Merge pull request #485 from cheshire-cat-ai/develop

version 1.1.2

core/cat/db/crud.py

@@ -20,7 +20,7 @@ def get_settings_by_category(category: str) -> List[Dict]:
 def create_setting(payload: models.Setting) -> Dict:
 
     # Missing fields (setting_id, updated_at) are filled automatically by pydantic
-    get_db().insert(payload.dict())
+    get_db().insert(payload.model_dump())
 
     # retrieve the record we just created
     new_record = get_setting_by_id(payload.setting_id)

core/cat/db/models.py

@@ -16,7 +16,7 @@ def generate_timestamp():
 class SettingBody(BaseModel):
     name: str
     value: Union[Dict, List]
-    category: Optional[str]
+    category: Optional[str] = None
 
 # actual setting class, with additional auto generated id and update time
 class Setting(SettingBody):

core/cat/factory/embedder.py

@@ -1,67 +1,78 @@
+from typing import Type
 import langchain
-from pydantic import PyObject, BaseSettings
+from pydantic import BaseModel, ConfigDict
 
 from cat.factory.custom_embedder import DumbEmbedder, CustomOpenAIEmbeddings
 
 
 # Base class to manage LLM configuration.
-class EmbedderSettings(BaseSettings):
+class EmbedderSettings(BaseModel):
     # class instantiating the embedder
-    _pyclass: None
+    _pyclass: Type = None
+
+    # This is related to pydantic, because "model_*" attributes are protected.
+    # We deactivate the protection because langchain relies on several "model_*" named attributes
+    model_config = ConfigDict(
+        protected_namespaces=()
+    )
 
     # instantiate an Embedder from configuration
     @classmethod
     def get_embedder_from_config(cls, config):
         if cls._pyclass is None:
             raise Exception(
-                "Embedder configuration class has self._pyclass = None. Should be a valid Embedder class"
+                "Embedder configuration class has self._pyclass==None. Should be a valid Embedder class"
             )
-        return cls._pyclass(**config)
+        return cls._pyclass.default(**config)
 
 
 class EmbedderFakeConfig(EmbedderSettings):
     size: int = 128
-    _pyclass: PyObject = langchain.embeddings.FakeEmbeddings
+    _pyclass: Type = langchain.embeddings.FakeEmbeddings
 
-    class Config:
-        schema_extra = {
+    model_config = ConfigDict(
+        json_schema_extra = {
             "humanReadableName": "Default Embedder",
             "description": "Configuration for default embedder. It just outputs random numbers.",
         }
+    )
 
 
 class EmbedderDumbConfig(EmbedderSettings):
 
-    _pyclass = PyObject = DumbEmbedder
+    _pyclass: Type = DumbEmbedder
 
-    class Config:
-        schema_extra = {
+    model_config = ConfigDict(
+        json_schema_extra = {
             "humanReadableName": "Dumb Embedder",
             "description": "Configuration for default embedder. It encodes the pairs of characters",
         }
+    )
 
 
 class EmbedderLlamaCppConfig(EmbedderSettings):
     url: str
-    _pyclass = PyObject = CustomOpenAIEmbeddings
+    _pyclass: Type = CustomOpenAIEmbeddings
 
-    class Config:
-        schema_extra = {
+    model_config = ConfigDict(
+        json_schema_extra = {
             "humanReadableName": "Self-hosted llama-cpp-python embedder",
             "description": "Self-hosted llama-cpp-python embedder",
         }
+    )
 
 
 class EmbedderOpenAIConfig(EmbedderSettings):
     openai_api_key: str
     model: str = "text-embedding-ada-002"
-    _pyclass: PyObject = langchain.embeddings.OpenAIEmbeddings
+    _pyclass: Type = langchain.embeddings.OpenAIEmbeddings
 
-    class Config:
-        schema_extra = {
+    model_config = ConfigDict(
+        json_schema_extra = {
             "humanReadableName": "OpenAI Embedder",
             "description": "Configuration for OpenAI embeddings",
         }
+    )
 
 
 # https://python.langchain.com/en/latest/_modules/langchain/embeddings/openai.html#OpenAIEmbeddings
@@ -73,37 +84,40 @@ class EmbedderAzureOpenAIConfig(EmbedderSettings):
     openai_api_version: str
     deployment: str
 
-    _pyclass: PyObject = langchain.embeddings.OpenAIEmbeddings
+    _pyclass: Type = langchain.embeddings.OpenAIEmbeddings
 
-    class Config:
-        schema_extra = {
+    model_config = ConfigDict(
+        json_schema_extra = {
             "humanReadableName": "Azure OpenAI Embedder",
             "description": "Configuration for Azure OpenAI embeddings",
         }
+    )
 
 
 class EmbedderCohereConfig(EmbedderSettings):
     cohere_api_key: str
     model: str = "embed-multilingual-v2.0"
-    _pyclass: PyObject = langchain.embeddings.CohereEmbeddings
+    _pyclass: Type = langchain.embeddings.CohereEmbeddings
 
-    class Config:
-        schema_extra = {
+    model_config = ConfigDict(
+        json_schema_extra = {
             "humanReadableName": "Cohere Embedder",
             "description": "Configuration for Cohere embeddings",
         }
+    )
 
 
 class EmbedderHuggingFaceHubConfig(EmbedderSettings):
     repo_id: str = "sentence-transformers/all-MiniLM-L12-v2"
     huggingfacehub_api_token: str
-    _pyclass: PyObject = langchain.embeddings.HuggingFaceHubEmbeddings
+    _pyclass: Type = langchain.embeddings.HuggingFaceHubEmbeddings
 
-    class Config:
-        schema_extra = {
+    model_config = ConfigDict(
+        json_schema_extra = {
             "humanReadableName": "HuggingFace Hub Embedder",
             "description": "Configuration for HuggingFace Hub embeddings",
         }
+    )
 
 
 SUPPORTED_EMDEDDING_MODELS = [
@@ -120,7 +134,7 @@ class Config:
 # EMBEDDER_SCHEMAS contains metadata to let any client know which fields are required to create the language embedder.
 EMBEDDER_SCHEMAS = {}
 for config_class in SUPPORTED_EMDEDDING_MODELS:
-    schema = config_class.schema()
+    schema = config_class.model_json_schema()
 
     # useful for clients in order to call the correct config endpoints
     schema["languageEmbedderName"] = schema["title"]