openlit 1.34.8__py3-none-any.whl → 1.34.10__py3-none-any.whl

openlit/instrumentation/elevenlabs/elevenlabs.py

@@ -1,145 +1,55 @@
 """
-Module for monitoring Ollama API calls.
+Module for monitoring ElevenLabs API calls.
 """
 
-import logging
 import time
-from opentelemetry.trace import SpanKind, Status, StatusCode
-from opentelemetry.sdk.resources import SERVICE_NAME, TELEMETRY_SDK_NAME, DEPLOYMENT_ENVIRONMENT
-from openlit.__helpers import (
-    get_audio_model_cost,
-    handle_exception,
-    create_metrics_attributes,
-)
+from opentelemetry.trace import SpanKind
+from openlit.__helpers import handle_exception
+from openlit.instrumentation.elevenlabs.utils import process_audio_response
 from openlit.semcov import SemanticConvention
 
-# Initialize logger for logging potential issues and operations
-logger = logging.getLogger(__name__)
-
 def generate(gen_ai_endpoint, version, environment, application_name,
-                 tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+    tracer, pricing_info, capture_message_content, metrics, disable_metrics):
     """
-    Generates a telemetry wrapper for creating speech audio to collect metrics.
-    
-    Args:
-        version: Version of the monitoring package.
-        environment: Deployment environment (e.g., production, staging).
-        application_name: Name of the application using the ElevenLabs API.
-        tracer: OpenTelemetry tracer for creating spans.
-        pricing_info: Information used for calculating the cost of generating speech audio.
-        capture_message_content: Flag indicating whether to trace the input text and generated audio.
-    
-    Returns:
-        A function that wraps the speech audio creation method to add telemetry.
+    Generates a telemetry wrapper for GenAI function call
     """
 
     def wrapper(wrapped, instance, args, kwargs):
         """
-        Wraps the 'generate' API call to add telemetry.
-
-        This collects metrics such as execution time, cost, and handles errors
-        gracefully, adding details to the trace for observability.
-
-        Args:
-            wrapped: The original 'generate' method to be wrapped.
-            instance: The instance of the class where the original method is defined.
-            args: Positional arguments for the 'generate' method.
-            kwargs: Keyword arguments for the 'generate' method.
-
-        Returns:
-            The response from the original 'generate' method.
+        Wraps the GenAI function call.
         """
 
         server_address, server_port = "api.elevenlabs.io", 443
-        request_model = kwargs.get('model', kwargs.get('model_id', 'eleven_multilingual_v2'))
+        request_model = kwargs.get("model", kwargs.get("model_id", "eleven_multilingual_v2"))
 
-        span_name = f'{SemanticConvention.GEN_AI_OPERATION_TYPE_AUDIO} {request_model}'
+        span_name = f"{SemanticConvention.GEN_AI_OPERATION_TYPE_AUDIO} {request_model}"
 
-        with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
+        with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
             start_time = time.time()
             response = wrapped(*args, **kwargs)
-            end_time = time.time()
 
             try:
-                # Calculate cost of the operation
-                cost = get_audio_model_cost(request_model,
-                                            pricing_info, kwargs.get('text', ''))
-
-                # Set Span attributes
-                span.set_attribute(TELEMETRY_SDK_NAME, 'openlit')
-                span.set_attribute(SemanticConvention.GEN_AI_OPERATION,
-                                    SemanticConvention.GEN_AI_OPERATION_TYPE_AUDIO)
-                span.set_attribute(SemanticConvention.GEN_AI_SYSTEM,
-                                    SemanticConvention.GEN_AI_SYSTEM_ASSEMBLYAI)
-                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MODEL,
-                                    request_model)
-                span.set_attribute(SemanticConvention.SERVER_ADDRESS,
-                                    server_address)
-                span.set_attribute(SemanticConvention.SERVER_PORT,
-                                    server_port)
-                span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_MODEL,
-                                    request_model)
-                span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MODEL,
-                                    request_model)
-                span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE,
-                                    'audio')
-
-                # Set Span attributes (Extras)
-                if gen_ai_endpoint == 'elevenlabs.generate':
-                    if isinstance(kwargs.get('voice', 'Rachel'), str):
-                        span.set_attribute(SemanticConvention.GEN_AI_REQUEST_AUDIO_VOICE,
-                                        kwargs.get('voice', 'Rachel'))
-                else:
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_AUDIO_VOICE,
-                                        kwargs.get('voice_id', ''))
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_AUDIO_RESPONSE_FORMAT,
-                                        kwargs.get('output_format', 'mp3'))
-                    span.set_attribute(SemanticConvention.GEN_AI_REQUEST_AUDIO_SETTINGS,
-                                        str(kwargs.get('voice_settings', '')))
-                    span.set_attribute(DEPLOYMENT_ENVIRONMENT,
-                                        environment)
-                    span.set_attribute(SERVICE_NAME,
-                                        application_name)
-                    span.set_attribute(SemanticConvention.GEN_AI_USAGE_COST,
-                                        cost)
-                    span.set_attribute(SemanticConvention.GEN_AI_SDK_VERSION,
-                                        version)
-                if capture_message_content:
-                    span.add_event(
-                        name=SemanticConvention.GEN_AI_CONTENT_PROMPT_EVENT,
-                        attributes={
-                            SemanticConvention.GEN_AI_CONTENT_PROMPT: str(kwargs.get('text', '')),
-                        },
-                    )
-
-                span.set_status(Status(StatusCode.OK))
-
-                if disable_metrics is False:
-                    attributes = create_metrics_attributes(
-                        service_name=application_name,
-                        deployment_environment=environment,
-                        operation=SemanticConvention.GEN_AI_OPERATION_TYPE_AUDIO,
-                        system=SemanticConvention.GEN_AI_SYSTEM_ELEVENLABS,
-                        request_model=request_model,
-                        server_address=server_address,
-                        server_port=server_port,
-                        response_model=request_model,
-                    )
-
-                    metrics['genai_client_operation_duration'].record(
-                        end_time - start_time, attributes
-                    )
-                    metrics['genai_requests'].add(1, attributes)
-                    metrics['genai_cost'].record(cost, attributes)
-
-                # Return original response
-                return response
+                response = process_audio_response(
+                    response=response,
+                    gen_ai_endpoint=gen_ai_endpoint,
+                    pricing_info=pricing_info,
+                    server_port=server_port,
+                    server_address=server_address,
+                    environment=environment,
+                    application_name=application_name,
+                    metrics=metrics,
+                    start_time=start_time,
+                    span=span,
+                    args=args,
+                    kwargs=kwargs,
+                    capture_message_content=capture_message_content,
+                    disable_metrics=disable_metrics,
+                    version=version
+                )
 
             except Exception as e:
                 handle_exception(span, e)
-                logger.error('Error in trace creation: %s', e)
 
-                # Return original response
-                return response
+            return response
 
     return wrapper

openlit/instrumentation/elevenlabs/utils.py

@@ -0,0 +1,133 @@
+"""
+ElevenLabs OpenTelemetry instrumentation utility functions
+"""
+import time
+
+from opentelemetry.sdk.resources import SERVICE_NAME, TELEMETRY_SDK_NAME, DEPLOYMENT_ENVIRONMENT
+from opentelemetry.trace import Status, StatusCode
+
+from openlit.__helpers import (
+    get_audio_model_cost,
+    create_metrics_attributes,
+)
+from openlit.semcov import SemanticConvention
+
+def format_content(text):
+    """
+    Process text input to extract content.
+    """
+    return str(text) if text else ""
+
+def common_span_attributes(scope, gen_ai_operation, gen_ai_system, server_address, server_port,
+    request_model, response_model, environment, application_name, is_stream, tbt, ttft, version):
+    """
+    Set common span attributes for both chat and RAG operations.
+    """
+
+    scope._span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
+    scope._span.set_attribute(SemanticConvention.GEN_AI_OPERATION, gen_ai_operation)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_SYSTEM, gen_ai_system)
+    scope._span.set_attribute(SemanticConvention.SERVER_ADDRESS, server_address)
+    scope._span.set_attribute(SemanticConvention.SERVER_PORT, server_port)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_MODEL, request_model)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_RESPONSE_MODEL, scope._response_model)
+    scope._span.set_attribute(DEPLOYMENT_ENVIRONMENT, environment)
+    scope._span.set_attribute(SERVICE_NAME, application_name)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_IS_STREAM, is_stream)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_SERVER_TBT, scope._tbt)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_SERVER_TTFT, scope._ttft)
+    scope._span.set_attribute(SemanticConvention.GEN_AI_SDK_VERSION, version)
+
+def record_audio_metrics(metrics, gen_ai_operation, gen_ai_system, server_address, server_port,
+    request_model, response_model, environment, application_name, start_time, end_time, cost):
+    """
+    Record audio generation metrics for the operation.
+    """
+
+    attributes = create_metrics_attributes(
+        operation=gen_ai_operation,
+        system=gen_ai_system,
+        server_address=server_address,
+        server_port=server_port,
+        request_model=request_model,
+        response_model=response_model,
+        service_name=application_name,
+        deployment_environment=environment,
+    )
+    metrics["genai_client_operation_duration"].record(end_time - start_time, attributes)
+    metrics["genai_requests"].add(1, attributes)
+    metrics["genai_cost"].record(cost, attributes)
+
+def common_audio_logic(scope, gen_ai_endpoint, pricing_info, environment, application_name,
+    metrics, capture_message_content, disable_metrics, version):
+    """
+    Process audio generation request and generate Telemetry
+    """
+
+    text = format_content(scope._kwargs.get("text", ""))
+    request_model = scope._kwargs.get("model", scope._kwargs.get("model_id", "eleven_multilingual_v2"))
+    is_stream = False  # ElevenLabs audio generation is not streaming
+
+    cost = get_audio_model_cost(request_model, pricing_info, text)
+
+    # Common Span Attributes
+    common_span_attributes(scope,
+        SemanticConvention.GEN_AI_OPERATION_TYPE_AUDIO, SemanticConvention.GEN_AI_SYSTEM_ELEVENLABS,
+        scope._server_address, scope._server_port, request_model, request_model,
+        environment, application_name, is_stream, scope._tbt, scope._ttft, version)
+
+    # Span Attributes for Cost and Tokens
+    scope._span.set_attribute(SemanticConvention.GEN_AI_USAGE_COST, cost)
+
+    # Span Attributes for Response parameters
+    scope._span.set_attribute(SemanticConvention.GEN_AI_OUTPUT_TYPE, scope._kwargs.get("output_format", "mp3_44100_128"))
+
+    # Audio-specific span attributes
+    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_AUDIO_VOICE, scope._kwargs.get("voice_id", ""))
+    scope._span.set_attribute(SemanticConvention.GEN_AI_REQUEST_AUDIO_SETTINGS, str(scope._kwargs.get("voice_settings", "")))
+
+    # Span Attributes for Content
+    if capture_message_content:
+        scope._span.set_attribute(SemanticConvention.GEN_AI_CONTENT_PROMPT, text)
+
+        # To be removed once the change to span_attributes (from span events) is complete
+        scope._span.add_event(
+            name=SemanticConvention.GEN_AI_CONTENT_PROMPT_EVENT,
+            attributes={
+                SemanticConvention.GEN_AI_CONTENT_PROMPT: text,
+            },
+        )
+
+    scope._span.set_status(Status(StatusCode.OK))
+
+    # Metrics
+    if not disable_metrics:
+        record_audio_metrics(metrics, SemanticConvention.GEN_AI_OPERATION_TYPE_AUDIO, SemanticConvention.GEN_AI_SYSTEM_ELEVENLABS,
+            scope._server_address, scope._server_port, request_model, request_model, environment,
+            application_name, scope._start_time, scope._end_time, cost)
+
+def process_audio_response(response, gen_ai_endpoint, pricing_info, server_port, server_address,
+    environment, application_name, metrics, start_time, span, args, kwargs, capture_message_content=False,
+    disable_metrics=False, version="1.0.0"):
+    """
+    Process audio generation request and generate Telemetry
+    """
+
+    scope = type("GenericScope", (), {})()
+
+    scope._start_time = start_time
+    scope._end_time = time.time()
+    scope._span = span
+    scope._server_address, scope._server_port = server_address, server_port
+    scope._kwargs = kwargs
+    scope._args = args
+
+    # Initialize streaming and timing values for ElevenLabs audio generation
+    scope._response_model = kwargs.get("model", kwargs.get("model_id", "eleven_multilingual_v2"))
+    scope._tbt = 0.0
+    scope._ttft = scope._end_time - scope._start_time
+
+    common_audio_logic(scope, gen_ai_endpoint, pricing_info, environment, application_name,
+        metrics, capture_message_content, disable_metrics, version)
+
+    return response

openlit/instrumentation/gpt4all/utils.py

@@ -109,7 +109,7 @@ def record_embedding_metrics(metrics, gen_ai_operation, gen_ai_system, server_ad
     metrics["genai_prompt_tokens"].add(input_tokens, attributes)
     metrics["genai_cost"].record(cost, attributes)
 
-def common_generate_logic(scope, pricing_info, environment, application_name, metrics,
+def common_t2s_logic(scope, pricing_info, environment, application_name, metrics,
     capture_message_content, disable_metrics, version, is_stream):
     """
     Process generate request and generate Telemetry
@@ -228,7 +228,7 @@ def process_streaming_generate_response(scope, pricing_info, environment, applic
     """
     Process generate request and generate Telemetry
     """
-    common_generate_logic(scope, pricing_info, environment, application_name, metrics,
+    common_t2s_logic(scope, pricing_info, environment, application_name, metrics,
         capture_message_content, disable_metrics, version, is_stream=True)
 
 def process_generate_response(response, request_model, pricing_info, server_port, server_address,
@@ -252,7 +252,7 @@ def process_generate_response(response, request_model, pricing_info, server_port
     scope._args = args
     scope._tools = None
 
-    common_generate_logic(scope, pricing_info, environment, application_name, metrics,
+    common_t2s_logic(scope, pricing_info, environment, application_name, metrics,
         capture_message_content, disable_metrics, version, is_stream=False)
 
     return response

openlit/instrumentation/ollama/__init__.py

@@ -41,7 +41,7 @@ def _dispatch_async(async_chat_wrap, async_emb_wrap):
 
 class OllamaInstrumentor(BaseInstrumentor):
     """
-    An instrumentor for Ollama's client library.
+    An instrumentor for Ollama client library.
     """
 
     def instrumentation_dependencies(self) -> Collection[str]:
@@ -51,7 +51,6 @@ class OllamaInstrumentor(BaseInstrumentor):
         application_name = kwargs.get("application_name", "default_application")
         environment = kwargs.get("environment", "default_environment")
         tracer = kwargs.get("tracer")
-        event_provider = kwargs.get("event_provider")
         metrics = kwargs.get("metrics_dict")
         pricing_info = kwargs.get("pricing_info", {})
         capture_message_content = kwargs.get("capture_message_content", False)
@@ -61,22 +60,22 @@ class OllamaInstrumentor(BaseInstrumentor):
         # Build wrapper factories for chat and embeddings
         sync_chat_wrap = chat(
             version, environment, application_name,
-            tracer, event_provider, pricing_info,
+            tracer, pricing_info,
             capture_message_content, metrics, disable_metrics
         )
         sync_emb_wrap = embeddings(
             version, environment, application_name,
-            tracer, event_provider, pricing_info,
+            tracer, pricing_info,
             capture_message_content, metrics, disable_metrics
         )
         async_chat_wrap = async_chat(
             version, environment, application_name,
-            tracer, event_provider, pricing_info,
+            tracer, pricing_info,
             capture_message_content, metrics, disable_metrics
         )
         async_emb_wrap = async_embeddings(
             version, environment, application_name,
-            tracer, event_provider, pricing_info,
+            tracer, pricing_info,
             capture_message_content, metrics, disable_metrics
         )
 

openlit/instrumentation/ollama/async_ollama.py

@@ -2,7 +2,6 @@
 Module for monitoring Ollama API calls.
 """
 
-import logging
 import time
 from opentelemetry.trace import SpanKind
 from openlit.__helpers import (
@@ -17,12 +16,10 @@ from openlit.instrumentation.ollama.utils import (
 )
 from openlit.semcov import SemanticConvention
 
-logger = logging.getLogger(__name__)
-
 def async_chat(version, environment, application_name,
-            tracer, event_provider, pricing_info, capture_message_content, metrics, disable_metrics):
+            tracer, pricing_info, capture_message_content, metrics, disable_metrics):
     """
-    Generates a telemetry wrapper for GenAI function call
+    Generates a telemetry wrapper for Ollama async chat function call
     """
 
     class TracedAsyncStream:
@@ -38,7 +35,7 @@ def async_chat(version, environment, application_name,
                 kwargs,
                 server_address,
                 server_port,
-                **args,
+                args,
             ):
             self.__wrapped__ = wrapped
             self._span = span
@@ -48,11 +45,11 @@ def async_chat(version, environment, application_name,
             self._tool_calls = []
             self._input_tokens = 0
             self._output_tokens = 0
-            self._response_role = ''
+            self._response_role = ""
             self._span_name = span_name
             self._args = args
             self._kwargs = kwargs
-            self._start_time = time.time()
+            self._start_time = time.monotonic()
             self._end_time = None
             self._timestamps = []
             self._ttft = 0
@@ -81,56 +78,101 @@ def async_chat(version, environment, application_name,
                 return chunk
             except StopAsyncIteration:
                 try:
-                    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,
                             environment=environment,
                             application_name=application_name,
                             metrics=metrics,
-                            event_provider=event_provider,
                             capture_message_content=capture_message_content,
                             disable_metrics=disable_metrics,
                             version=version
                         )
                 except Exception as e:
                     handle_exception(self._span, e)
-                    logger.error("Error in trace creation: %s", e)
+
                 raise
 
     async def wrapper(wrapped, instance, args, kwargs):
         """
-        Wraps the GenAI function call.
+        Wraps the Ollama async chat function call.
         """
 
         streaming = kwargs.get("stream", False)
 
         server_address, server_port = set_server_address_and_port(instance, "127.0.0.1", 11434)
-        json_body = kwargs.get("json", {}) or {}
-        request_model = json_body.get("model") or kwargs.get("model")
+        request_model = kwargs.get("model")
 
         span_name = f"{SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT} {request_model}"
 
-        # pylint: disable=no-else-return
         if streaming:
             awaited_wrapped = await wrapped(*args, **kwargs)
             span = tracer.start_span(span_name, kind=SpanKind.CLIENT)
-            return TracedAsyncStream(awaited_wrapped, span, span_name, kwargs, server_address, server_port)
+            return TracedAsyncStream(awaited_wrapped, span, span_name, kwargs, server_address, server_port, args)
 
         else:
-            with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
-                start_time = time.time()
+            with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
+                start_time = time.monotonic()
+
+                try:
+                    response = await wrapped(*args, **kwargs)
+
+                    response = process_chat_response(
+                        response=response,
+                        gen_ai_endpoint="ollama.chat",
+                        pricing_info=pricing_info,
+                        server_port=server_port,
+                        server_address=server_address,
+                        environment=environment,
+                        application_name=application_name,
+                        metrics=metrics,
+                        start_time=start_time,
+                        span=span,
+                        capture_message_content=capture_message_content,
+                        disable_metrics=disable_metrics,
+                        version=version,
+                        **kwargs
+                    )
+
+                except Exception as e:
+                    handle_exception(span, e)
+
+            return response
+
+    return wrapper
+
+def async_embeddings(version, environment, application_name,
+              tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+    """
+    Generates a telemetry wrapper for Ollama async embeddings function call
+    """
+
+    async def wrapper(wrapped, instance, args, kwargs):
+        """
+        Wraps the Ollama async embeddings function call.
+        """
+
+        server_address, server_port = set_server_address_and_port(instance, "127.0.0.1", 11434)
+        request_model = kwargs.get("model")
+
+        span_name = f"{SemanticConvention.GEN_AI_OPERATION_TYPE_EMBEDDING} {request_model}"
+
+        with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
+            start_time = time.monotonic()
+
+            try:
                 response = await wrapped(*args, **kwargs)
-                response = process_chat_response(
+
+                response = process_embedding_response(
                     response=response,
-                    request_model=request_model,
+                    gen_ai_endpoint="ollama.embeddings",
                     pricing_info=pricing_info,
                     server_port=server_port,
                     server_address=server_address,
                     environment=environment,
                     application_name=application_name,
                     metrics=metrics,
-                    event_provider=event_provider,
                     start_time=start_time,
                     span=span,
                     capture_message_content=capture_message_content,
@@ -139,47 +181,8 @@ def async_chat(version, environment, application_name,
                     **kwargs
                 )
 
-            return response
-
-    return wrapper
-
-def async_embeddings(version, environment, application_name,
-              tracer, event_provider, pricing_info, capture_message_content, metrics, disable_metrics):
-    """
-    Generates a telemetry wrapper for GenAI function call
-    """
-
-    async def wrapper(wrapped, instance, args, kwargs):
-        """
-        Wraps the GenAI function call.
-        """
-
-        server_address, server_port = set_server_address_and_port(instance, '127.0.0.1', 11434)
-        json_body = kwargs.get('json', {}) or {}
-        request_model = json_body.get('model') or kwargs.get('model')
-
-        span_name = f'{SemanticConvention.GEN_AI_OPERATION_TYPE_EMBEDDING} {request_model}'
-
-        with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
-            start_time = time.time()
-            response = await wrapped(*args, **kwargs)
-            response = process_embedding_response(
-                response=response,
-                request_model=request_model,
-                pricing_info=pricing_info,
-                server_port=server_port,
-                server_address=server_address,
-                environment=environment,
-                application_name=application_name,
-                metrics=metrics,
-                event_provider=event_provider,
-                start_time=start_time,
-                span=span,
-                capture_message_content=capture_message_content,
-                disable_metrics=disable_metrics,
-                version=version,
-                **kwargs
-            )
+            except Exception as e:
+                handle_exception(span, e)
 
         return response