7210)
+- (langchain) Use `gen_ai.tool.definitions` attribute when data collection is enabled by @ericapisani in [#7204](https://github.com/getsentry/sentry-python/pull/7204)
+- (langgraph,huggingface_hub) Gate `gen_ai.response.tool_calls` on outputs, not inputs by @ericapisani in [#7208](https://github.com/getsentry/sentry-python/pull/7208)
+- (openai) Gate `gen_ai.response.tool_calls` on outputs, not inputs by @ericapisani in [#7205](https://github.com/getsentry/sentry-python/pull/7205)
+- (openai_agents) Gate `gen_ai.response.tool_calls` on outputs, not inputs by @ericapisani in [#7209](https://github.com/getsentry/sentry-python/pull/7209)
+
+
 ## 2.68.0
 
 ### Important

docs/conf.py

@@ -31,7 +31,7 @@
 copyright = "2019-{}, Sentry Team and Contributors".format(datetime.now().year)
 author = "Sentry Team and Contributors"
 
-release = "2.68.0"
+release = "2.68.1"
 version = ".".join(release.split(".")[:2])  # The short X.Y version.
 
 

scripts/populate_tox/config.py

@@ -23,7 +23,7 @@
     "anthropic": {
         "package": "anthropic",
         "deps": {
-            "*": ["pytest-asyncio"],
+            "*": ["pytest-asyncio", "httpx"],
             "<0.50": ["httpx<0.28.0"],
             # tokenizers dropped Python 3.8 support, but didn't update package metadata.
             # https://github.com/huggingface/tokenizers/commit/f4c9fd7f402fc794df8f1b547a95ee5305f9fe62
@@ -249,7 +249,7 @@
         "package": "langchain",
         "integration_name": "langchain",
         "deps": {
-            "*": ["pytest-asyncio", "openai", "tiktoken", "langchain-openai"],
+            "*": ["pytest-asyncio", "openai", "tiktoken", "langchain-openai", "httpx"],
             "<=0.1": ["httpx<0.28.0"],
             ">=0.3": ["langchain-community"],
             ">=1.0": ["langchain-classic"],
@@ -263,7 +263,7 @@
         "package": "langchain",
         "integration_name": "langchain",
         "deps": {
-            "*": ["pytest-asyncio", "openai", "langchain-openai"],
+            "*": ["pytest-asyncio", "openai", "langchain-openai", "httpx"],
             "<=0.1": ["httpx<0.28.0"],
             ">=0.3": ["langchain-community"],
             ">=1.0": ["langchain-classic"],
@@ -283,7 +283,7 @@
     "litellm": {
         "package": "litellm",
         "deps": {
-            "*": ["anthropic", "google-genai", "pytest-asyncio"],
+            "*": ["anthropic", "google-genai", "pytest-asyncio", "httpx"],
         },
         "include": "<=1.96",  # https://github.com/BerriAI/litellm/issues/36384
     },
@@ -323,7 +323,7 @@
         "package": "openai",
         "integration_name": "openai",
         "deps": {
-            "*": ["pytest-asyncio", "tiktoken"],
+            "*": ["pytest-asyncio", "tiktoken", "httpx"],
             "<1.55": ["httpx<0.28"],
         },
         "python": {
@@ -335,7 +335,7 @@
         "package": "openai",
         "integration_name": "openai",
         "deps": {
-            "*": ["pytest-asyncio"],
+            "*": ["pytest-asyncio", "httpx"],
             "<1.55": ["httpx<0.28"],
         },
         "python": {
@@ -346,7 +346,9 @@
     "openai_agents": {
         "package": "openai-agents",
         "deps": {
-            "*": ["pytest-asyncio"],
+            "*": ["pytest-asyncio", "httpx"],
+            # https://github.com/openai/openai-python/commit/f16fbbd2bd25dc1ff150b5f78dbd15ff6bab6d91 makes `cache_write_tokens` required.
+            "<=0.18.0": ["openai<2.45"],
         },
         "python": ">=3.10",
     },

scripts/populate_tox/package_dependencies.jsonl

@@ -648,13 +648,6 @@ def _record_lost_event(
 
             self.session_flusher = SessionFlusher(capture_func=_capture_envelope)
 
-            if self.options.get("enable_logs", False) or self.options[
-                "_experiments"
-            ].get("enable_logs", False):
-                logger.warning(
-                    "The enable_logs option has no effect and will be removed in the next major."
-                )
-
             self.log_batcher = LogBatcher(
                 capture_func=_capture_envelope,
                 record_lost_func=_record_lost_event,

scripts/populate_tox/releases.jsonl

@@ -1840,4 +1840,4 @@ def _get_default_options() -> "dict[str, Any]":
 del _get_default_options
 
 
-VERSION = "2.68.0"
+VERSION = "2.68.1"

sentry_sdk/client.py

@@ -89,38 +89,47 @@ def extract_into_event(self, event: "Event") -> None:
         content_length = self.content_length()
         request_info = event.get("request", {})
 
+        # Prior to data collection being implemented we unconditionally attached
+        # the request body, which is why we default to True here.
+        attach_request_body = True
+
         if has_data_collection_enabled(client.options):
             cookies = _apply_key_value_collection_filtering(
                 items=dict(self.cookies()),
                 behaviour=client.options["data_collection"]["cookies"],
             )
             if cookies:
                 request_info["cookies"] = cookies
+
+            attach_request_body = (
+                "incoming_request" in client.options["data_collection"]["http_bodies"]
+            )
         elif should_send_default_pii():
             request_info["cookies"] = dict(self.cookies())
 
-        if not request_body_within_bounds(client, content_length):
-            data = AnnotatedValue.removed_because_over_size_limit()
-        else:
-            # First read the raw body data
-            # It is important to read this first because if it is Django
-            # it will cache the body and then we can read the cached version
-            # again in parsed_body() (or json() or wherever).
-            raw_data = None
-            try:
-                raw_data = self.raw_data()
-            except _RAW_DATA_EXCEPTIONS:
-                # If DjangoRestFramework is used it already read the body for us
-                # so reading it here will fail. We can ignore this.
-                pass
-
-            parsed_body = self.parsed_body()
-            if parsed_body is not None:
-                data = parsed_body
-            elif raw_data:
-                data = AnnotatedValue.removed_because_raw_data()
+        if attach_request_body:
+            if not request_body_within_bounds(client, content_length):
+                data = AnnotatedValue.removed_because_over_size_limit()
             else:
-                data = None
+                # First read the raw body data
+                # It is important to read this first because if it is Django
+                # it will cache the body and then we can read the cached version
+                # again in parsed_body() (or json() or wherever).
+                raw_data = None
+                try:
+                    raw_data = self.raw_data()
+                except _RAW_DATA_EXCEPTIONS:
+                    # If DjangoRestFramework is used it already read the body for us
+                    # so reading it here will fail. We can ignore this.
+                    pass
+
+                parsed_body = self.parsed_body()
+                if parsed_body is not None:
+                    data = parsed_body
+                elif raw_data:
+                    data = AnnotatedValue.removed_because_raw_data()
+                else:
+                    data = None
 
         if data is not None:
             request_info["data"] = data

sentry_sdk/consts.py

@@ -608,16 +608,13 @@ def _set_output_data(
         set_on_span(SPANDATA.GEN_AI_RESPONSE_FINISH_REASONS, [finish_reason])
 
     client = sentry_sdk.get_client()
-    record_inputs = False
     record_outputs = False
     if has_data_collection_enabled(client.options):
-        record_inputs = client.options["data_collection"]["gen_ai"]["inputs"]
         record_outputs = client.options["data_collection"]["gen_ai"]["outputs"]
     elif should_send_default_pii() and integration.include_prompts:
-        record_inputs = True
         record_outputs = True
 
-    if record_inputs or record_outputs:
+    if record_outputs:
         output_messages: "dict[str, list[Any]]" = {
             "response": [],
             "tool": [],
@@ -629,15 +626,15 @@ def _set_output_data(
             elif output["type"] == "tool_use":
                 output_messages["tool"].append(output)
 
-        if record_inputs and len(output_messages["tool"]) > 0:
+        if len(output_messages["tool"]) > 0:
             set_data_normalized(
                 span,
                 SPANDATA.GEN_AI_RESPONSE_TOOL_CALLS,
                 output_messages["tool"],
                 unpack=False,
             )
 
-        if record_outputs and len(output_messages["response"]) > 0:
+        if len(output_messages["response"]) > 0:
             set_data_normalized(
                 span, SPANDATA.GEN_AI_RESPONSE_TEXT, output_messages["response"]
             )