openlit 1.34.25__tar.gz → 1.34.26__tar.gz

{openlit-1.34.25 → openlit-1.34.26}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: openlit
-Version: 1.34.25
+Version: 1.34.26
 Summary: OpenTelemetry-native Auto instrumentation library for monitoring LLM Applications and GPUs, facilitating the integration of observability into your GenAI-driven projects
 License: Apache-2.0
 Keywords: OpenTelemetry,otel,otlp,llm,tracing,openai,anthropic,claude,cohere,llm monitoring,observability,monitoring,gpt,Generative AI,chatGPT,gpu

{openlit-1.34.25 → openlit-1.34.26}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "openlit"
-version = "1.34.25"
+version = "1.34.26"
 description = "OpenTelemetry-native Auto instrumentation library for monitoring LLM Applications and GPUs, facilitating the integration of observability into your GenAI-driven projects"
 authors = ["OpenLIT"]
 license = "Apache-2.0"

{openlit-1.34.25 → openlit-1.34.26}/src/openlit/instrumentation/chroma/__init__.py

@@ -13,7 +13,7 @@ _instruments = ("chromadb >= 0.4.0",)
 
 class ChromaInstrumentor(BaseInstrumentor):
     """
-    An instrumentor for ChromaDB's client library.
+    An instrumentor for ChromaDB client library.
     """
 
     def instrumentation_dependencies(self) -> Collection[str]:

{openlit-1.34.25 → openlit-1.34.26}/src/openlit/instrumentation/chroma/chroma.py

@@ -5,13 +5,11 @@ Module for monitoring ChromaDB API calls.
 import time
 from opentelemetry.trace import SpanKind
 from opentelemetry import context as context_api
-from openlit.__helpers import (
-    handle_exception,
-    set_server_address_and_port,
-)
+from openlit.__helpers import handle_exception
 from openlit.instrumentation.chroma.utils import (
     process_vectordb_response,
     DB_OPERATION_MAP,
+    set_server_address_and_port,
 )
 
 def general_wrap(gen_ai_endpoint, version, environment, application_name,
@@ -29,7 +27,7 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
             return wrapped(*args, **kwargs)
 
         # Get server address and port using the standard helper
-        server_address, server_port = set_server_address_and_port(instance, "localhost", 8000)
+        server_address, server_port = set_server_address_and_port(instance)
 
         db_operation = DB_OPERATION_MAP.get(gen_ai_endpoint, "unknown")
         if db_operation == "create_collection":
@@ -39,10 +37,10 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
         span_name = f"{db_operation} {namespace}"
 
         with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
-            try:
-                start_time = time.time()
-                response = wrapped(*args, **kwargs)
+            start_time = time.time()
+            response = wrapped(*args, **kwargs)
 
+            try:
                 # Process response and generate telemetry
                 response = process_vectordb_response(
                     response, db_operation, server_address, server_port,

{openlit-1.34.25 → openlit-1.34.26}/src/openlit/instrumentation/chroma/utils.py

@@ -2,7 +2,7 @@
 ChromaDB OpenTelemetry instrumentation utility functions
 """
 import time
-
+from urllib.parse import urlparse
 from opentelemetry.trace import Status, StatusCode
 
 from openlit.__helpers import (
@@ -29,6 +29,51 @@ def object_count(obj):
     """
     return len(obj) if obj else 0
 
+def set_server_address_and_port(instance):
+    """
+    Extracts server address and port from ChromaDB client instance.
+    
+    Args:
+        instance: ChromaDB client instance
+        
+    Returns:
+        tuple: (server_address, server_port)
+    """
+    server_address = "localhost"
+    server_port = 8000
+
+    # Try getting base_url from multiple potential attributes
+    base_client = getattr(instance, "_client", None)
+    base_url = getattr(base_client, "base_url", None)
+
+    if not base_url:
+        # Attempt to get endpoint from instance._config.endpoint
+        config = getattr(instance, "_config", None)
+        base_url = getattr(config, "endpoint", None)
+
+    if not base_url:
+        # Attempt to get server_url from instance.sdk_configuration.server_url
+        config = getattr(instance, "sdk_configuration", None)
+        base_url = getattr(config, "server_url", None)
+
+    if base_url:
+        if isinstance(base_url, str):
+            # Check if it a full URL or just a hostname
+            if base_url.startswith(("http://", "https://")):
+                try:
+                    url = urlparse(base_url)
+                    if url.hostname:
+                        server_address = url.hostname
+                    if url.port:
+                        server_port = url.port
+                except Exception:
+                    pass
+            else:
+                # Just a hostname
+                server_address = base_url
+
+    return server_address, server_port
+
 def common_vectordb_logic(scope, environment, application_name,
     metrics, capture_message_content, disable_metrics, version, instance=None, endpoint=None):
     """
@@ -120,8 +165,8 @@ def common_vectordb_logic(scope, environment, application_name,
             # Extract response metrics if available
             if scope._response:
                 # Get number of results returned
-                if hasattr(scope._response, 'get') and scope._response.get('ids'):
-                    returned_rows = object_count(scope._response['ids'][0]) if scope._response['ids'] else 0
+                if hasattr(scope._response, "get") and scope._response.get("ids"):
+                    returned_rows = object_count(scope._response["ids"][0]) if scope._response["ids"] else 0
                     scope._span.set_attribute(SemanticConvention.DB_RESPONSE_RETURNED_ROWS, returned_rows)
 
             scope._span.set_attribute(SemanticConvention.DB_QUERY_SUMMARY,

openlit-1.34.26/src/openlit/instrumentation/pinecone/__init__.py

@@ -0,0 +1,78 @@
+"""Initializer of Auto Instrumentation of Pinecone Functions"""
+
+from typing import Collection
+import importlib.metadata
+from opentelemetry.instrumentation.instrumentor import BaseInstrumentor
+from wrapt import wrap_function_wrapper
+
+from openlit.instrumentation.pinecone.pinecone import general_wrap
+from openlit.instrumentation.pinecone.async_pinecone import async_general_wrap
+
+_instruments = ("pinecone >= 7.3.0",)
+
+# Pinecone sync operations
+PINECONE_SYNC_OPERATIONS = [
+    ("pinecone.pinecone", "Pinecone.create_index", "pinecone.create_collection"),
+    ("pinecone.pinecone", "Pinecone.create_index_for_model", "pinecone.create_collection"),
+    ("pinecone.db_data.index", "Index.upsert", "pinecone.upsert"),
+    ("pinecone.db_data.index", "Index.upsert_records", "pinecone.upsert_records"),
+    ("pinecone.db_data.index", "Index.query", "pinecone.query"),
+    ("pinecone.db_data.index", "Index.search", "pinecone.search"),
+    ("pinecone.db_data.index", "Index.fetch", "pinecone.fetch"),
+    ("pinecone.db_data.index", "Index.search_records", "pinecone.search_records"),
+    ("pinecone.db_data.index", "Index.update", "pinecone.update"),
+    ("pinecone.db_data.index", "Index.delete", "pinecone.delete"),
+]
+
+# Pinecone async operations
+PINECONE_ASYNC_OPERATIONS = [
+    ("pinecone.pinecone_asyncio", "PineconeAsyncio.create_index", "pinecone.create_index"),
+    ("pinecone.pinecone_asyncio", "PineconeAsyncio.create_index_for_model", "pinecone.create_index"),
+    ("pinecone.db_data.index_asyncio", "_IndexAsyncio.upsert", "pinecone.upsert"),
+    ("pinecone.db_data.index_asyncio", "_IndexAsyncio.upsert_records", "pinecone.upsert_records"),
+    ("pinecone.db_data.index_asyncio", "_IndexAsyncio.query", "pinecone.query"),
+    ("pinecone.db_data.index_asyncio", "_IndexAsyncio.search", "pinecone.search"),
+    ("pinecone.db_data.index_asyncio", "_IndexAsyncio.fetch", "pinecone.fetch"),
+    ("pinecone.db_data.index_asyncio", "_IndexAsyncio.search_records", "pinecone.search_records"),
+    ("pinecone.db_data.index_asyncio", "_IndexAsyncio.update", "pinecone.update"),
+    ("pinecone.db_data.index_asyncio", "_IndexAsyncio.delete", "pinecone.delete"),
+]
+
+class PineconeInstrumentor(BaseInstrumentor):
+    """
+    An instrumentor for Pinecone client library.
+    """
+
+    def instrumentation_dependencies(self) -> Collection[str]:
+        return _instruments
+
+    def _instrument(self, **kwargs):
+        version = importlib.metadata.version("pinecone")
+        environment = kwargs.get("environment", "default")
+        application_name = kwargs.get("application_name", "default")
+        tracer = kwargs.get("tracer")
+        pricing_info = kwargs.get("pricing_info", {})
+        capture_message_content = kwargs.get("capture_message_content", False)
+        metrics = kwargs.get("metrics_dict")
+        disable_metrics = kwargs.get("disable_metrics")
+
+        # Wrap sync operations
+        for module, class_method, endpoint in PINECONE_SYNC_OPERATIONS:
+            wrap_function_wrapper(
+                module,
+                class_method,
+                general_wrap(endpoint, version, environment, application_name,
+                    tracer, pricing_info, capture_message_content, metrics, disable_metrics),
+            )
+
+        # Wrap async operations
+        for module, class_method, endpoint in PINECONE_ASYNC_OPERATIONS:
+            wrap_function_wrapper(
+                module,
+                class_method,
+                async_general_wrap(endpoint, version, environment, application_name,
+                    tracer, pricing_info, capture_message_content, metrics, disable_metrics),
+            )
+
+    def _uninstrument(self, **kwargs):
+        pass

{openlit-1.34.25 → openlit-1.34.26}/src/openlit/instrumentation/pinecone/async_pinecone.py

@@ -5,13 +5,11 @@ Module for monitoring Pinecone async API calls.
 import time
 from opentelemetry.trace import SpanKind
 from opentelemetry import context as context_api
-from openlit.__helpers import (
-    handle_exception,
-    set_server_address_and_port,
-)
+from openlit.__helpers import handle_exception
 from openlit.instrumentation.pinecone.utils import (
     process_vectordb_response,
     DB_OPERATION_MAP,
+    set_server_address_and_port,
 )
 
 def async_general_wrap(gen_ai_endpoint, version, environment, application_name,
@@ -29,7 +27,7 @@ def async_general_wrap(gen_ai_endpoint, version, environment, application_name,
             return await wrapped(*args, **kwargs)
 
         # Get server address and port using the standard helper
-        server_address, server_port = set_server_address_and_port(instance, "pinecone.io", 443)
+        server_address, server_port = set_server_address_and_port(instance)
 
         db_operation = DB_OPERATION_MAP.get(gen_ai_endpoint, "unknown")
         if db_operation == "create_collection":
@@ -39,10 +37,10 @@ def async_general_wrap(gen_ai_endpoint, version, environment, application_name,
         span_name = f"{db_operation} {namespace}"
 
         with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
-            try:
-                start_time = time.time()
-                response = await wrapped(*args, **kwargs)
+            start_time = time.time()
+            response = await wrapped(*args, **kwargs)
 
+            try:
                 # Process response and generate telemetry
                 response = process_vectordb_response(
                     response, db_operation, server_address, server_port,

{openlit-1.34.25 → openlit-1.34.26}/src/openlit/instrumentation/pinecone/pinecone.py

@@ -5,13 +5,11 @@ Module for monitoring Pinecone API calls.
 import time
 from opentelemetry.trace import SpanKind
 from opentelemetry import context as context_api
-from openlit.__helpers import (
-    handle_exception,
-    set_server_address_and_port,
-)
+from openlit.__helpers import handle_exception
 from openlit.instrumentation.pinecone.utils import (
     process_vectordb_response,
     DB_OPERATION_MAP,
+    set_server_address_and_port,
 )
 
 def general_wrap(gen_ai_endpoint, version, environment, application_name,
@@ -29,7 +27,7 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
             return wrapped(*args, **kwargs)
 
         # Get server address and port using the standard helper
-        server_address, server_port = set_server_address_and_port(instance, "pinecone.io", 443)
+        server_address, server_port = set_server_address_and_port(instance)
 
         db_operation = DB_OPERATION_MAP.get(gen_ai_endpoint, "unknown")
         if db_operation == "create_collection":
@@ -39,10 +37,10 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
         span_name = f"{db_operation} {namespace}"
 
         with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
-            try:
-                start_time = time.time()
-                response = wrapped(*args, **kwargs)
+            start_time = time.time()
+            response = wrapped(*args, **kwargs)
 
+            try:
                 # Process response and generate telemetry
                 response = process_vectordb_response(
                     response, db_operation, server_address, server_port,

{openlit-1.34.25 → openlit-1.34.26}/src/openlit/instrumentation/pinecone/utils.py

@@ -2,7 +2,7 @@
 Pinecone OpenTelemetry instrumentation utility functions
 """
 import time
-
+from urllib.parse import urlparse
 from opentelemetry.trace import Status, StatusCode
 
 from openlit.__helpers import (
@@ -30,6 +30,46 @@ def object_count(obj):
     """
     return len(obj) if obj else 0
 
+def set_server_address_and_port(instance):
+    """
+    Extracts server address and port from Pinecone client instance.
+    
+    Args:
+        instance: Pinecone client instance
+        
+    Returns:
+        tuple: (server_address, server_port)
+    """
+    server_address = "pinecone.io"
+    server_port = 443
+
+    # Try getting base_url from multiple potential attributes
+    base_client = getattr(instance, "_client", None)
+    base_url = getattr(base_client, "base_url", None)
+
+    if not base_url:
+        # Attempt to get host from instance.config.host (used by Pinecone)
+        config = getattr(instance, "config", None)
+        base_url = getattr(config, "host", None)
+
+    if base_url:
+        if isinstance(base_url, str):
+            # Check if its a full URL or just a hostname
+            if base_url.startswith(("http://", "https://")):
+                try:
+                    url = urlparse(base_url)
+                    if url.hostname:
+                        server_address = url.hostname
+                    if url.port:
+                        server_port = url.port
+                except Exception:
+                    pass
+            else:
+                # Just a hostname
+                server_address = base_url
+
+    return server_address, server_port
+
 def common_vectordb_logic(scope, environment, application_name,
     metrics, capture_message_content, disable_metrics, version, instance=None):
     """

openlit-1.34.26/src/openlit/instrumentation/qdrant/__init__.py

@@ -0,0 +1,80 @@
+"""
+OpenLIT Qdrant Instrumentation
+"""
+
+from typing import Collection
+import importlib.metadata
+from opentelemetry.instrumentation.instrumentor import BaseInstrumentor
+from wrapt import wrap_function_wrapper
+
+from openlit.instrumentation.qdrant.qdrant import general_wrap
+from openlit.instrumentation.qdrant.async_qdrant import async_general_wrap
+
+_instruments = ("qdrant-client >= 1.9.0",)
+
+# Operations to wrap for both sync and async clients
+QDRANT_OPERATIONS = [
+    ("create_collection", "qdrant.create_collection"),
+    ("delete_collection", "qdrant.delete_collection"),
+    ("update_collection", "qdrant.update_collection"),
+    ("upload_collection", "qdrant.upload_collection"),
+    ("upsert", "qdrant.upsert"),
+    ("set_payload", "qdrant.set_payload"),
+    ("overwrite_payload", "qdrant.overwrite_payload"),
+    ("clear_payload", "qdrant.clear_payload"),
+    ("delete_payload", "qdrant.delete_payload"),
+    ("upload_points", "qdrant.upload_points"),
+    ("update_vectors", "qdrant.update_vectors"),
+    ("delete_vectors", "qdrant.delete_vectors"),
+    ("delete", "qdrant.delete"),
+    ("retrieve", "qdrant.retrieve"),
+    ("scroll", "qdrant.scroll"),
+    ("search", "qdrant.search"),
+    ("search_groups", "qdrant.search_groups"),
+    ("recommend", "qdrant.recommend"),
+    ("create_payload_index", "qdrant.create_payload_index"),
+    ("query_points", "qdrant.query_points"),
+]
+
+class QdrantInstrumentor(BaseInstrumentor):
+    """
+    An instrumentor for Qdrant client library.
+    """
+
+    def instrumentation_dependencies(self) -> Collection[str]:
+        return _instruments
+
+    def _instrument(self, **kwargs):
+        version = importlib.metadata.version("qdrant-client")
+        environment = kwargs.get("environment", "default")
+        application_name = kwargs.get("application_name", "default")
+        tracer = kwargs.get("tracer")
+        pricing_info = kwargs.get("pricing_info", {})
+        capture_message_content = kwargs.get("capture_message_content", False)
+        metrics = kwargs.get("metrics_dict")
+        disable_metrics = kwargs.get("disable_metrics")
+
+        # Wrap sync operations
+        for method_name, endpoint in QDRANT_OPERATIONS:
+            wrap_function_wrapper(
+                "qdrant_client",
+                f"QdrantClient.{method_name}",
+                general_wrap(
+                    endpoint, version, environment, application_name, tracer,
+                    pricing_info, capture_message_content, metrics, disable_metrics
+                ),
+            )
+
+        # Wrap async operations
+        for method_name, endpoint in QDRANT_OPERATIONS:
+            wrap_function_wrapper(
+                "qdrant_client",
+                f"AsyncQdrantClient.{method_name}",
+                async_general_wrap(
+                    endpoint, version, environment, application_name, tracer,
+                    pricing_info, capture_message_content, metrics, disable_metrics
+                ),
+            )
+
+    def _uninstrument(self, **kwargs):
+        pass

openlit-1.34.26/src/openlit/instrumentation/qdrant/async_qdrant.py

@@ -0,0 +1,55 @@
+# pylint: disable=duplicate-code, broad-exception-caught, too-many-statements, unused-argument, possibly-used-before-assignment, too-many-branches
+"""
+Module for monitoring Qdrant async operations.
+"""
+
+import time
+from opentelemetry.trace import SpanKind
+from opentelemetry import context as context_api
+from openlit.__helpers import handle_exception
+from openlit.instrumentation.qdrant.utils import DB_OPERATION_MAP, process_qdrant_response, set_server_address_and_port
+
+def async_general_wrap(gen_ai_endpoint, version, environment, application_name, tracer,
+    pricing_info, capture_message_content, metrics, disable_metrics):
+    """
+    Generates a telemetry wrapper for Pinecone function calls.
+    """
+
+    async def wrapper(wrapped, instance, args, kwargs):
+        """
+        Wraps the Qdrant async operation with tracing and logging.
+        """
+        # CRITICAL: Suppression check
+        if context_api.get_value(context_api._SUPPRESS_INSTRUMENTATION_KEY):
+            return await wrapped(*args, **kwargs)
+
+        # Extract operation from endpoint
+        db_operation = DB_OPERATION_MAP.get(gen_ai_endpoint, "unknown")
+
+        # Server address calculation
+        server_address, server_port = set_server_address_and_port(instance)
+
+        # Span naming: use operation + collection
+        collection_name = kwargs.get("collection_name", "unknown")
+        span_name = f"{db_operation} {collection_name}"
+
+        # CRITICAL: Use tracer.start_as_current_span() for proper context
+        with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
+            start_time = time.time()
+            response = await wrapped(*args, **kwargs)
+
+            try:
+                # Process response with endpoint information
+                response = process_qdrant_response(
+                    response, db_operation, server_address, server_port,
+                    environment, application_name, metrics, start_time, span,
+                    capture_message_content, disable_metrics, version,
+                    instance=instance, args=args, endpoint=gen_ai_endpoint, **kwargs
+                )
+
+            except Exception as e:
+                handle_exception(span, e)
+
+            return response
+
+    return wrapper

openlit-1.34.26/src/openlit/instrumentation/qdrant/qdrant.py

@@ -0,0 +1,55 @@
+# pylint: disable=duplicate-code, broad-exception-caught, too-many-statements, unused-argument, possibly-used-before-assignment, too-many-branches
+"""
+Module for monitoring Qdrant operations.
+"""
+
+import time
+from opentelemetry.trace import SpanKind
+from opentelemetry import context as context_api
+from openlit.__helpers import handle_exception
+from openlit.instrumentation.qdrant.utils import DB_OPERATION_MAP, process_qdrant_response, set_server_address_and_port
+
+def general_wrap(gen_ai_endpoint, version, environment, application_name, tracer,
+    pricing_info, capture_message_content, metrics, disable_metrics):
+    """
+    Generates a telemetry wrapper for Pinecone function calls.
+    """
+
+    def wrapper(wrapped, instance, args, kwargs):
+        """
+        Wraps the Qdrant operation with tracing and logging.
+        """
+        # CRITICAL: Suppression check
+        if context_api.get_value(context_api._SUPPRESS_INSTRUMENTATION_KEY):
+            return wrapped(*args, **kwargs)
+
+        # Extract operation from endpoint
+        db_operation = DB_OPERATION_MAP.get(gen_ai_endpoint, "unknown")
+
+        # Server address calculation
+        server_address, server_port = set_server_address_and_port(instance)
+
+        # Span naming: use operation + collection
+        collection_name = kwargs.get("collection_name", "unknown")
+        span_name = f"{db_operation} {collection_name}"
+
+        # CRITICAL: Use tracer.start_as_current_span() for proper context
+        with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
+            start_time = time.time()
+            response = wrapped(*args, **kwargs)
+
+            try:
+                # Process response with endpoint information
+                response = process_qdrant_response(
+                    response, db_operation, server_address, server_port,
+                    environment, application_name, metrics, start_time, span,
+                    capture_message_content, disable_metrics, version,
+                    instance=instance, args=args, endpoint=gen_ai_endpoint, **kwargs
+                )
+
+            except Exception as e:
+                handle_exception(span, e)
+
+            return response
+
+    return wrapper