Merge pull request #2517 from pipecat-ai/aleix/unify-get-messages-for-logging

unify get_messages_for_logging()

CHANGELOG.md

@@ -100,6 +100,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Fixed
 
+- Fixed some `get_messages_for_logging()` that were returning a JSON string
+  instead of a list.
+
 - Fixed a `DailyTransport` issue that prevented DTMF tones from being sent.
 
 - Fixed a missing import in `SentryMetrics`.

src/pipecat/adapters/base_llm_adapter.py

@@ -11,7 +11,7 @@ adapters that handle tool format conversion and standardization.
 """
 
 from abc import ABC, abstractmethod
-from typing import Any, Generic, List, TypeVar, Union, cast
+from typing import Any, Dict, Generic, List, TypeVar
 
 from loguru import logger
 
@@ -63,7 +63,7 @@ class BaseLLMAdapter(ABC, Generic[TLLMInvocationParams]):
         pass
 
     @abstractmethod
-    def get_messages_for_logging(self, context: LLMContext) -> List[dict[str, Any]]:
+    def get_messages_for_logging(self, context: LLMContext) -> List[Dict[str, Any]]:
         """Get messages from a universal LLM context in a format ready for logging about this provider.
 
         Args:

src/pipecat/adapters/services/anthropic_adapter.py

@@ -43,7 +43,7 @@ class AnthropicLLMAdapter(BaseLLMAdapter[AnthropicLLMInvocationParams]):
         """
         raise NotImplementedError("Universal LLMContext is not yet supported for Anthropic.")
 
-    def get_messages_for_logging(self, context) -> List[dict[str, Any]]:
+    def get_messages_for_logging(self, context) -> List[Dict[str, Any]]:
         """Get messages from a universal LLM context in a format ready for logging about Anthropic.
 
         Removes or truncates sensitive data like image content for safe logging.

src/pipecat/adapters/services/aws_nova_sonic_adapter.py

@@ -44,7 +44,7 @@ class AWSNovaSonicLLMAdapter(BaseLLMAdapter[AWSNovaSonicLLMInvocationParams]):
         """
         raise NotImplementedError("Universal LLMContext is not yet supported for AWS Nova Sonic.")
 
-    def get_messages_for_logging(self, context) -> List[dict[str, Any]]:
+    def get_messages_for_logging(self, context) -> List[Dict[str, Any]]:
         """Get messages from a universal LLM context in a format ready for logging about AWS Nova Sonic.
 
         Removes or truncates sensitive data like image content for safe logging.

src/pipecat/adapters/services/bedrock_adapter.py

@@ -43,7 +43,7 @@ class AWSBedrockLLMAdapter(BaseLLMAdapter[AWSBedrockLLMInvocationParams]):
         """
         raise NotImplementedError("Universal LLMContext is not yet supported for AWS Bedrock.")
 
-    def get_messages_for_logging(self, context) -> List[dict[str, Any]]:
+    def get_messages_for_logging(self, context) -> List[Dict[str, Any]]:
         """Get messages from a universal LLM context in a format ready for logging about AWS Bedrock.
 
         Removes or truncates sensitive data like image content for safe logging.

src/pipecat/adapters/services/gemini_adapter.py

@@ -91,7 +91,7 @@ class GeminiLLMAdapter(BaseLLMAdapter[GeminiLLMInvocationParams]):
 
         return formatted_standard_tools + custom_gemini_tools
 
-    def get_messages_for_logging(self, context: LLMContext) -> List[dict[str, Any]]:
+    def get_messages_for_logging(self, context: LLMContext) -> List[Dict[str, Any]]:
         """Get messages from a universal LLM context in a format ready for logging about Gemini.
 
         Removes or truncates sensitive data like image content for safe logging.

src/pipecat/adapters/services/open_ai_adapter.py

@@ -8,7 +8,7 @@
 
 import copy
 import json
-from typing import Any, List, TypedDict
+from typing import Any, Dict, List, TypedDict
 
 from openai._types import NOT_GIVEN as OPEN_AI_NOT_GIVEN
 from openai._types import NotGiven as OpenAINotGiven
@@ -79,7 +79,7 @@ class OpenAILLMAdapter(BaseLLMAdapter[OpenAILLMInvocationParams]):
             for func in functions_schema
         ]
 
-    def get_messages_for_logging(self, context: LLMContext) -> List[dict[str, Any]]:
+    def get_messages_for_logging(self, context: LLMContext) -> List[Dict[str, Any]]:
         """Get messages from a universal LLM context in a format ready for logging about OpenAI.
 
         Removes or truncates sensitive data like image content for safe logging.
@@ -102,7 +102,7 @@ class OpenAILLMAdapter(BaseLLMAdapter[OpenAILLMInvocationParams]):
             if "mime_type" in msg and msg["mime_type"].startswith("image/"):
                 msg["data"] = "..."
             msgs.append(msg)
-        return json.dumps(msgs, ensure_ascii=False)
+        return msgs
 
     def _get_messages(self, context: LLMContext) -> List[LLMContextMessage]:
         return context.get_messages("openai")

src/pipecat/adapters/services/open_ai_realtime_adapter.py

@@ -6,7 +6,7 @@
 
 """OpenAI Realtime LLM adapter for Pipecat."""
 
-from typing import Any, Dict, List, TypedDict, Union
+from typing import Any, Dict, List, TypedDict
 
 from pipecat.adapters.base_llm_adapter import BaseLLMAdapter
 from pipecat.adapters.schemas.function_schema import FunctionSchema
@@ -43,7 +43,7 @@ class OpenAIRealtimeLLMAdapter(BaseLLMAdapter):
         """
         raise NotImplementedError("Universal LLMContext is not yet supported for OpenAI Realtime.")
 
-    def get_messages_for_logging(self, context) -> List[dict[str, Any]]:
+    def get_messages_for_logging(self, context) -> List[Dict[str, Any]]:
         """Get messages from a universal LLM context in a format ready for logging about OpenAI Realtime.
 
         Removes or truncates sensitive data like image content for safe logging.

src/pipecat/processors/aggregators/openai_llm_context.py

@@ -15,7 +15,7 @@ import copy
 import io
 import json
 from dataclasses import dataclass
-from typing import Any, List, Optional
+from typing import Any, Dict, List, Optional
 
 from openai._types import NOT_GIVEN, NotGiven
 from openai.types.chat import (
@@ -183,13 +183,13 @@ class OpenAILLMContext:
         """
         return json.dumps(self._messages, cls=CustomEncoder, ensure_ascii=False, indent=2)
 
-    def get_messages_for_logging(self) -> str:
+    def get_messages_for_logging(self) -> List[Dict[str, Any]]:
         """Get sanitized messages suitable for logging.
 
         Removes or truncates sensitive data like image content for safe logging.
 
         Returns:
-            JSON string with sanitized message content for logging.
+            List of messages in a format ready for logging.
         """
         msgs = []
         for message in self.messages:
@@ -203,7 +203,7 @@ class OpenAILLMContext:
             if "mime_type" in msg and msg["mime_type"].startswith("image/"):
                 msg["data"] = "..."
             msgs.append(msg)
-        return json.dumps(msgs, ensure_ascii=False)
+        return msgs
 
     def from_standard_message(self, message):
         """Convert from OpenAI message format to OpenAI message format (passthrough).

src/pipecat/services/anthropic/llm.py

@@ -295,7 +295,7 @@ class AnthropicLLMService(LLMService):
             await self.start_processing_metrics()
 
             logger.debug(
-                f"{self}: Generating chat [{context.system}] | [{context.get_messages_for_logging()}]"
+                f"{self}: Generating chat [{context.system}] | {context.get_messages_for_logging()}"
             )
 
             messages = context.messages
@@ -933,13 +933,13 @@ class AnthropicLLMContext(OpenAILLMContext):
             messages.insert(0, {"role": "system", "content": self.system})
         return messages
 
-    def get_messages_for_logging(self) -> str:
+    def get_messages_for_logging(self) -> List[Dict[str, Any]]:
         """Get messages formatted for logging with sensitive data redacted.
 
         Replaces image data with placeholder text for cleaner logs.
 
         Returns:
-            JSON string representation of messages for logging.
+            List of messages in a format ready for logging.
         """
         msgs = []
         for message in self.messages:
@@ -950,7 +950,7 @@ class AnthropicLLMContext(OpenAILLMContext):
                         if item["type"] == "image":
                             item["source"]["data"] = "..."
             msgs.append(msg)
-        return json.dumps(msgs)
+        return msgs
 
 
 class AnthropicUserContextAggregator(LLMUserContextAggregator):

src/pipecat/services/aws/llm.py

@@ -556,11 +556,11 @@ class AWSBedrockLLMContext(OpenAILLMContext):
             messages.insert(0, {"role": "system", "content": self.system})
         return messages
 
-    def get_messages_for_logging(self) -> str:
+    def get_messages_for_logging(self) -> List[Dict[str, Any]]:
         """Get messages formatted for logging with sensitive data redacted.
 
         Returns:
-            JSON string representation of messages with image data redacted.
+            List of messages in a format ready for logging.
         """
         msgs = []
         for message in self.messages:
@@ -571,7 +571,7 @@ class AWSBedrockLLMContext(OpenAILLMContext):
                         if item.get("image"):
                             item["source"]["bytes"] = "..."
             msgs.append(msg)
-        return json.dumps(msgs)
+        return msgs
 
 
 class AWSBedrockUserContextAggregator(LLMUserContextAggregator):

src/pipecat/services/google/llm.py

@@ -292,11 +292,11 @@ class GoogleLLMContext(OpenAILLMContext):
         # Add the converted messages to our existing messages
         self._messages.extend(converted_messages)
 
-    def get_messages_for_logging(self):
+    def get_messages_for_logging(self) -> List[Dict[str, Any]]:
         """Get messages formatted for logging with sensitive data redacted.
 
         Returns:
-            List of message dictionaries with inline data redacted.
+            List of messages in a format ready for logging.
         """
         msgs = []
         for message in self.messages:
@@ -858,8 +858,8 @@ class GoogleLLMService(LLMService):
         self, context: OpenAILLMContext
     ) -> AsyncIterator[GenerateContentResponse]:
         logger.debug(
-            # f"{self}: Generating chat [{self._system_instruction}] | [{context.get_messages_for_logging()}]"
+            # f"{self}: Generating chat [{self._system_instruction}] | {context.get_messages_for_logging()}"
-            f"{self}: Generating chat from OpenAI context [{context.get_messages_for_logging()}]"
+            f"{self}: Generating chat from OpenAI context {context.get_messages_for_logging()}"
         )
 
         params = GeminiLLMInvocationParams(
@@ -875,8 +875,8 @@ class GoogleLLMService(LLMService):
     ) -> AsyncIterator[GenerateContentResponse]:
         adapter = self.get_llm_adapter()
         logger.debug(
-            # f"{self}: Generating chat [{self._system_instruction}] | [{context.get_messages_for_logging()}]"
+            # f"{self}: Generating chat [{self._system_instruction}] | {context.get_messages_for_logging()}"
-            f"{self}: Generating chat from universal context [{adapter.get_messages_for_logging(context)}]"
+            f"{self}: Generating chat from universal context {adapter.get_messages_for_logging(context)}"
         )
 
         params: GeminiLLMInvocationParams = adapter.get_llm_invocation_params(context)

src/pipecat/services/openai/base_llm.py

@@ -279,7 +279,7 @@ class BaseOpenAILLMService(LLMService):
         self, context: OpenAILLMContext
     ) -> AsyncStream[ChatCompletionChunk]:
         logger.debug(
-            f"{self}: Generating chat from OpenAI context [{context.get_messages_for_logging()}]"
+            f"{self}: Generating chat from OpenAI context {context.get_messages_for_logging()}"
         )
 
         messages: List[ChatCompletionMessageParam] = context.get_messages()
@@ -311,7 +311,7 @@ class BaseOpenAILLMService(LLMService):
     ) -> AsyncStream[ChatCompletionChunk]:
         adapter = self.get_llm_adapter()
         logger.debug(
-            f"{self}: Generating chat from universal context [{adapter.get_messages_for_logging(context)}]"
+            f"{self}: Generating chat from universal context {adapter.get_messages_for_logging(context)}"
         )
 
         params: OpenAILLMInvocationParams = adapter.get_llm_invocation_params(context)