openlit 1.34.29__py3-none-any.whl → 1.34.31__py3-none-any.whl

openlit/instrumentation/bedrock/bedrock.py

@@ -4,10 +4,7 @@ Module for monitoring Amazon Bedrock API calls.
 
 import time
 from opentelemetry.trace import SpanKind
-from openlit.__helpers import (
-    handle_exception,
-    set_server_address_and_port
-)
+from openlit.__helpers import handle_exception, set_server_address_and_port
 from openlit.instrumentation.bedrock.utils import (
     process_chunk,
     process_chat_response,
@@ -15,7 +12,17 @@ from openlit.instrumentation.bedrock.utils import (
 )
 from openlit.semcov import SemanticConvention
 
-def converse(version, environment, application_name, tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+
+def converse(
+    version,
+    environment,
+    application_name,
+    tracer,
+    pricing_info,
+    capture_message_content,
+    metrics,
+    disable_metrics,
+):
     """
     Generates a telemetry wrapper for AWS Bedrock converse calls.
     """
@@ -30,10 +37,14 @@ def converse(version, environment, application_name, tracer, pricing_info, captu
             Wraps the individual converse method call.
             """
 
-            server_address, server_port = set_server_address_and_port(instance, "aws.amazon.com", 443)
+            server_address, server_port = set_server_address_and_port(
+                instance, "aws.amazon.com", 443
+            )
             request_model = method_kwargs.get("modelId", "amazon.titan-text-express-v1")
 
-            span_name = f"{SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT} {request_model}"
+            span_name = (
+                f"{SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT} {request_model}"
+            )
 
             with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
                 start_time = time.time()
@@ -56,7 +67,7 @@ def converse(version, environment, application_name, tracer, pricing_info, captu
                         disable_metrics=disable_metrics,
                         version=version,
                         llm_config=llm_config,
-                        **method_kwargs
+                        **method_kwargs,
                     )
 
                 except Exception as e:
@@ -70,13 +81,25 @@ def converse(version, environment, application_name, tracer, pricing_info, captu
         # Replace the original method with the instrumented one
         if kwargs.get("service_name") == "bedrock-runtime":
             original_invoke_model = client.converse
-            client.converse = lambda *args, **kwargs: converse_wrapper(original_invoke_model, *args, **kwargs)
+            client.converse = lambda *args, **kwargs: converse_wrapper(
+                original_invoke_model, *args, **kwargs
+            )
 
         return client
 
     return wrapper
 
-def converse_stream(version, environment, application_name, tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+
+def converse_stream(
+    version,
+    environment,
+    application_name,
+    tracer,
+    pricing_info,
+    capture_message_content,
+    metrics,
+    disable_metrics,
+):
     """
     Generates a telemetry wrapper for AWS Bedrock converse_stream calls.
     """
@@ -87,15 +110,15 @@ def converse_stream(version, environment, application_name, tracer, pricing_info
         """
 
         def __init__(
-                self,
-                wrapped_response,
-                span,
-                span_name,
-                kwargs,
-                server_address,
-                server_port,
-                **args,
-            ):
+            self,
+            wrapped_response,
+            span,
+            span_name,
+            kwargs,
+            server_address,
+            server_port,
+            **args,
+        ):
             self.__wrapped_response = wrapped_response
             # Extract the actual stream iterator from the response
             if isinstance(wrapped_response, dict) and "stream" in wrapped_response:
@@ -159,7 +182,9 @@ def converse_stream(version, environment, application_name, tracer, pricing_info
             except StopIteration:
                 try:
                     llm_config = self._kwargs.get("inferenceConfig", {})
-                    with tracer.start_as_current_span(self._span_name, kind=SpanKind.CLIENT) as self._span:
+                    with tracer.start_as_current_span(
+                        self._span_name, kind=SpanKind.CLIENT
+                    ) as self._span:
                         process_streaming_chat_response(
                             self,
                             pricing_info=pricing_info,
@@ -169,7 +194,7 @@ def converse_stream(version, environment, application_name, tracer, pricing_info
                             capture_message_content=capture_message_content,
                             disable_metrics=disable_metrics,
                             version=version,
-                            llm_config=llm_config
+                            llm_config=llm_config,
                         )
 
                 except Exception as e:
@@ -187,16 +212,27 @@ def converse_stream(version, environment, application_name, tracer, pricing_info
             Wraps the individual converse_stream method call.
             """
 
-            server_address, server_port = set_server_address_and_port(instance, "aws.amazon.com", 443)
+            server_address, server_port = set_server_address_and_port(
+                instance, "aws.amazon.com", 443
+            )
             request_model = method_kwargs.get("modelId", "amazon.titan-text-express-v1")
 
-            span_name = f"{SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT} {request_model}"
+            span_name = (
+                f"{SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT} {request_model}"
+            )
 
             # Get the streaming response
             stream_response = original_method(*method_args, **method_kwargs)
             span = tracer.start_span(span_name, kind=SpanKind.CLIENT)
 
-            return TracedSyncStream(stream_response, span, span_name, method_kwargs, server_address, server_port)
+            return TracedSyncStream(
+                stream_response,
+                span,
+                span_name,
+                method_kwargs,
+                server_address,
+                server_port,
+            )
 
         # Get the original client instance from the wrapper
         client = wrapped(*args, **kwargs)
@@ -204,7 +240,9 @@ def converse_stream(version, environment, application_name, tracer, pricing_info
         # Replace the original method with the instrumented one
         if kwargs.get("service_name") == "bedrock-runtime":
             original_stream_model = client.converse_stream
-            client.converse_stream = lambda *args, **kwargs: converse_stream_wrapper(original_stream_model, *args, **kwargs)
+            client.converse_stream = lambda *args, **kwargs: converse_stream_wrapper(
+                original_stream_model, *args, **kwargs
+            )
 
         return client
 

openlit/instrumentation/bedrock/utils.py

@@ -1,6 +1,7 @@
 """
 AWS Bedrock OpenTelemetry instrumentation utility functions
 """
+
 import time
 
 from opentelemetry.trace import Status, StatusCode
@@ -12,10 +13,11 @@ from openlit.__helpers import (
     get_chat_model_cost,
     record_completion_metrics,
     common_span_attributes,
-    handle_exception
+    handle_exception,
 )
 from openlit.semcov import SemanticConvention
 
+
 def format_content(messages):
     """
     Format the messages into a string for span events.
@@ -52,6 +54,7 @@ def format_content(messages):
 
     return "\n".join(formatted_messages)
 
+
 def process_chunk(self, chunk):
     """
     Process a chunk of response data and update state.
@@ -92,8 +95,19 @@ def process_chunk(self, chunk):
         self._output_tokens = usage.get("outputTokens", 0)
         self._end_time = end_time
 
-def common_chat_logic(scope, pricing_info, environment, application_name, metrics,
-                        capture_message_content, disable_metrics, version, llm_config, is_stream):
+
+def common_chat_logic(
+    scope,
+    pricing_info,
+    environment,
+    application_name,
+    metrics,
+    capture_message_content,
+    disable_metrics,
+    version,
+    llm_config,
+    is_stream,
+):
     """
     Process chat request and generate Telemetry
     """
@@ -105,13 +119,26 @@ def common_chat_logic(scope, pricing_info, environment, application_name, metric
     formatted_messages = format_content(scope._kwargs.get("messages", []))
     request_model = scope._kwargs.get("modelId", "amazon.titan-text-express-v1")
 
-    cost = get_chat_model_cost(request_model, pricing_info, scope._input_tokens, scope._output_tokens)
+    cost = get_chat_model_cost(
+        request_model, pricing_info, scope._input_tokens, scope._output_tokens
+    )
 
     # Common Span Attributes
-    common_span_attributes(scope,
-        SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT, SemanticConvention.GEN_AI_SYSTEM_AWS_BEDROCK,
-        scope._server_address, scope._server_port, request_model, scope._response_model,
-        environment, application_name, is_stream, scope._tbt, scope._ttft, version)
+    common_span_attributes(
+        scope,
+        SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT,
+        SemanticConvention.GEN_AI_SYSTEM_AWS_BEDROCK,
+        scope._server_address,
+        scope._server_port,
+        request_model,
+        scope._response_model,
+        environment,
+        application_name,
+        is_stream,
+        scope._tbt,
+        scope._ttft,
+        version,
+    )
 
     # Bedrock-specific attributes from llm_config
     bedrock_attributes = [
@@ -132,19 +159,35 @@ def common_chat_logic(scope, pricing_info, environment, application_name, metric
 
     # Span Attributes for Response parameters
     scope._span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_ID, scope._response_id)
-    scope._span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_FINISH_REASON, [scope._finish_reason])
-    scope._span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE, "text" if isinstance(scope._llmresponse, str) else "json")
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_RESPONSE_FINISH_REASON, [scope._finish_reason]
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_OUTPUT_TYPE,
+        "text" if isinstance(scope._llmresponse, str) else "json",
+    )
 
     # Span Attributes for Cost and Tokens
-    scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS, scope._input_tokens)
-    scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_OUTPUT_TOKENS, scope._output_tokens)
-    scope._span.set_attribute(SemanticConvention.GEN_AI_CLIENT_TOKEN_USAGE, scope._input_tokens + scope._output_tokens)
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS, scope._input_tokens
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_USAGE_OUTPUT_TOKENS, scope._output_tokens
+    )
+    scope._span.set_attribute(
+        SemanticConvention.GEN_AI_CLIENT_TOKEN_USAGE,
+        scope._input_tokens + scope._output_tokens,
+    )
     scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_COST, cost)
 
     # Span Attributes for Content
     if capture_message_content:
-        scope._span.set_attribute(SemanticConvention.GEN_AI_CONTENT_PROMPT, formatted_messages)
-        scope._span.set_attribute(SemanticConvention.GEN_AI_CONTENT_COMPLETION, scope._llmresponse)
+        scope._span.set_attribute(
+            SemanticConvention.GEN_AI_CONTENT_PROMPT, formatted_messages
+        )
+        scope._span.set_attribute(
+            SemanticConvention.GEN_AI_CONTENT_COMPLETION, scope._llmresponse
+        )
 
         # To be removed once the change to span_attributes (from span events) is complete
         scope._span.add_event(
@@ -164,13 +207,37 @@ def common_chat_logic(scope, pricing_info, environment, application_name, metric
 
     # Record metrics
     if not disable_metrics:
-        record_completion_metrics(metrics, SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT, SemanticConvention.GEN_AI_SYSTEM_AWS_BEDROCK,
-            scope._server_address, scope._server_port, request_model, scope._response_model, environment,
-            application_name, scope._start_time, scope._end_time, scope._input_tokens, scope._output_tokens,
-            cost, scope._tbt, scope._ttft)
+        record_completion_metrics(
+            metrics,
+            SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT,
+            SemanticConvention.GEN_AI_SYSTEM_AWS_BEDROCK,
+            scope._server_address,
+            scope._server_port,
+            request_model,
+            scope._response_model,
+            environment,
+            application_name,
+            scope._start_time,
+            scope._end_time,
+            scope._input_tokens,
+            scope._output_tokens,
+            cost,
+            scope._tbt,
+            scope._ttft,
+        )
 
-def process_streaming_chat_response(scope, pricing_info, environment, application_name, metrics,
-                                    capture_message_content=False, disable_metrics=False, version="", llm_config=None):
+
+def process_streaming_chat_response(
+    scope,
+    pricing_info,
+    environment,
+    application_name,
+    metrics,
+    capture_message_content=False,
+    disable_metrics=False,
+    version="",
+    llm_config=None,
+):
     """
     Process streaming chat response and generate telemetry.
     """
@@ -179,15 +246,40 @@ def process_streaming_chat_response(scope, pricing_info, environment, applicatio
         if llm_config is None:
             llm_config = {}
 
-        common_chat_logic(scope, pricing_info, environment, application_name, metrics,
-                            capture_message_content, disable_metrics, version, llm_config, is_stream=True)
+        common_chat_logic(
+            scope,
+            pricing_info,
+            environment,
+            application_name,
+            metrics,
+            capture_message_content,
+            disable_metrics,
+            version,
+            llm_config,
+            is_stream=True,
+        )
     except Exception as e:
         handle_exception(scope._span, e)
         raise
 
-def process_chat_response(response, request_model, pricing_info, server_port, server_address, environment,
-                          application_name, metrics, start_time, span, capture_message_content=False,
-                          disable_metrics=False, version="1.0.0", llm_config=None, **kwargs):
+
+def process_chat_response(
+    response,
+    request_model,
+    pricing_info,
+    server_port,
+    server_address,
+    environment,
+    application_name,
+    metrics,
+    start_time,
+    span,
+    capture_message_content=False,
+    disable_metrics=False,
+    version="1.0.0",
+    llm_config=None,
+    **kwargs,
+):
     """
     Process non-streaming chat response and generate telemetry.
     """
@@ -202,8 +294,15 @@ def process_chat_response(response, request_model, pricing_info, server_port, se
         scope._start_time = start_time
         scope._end_time = time.time()
         scope._span = span
-        scope._llmresponse = response_dict.get("output", {}).get("message", {}).get("content", [{}])[0].get("text", "")
-        scope._response_role = response_dict.get("output", {}).get("message", {}).get("role", "assistant")
+        scope._llmresponse = (
+            response_dict.get("output", {})
+            .get("message", {})
+            .get("content", [{}])[0]
+            .get("text", "")
+        )
+        scope._response_role = (
+            response_dict.get("output", {}).get("message", {}).get("role", "assistant")
+        )
         scope._input_tokens = response_dict.get("usage", {}).get("inputTokens", 0)
         scope._output_tokens = response_dict.get("usage", {}).get("outputTokens", 0)
         scope._response_model = request_model
@@ -214,8 +313,18 @@ def process_chat_response(response, request_model, pricing_info, server_port, se
         scope._server_address, scope._server_port = server_address, server_port
         scope._kwargs = kwargs
 
-        common_chat_logic(scope, pricing_info, environment, application_name, metrics,
-                            capture_message_content, disable_metrics, version, llm_config, is_stream=False)
+        common_chat_logic(
+            scope,
+            pricing_info,
+            environment,
+            application_name,
+            metrics,
+            capture_message_content,
+            disable_metrics,
+            version,
+            llm_config,
+            is_stream=False,
+        )
 
         return response
     except Exception as e:

openlit/instrumentation/chroma/__init__.py

@@ -11,6 +11,7 @@ from openlit.instrumentation.chroma.chroma import general_wrap
 
 _instruments = ("chromadb >= 0.4.0",)
 
+
 class ChromaInstrumentor(BaseInstrumentor):
     """
     An instrumentor for ChromaDB client library.
@@ -33,57 +34,129 @@ class ChromaInstrumentor(BaseInstrumentor):
         wrap_function_wrapper(
             "chromadb.db",
             "DB.create_collection",
-            general_wrap("chroma.create_collection", version, environment, application_name,
-                tracer, pricing_info, capture_message_content, metrics, disable_metrics),
+            general_wrap(
+                "chroma.create_collection",
+                version,
+                environment,
+                application_name,
+                tracer,
+                pricing_info,
+                capture_message_content,
+                metrics,
+                disable_metrics,
+            ),
         )
 
         wrap_function_wrapper(
             "chromadb",
             "Collection.add",
-            general_wrap("chroma.add", version, environment, application_name,
-                tracer, pricing_info, capture_message_content, metrics, disable_metrics),
+            general_wrap(
+                "chroma.add",
+                version,
+                environment,
+                application_name,
+                tracer,
+                pricing_info,
+                capture_message_content,
+                metrics,
+                disable_metrics,
+            ),
         )
 
         wrap_function_wrapper(
             "chromadb",
             "Collection.get",
-            general_wrap("chroma.get", version, environment, application_name,
-                tracer, pricing_info, capture_message_content, metrics, disable_metrics),
+            general_wrap(
+                "chroma.get",
+                version,
+                environment,
+                application_name,
+                tracer,
+                pricing_info,
+                capture_message_content,
+                metrics,
+                disable_metrics,
+            ),
         )
 
         wrap_function_wrapper(
             "chromadb",
             "Collection.peek",
-            general_wrap("chroma.peek", version, environment, application_name,
-                tracer, pricing_info, capture_message_content, metrics, disable_metrics),
+            general_wrap(
+                "chroma.peek",
+                version,
+                environment,
+                application_name,
+                tracer,
+                pricing_info,
+                capture_message_content,
+                metrics,
+                disable_metrics,
+            ),
         )
 
         wrap_function_wrapper(
             "chromadb",
             "Collection.query",
-            general_wrap("chroma.query", version, environment, application_name,
-                tracer, pricing_info, capture_message_content, metrics, disable_metrics),
+            general_wrap(
+                "chroma.query",
+                version,
+                environment,
+                application_name,
+                tracer,
+                pricing_info,
+                capture_message_content,
+                metrics,
+                disable_metrics,
+            ),
         )
 
         wrap_function_wrapper(
             "chromadb",
             "Collection.update",
-            general_wrap("chroma.update", version, environment, application_name,
-                tracer, pricing_info, capture_message_content, metrics, disable_metrics),
+            general_wrap(
+                "chroma.update",
+                version,
+                environment,
+                application_name,
+                tracer,
+                pricing_info,
+                capture_message_content,
+                metrics,
+                disable_metrics,
+            ),
         )
 
         wrap_function_wrapper(
             "chromadb",
             "Collection.upsert",
-            general_wrap("chroma.upsert", version, environment, application_name,
-                tracer, pricing_info, capture_message_content, metrics, disable_metrics),
+            general_wrap(
+                "chroma.upsert",
+                version,
+                environment,
+                application_name,
+                tracer,
+                pricing_info,
+                capture_message_content,
+                metrics,
+                disable_metrics,
+            ),
         )
 
         wrap_function_wrapper(
             "chromadb",
             "Collection.delete",
-            general_wrap("chroma.delete", version, environment, application_name,
-                tracer, pricing_info, capture_message_content, metrics, disable_metrics),
+            general_wrap(
+                "chroma.delete",
+                version,
+                environment,
+                application_name,
+                tracer,
+                pricing_info,
+                capture_message_content,
+                metrics,
+                disable_metrics,
+            ),
         )
 
     def _uninstrument(self, **kwargs):

openlit/instrumentation/chroma/chroma.py

@@ -12,8 +12,18 @@ from openlit.instrumentation.chroma.utils import (
     set_server_address_and_port,
 )
 
-def general_wrap(gen_ai_endpoint, version, environment, application_name,
-    tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+
+def general_wrap(
+    gen_ai_endpoint,
+    version,
+    environment,
+    application_name,
+    tracer,
+    pricing_info,
+    capture_message_content,
+    metrics,
+    disable_metrics,
+):
     """
     Generates a telemetry wrapper for ChromaDB function calls.
     """
@@ -43,9 +53,22 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
             try:
                 # Process response and generate telemetry
                 response = process_vectordb_response(
-                    response, db_operation, server_address, server_port,
-                    environment, application_name, metrics, start_time, span,
-                    capture_message_content, disable_metrics, version, instance, args, endpoint=gen_ai_endpoint, **kwargs
+                    response,
+                    db_operation,
+                    server_address,
+                    server_port,
+                    environment,
+                    application_name,
+                    metrics,
+                    start_time,
+                    span,
+                    capture_message_content,
+                    disable_metrics,
+                    version,
+                    instance,
+                    args,
+                    endpoint=gen_ai_endpoint,
+                    **kwargs,
                 )
 
             except Exception as e:
@@ -53,5 +76,4 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
 
             return response
 
-
     return wrapper