rasa-pro 3.11.0__py3-none-any.whl → 3.11.0a1__py3-none-any.whl

rasa/shared/importers/importer.py

@@ -1,18 +1,7 @@
 import logging
 from abc import ABC, abstractmethod
 from functools import reduce
-from typing import (
-    Any,
-    Dict,
-    List,
-    Optional,
-    Set,
-    Text,
-    Tuple,
-    Type,
-    Union,
-    cast,
-)
+from typing import Any, Dict, List, Optional, Set, Text, Tuple, Type, Union, cast
 
 import importlib_resources
 
@@ -178,7 +167,6 @@ class TrainingDataImporter(ABC):
         domain_path: Optional[Text] = None,
         training_data_paths: Optional[List[Text]] = None,
         args: Optional[Dict[Text, Any]] = None,
-        expand_env_vars: bool = True,
     ) -> "TrainingDataImporter":
         """Loads a `TrainingDataImporter` instance from a dictionary."""
         from rasa.shared.importers.rasa import RasaFileImporter
@@ -194,12 +182,7 @@ class TrainingDataImporter(ABC):
         importers = [importer for importer in importers if importer]
         if not importers:
             importers = [
-                RasaFileImporter(
-                    config_path,
-                    domain_path,
-                    training_data_paths,
-                    expand_env_vars=expand_env_vars,
-                )
+                RasaFileImporter(config_path, domain_path, training_data_paths)
             ]
 
         return E2EImporter(

rasa/shared/importers/rasa.py

@@ -29,9 +29,7 @@ class RasaFileImporter(TrainingDataImporter):
         config_file: Optional[Text] = None,
         domain_path: Optional[Text] = None,
         training_data_paths: Optional[Union[List[Text], Text]] = None,
-        expand_env_vars: bool = True,
     ):
-        self.expand_env_vars = expand_env_vars
         self._domain_path = domain_path
 
         self._nlu_files = rasa.shared.data.get_data_files(
@@ -56,9 +54,7 @@ class RasaFileImporter(TrainingDataImporter):
             logger.debug("No configuration file was provided to the RasaFileImporter.")
             return {}
 
-        config = read_model_configuration(
-            self.config_file, expand_env_vars=self.expand_env_vars
-        )
+        config = read_model_configuration(self.config_file)
         return config
 
     def get_config_file_for_auto_config(self) -> Optional[Text]:

rasa/shared/nlu/training_data/features.py

@@ -1,133 +1,15 @@
 from __future__ import annotations
-
-import itertools
-from dataclasses import dataclass
 from typing import Iterable, Union, Text, Optional, List, Any, Tuple, Dict, Set
+import itertools
 
 import numpy as np
 import scipy.sparse
-from safetensors.numpy import save_file, load_file
 
-import rasa.shared.nlu.training_data.util
 import rasa.shared.utils.io
+import rasa.shared.nlu.training_data.util
 from rasa.shared.nlu.constants import FEATURE_TYPE_SEQUENCE, FEATURE_TYPE_SENTENCE
 
 
-@dataclass
-class FeatureMetadata:
-    data_type: str
-    attribute: str
-    origin: Union[str, List[str]]
-    is_sparse: bool
-    shape: tuple
-    safetensors_key: str
-
-
-def save_features(
-    features_dict: Dict[Text, List[Features]], file_name: str
-) -> Dict[str, Any]:
-    """Save a dictionary of Features lists to disk using safetensors.
-
-    Args:
-        features_dict: Dictionary mapping strings to lists of Features objects
-        file_name: File to save the features to
-
-    Returns:
-        The metadata to reconstruct the features.
-    """
-    # All tensors are stored in a single safetensors file
-    tensors_to_save = {}
-    # Metadata will be stored separately
-    metadata = {}
-
-    for key, features_list in features_dict.items():
-        feature_metadata_list = []
-
-        for idx, feature in enumerate(features_list):
-            # Create a unique key for this tensor in the safetensors file
-            safetensors_key = f"{key}_{idx}"
-
-            # Convert sparse matrices to dense if needed
-            if feature.is_sparse():
-                # For sparse matrices, use the COO format
-                coo = feature.features.tocoo()  # type:ignore[union-attr]
-                # Save data, row indices and col indices separately
-                tensors_to_save[f"{safetensors_key}_data"] = coo.data
-                tensors_to_save[f"{safetensors_key}_row"] = coo.row
-                tensors_to_save[f"{safetensors_key}_col"] = coo.col
-            else:
-                tensors_to_save[safetensors_key] = feature.features
-
-            # Store metadata
-            metadata_item = FeatureMetadata(
-                data_type=feature.type,
-                attribute=feature.attribute,
-                origin=feature.origin,
-                is_sparse=feature.is_sparse(),
-                shape=feature.features.shape,
-                safetensors_key=safetensors_key,
-            )
-            feature_metadata_list.append(vars(metadata_item))
-
-        metadata[key] = feature_metadata_list
-
-    # Save tensors
-    save_file(tensors_to_save, file_name)
-
-    return metadata
-
-
-def load_features(
-    filename: str, metadata: Dict[str, Any]
-) -> Dict[Text, List[Features]]:
-    """Load Features dictionary from disk.
-
-    Args:
-        filename: File name of the safetensors file.
-        metadata: Metadata to reconstruct the features.
-
-    Returns:
-        Dictionary mapping strings to lists of Features objects
-    """
-    # Load tensors
-    tensors = load_file(filename)
-
-    # Reconstruct the features dictionary
-    features_dict: Dict[Text, List[Features]] = {}
-
-    for key, feature_metadata_list in metadata.items():
-        features_list = []
-
-        for meta in feature_metadata_list:
-            safetensors_key = meta["safetensors_key"]
-
-            if meta["is_sparse"]:
-                # Reconstruct sparse matrix from COO format
-                data = tensors[f"{safetensors_key}_data"]
-                row = tensors[f"{safetensors_key}_row"]
-                col = tensors[f"{safetensors_key}_col"]
-
-                features_matrix = scipy.sparse.coo_matrix(
-                    (data, (row, col)), shape=tuple(meta["shape"])
-                ).tocsr()  # Convert back to CSR format
-            else:
-                features_matrix = tensors[safetensors_key]
-
-            # Reconstruct Features object
-            features = Features(
-                features=features_matrix,
-                feature_type=meta["data_type"],
-                attribute=meta["attribute"],
-                origin=meta["origin"],
-            )
-
-            features_list.append(features)
-
-        features_dict[key] = features_list
-
-    return features_dict
-
-
 class Features:
     """Stores the features produced by any featurizer."""
 

rasa/shared/nlu/training_data/formats/rasa_yaml.py

@@ -1,18 +1,7 @@
 import logging
 from collections import OrderedDict
 from pathlib import Path
-from typing import (
-    ClassVar,
-    Text,
-    Any,
-    List,
-    Dict,
-    Tuple,
-    Union,
-    Iterator,
-    Optional,
-    Callable,
-)
+from typing import Text, Any, List, Dict, Tuple, Union, Iterator, Optional, Callable
 
 import rasa.shared.data
 from rasa.shared.core.domain import Domain
@@ -66,8 +55,6 @@ STRIP_SYMBOLS = "\n\r "
 class RasaYAMLReader(TrainingDataReader):
     """Reads YAML training data and creates a TrainingData object."""
 
-    expand_env_vars: ClassVar[bool] = True
-
     def __init__(self) -> None:
         super().__init__()
         self.training_examples: List[Message] = []
@@ -82,9 +69,7 @@ class RasaYAMLReader(TrainingDataReader):
         If the string is not in the right format, an exception will be raised.
         """
         try:
-            validate_raw_yaml_using_schema_file_with_responses(
-                string, NLU_SCHEMA_FILE, expand_env_vars=self.expand_env_vars
-            )
+            validate_raw_yaml_using_schema_file_with_responses(string, NLU_SCHEMA_FILE)
         except YamlException as e:
             e.filename = self.filename
             raise e
@@ -103,7 +88,7 @@ class RasaYAMLReader(TrainingDataReader):
         """
         self.validate(string)
 
-        yaml_content = read_yaml(string, expand_env_vars=self.expand_env_vars)
+        yaml_content = read_yaml(string)
 
         if not validate_training_data_format_version(yaml_content, self.filename):
             return TrainingData()

rasa/shared/providers/_configs/azure_openai_client_config.py

@@ -107,7 +107,8 @@ class AzureOpenAIClientConfig:
 
     @classmethod
     def from_dict(cls, config: dict) -> "AzureOpenAIClientConfig":
-        """Initializes a dataclass from the passed config.
+        """
+        Initializes a dataclass from the passed config.
 
         Args:
             config: (dict) The config from which to initialize.
@@ -174,10 +175,7 @@ def is_azure_openai_config(config: dict) -> bool:
 
     # Case: Configuration contains `deployment` key
     # (specific to Azure OpenAI configuration)
-    if (
-        config.get(DEPLOYMENT_CONFIG_KEY) is not None
-        and config.get(PROVIDER_CONFIG_KEY) is None
-    ):
+    if config.get(DEPLOYMENT_CONFIG_KEY) is not None:
         return True
 
     return False

rasa/shared/providers/_configs/openai_client_config.py

@@ -19,8 +19,8 @@ from rasa.shared.constants import (
     REQUEST_TIMEOUT_CONFIG_KEY,
     TIMEOUT_CONFIG_KEY,
     PROVIDER_CONFIG_KEY,
-    OPENAI_API_TYPE,
     OPENAI_PROVIDER,
+    OPENAI_API_TYPE,
 )
 from rasa.shared.providers._configs.utils import (
     resolve_aliases,

rasa/shared/providers/_configs/self_hosted_llm_client_config.py

@@ -23,7 +23,6 @@ from rasa.shared.constants import (
     SELF_HOSTED_PROVIDER,
     USE_CHAT_COMPLETIONS_ENDPOINT_CONFIG_KEY,
 )
-
 from rasa.shared.providers._configs.utils import (
     raise_deprecation_warnings,
     resolve_aliases,

rasa/shared/providers/_configs/utils.py

@@ -99,19 +99,3 @@ def validate_forbidden_keys(config: dict, forbidden_keys: list) -> None:
             config=config,
         )
         raise ValueError(message)
-
-
-def get_provider_prefixed_model_name(provider: str, model: str) -> str:
-    """
-    Returns the model name with the provider prefixed.
-
-    Args:
-        provider: The provider of the model.
-        model: The model name.
-
-    Returns:
-        The model name with the provider prefixed.
-    """
-    if model and f"{provider}/" not in model:
-        return f"{provider}/{model}"
-    return model

rasa/shared/providers/embedding/_base_litellm_embedding_client.py

@@ -1,12 +1,10 @@
-import logging
 from abc import abstractmethod
 from typing import Any, Dict, List
 
 import litellm
+import logging
 import structlog
 from litellm import aembedding, embedding, validate_environment
-
-from rasa.shared.constants import API_BASE_CONFIG_KEY, API_KEY
 from rasa.shared.exceptions import (
     ProviderClientAPIException,
     ProviderClientValidationError,
@@ -19,7 +17,7 @@ from rasa.shared.providers.embedding.embedding_response import (
     EmbeddingResponse,
     EmbeddingUsage,
 )
-from rasa.shared.utils.io import suppress_logs, resolve_environment_variables
+from rasa.shared.utils.io import suppress_logs
 
 structlogger = structlog.get_logger()
 
@@ -27,7 +25,8 @@ _VALIDATE_ENVIRONMENT_MISSING_KEYS_KEY = "missing_keys"
 
 
 class _BaseLiteLLMEmbeddingClient:
-    """An abstract base class for LiteLLM embedding clients.
+    """
+    An abstract base class for LiteLLM embedding clients.
 
     This class defines the interface and common functionality for all clients
     based on LiteLLM.
@@ -82,14 +81,11 @@ class _BaseLiteLLMEmbeddingClient:
             ProviderClientValidationError if validation fails.
         """
         self._validate_environment_variables()
+        self._validate_api_key_not_in_config()
 
     def _validate_environment_variables(self) -> None:
         """Validate that the required environment variables are set."""
-        validation_info = validate_environment(
-            self._litellm_model_name,
-            api_key=self._litellm_extra_parameters.get(API_KEY),
-            api_base=self._litellm_extra_parameters.get(API_BASE_CONFIG_KEY),
-        )
+        validation_info = validate_environment(self._litellm_model_name)
         if missing_environment_variables := validation_info.get(
             _VALIDATE_ENVIRONMENT_MISSING_KEYS_KEY
         ):
@@ -104,8 +100,21 @@ class _BaseLiteLLMEmbeddingClient:
             )
             raise ProviderClientValidationError(event_info)
 
+    def _validate_api_key_not_in_config(self) -> None:
+        if "api_key" in self._litellm_extra_parameters:
+            event_info = (
+                "API Key is set through `api_key` extra parameter."
+                "Set API keys through environment variables."
+            )
+            structlogger.error(
+                "base_litellm_client.validate_api_key_not_in_config",
+                event_info=event_info,
+            )
+            raise ProviderClientValidationError(event_info)
+
     def validate_documents(self, documents: List[str]) -> None:
-        """Validates a list of documents to ensure they are suitable for embedding.
+        """
+        Validates a list of documents to ensure they are suitable for embedding.
 
         Args:
             documents: List of documents to be validated.
@@ -121,7 +130,8 @@ class _BaseLiteLLMEmbeddingClient:
 
     @suppress_logs(log_level=logging.WARNING)
     def embed(self, documents: List[str]) -> EmbeddingResponse:
-        """Embeds a list of documents synchronously.
+        """
+        Embeds a list of documents synchronously.
 
         Args:
             documents: List of documents to be embedded.
@@ -134,8 +144,7 @@ class _BaseLiteLLMEmbeddingClient:
         """
         self.validate_documents(documents)
         try:
-            arguments = resolve_environment_variables(self._embedding_fn_args)
-            response = embedding(input=documents, **arguments)
+            response = embedding(input=documents, **self._embedding_fn_args)
             return self._format_response(response)
         except Exception as e:
             raise ProviderClientAPIException(
@@ -144,7 +153,8 @@ class _BaseLiteLLMEmbeddingClient:
 
     @suppress_logs(log_level=logging.WARNING)
     async def aembed(self, documents: List[str]) -> EmbeddingResponse:
-        """Embeds a list of documents asynchronously.
+        """
+        Embeds a list of documents asynchronously.
 
         Args:
             documents: List of documents to be embedded.
@@ -157,8 +167,7 @@ class _BaseLiteLLMEmbeddingClient:
         """
         self.validate_documents(documents)
         try:
-            arguments = resolve_environment_variables(self._embedding_fn_args)
-            response = await aembedding(input=documents, **arguments)
+            response = await aembedding(input=documents, **self._embedding_fn_args)
             return self._format_response(response)
         except Exception as e:
             raise ProviderClientAPIException(
@@ -173,6 +182,7 @@ class _BaseLiteLLMEmbeddingClient:
         Raises:
             ValueError: If any response data is None.
         """
+
         # If data is not available (None), raise a ValueError
         if response.data is None:
             message = (
@@ -229,7 +239,8 @@ class _BaseLiteLLMEmbeddingClient:
 
     @staticmethod
     def _ensure_certificates() -> None:
-        """Configures SSL certificates for LiteLLM. This method is invoked during
+        """
+        Configures SSL certificates for LiteLLM. This method is invoked during
         client initialization.
 
         LiteLLM may utilize `openai` clients or other providers that require

rasa/shared/providers/embedding/azure_openai_embedding_client.py

@@ -1,6 +1,5 @@
-import os
 from typing import Any, Dict, List, Optional
-
+import os
 import structlog
 
 from rasa.shared.constants import (
@@ -43,7 +42,6 @@ class AzureOpenAIEmbeddingClient(_BaseLiteLLMEmbeddingClient):
             If not provided, it will be set via environment variable.
         kwargs (Optional[Dict[str, Any]]): Optional configuration parameters specific
             to the embedding model deployment.
-
     Raises:
         ProviderClientValidationError: If validation of the client setup fails.
         DeprecationWarning: If deprecated environment variables are used for
@@ -62,7 +60,6 @@ class AzureOpenAIEmbeddingClient(_BaseLiteLLMEmbeddingClient):
         super().__init__()  # type: ignore
         self._deployment = deployment
         self._model = model
-        self._extra_parameters = kwargs or {}
 
         # Set api_base with the following priority:
         # parameter -> Azure Env Var -> (deprecated) OpenAI Env Var
@@ -84,55 +81,17 @@ class AzureOpenAIEmbeddingClient(_BaseLiteLLMEmbeddingClient):
         # Litellm does not support use of OPENAI_API_KEY, so we need to map it
         # because of backward compatibility. However, we're first looking at
         # AZURE_API_KEY.
-        self._api_key_env_var = self._resolve_api_key_env_var()
+        self._api_key = os.environ.get(AZURE_API_KEY_ENV_VAR) or os.environ.get(
+            OPENAI_API_KEY_ENV_VAR
+        )
 
+        self._extra_parameters = kwargs or {}
         self.validate_client_setup()
 
-    def _resolve_api_key_env_var(self) -> str:
-        """Resolves the environment variable to use for the API key.
-
-        Returns:
-            str: The env variable in dollar syntax format to use for the API key.
-        """
-        if API_KEY in self._extra_parameters:
-            # API key is set to an env var in the config itself
-            # in case the model is defined in the endpoints.yml
-            return self._extra_parameters[API_KEY]
-
-        if os.getenv(AZURE_API_KEY_ENV_VAR) is not None:
-            return "${AZURE_API_KEY}"
-
-        if os.getenv(OPENAI_API_KEY_ENV_VAR) is not None:
-            # API key can be set through OPENAI_API_KEY too,
-            # because of the backward compatibility
-            raise_deprecation_warning(
-                message=(
-                    f"Usage of '{OPENAI_API_KEY_ENV_VAR}' environment variable "
-                    "for setting the API key of "
-                    "Azure OpenAI client is deprecated and will "
-                    "be removed in 4.0.0. Please "
-                    f"use '{AZURE_API_KEY_ENV_VAR}' instead."
-                )
-            )
-            return "${OPENAI_API_KEY}"
-
-        structlogger.error(
-            "azure_openai_embedding_client.api_key_not_set",
-            event_info=(
-                "API key not set, it is required for API calls. "
-                f"Set it either via the environment variable "
-                f"'{AZURE_API_KEY_ENV_VAR}' or directly"
-                f"via the config key '{API_KEY}'."
-            ),
-        )
-        raise ProviderClientValidationError(
-            f"Missing required environment variable/config key '{API_KEY}' for "
-            f"API calls."
-        )
-
     @classmethod
     def from_config(cls, config: Dict[str, Any]) -> "AzureOpenAIEmbeddingClient":
-        """Initializes the client from given configuration.
+        """
+        Initializes the client from given configuration.
 
         Args:
             config (Dict[str, Any]): Configuration.
@@ -183,7 +142,8 @@ class AzureOpenAIEmbeddingClient(_BaseLiteLLMEmbeddingClient):
 
     @property
     def model(self) -> Optional[str]:
-        """Returns the name of the model deployed on Azure. If model name is not
+        """
+        Returns the name of the model deployed on Azure. If model name is not
         provided, returns "N/A".
         """
         return self._model
@@ -210,7 +170,8 @@ class AzureOpenAIEmbeddingClient(_BaseLiteLLMEmbeddingClient):
 
     @property
     def _litellm_extra_parameters(self) -> Dict[str, Any]:
-        """Returns the model parameters for the azure openai embedding client.
+        """
+        Returns the model parameters for the azure openai embedding client.
 
         Returns:
             Dictionary containing the model parameters.
@@ -225,7 +186,7 @@ class AzureOpenAIEmbeddingClient(_BaseLiteLLMEmbeddingClient):
             "api_base": self.api_base,
             "api_type": self.api_type,
             "api_version": self.api_version,
-            "api_key": self._api_key_env_var,
+            "api_key": self._api_key,
         }
 
     @property
@@ -236,9 +197,8 @@ class AzureOpenAIEmbeddingClient(_BaseLiteLLMEmbeddingClient):
         return self.deployment
 
     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.
@@ -254,6 +214,13 @@ class AzureOpenAIEmbeddingClient(_BaseLiteLLMEmbeddingClient):
                 "current_value": self.api_base,
                 "new_env_key": AZURE_API_BASE_ENV_VAR,
             },
+            {
+                "param_name": "API key",
+                "config_key": API_KEY,
+                "deprecated_env_key": OPENAI_API_KEY_ENV_VAR,
+                "current_value": self._api_key,
+                "new_env_key": AZURE_API_KEY_ENV_VAR,
+            },
             {
                 "param_name": "API version",
                 "config_key": API_VERSION_CONFIG_KEY,

rasa/shared/providers/embedding/default_litellm_embedding_client.py

@@ -1,13 +1,8 @@
 from typing import Any, Dict
 
-from rasa.shared.constants import (
-    AWS_BEDROCK_PROVIDER,
-    AWS_SAGEMAKER_PROVIDER,
-)
 from rasa.shared.providers._configs.default_litellm_client_config import (
     DefaultLiteLLMClientConfig,
 )
-from rasa.shared.providers._utils import validate_aws_setup_for_litellm_clients
 from rasa.shared.providers.embedding._base_litellm_embedding_client import (
     _BaseLiteLLMEmbeddingClient,
 )
@@ -105,22 +100,3 @@ class DefaultLiteLLMEmbeddingClient(_BaseLiteLLMEmbeddingClient):
             "model": self._litellm_model_name,
             **self._litellm_extra_parameters,
         }
-
-    def validate_client_setup(self) -> None:
-        # TODO: Temporarily disable environment variable validation for AWS setup
-        #       (Bedrock and SageMaker) until resolved by either:
-        #       1. An update from the LiteLLM package addressing the issue.
-        #       2. The implementation of a Bedrock client on our end.
-        #       ---
-        #       This fix ensures a consistent user experience for Bedrock (and
-        #       SageMaker) in Rasa by allowing AWS secrets to be provided as extra
-        #       parameters without triggering validation errors due to missing AWS
-        #       environment variables.
-        if self.provider.lower() in [AWS_BEDROCK_PROVIDER, AWS_SAGEMAKER_PROVIDER]:
-            validate_aws_setup_for_litellm_clients(
-                self._litellm_model_name,
-                self._litellm_extra_parameters,
-                "default_litellm_embedding_client",
-            )
-        else:
-            super().validate_client_setup()