rasa-pro 3.15.0a1__py3-none-any.whl → 3.15.0a3__py3-none-any.whl

rasa/shared/providers/_configs/azure_openai_client_config.py

@@ -167,8 +167,9 @@ class OAuthConfigWrapper(OAuth, BaseModel):
 
 @dataclass
 class AzureOpenAIClientConfig:
-    """Parses configuration for Azure OpenAI client, resolves aliases and
-    raises deprecation warnings.
+    """Parses configuration for Azure OpenAI client.
+
+    Resolves aliases and raises deprecation warnings.
 
     Raises:
         ValueError: Raised in cases of invalid configuration:
@@ -301,9 +302,7 @@ class AzureOpenAIClientConfig:
 
 
 def is_azure_openai_config(config: dict) -> bool:
-    """Check whether the configuration is meant to configure
-    an Azure OpenAI client.
-    """
+    """Check whether the configuration is meant to configure an Azure OpenAI client."""
     # Resolve any aliases that are specific to Azure OpenAI configuration
     config = AzureOpenAIClientConfig.resolve_config_aliases(config)
 

rasa/shared/providers/_configs/default_litellm_client_config.py

@@ -40,8 +40,9 @@ FORBIDDEN_KEYS = [
 
 @dataclass
 class DefaultLiteLLMClientConfig:
-    """Parses configuration for default LiteLLM client, resolves aliases and
-    raises deprecation warnings.
+    """Parses configuration for default LiteLLM client.
+
+    Resolves aliases and raises deprecation warnings.
 
     Raises:
         ValueError: Raised in cases of invalid configuration:
@@ -72,8 +73,7 @@ class DefaultLiteLLMClientConfig:
 
     @classmethod
     def from_dict(cls, config: dict) -> DefaultLiteLLMClientConfig:
-        """
-        Initializes a dataclass from the passed config.
+        """Initializes a dataclass from the passed config.
 
         Args:
             config: (dict) The config from which to initialize.

rasa/shared/providers/_configs/litellm_router_client_config.py

@@ -38,8 +38,9 @@ _LITELLM_UNSUPPORTED_KEYS = [
 
 @dataclass
 class LiteLLMRouterClientConfig:
-    """Parses configuration for a LiteLLM Router client. The configuration is expected
-    to be in the following format:
+    """Parses configuration for a LiteLLM Router client.
+
+    The configuration is expected to be in the following format:
 
     {
         "id": "model_group_id",

rasa/shared/providers/_configs/openai_client_config.py

@@ -64,8 +64,9 @@ FORBIDDEN_KEYS = [
 
 @dataclass
 class OpenAIClientConfig:
-    """Parses configuration for Azure OpenAI client, resolves aliases and
-    raises deprecation warnings.
+    """Parses configuration for OpenAI client.
+
+    Resolves aliases and raises deprecation warnings.
 
     Raises:
         ValueError: Raised in cases of invalid configuration:
@@ -118,8 +119,7 @@ class OpenAIClientConfig:
 
     @classmethod
     def from_dict(cls, config: dict) -> OpenAIClientConfig:
-        """
-        Initializes a dataclass from the passed config.
+        """Initializes a dataclass from the passed config.
 
         Args:
             config: (dict) The config from which to initialize.
@@ -168,9 +168,7 @@ class OpenAIClientConfig:
 
 
 def is_openai_config(config: dict) -> bool:
-    """Check whether the configuration is meant to configure
-    an OpenAI client.
-    """
+    """Check whether the configuration is meant to configure an OpenAI client."""
     # Process the config to handle all the aliases
     config = OpenAIClientConfig.resolve_config_aliases(config)
 

rasa/shared/providers/_configs/rasa_llm_client_config.py

@@ -22,8 +22,9 @@ structlogger = structlog.get_logger()
 
 @dataclass
 class RasaLLMClientConfig:
-    """Parses configuration for a Rasa Hosted LiteLLM client,
-    checks required keys present.
+    """Parses configuration for a Rasa Hosted LiteLLM client.
+
+    Checks required keys present.
 
     Raises:
         ValueError: Raised in cases of invalid configuration:
@@ -40,8 +41,7 @@ class RasaLLMClientConfig:
 
     @classmethod
     def from_dict(cls, config: dict) -> RasaLLMClientConfig:
-        """
-        Initializes a dataclass from the passed config.
+        """Initializes a dataclass from the passed config.
 
         Args:
             config: (dict) The config from which to initialize.

rasa/shared/providers/_configs/self_hosted_llm_client_config.py

@@ -61,8 +61,9 @@ FORBIDDEN_KEYS = [
 
 @dataclass
 class SelfHostedLLMClientConfig:
-    """Parses configuration for Self Hosted LiteLLM client, resolves aliases and
-    raises deprecation warnings.
+    """Parses configuration for Self Hosted LiteLLM client.
+
+    Resolves aliases and raises deprecation warnings.
 
     Raises:
         ValueError: Raised in cases of invalid configuration:
@@ -116,8 +117,7 @@ class SelfHostedLLMClientConfig:
 
     @classmethod
     def from_dict(cls, config: dict) -> SelfHostedLLMClientConfig:
-        """
-        Initializes a dataclass from the passed config.
+        """Initializes a dataclass from the passed config.
 
         Args:
             config: (dict) The config from which to initialize.

rasa/shared/providers/llm/_base_litellm_client.py

@@ -1,12 +1,14 @@
 from __future__ import annotations
 
+import asyncio
 import logging
 from abc import abstractmethod
-from typing import Any, Dict, List, Union, cast
+from typing import Any, Dict, List, NoReturn, Union, cast
 
 import structlog
 from litellm import acompletion, completion, validate_environment
 
+from rasa.core.constants import DEFAULT_REQUEST_TIMEOUT
 from rasa.shared.constants import (
     _VALIDATE_ENVIRONMENT_MISSING_KEYS_KEY,
     API_BASE_CONFIG_KEY,
@@ -57,26 +59,24 @@ class _BaseLiteLLMClient:
     @property
     @abstractmethod
     def config(self) -> dict:
-        """Returns the configuration for that the llm client
-        in dictionary form.
-        """
+        """Returns the configuration for that the llm client in dictionary form."""
         pass
 
     @property
     @abstractmethod
     def _litellm_model_name(self) -> str:
-        """Returns the value of LiteLLM's model parameter to be used in
-        completion/acompletion in LiteLLM format:
+        """Returns the value of LiteLLM's model parameter.
 
+        To be used in completion/acompletion in LiteLLM format:
         <provider>/<model or deployment name>
         """
         pass
 
     @property
     def _litellm_extra_parameters(self) -> Dict[str, Any]:
-        """Returns a dictionary of extra parameters which include model
-        parameters as well as LiteLLM specific input parameters.
+        """Returns a dictionary of extra parameters.
 
+        Includes model parameters as well as LiteLLM specific input parameters.
         By default, this returns an empty dictionary (no extra parameters).
         """
         return {}
@@ -96,8 +96,9 @@ class _BaseLiteLLMClient:
         }
 
     def validate_client_setup(self) -> None:
-        """Perform client validation. By default only environment variables
-        are validated.
+        """Perform client validation.
+
+        By default only environment variables are validated.
 
         Raises:
             ProviderClientValidationError if validation fails.
@@ -188,10 +189,17 @@ class _BaseLiteLLMClient:
             arguments = cast(
                 Dict[str, Any], resolve_environment_variables(self._completion_fn_args)
             )
-            response = await acompletion(
-                messages=formatted_messages, **{**arguments, **kwargs}
+
+            timeout = self._litellm_extra_parameters.get(
+                "timeout", DEFAULT_REQUEST_TIMEOUT
+            )
+            response = await asyncio.wait_for(
+                acompletion(messages=formatted_messages, **{**arguments, **kwargs}),
+                timeout=timeout,
             )
             return self._format_response(response)
+        except asyncio.TimeoutError:
+            self._handle_timeout_error()
         except Exception as e:
             message = ""
             from rasa.shared.providers.llm.self_hosted_llm_client import (
@@ -211,6 +219,25 @@ class _BaseLiteLLMClient:
                 )
             raise ProviderClientAPIException(e, message) from e
 
+    def _handle_timeout_error(self) -> NoReturn:
+        """Handle asyncio.TimeoutError and raise ProviderClientAPIException.
+
+        Raises:
+            ProviderClientAPIException: Always raised with formatted timeout error.
+        """
+        timeout = self._litellm_extra_parameters.get("timeout", DEFAULT_REQUEST_TIMEOUT)
+        error_message = (
+            f"APITimeoutError - Request timed out. Error_str: "
+            f"Request timed out. - timeout value={timeout:.6f}, "
+            f"time taken={timeout:.6f} seconds"
+        )
+        # nosemgrep: semgrep.rules.pii-positional-arguments-in-logging
+        # Error message contains only numeric timeout values, not PII
+        structlogger.error(
+            f"{self.__class__.__name__.lower()}.llm.timeout", error=error_message
+        )
+        raise ProviderClientAPIException(asyncio.TimeoutError(error_message)) from None
+
     def _get_formatted_messages(
         self, messages: Union[List[dict], List[str], str]
     ) -> List[Dict[str, str]]:
@@ -312,8 +339,9 @@ class _BaseLiteLLMClient:
 
     @staticmethod
     def _ensure_certificates() -> None:
-        """Configures SSL certificates for LiteLLM. This method is invoked during
-        client initialization.
+        """Configures SSL certificates for LiteLLM.
+
+        This method is invoked during client initialization.
 
         LiteLLM may utilize `openai` clients or other providers that require
         SSL verification settings through the `SSL_VERIFY` / `SSL_CERTIFICATE`

rasa/shared/providers/llm/litellm_router_llm_client.py

@@ -1,10 +1,12 @@
 from __future__ import annotations
 
+import asyncio
 import logging
 from typing import Any, Dict, List, Union
 
 import structlog
 
+from rasa.core.constants import DEFAULT_REQUEST_TIMEOUT
 from rasa.shared.exceptions import ProviderClientAPIException
 from rasa.shared.providers._configs.litellm_router_client_config import (
     LiteLLMRouterClientConfig,
@@ -79,13 +81,14 @@ class LiteLLMRouterLLMClient(_BaseLiteLLMRouterClient, _BaseLiteLLMClient):
 
     @suppress_logs(log_level=logging.WARNING)
     def _text_completion(self, prompt: Union[List[str], str]) -> LLMResponse:
-        """
-        Synchronously generate completions for given prompt.
+        """Synchronously generate completions for given prompt.
 
         Args:
             prompt: Prompt to generate the completion for.
+
         Returns:
             List of message completions.
+
         Raises:
             ProviderClientAPIException: If the API request fails.
         """
@@ -103,21 +106,30 @@ class LiteLLMRouterLLMClient(_BaseLiteLLMRouterClient, _BaseLiteLLMClient):
 
     @suppress_logs(log_level=logging.WARNING)
     async def _atext_completion(self, prompt: Union[List[str], str]) -> LLMResponse:
-        """
-        Asynchronously generate completions for given prompt.
+        """Asynchronously generate completions for given prompt.
 
         Args:
             prompt: Prompt to generate the completion for.
+
         Returns:
             List of message completions.
+
         Raises:
             ProviderClientAPIException: If the API request fails.
         """
         try:
-            response = await self.router_client.atext_completion(
-                prompt=prompt, **self._completion_fn_args
+            timeout = self._litellm_extra_parameters.get(
+                "timeout", DEFAULT_REQUEST_TIMEOUT
+            )
+            response = await asyncio.wait_for(
+                self.router_client.atext_completion(
+                    prompt=prompt, **self._completion_fn_args
+                ),
+                timeout=timeout,
             )
             return self._format_text_completion_response(response)
+        except asyncio.TimeoutError:
+            self._handle_timeout_error()
         except Exception as e:
             raise ProviderClientAPIException(e)
 
@@ -125,8 +137,7 @@ class LiteLLMRouterLLMClient(_BaseLiteLLMRouterClient, _BaseLiteLLMClient):
     def completion(
         self, messages: Union[List[dict], List[str], str], **kwargs: Any
     ) -> LLMResponse:
-        """
-        Synchronously generate completions for given list of messages.
+        """Synchronously generate completions for given list of messages.
 
         Method overrides the base class method to call the appropriate
         completion method based on the configuration. If the chat completions
@@ -143,8 +154,10 @@ class LiteLLMRouterLLMClient(_BaseLiteLLMRouterClient, _BaseLiteLLMClient):
                     as a user message.
                 - a single message as a string which will be formatted as user message.
             **kwargs: Additional parameters to pass to the completion call.
+
         Returns:
             List of message completions.
+
         Raises:
             ProviderClientAPIException: If the API request fails.
         """
@@ -163,8 +176,7 @@ class LiteLLMRouterLLMClient(_BaseLiteLLMRouterClient, _BaseLiteLLMClient):
     async def acompletion(
         self, messages: Union[List[dict], List[str], str], **kwargs: Any
     ) -> LLMResponse:
-        """
-        Asynchronously generate completions for given list of messages.
+        """Asynchronously generate completions for given list of messages.
 
         Method overrides the base class method to call the appropriate
         completion method based on the configuration. If the chat completions
@@ -181,8 +193,10 @@ class LiteLLMRouterLLMClient(_BaseLiteLLMRouterClient, _BaseLiteLLMClient):
                     as a user message.
                 - a single message as a string which will be formatted as user message.
             **kwargs: Additional parameters to pass to the completion call.
+
         Returns:
             List of message completions.
+
         Raises:
             ProviderClientAPIException: If the API request fails.
         """
@@ -190,19 +204,28 @@ class LiteLLMRouterLLMClient(_BaseLiteLLMRouterClient, _BaseLiteLLMClient):
             return await self._atext_completion(messages)
         try:
             formatted_messages = self._get_formatted_messages(messages)
-            response = await self.router_client.acompletion(
-                messages=formatted_messages, **{**self._completion_fn_args, **kwargs}
+            timeout = self._litellm_extra_parameters.get(
+                "timeout", DEFAULT_REQUEST_TIMEOUT
+            )
+            response = await asyncio.wait_for(
+                self.router_client.acompletion(
+                    messages=formatted_messages,
+                    **{**self._completion_fn_args, **kwargs},
+                ),
+                timeout=timeout,
             )
             return self._format_response(response)
+        except asyncio.TimeoutError:
+            self._handle_timeout_error()
         except Exception as e:
             raise ProviderClientAPIException(e)
 
     @property
     def _completion_fn_args(self) -> Dict[str, Any]:
-        """Returns the completion arguments for invoking a call through
-        LiteLLM's completion functions.
-        """
+        """Returns the completion arguments.
 
+        For invoking a call through LiteLLM's completion functions.
+        """
         return {
             **self._litellm_extra_parameters,
             LITE_LLM_MODEL_FIELD: self.model_group_id,

rasa/shared/providers/llm/self_hosted_llm_client.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+import asyncio
 import logging
 import os
 from typing import Any, Dict, List, Optional, Union
@@ -7,6 +8,7 @@ from typing import Any, Dict, List, Optional, Union
 import structlog
 from litellm import atext_completion, text_completion
 
+from rasa.core.constants import DEFAULT_REQUEST_TIMEOUT
 from rasa.shared.constants import (
     API_KEY,
     SELF_HOSTED_VLLM_API_KEY_ENV_VAR,
@@ -28,7 +30,7 @@ structlogger = structlog.get_logger()
 
 
 class SelfHostedLLMClient(_BaseLiteLLMClient):
-    """A client for interfacing with Self Hosted LLM endpoints that uses
+    """A client for interfacing with Self Hosted LLM endpoints.
 
     Parameters:
         model (str): The model or deployment name.
@@ -95,8 +97,7 @@ class SelfHostedLLMClient(_BaseLiteLLMClient):
 
     @property
     def provider(self) -> str:
-        """
-        Returns the provider name for the self hosted llm client.
+        """Returns the provider name for the self hosted llm client.
 
         Returns:
             String representing the provider name.
@@ -105,8 +106,7 @@ class SelfHostedLLMClient(_BaseLiteLLMClient):
 
     @property
     def model(self) -> str:
-        """
-        Returns the model name for the self hosted llm client.
+        """Returns the model name for the self hosted llm client.
 
         Returns:
             String representing the model name.
@@ -115,8 +115,7 @@ class SelfHostedLLMClient(_BaseLiteLLMClient):
 
     @property
     def api_base(self) -> str:
-        """
-        Returns the base URL for the API endpoint.
+        """Returns the base URL for the API endpoint.
 
         Returns:
             String representing the base URL.
@@ -125,8 +124,7 @@ class SelfHostedLLMClient(_BaseLiteLLMClient):
 
     @property
     def api_type(self) -> Optional[str]:
-        """
-        Returns the type of the API endpoint. Currently only OpenAI is supported.
+        """Returns the type of the API endpoint. Currently only OpenAI is supported.
 
         Returns:
             String representing the API type.
@@ -135,8 +133,7 @@ class SelfHostedLLMClient(_BaseLiteLLMClient):
 
     @property
     def api_version(self) -> Optional[str]:
-        """
-        Returns the version of the API endpoint.
+        """Returns the version of the API endpoint.
 
         Returns:
             String representing the API version.
@@ -145,8 +142,8 @@ class SelfHostedLLMClient(_BaseLiteLLMClient):
 
     @property
     def config(self) -> Dict:
-        """
-        Returns the configuration for the self hosted llm client.
+        """Returns the configuration for the self hosted llm client.
+
         Returns:
             Dictionary containing the configuration.
         """
@@ -163,9 +160,9 @@ class SelfHostedLLMClient(_BaseLiteLLMClient):
 
     @property
     def _litellm_model_name(self) -> str:
-        """Returns the value of LiteLLM's model parameter to be used in
-        completion/acompletion in LiteLLM format:
+        """Returns the value of LiteLLM's model parameter.
 
+        To be used in completion/acompletion in LiteLLM format:
         <hosted_vllm>/<model or deployment name>
         """
         if self.model and f"{SELF_HOSTED_VLLM_PREFIX}/" not in self.model:
@@ -174,15 +171,17 @@ class SelfHostedLLMClient(_BaseLiteLLMClient):
 
     @property
     def _litellm_extra_parameters(self) -> Dict[str, Any]:
-        """Returns optional configuration parameters specific
-        to the client provider and deployed model.
+        """Returns optional configuration parameters.
+
+        Specific to the client provider and deployed model.
         """
         return self._extra_parameters
 
     @property
     def _completion_fn_args(self) -> Dict[str, Any]:
-        """Returns the completion arguments for invoking a call through
-        LiteLLM's completion functions.
+        """Returns the completion arguments.
+
+        For invoking a call through LiteLLM's completion functions.
         """
         fn_args = super()._completion_fn_args
         fn_args.update(
@@ -195,13 +194,14 @@ class SelfHostedLLMClient(_BaseLiteLLMClient):
 
     @suppress_logs(log_level=logging.WARNING)
     def _text_completion(self, prompt: Union[List[str], str]) -> LLMResponse:
-        """
-        Synchronously generate completions for given prompt.
+        """Synchronously generate completions for given prompt.
 
         Args:
             prompt: Prompt to generate the completion for.
+
         Returns:
             List of message completions.
+
         Raises:
             ProviderClientAPIException: If the API request fails.
         """
@@ -213,26 +213,28 @@ class SelfHostedLLMClient(_BaseLiteLLMClient):
 
     @suppress_logs(log_level=logging.WARNING)
     async def _atext_completion(self, prompt: Union[List[str], str]) -> LLMResponse:
-        """
-        Asynchronously generate completions for given prompt.
+        """Asynchronously generate completions for given prompt.
 
         Args:
-            messages: The message can be,
-                - a list of preformatted messages. Each message should be a dictionary
-                    with the following keys:
-                    - content: The message content.
-                    - role: The role of the message (e.g. user or system).
-                - a list of messages. Each message is a string and will be formatted
-                    as a user message.
-                - a single message as a string which will be formatted as user message.
+            prompt: Prompt to generate the completion for.
+
         Returns:
             List of message completions.
+
         Raises:
             ProviderClientAPIException: If the API request fails.
         """
         try:
-            response = await atext_completion(prompt=prompt, **self._completion_fn_args)
+            timeout = self._litellm_extra_parameters.get(
+                "timeout", DEFAULT_REQUEST_TIMEOUT
+            )
+            response = await asyncio.wait_for(
+                atext_completion(prompt=prompt, **self._completion_fn_args),
+                timeout=timeout,
+            )
             return self._format_text_completion_response(response)
+        except asyncio.TimeoutError:
+            self._handle_timeout_error()
         except Exception as e:
             raise ProviderClientAPIException(e)
 

rasa/shared/utils/configs.py

@@ -8,8 +8,7 @@ structlogger = structlog.get_logger()
 
 
 def resolve_aliases(config: dict, deprecated_alias_mapping: dict) -> dict:
-    """
-    Resolve aliases in the configuration to standard keys.
+    """Resolve aliases in the configuration to standard keys.
 
     Args:
         config: Dictionary containing the configuration.
@@ -37,13 +36,13 @@ def raise_deprecation_warnings(
     deprecated_alias_mapping: dict,
     source: Optional[str] = None,
 ) -> None:
-    """
-    Raises warnings for deprecated keys in the configuration.
+    """Raises warnings for deprecated keys in the configuration.
 
     Args:
         config: Dictionary containing the configuration.
         deprecated_alias_mapping: Dictionary mapping deprecated keys to
             their standard keys.
+        source: Optional source context for the deprecation warning.
 
     Raises:
         DeprecationWarning: If any deprecated key is found in the config.
@@ -61,8 +60,7 @@ def raise_deprecation_warnings(
 
 
 def validate_required_keys(config: dict, required_keys: list) -> None:
-    """
-    Validates that the passed config contains all the required keys.
+    """Validates that the passed config contains all the required keys.
 
     Args:
         config: Dictionary containing the configuration.
@@ -84,8 +82,7 @@ def validate_required_keys(config: dict, required_keys: list) -> None:
 
 
 def validate_forbidden_keys(config: dict, forbidden_keys: list) -> None:
-    """
-    Validates that the passed config doesn't contain any forbidden keys.
+    """Validates that the passed config doesn't contain any forbidden keys.
 
     Args:
         config: Dictionary containing the configuration.

rasa/utils/endpoints.py

@@ -10,6 +10,7 @@ import structlog
 from aiohttp.client_exceptions import ContentTypeError
 from sanic.request import Request
 
+from rasa.core.actions.constants import MISSING_DOMAIN_MARKER
 from rasa.core.constants import DEFAULT_REQUEST_TIMEOUT
 from rasa.shared.exceptions import FileNotFoundException
 from rasa.shared.utils.yaml import read_config_file
@@ -224,6 +225,11 @@ class EndpointConfig:
                 ssl=sslcontext,
                 **kwargs,
             ) as response:
+                if response.status == 449:
+                    # Return a special marker that HTTPCustomActionExecutor can detect
+                    # This avoids raising an exception for this expected case
+                    return {MISSING_DOMAIN_MARKER: True}
+
                 if response.status >= 400:
                     raise ClientResponseError(
                         response.status,

rasa/version.py

@@ -1,3 +1,3 @@
 # this file will automatically be changed,
 # do not add anything but the version number here!
-__version__ = "3.15.0a1"
+__version__ = "3.15.0a3"