Merge branch 'master' into dependabot/github_actions/actions/create-github-app-token-1.11.2

Author: sentrivana

sentry_sdk/_types.py

@@ -1,10 +1,88 @@
-from typing import TYPE_CHECKING
+from typing import TYPE_CHECKING, TypeVar, Union
 
 
 # Re-exported for compat, since code out there in the wild might use this variable.
 MYPY = TYPE_CHECKING
 
 
+SENSITIVE_DATA_SUBSTITUTE = "[Filtered]"
+
+
+class AnnotatedValue:
+    """
+    Meta information for a data field in the event payload.
+    This is to tell Relay that we have tampered with the fields value.
+    See:
+    https://github.com/getsentry/relay/blob/be12cd49a0f06ea932ed9b9f93a655de5d6ad6d1/relay-general/src/types/meta.rs#L407-L423
+    """
+
+    __slots__ = ("value", "metadata")
+
+    def __init__(self, value, metadata):
+        # type: (Optional[Any], Dict[str, Any]) -> None
+        self.value = value
+        self.metadata = metadata
+
+    def __eq__(self, other):
+        # type: (Any) -> bool
+        if not isinstance(other, AnnotatedValue):
+            return False
+
+        return self.value == other.value and self.metadata == other.metadata
+
+    @classmethod
+    def removed_because_raw_data(cls):
+        # type: () -> AnnotatedValue
+        """The value was removed because it could not be parsed. This is done for request body values that are not json nor a form."""
+        return AnnotatedValue(
+            value="",
+            metadata={
+                "rem": [  # Remark
+                    [
+                        "!raw",  # Unparsable raw data
+                        "x",  # The fields original value was removed
+                    ]
+                ]
+            },
+        )
+
+    @classmethod
+    def removed_because_over_size_limit(cls):
+        # type: () -> AnnotatedValue
+        """The actual value was removed because the size of the field exceeded the configured maximum size (specified with the max_request_body_size sdk option)"""
+        return AnnotatedValue(
+            value="",
+            metadata={
+                "rem": [  # Remark
+                    [
+                        "!config",  # Because of configured maximum size
+                        "x",  # The fields original value was removed
+                    ]
+                ]
+            },
+        )
+
+    @classmethod
+    def substituted_because_contains_sensitive_data(cls):
+        # type: () -> AnnotatedValue
+        """The actual value was removed because it contained sensitive information."""
+        return AnnotatedValue(
+            value=SENSITIVE_DATA_SUBSTITUTE,
+            metadata={
+                "rem": [  # Remark
+                    [
+                        "!config",  # Because of SDK configuration (in this case the config is the hard coded removal of certain django cookies)
+                        "s",  # The fields original value was substituted
+                    ]
+                ]
+            },
+        )
+
+
+T = TypeVar("T")
+Annotated = Union[AnnotatedValue, T]
+
+
 if TYPE_CHECKING:
     from collections.abc import Container, MutableMapping, Sequence
 
@@ -19,7 +97,6 @@
     from typing import Optional
     from typing import Tuple
     from typing import Type
-    from typing import Union
     from typing_extensions import Literal, TypedDict
 
     class SDKInfo(TypedDict):
@@ -101,7 +178,7 @@ class SDKInfo(TypedDict):
             "request": dict[str, object],
             "sdk": Mapping[str, object],
             "server_name": str,
-            "spans": list[dict[str, object]],
+            "spans": Annotated[list[dict[str, object]]],
             "stacktrace": dict[
                 str, object
             ],  # We access this key in the code, but I am unsure whether we ever set it
@@ -118,6 +195,7 @@ class SDKInfo(TypedDict):
             "transaction_info": Mapping[str, Any],  # TODO: We can expand on this type
             "type": Literal["check_in", "transaction"],
             "user": dict[str, object],
+            "_dropped_spans": int,
             "_metrics_summary": dict[str, object],
         },
         total=False,

sentry_sdk/client.py

@@ -5,11 +5,12 @@
 from collections.abc import Mapping
 from datetime import datetime, timezone
 from importlib import import_module
-from typing import cast, overload
+from typing import TYPE_CHECKING, List, Dict, cast, overload
 import warnings
 
 from sentry_sdk._compat import PY37, check_uwsgi_thread_support
 from sentry_sdk.utils import (
+    AnnotatedValue,
     ContextVar,
     capture_internal_exceptions,
     current_stacktrace,
@@ -45,12 +46,9 @@
 from sentry_sdk.monitor import Monitor
 from sentry_sdk.spotlight import setup_spotlight
 
-from typing import TYPE_CHECKING
-
 if TYPE_CHECKING:
     from typing import Any
     from typing import Callable
-    from typing import Dict
     from typing import Optional
     from typing import Sequence
     from typing import Type
@@ -483,12 +481,14 @@ def _prepare_event(
     ):
         # type: (...) -> Optional[Event]
 
+        previous_total_spans = None  # type: Optional[int]
+
         if event.get("timestamp") is None:
             event["timestamp"] = datetime.now(timezone.utc)
 
         if scope is not None:
             is_transaction = event.get("type") == "transaction"
-            spans_before = len(event.get("spans", []))
+            spans_before = len(cast(List[Dict[str, object]], event.get("spans", [])))
             event_ = scope.apply_to_event(event, hint, self.options)
 
             # one of the event/error processors returned None
@@ -507,13 +507,18 @@ def _prepare_event(
                 return None
 
             event = event_
-
-            spans_delta = spans_before - len(event.get("spans", []))
+            spans_delta = spans_before - len(
+                cast(List[Dict[str, object]], event.get("spans", []))
+            )
             if is_transaction and spans_delta > 0 and self.transport is not None:
                 self.transport.record_lost_event(
                     "event_processor", data_category="span", quantity=spans_delta
                 )
 
+            dropped_spans = event.pop("_dropped_spans", 0) + spans_delta  # type: int
+            if dropped_spans > 0:
+                previous_total_spans = spans_before + dropped_spans
+
         if (
             self.options["attach_stacktrace"]
             and "exception" not in event
@@ -561,6 +566,11 @@ def _prepare_event(
             if event_scrubber:
                 event_scrubber.scrub_event(event)
 
+        if previous_total_spans is not None:
+            event["spans"] = AnnotatedValue(
+                event.get("spans", []), {"len": previous_total_spans}
+            )
+
         # Postprocess the event here so that annotated types do
         # generally not surface in before_send
         if event is not None:
@@ -598,7 +608,7 @@ def _prepare_event(
             and event.get("type") == "transaction"
         ):
             new_event = None
-            spans_before = len(event.get("spans", []))
+            spans_before = len(cast(List[Dict[str, object]], event.get("spans", [])))
             with capture_internal_exceptions():
                 new_event = before_send_transaction(event, hint or {})
             if new_event is None:

sentry_sdk/consts.py

@@ -528,6 +528,7 @@ def __init__(
         profiles_sample_rate=None,  # type: Optional[float]
         profiles_sampler=None,  # type: Optional[TracesSampler]
         profiler_mode=None,  # type: Optional[ProfilerMode]
+        profile_session_sample_rate=None,  # type: Optional[float]
         auto_enabling_integrations=True,  # type: bool
         disabled_integrations=None,  # type: Optional[Sequence[sentry_sdk.integrations.Integration]]
         auto_session_tracking=True,  # type: bool

sentry_sdk/integrations/grpc/__init__.py

@@ -81,7 +81,7 @@ def _wrap_channel_async(func: Callable[P, AsyncChannel]) -> Callable[P, AsyncCha
     "Wrapper for asynchronous secure and insecure channel."
 
     @wraps(func)
-    def patched_channel(
+    def patched_channel(  # type: ignore
         *args: P.args,
         interceptors: Optional[Sequence[grpc.aio.ClientInterceptor]] = None,
         **kwargs: P.kwargs,
@@ -100,7 +100,7 @@ def _wrap_sync_server(func: Callable[P, Server]) -> Callable[P, Server]:
     """Wrapper for synchronous server."""
 
     @wraps(func)
-    def patched_server(
+    def patched_server(  # type: ignore
         *args: P.args,
         interceptors: Optional[Sequence[grpc.ServerInterceptor]] = None,
         **kwargs: P.kwargs,
@@ -121,7 +121,7 @@ def _wrap_async_server(func: Callable[P, AsyncServer]) -> Callable[P, AsyncServe
     """Wrapper for asynchronous server."""
 
     @wraps(func)
-    def patched_aio_server(
+    def patched_aio_server(  # type: ignore
         *args: P.args,
         interceptors: Optional[Sequence[grpc.ServerInterceptor]] = None,
         **kwargs: P.kwargs,

sentry_sdk/integrations/socket.py

@@ -27,15 +27,19 @@ def setup_once():
 
 
 def _get_span_description(host, port):
-    # type: (Union[bytes, str, None], Union[str, int, None]) -> str
+    # type: (Union[bytes, str, None], Union[bytes, str, int, None]) -> str
 
     try:
         host = host.decode()  # type: ignore
     except (UnicodeDecodeError, AttributeError):
         pass
 
-    description = "%s:%s" % (host, port)  # type: ignore
+    try:
+        port = port.decode()  # type: ignore
+    except (UnicodeDecodeError, AttributeError):
+        pass
 
+    description = "%s:%s" % (host, port)  # type: ignore
     return description
 
 
@@ -74,7 +78,7 @@ def _patch_getaddrinfo():
     real_getaddrinfo = socket.getaddrinfo
 
     def getaddrinfo(host, port, family=0, type=0, proto=0, flags=0):
-        # type: (Union[bytes, str, None], Union[str, int, None], int, int, int, int) -> List[Tuple[AddressFamily, SocketKind, int, str, Union[Tuple[str, int], Tuple[str, int, int, int]]]]
+        # type: (Union[bytes, str, None], Union[bytes, str, int, None], int, int, int, int) -> List[Tuple[AddressFamily, SocketKind, int, str, Union[Tuple[str, int], Tuple[str, int, int, int], Tuple[int, bytes]]]]
         integration = sentry_sdk.get_client().get_integration(SocketIntegration)
         if integration is None:
             return real_getaddrinfo(host, port, family, type, proto, flags)
@@ -89,4 +93,4 @@ def getaddrinfo(host, port, family=0, type=0, proto=0, flags=0):
 
             return real_getaddrinfo(host, port, family, type, proto, flags)
 
-    socket.getaddrinfo = getaddrinfo  # type: ignore
+    socket.getaddrinfo = getaddrinfo

sentry_sdk/integrations/tornado.py

@@ -79,7 +79,7 @@ async def sentry_execute_request_handler(self, *args, **kwargs):
         else:
 
             @coroutine  # type: ignore
-            def sentry_execute_request_handler(self, *args, **kwargs):
+            def sentry_execute_request_handler(self, *args, **kwargs):  # type: ignore
                 # type: (RequestHandler, *Any, **Any) -> Any
                 with _handle_request_impl(self):
                     result = yield from old_execute(self, *args, **kwargs)