Merge branch 'potel-base' into ivana/potel/slim-down-attrs

Author: sentrivana

MIGRATION_GUIDE.md

@@ -27,6 +27,7 @@ Looking to upgrade from Sentry SDK 2.x to 3.x? Here's a comprehensive list of wh
 - The `sampling_context` argument of `traces_sampler` and `profiles_sampler` now additionally contains all span attributes known at span start.
 - We updated how we handle `ExceptionGroup`s. You will now get more data if ExceptionGroups are appearing in chained exceptions. It could happen that after updating the SDK the grouping of issues change because of this. So eventually you will see the same exception in two Sentry issues (one from before the update, one from after the update)
 - The integration for Python `logging` module does not send Sentry issues by default anymore when calling `logging.error()`, `logging.critical()` or `logging.exception()`. If you want to preserve the old behavior use `sentry_sdk.init(integrations=[LoggingIntegration(event_level="ERROR")])`.
+- The `SentrySpanProcessor` and `SentryPropagator` are exported from `sentry_sdk.opentelemetry` instead of `sentry_sdk.integrations.opentelemetry`.
 - The integration-specific content of the `sampling_context` argument of `traces_sampler` and `profiles_sampler` now looks different.
   - The Celery integration doesn't add the `celery_job` dictionary anymore. Instead, the individual keys are now available as:
 
@@ -137,6 +138,7 @@ Looking to upgrade from Sentry SDK 2.x to 3.x? Here's a comprehensive list of wh
 - The `enable_tracing` `init` option has been removed. Configure `traces_sample_rate` directly.
 - The `propagate_traces` `init` option has been removed. Use `trace_propagation_targets` instead.
 - The `custom_sampling_context` parameter of `start_transaction` has been removed. Use `attributes` instead to set key-value pairs of data that should be accessible in the traces sampler. Note that span attributes need to conform to the [OpenTelemetry specification](https://opentelemetry.io/docs/concepts/signals/traces/#attributes), meaning only certain types can be set as values.
+- `set_measurement` has been removed.
 - The PyMongo integration no longer sets tags. The data is still accessible via span attributes.
 - The PyMongo integration doesn't set `operation_ids` anymore. The individual IDs (`operation_id`, `request_id`, `session_id`) are now accessible as separate span attributes.
 - `sentry_sdk.metrics` and associated metrics APIs have been removed as Sentry no longer accepts metrics data in this form. See https://sentry.zendesk.com/hc/en-us/articles/26369339769883-Upcoming-API-Changes-to-Metrics

sentry_sdk/__init__.py

@@ -36,7 +36,6 @@
     "set_context",
     "set_extra",
     "set_level",
-    "set_measurement",
     "set_tag",
     "set_tags",
     "set_user",

sentry_sdk/_types.py

@@ -107,7 +107,6 @@ def substituted_because_contains_sensitive_data(cls):
     from typing import Callable
     from typing import Dict
     from typing import Mapping
-    from typing import NotRequired
     from typing import Optional
     from typing import Type
     from typing_extensions import Literal, TypedDict
@@ -120,45 +119,6 @@ class SDKInfo(TypedDict):
     # "critical" is an alias of "fatal" recognized by Relay
     LogLevelStr = Literal["fatal", "critical", "error", "warning", "info", "debug"]
 
-    DurationUnit = Literal[
-        "nanosecond",
-        "microsecond",
-        "millisecond",
-        "second",
-        "minute",
-        "hour",
-        "day",
-        "week",
-    ]
-
-    InformationUnit = Literal[
-        "bit",
-        "byte",
-        "kilobyte",
-        "kibibyte",
-        "megabyte",
-        "mebibyte",
-        "gigabyte",
-        "gibibyte",
-        "terabyte",
-        "tebibyte",
-        "petabyte",
-        "pebibyte",
-        "exabyte",
-        "exbibyte",
-    ]
-
-    FractionUnit = Literal["ratio", "percent"]
-    MeasurementUnit = Union[DurationUnit, InformationUnit, FractionUnit, str]
-
-    MeasurementValue = TypedDict(
-        "MeasurementValue",
-        {
-            "value": float,
-            "unit": NotRequired[Optional[MeasurementUnit]],
-        },
-    )
-
     Event = TypedDict(
         "Event",
         {
@@ -180,7 +140,6 @@ class SDKInfo(TypedDict):
             "level": LogLevelStr,
             "logentry": Mapping[str, object],
             "logger": str,
-            "measurements": dict[str, MeasurementValue],
             "message": str,
             "modules": dict[str, str],
             "monitor_config": Mapping[str, object],

sentry_sdk/ai/monitoring.py

@@ -106,14 +106,14 @@ def record_token_usage(
     if ai_pipeline_name:
         span.set_attribute("ai.pipeline.name", ai_pipeline_name)
     if prompt_tokens is not None:
-        span.set_measurement("ai_prompt_tokens_used", value=prompt_tokens)
+        span.set_attribute("ai.prompt_tokens.used", prompt_tokens)
     if completion_tokens is not None:
-        span.set_measurement("ai_completion_tokens_used", value=completion_tokens)
+        span.set_attribute("ai.completion_tokens.used", completion_tokens)
     if (
         total_tokens is None
         and prompt_tokens is not None
         and completion_tokens is not None
     ):
         total_tokens = prompt_tokens + completion_tokens
     if total_tokens is not None:
-        span.set_measurement("ai_total_tokens_used", total_tokens)
+        span.set_attribute("ai.total_tokens.used", total_tokens)

sentry_sdk/api.py

@@ -59,7 +59,6 @@
     "set_context",
     "set_extra",
     "set_level",
-    "set_measurement",
     "set_tag",
     "set_tags",
     "set_user",
@@ -287,13 +286,6 @@ def start_transaction(
     )
 
 
-def set_measurement(name, value, unit=""):
-    # type: (str, float, sentry_sdk._types.MeasurementUnit) -> None
-    transaction = get_current_scope().root_span
-    if transaction is not None:
-        transaction.set_measurement(name, value, unit)
-
-
 def get_current_span(scope=None):
     # type: (Optional[Scope]) -> Optional[sentry_sdk.tracing.Span]
     """

sentry_sdk/opentelemetry/consts.py

@@ -26,7 +26,6 @@ class SentrySpanAttribute:
     DESCRIPTION = "sentry.description"
     OP = "sentry.op"
     ORIGIN = "sentry.origin"
-    MEASUREMENT = "sentry.measurement"
     TAG = "sentry.tag"
     NAME = "sentry.name"
     SOURCE = "sentry.source"

sentry_sdk/opentelemetry/span_processor.py

@@ -308,10 +308,6 @@ def _common_span_transaction_attributes_as_json(self, span):
             "timestamp": convert_from_otel_timestamp(span.end_time),
         }  # type: Event
 
-        measurements = extract_span_attributes(span, SentrySpanAttribute.MEASUREMENT)
-        if measurements:
-            common_json["measurements"] = measurements
-
         tags = extract_span_attributes(span, SentrySpanAttribute.TAG)
         if tags:
             common_json["tags"] = tags

sentry_sdk/opentelemetry/utils.py

@@ -309,15 +309,7 @@ def extract_span_attributes(span, namespace):
     for attr, value in (span.attributes or {}).items():
         if attr.startswith(namespace):
             key = attr[len(namespace) + 1 :]
-
-            if namespace == SentrySpanAttribute.MEASUREMENT:
-                value = cast("tuple[str, str]", value)
-                extracted_attrs[key] = {
-                    "value": float(value[0]),
-                    "unit": value[1],
-                }
-            else:
-                extracted_attrs[key] = value
+            extracted_attrs[key] = value
 
     return extracted_attrs
 

sentry_sdk/tracing.py

@@ -65,7 +65,6 @@
     R = TypeVar("R")
 
     from sentry_sdk._types import (
-        MeasurementUnit,
         SamplingContext,
     )
 
@@ -150,10 +149,6 @@ def finish(
         # type: (...) -> None
         pass
 
-    def set_measurement(self, name, value, unit=""):
-        # type: (str, float, MeasurementUnit) -> None
-        pass
-
     def set_context(self, key, value):
         # type: (str, dict[str, Any]) -> None
         pass
@@ -540,13 +535,6 @@ def set_status(self, status):
         else:
             self._otel_span.set_status(Status(otel_status, otel_description))
 
-    def set_measurement(self, name, value, unit=""):
-        # type: (str, float, MeasurementUnit) -> None
-        # Stringify value here since OTel expects all seq items to be of one type
-        self.set_attribute(
-            f"{SentrySpanAttribute.MEASUREMENT}.{name}", (str(value), unit)
-        )
-
     def set_thread(self, thread_id, thread_name):
         # type: (Optional[int], Optional[str]) -> None
         if thread_id is not None:

tests/integrations/anthropic/test_anthropic.py

@@ -127,9 +127,9 @@ def test_nonstreaming_create_message(
         assert SPANDATA.AI_INPUT_MESSAGES not in span["data"]
         assert SPANDATA.AI_RESPONSES not in span["data"]
 
-    assert span["measurements"]["ai_prompt_tokens_used"]["value"] == 10
-    assert span["measurements"]["ai_completion_tokens_used"]["value"] == 20
-    assert span["measurements"]["ai_total_tokens_used"]["value"] == 30
+    assert span["data"]["ai.prompt_tokens.used"] == 10
+    assert span["data"]["ai.completion_tokens.used"] == 20
+    assert span["data"]["ai.total_tokens.used"] == 30
     assert span["data"]["ai.streaming"] is False
 
 
@@ -197,9 +197,9 @@ async def test_nonstreaming_create_message_async(
         assert SPANDATA.AI_INPUT_MESSAGES not in span["data"]
         assert SPANDATA.AI_RESPONSES not in span["data"]
 
-    assert span["measurements"]["ai_prompt_tokens_used"]["value"] == 10
-    assert span["measurements"]["ai_completion_tokens_used"]["value"] == 20
-    assert span["measurements"]["ai_total_tokens_used"]["value"] == 30
+    assert span["data"]["ai.prompt_tokens.used"] == 10
+    assert span["data"]["ai.completion_tokens.used"] == 20
+    assert span["data"]["ai.total_tokens.used"] == 30
     assert span["data"]["ai.streaming"] is False
 
 
@@ -299,9 +299,9 @@ def test_streaming_create_message(
         assert SPANDATA.AI_INPUT_MESSAGES not in span["data"]
         assert SPANDATA.AI_RESPONSES not in span["data"]
 
-    assert span["measurements"]["ai_prompt_tokens_used"]["value"] == 10
-    assert span["measurements"]["ai_completion_tokens_used"]["value"] == 30
-    assert span["measurements"]["ai_total_tokens_used"]["value"] == 40
+    assert span["data"]["ai.prompt_tokens.used"] == 10
+    assert span["data"]["ai.completion_tokens.used"] == 30
+    assert span["data"]["ai.total_tokens.used"] == 40
     assert span["data"]["ai.streaming"] is True
 
 
@@ -404,9 +404,9 @@ async def test_streaming_create_message_async(
         assert SPANDATA.AI_INPUT_MESSAGES not in span["data"]
         assert SPANDATA.AI_RESPONSES not in span["data"]
 
-    assert span["measurements"]["ai_prompt_tokens_used"]["value"] == 10
-    assert span["measurements"]["ai_completion_tokens_used"]["value"] == 30
-    assert span["measurements"]["ai_total_tokens_used"]["value"] == 40
+    assert span["data"]["ai.prompt_tokens.used"] == 10
+    assert span["data"]["ai.completion_tokens.used"] == 30
+    assert span["data"]["ai.total_tokens.used"] == 40
     assert span["data"]["ai.streaming"] is True
 
 
@@ -536,9 +536,9 @@ def test_streaming_create_message_with_input_json_delta(
         assert SPANDATA.AI_INPUT_MESSAGES not in span["data"]
         assert SPANDATA.AI_RESPONSES not in span["data"]
 
-    assert span["measurements"]["ai_prompt_tokens_used"]["value"] == 366
-    assert span["measurements"]["ai_completion_tokens_used"]["value"] == 51
-    assert span["measurements"]["ai_total_tokens_used"]["value"] == 417
+    assert span["data"]["ai.prompt_tokens.used"] == 366
+    assert span["data"]["ai.completion_tokens.used"] == 51
+    assert span["data"]["ai.total_tokens.used"] == 417
     assert span["data"]["ai.streaming"] is True
 
 
@@ -675,9 +675,9 @@ async def test_streaming_create_message_with_input_json_delta_async(
         assert SPANDATA.AI_INPUT_MESSAGES not in span["data"]
         assert SPANDATA.AI_RESPONSES not in span["data"]
 
-    assert span["measurements"]["ai_prompt_tokens_used"]["value"] == 366
-    assert span["measurements"]["ai_completion_tokens_used"]["value"] == 51
-    assert span["measurements"]["ai_total_tokens_used"]["value"] == 417
+    assert span["data"]["ai.prompt_tokens.used"] == 366
+    assert span["data"]["ai.completion_tokens.used"] == 51
+    assert span["data"]["ai.total_tokens.used"] == 417
     assert span["data"]["ai.streaming"] is True
 
 
@@ -822,11 +822,6 @@ def test_add_ai_data_to_span_with_input_json_delta(sentry_init, capture_events):
                 content_blocks=["{'test': 'data',", "'more': 'json'}"],
             )
 
-            # assert span._data.get("ai.streaming") is True
-            # assert span._measurements.get("ai_prompt_tokens_used")["value"] == 10
-            # assert span._measurements.get("ai_completion_tokens_used")["value"] == 20
-            # assert span._measurements.get("ai_total_tokens_used")["value"] == 30
-
     (event,) = events
 
     assert len(event["spans"]) == 1
@@ -836,6 +831,6 @@ def test_add_ai_data_to_span_with_input_json_delta(sentry_init, capture_events):
         [{"type": "text", "text": "{'test': 'data','more': 'json'}"}]
     )
     assert span["data"]["ai.streaming"] is True
-    assert span["measurements"]["ai_prompt_tokens_used"]["value"] == 10
-    assert span["measurements"]["ai_completion_tokens_used"]["value"] == 20
-    assert span["measurements"]["ai_total_tokens_used"]["value"] == 30
+    assert span["data"]["ai.prompt_tokens.used"] == 10
+    assert span["data"]["ai.completion_tokens.used"] == 20
+    assert span["data"]["ai.total_tokens.used"] == 30

tests/integrations/cohere/test_cohere.py

@@ -64,9 +64,9 @@ def test_nonstreaming_chat(
         assert "ai.input_messages" not in span["data"]
         assert "ai.responses" not in span["data"]
 
-    assert span["measurements"]["ai_completion_tokens_used"]["value"] == 10
-    assert span["measurements"]["ai_prompt_tokens_used"]["value"] == 20
-    assert span["measurements"]["ai_total_tokens_used"]["value"] == 30
+    assert span["data"]["ai.completion_tokens.used"] == 10
+    assert span["data"]["ai.prompt_tokens.used"] == 20
+    assert span["data"]["ai.total_tokens.used"] == 30
 
 
 # noinspection PyTypeChecker
@@ -136,9 +136,9 @@ def test_streaming_chat(sentry_init, capture_events, send_default_pii, include_p
         assert "ai.input_messages" not in span["data"]
         assert "ai.responses" not in span["data"]
 
-    assert span["measurements"]["ai_completion_tokens_used"]["value"] == 10
-    assert span["measurements"]["ai_prompt_tokens_used"]["value"] == 20
-    assert span["measurements"]["ai_total_tokens_used"]["value"] == 30
+    assert span["data"]["ai.completion_tokens.used"] == 10
+    assert span["data"]["ai.prompt_tokens.used"] == 20
+    assert span["data"]["ai.total_tokens.used"] == 30
 
 
 def test_bad_chat(sentry_init, capture_events):
@@ -200,8 +200,8 @@ def test_embed(sentry_init, capture_events, send_default_pii, include_prompts):
     else:
         assert "ai.input_messages" not in span["data"]
 
-    assert span["measurements"]["ai_prompt_tokens_used"]["value"] == 10
-    assert span["measurements"]["ai_total_tokens_used"]["value"] == 10
+    assert span["data"]["ai.prompt_tokens.used"] == 10
+    assert span["data"]["ai.total_tokens.used"] == 10
 
 
 def test_span_origin_chat(sentry_init, capture_events):

tests/integrations/huggingface_hub/test_huggingface_hub.py

@@ -74,7 +74,7 @@ def test_nonstreaming_chat_completion(
         assert "ai.responses" not in span["data"]
 
     if details_arg:
-        assert span["measurements"]["ai_total_tokens_used"]["value"] == 10
+        assert span["data"]["ai.total_tokens.used"] == 10
 
 
 @pytest.mark.parametrize(
@@ -133,7 +133,7 @@ def test_streaming_chat_completion(
         assert "ai.responses" not in span["data"]
 
     if details_arg:
-        assert span["measurements"]["ai_total_tokens_used"]["value"] == 10
+        assert span["data"]["ai.total_tokens.used"] == 10
 
 
 def test_bad_chat_completion(sentry_init, capture_events):

tests/integrations/langchain/test_langchain.py

@@ -179,12 +179,13 @@ def test_langchain_agent(
     assert len(list(x for x in tx["spans"] if x["op"] == "ai.run.langchain")) > 0
 
     if use_unknown_llm_type:
-        assert "ai_prompt_tokens_used" in chat_spans[0]["measurements"]
-        assert "ai_total_tokens_used" in chat_spans[0]["measurements"]
+        assert "ai.prompt_tokens.used" in chat_spans[0]["data"]
+        assert "ai.total_tokens.used" in chat_spans[0]["data"]
     else:
         # important: to avoid double counting, we do *not* measure
         # tokens used if we have an explicit integration (e.g. OpenAI)
-        assert "measurements" not in chat_spans[0]
+        assert "ai.prompt_tokens.used" not in chat_spans[0]["data"]
+        assert "ai.total_tokens.used" not in chat_spans[0]["data"]
 
     if send_default_pii and include_prompts:
         assert "You are very powerful" in chat_spans[0]["data"]["ai.input_messages"]