openlit 1.34.14__py3-none-any.whl → 1.34.16__py3-none-any.whl

openlit/instrumentation/azure_ai_inference/__init__.py

@@ -5,49 +5,65 @@ import importlib.metadata
 from opentelemetry.instrumentation.instrumentor import BaseInstrumentor
 from wrapt import wrap_function_wrapper
 from openlit.instrumentation.azure_ai_inference.azure_ai_inference import (
-    complete
+    complete,
+    embed
 )
 from openlit.instrumentation.azure_ai_inference.async_azure_ai_inference import (
-    async_complete
+    async_complete,
+    async_embed
 )
 
-_instruments = ('azure-ai-inference >= 1.0.0b4',)
+_instruments = ("azure-ai-inference >= 1.0.0b4",)
 
 class AzureAIInferenceInstrumentor(BaseInstrumentor):
     """
-    An instrumentor for azure-ai-inference's client library.
+    An instrumentor for azure-ai-inference client library.
     """
 
     def instrumentation_dependencies(self) -> Collection[str]:
         return _instruments
 
     def _instrument(self, **kwargs):
-        application_name = kwargs.get('application_name', 'default')
-        environment = kwargs.get('environment', 'default')
-        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)
-        disable_metrics = kwargs.get('disable_metrics')
-        version = importlib.metadata.version('azure-ai-inference')
-
-        # sync generate
+        application_name = kwargs.get("application_name", "default")
+        environment = kwargs.get("environment", "default")
+        tracer = kwargs.get("tracer")
+        metrics = kwargs.get("metrics_dict")
+        pricing_info = kwargs.get("pricing_info", {})
+        capture_message_content = kwargs.get("capture_message_content", False)
+        disable_metrics = kwargs.get("disable_metrics")
+        version = importlib.metadata.version("azure-ai-inference")
+
+        # sync chat completions
         wrap_function_wrapper(
-            'azure.ai.inference',
-            'ChatCompletionsClient.complete',
+            "azure.ai.inference",
+            "ChatCompletionsClient.complete",
             complete(version, environment, application_name,
-                  tracer, event_provider, pricing_info, capture_message_content, metrics, disable_metrics),
+                  tracer, pricing_info, capture_message_content, metrics, disable_metrics),
         )
 
-        # async generate
+        # async chat completions
         wrap_function_wrapper(
-            'azure.ai.inference.aio',
-            'ChatCompletionsClient.complete',
+            "azure.ai.inference.aio",
+            "ChatCompletionsClient.complete",
             async_complete(version, environment, application_name,
-                  tracer, event_provider, pricing_info, capture_message_content, metrics, disable_metrics),
+                  tracer, pricing_info, capture_message_content, metrics, disable_metrics),
+        )
+
+        # sync embeddings
+        wrap_function_wrapper(
+            "azure.ai.inference",
+            "EmbeddingsClient.embed",
+            embed(version, environment, application_name,
+                  tracer, pricing_info, capture_message_content, metrics, disable_metrics),
+        )
+
+        # async embeddings
+        wrap_function_wrapper(
+            "azure.ai.inference.aio",
+            "EmbeddingsClient.embed",
+            async_embed(version, environment, application_name,
+                  tracer, pricing_info, capture_message_content, metrics, disable_metrics),
         )
 
     def _uninstrument(self, **kwargs):
-        # Proper uninstrumentation logic to revert patched methods
         pass

openlit/instrumentation/azure_ai_inference/async_azure_ai_inference.py

@@ -13,6 +13,7 @@ from openlit.instrumentation.azure_ai_inference.utils import (
     process_chunk,
     process_chat_response,
     process_streaming_chat_response,
+    process_embedding_response,
 )
 from openlit.semcov import SemanticConvention
 
@@ -20,7 +21,7 @@ from openlit.semcov import SemanticConvention
 logger = logging.getLogger(__name__)
 
 def async_complete(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
     """
@@ -43,12 +44,15 @@ def async_complete(version, environment, application_name,
             self.__wrapped__ = wrapped
             self._span = span
             self._span_name = span_name
-            self._llmresponse = ''
-            self._response_id = ''
-            self._response_model = ''
-            self._finish_reason = ''
+            self._llmresponse = ""
+            self._response_id = ""
+            self._response_model = ""
+            self._finish_reason = ""
+            self._response_service_tier = ""
+            self._tools = None
             self._input_tokens = 0
             self._output_tokens = 0
+            self._reasoning_tokens = 0
 
             self._args = args
             self._kwargs = kwargs
@@ -64,53 +68,35 @@ def async_complete(version, environment, application_name,
             await self.__wrapped__.__aenter__()
             return self
 
-        async def __aexit__(self, exc_type, exc_value, traceback):
-            await self.__wrapped__.__aexit__(exc_type, exc_value, traceback)
+        async def __aexit__(self, exc_type, exc_val, exc_tb):
+            await self.__wrapped__.__aexit__(exc_type, exc_val, exc_tb)
+            process_streaming_chat_response(
+                self, pricing_info, environment, application_name, metrics,
+                capture_message_content, disable_metrics, version
+            )
 
         def __aiter__(self):
             return self
 
-        async def __getattr__(self, name):
-            """Delegate attribute access to the wrapped object."""
-            return getattr(await self.__wrapped__, name)
-
         async def __anext__(self):
-            try:
-                chunk = await self.__wrapped__.__anext__()
-                process_chunk(self, chunk)
-                return chunk
-            except StopAsyncIteration:
-                try:
-                    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
+            chunk = await self.__wrapped__.__anext__()
+            process_chunk(self, chunk)
+            return chunk
+
+        def __getattr__(self, name):
+            return getattr(self.__wrapped__, name)
 
     async def wrapper(wrapped, instance, args, kwargs):
         """
         Wraps the GenAI function call.
         """
 
-        streaming = kwargs.get('stream', False)
-        server_address, server_port = set_server_address_and_port(instance, 'models.github.ai', 443)
-        request_model = kwargs.get('model', 'gpt-4o')
+        streaming = kwargs.get("stream", False)
+        server_address, server_port = set_server_address_and_port(instance, "models.github.ai", 443)
+        request_model = kwargs.get("model", "gpt-4o")
 
-        span_name = f'{SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT} {request_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)
@@ -130,7 +116,6 @@ def async_complete(version, environment, application_name,
                     environment=environment,
                     application_name=application_name,
                     metrics=metrics,
-                    event_provider=event_provider,
                     start_time=start_time,
                     span=span,
                     capture_message_content=capture_message_content,
@@ -142,3 +127,48 @@ def async_complete(version, environment, application_name,
             return response
 
     return wrapper
+
+def async_embed(version, environment, application_name,
+                tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+    """
+    Generates a telemetry wrapper for GenAI embedding function call
+    """
+
+    async def wrapper(wrapped, instance, args, kwargs):
+        """
+        Wraps the GenAI embedding function call.
+        """
+
+        server_address, server_port = set_server_address_and_port(instance, "models.github.ai", 443)
+        request_model = kwargs.get("model", "text-embedding-3-small")
+
+        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)
+
+            try:
+                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,
+                    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

openlit/instrumentation/azure_ai_inference/azure_ai_inference.py

@@ -13,6 +13,7 @@ from openlit.instrumentation.azure_ai_inference.utils import (
     process_chunk,
     process_chat_response,
     process_streaming_chat_response,
+    process_embedding_response,
 )
 from openlit.semcov import SemanticConvention
 
@@ -20,7 +21,7 @@ from openlit.semcov import SemanticConvention
 logger = logging.getLogger(__name__)
 
 def complete(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
     """
@@ -43,12 +44,15 @@ def complete(version, environment, application_name,
             self.__wrapped__ = wrapped
             self._span = span
             self._span_name = span_name
-            self._llmresponse = ''
-            self._response_id = ''
-            self._response_model = ''
-            self._finish_reason = ''
+            self._llmresponse = ""
+            self._response_id = ""
+            self._response_model = ""
+            self._finish_reason = ""
+            self._response_service_tier = ""
+            self._tools = None
             self._input_tokens = 0
             self._output_tokens = 0
+            self._reasoning_tokens = 0
 
             self._args = args
             self._kwargs = kwargs
@@ -64,53 +68,35 @@ def complete(version, environment, application_name,
             self.__wrapped__.__enter__()
             return self
 
-        def __exit__(self, exc_type, exc_value, traceback):
-            self.__wrapped__.__exit__(exc_type, exc_value, traceback)
+        def __exit__(self, exc_type, exc_val, exc_tb):
+            self.__wrapped__.__exit__(exc_type, exc_val, exc_tb)
+            process_streaming_chat_response(
+                self, pricing_info, environment, application_name, metrics,
+                capture_message_content, disable_metrics, version
+            )
 
         def __iter__(self):
             return self
 
+        def __next__(self):
+            chunk = next(self.__wrapped__)
+            process_chunk(self, chunk)
+            return chunk
+
         def __getattr__(self, name):
-            """Delegate attribute access to the wrapped object."""
             return getattr(self.__wrapped__, name)
 
-        def __next__(self):
-            try:
-                chunk = self.__wrapped__.__next__()
-                process_chunk(self, chunk)
-                return chunk
-            except StopIteration:
-                try:
-                    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
-
     def wrapper(wrapped, instance, args, kwargs):
         """
         Wraps the GenAI function call.
         """
 
-        streaming = kwargs.get('stream', False)
-        server_address, server_port = set_server_address_and_port(instance, 'models.github.ai', 443)
-        request_model = kwargs.get('model', 'gpt-4o')
+        streaming = kwargs.get("stream", False)
+        server_address, server_port = set_server_address_and_port(instance, "models.github.ai", 443)
+        request_model = kwargs.get("model", "gpt-4o")
 
-        span_name = f'{SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT} {request_model}'
+        span_name = f"{SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT} {request_model}"
 
-        # pylint: disable=no-else-return
         if streaming:
             awaited_wrapped = wrapped(*args, **kwargs)
             span = tracer.start_span(span_name, kind=SpanKind.CLIENT)
@@ -130,7 +116,6 @@ def complete(version, environment, application_name,
                     environment=environment,
                     application_name=application_name,
                     metrics=metrics,
-                    event_provider=event_provider,
                     start_time=start_time,
                     span=span,
                     capture_message_content=capture_message_content,
@@ -142,3 +127,48 @@ def complete(version, environment, application_name,
             return response
 
     return wrapper
+
+def embed(version, environment, application_name,
+          tracer, pricing_info, capture_message_content, metrics, disable_metrics):
+    """
+    Generates a telemetry wrapper for GenAI embedding function call
+    """
+
+    def wrapper(wrapped, instance, args, kwargs):
+        """
+        Wraps the GenAI embedding function call.
+        """
+
+        server_address, server_port = set_server_address_and_port(instance, "models.github.ai", 443)
+        request_model = kwargs.get("model", "text-embedding-3-small")
+
+        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 = wrapped(*args, **kwargs)
+
+            try:
+                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,
+                    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