First commmit

src/rag_eval/__init__.py

@@ -0,0 +1,2 @@
+def main() -> None:
+    print("Hello from rag-eval!")

src/rag_eval/config.py

@@ -0,0 +1,121 @@
+import os
+
+from pydantic import BaseModel
+from pydantic_settings import (
+    BaseSettings,
+    PydanticBaseSettingsSource,
+    SettingsConfigDict,
+    YamlConfigSettingsSource,
+)
+
+CONFIG_FILE_PATH = os.getenv("CONFIG_YAML", "config.yaml")
+
+
+class IndexConfig(BaseModel):
+    name: str
+    endpoint: str
+    dimensions: int
+    machine_type: str = "e2-standard-16"
+    origin: str
+    destination: str
+    chunk_limit: int
+
+    @property
+    def deployment(self) -> str:
+        return self.name.replace("-", "_") + "_deployed"
+    
+    @property
+    def data(self) -> str:
+        return self.destination + self.name
+
+class AgentConfig(BaseModel):
+    name: str
+    instructions: str
+    language_model: str
+    embedding_model: str
+    thinking: int
+
+
+class BigQueryConfig(BaseModel):
+    dataset_id: str
+    project_id: str | None = None
+    table_ids: dict[str, str]
+
+
+class Settings(BaseSettings):
+    project_id: str
+    location: str
+    service_account: str
+    
+    # Flattened fields from nested models
+    agent_name: str
+    agent_instructions: str
+    agent_language_model: str
+    agent_embedding_model: str
+    agent_thinking: int
+
+    index_name: str
+    index_endpoint: str
+    index_dimensions: int
+    index_machine_type: str = "e2-standard-16"
+    index_origin: str
+    index_destination: str
+    index_chunk_limit: int
+
+    bigquery_dataset_id: str
+    bigquery_project_id: str | None = None
+    bigquery_table_ids: dict[str, str]
+
+    bucket: str
+    base_image: str
+    dialogflow_agent_id: str
+    processing_image: str
+    
+    model_config = SettingsConfigDict(yaml_file=CONFIG_FILE_PATH)
+
+    @property
+    def agent(self) -> AgentConfig:
+        return AgentConfig(
+            name=self.agent_name,
+            instructions=self.agent_instructions,
+            language_model=self.agent_language_model,
+            embedding_model=self.agent_embedding_model,
+            thinking=self.agent_thinking,
+        )
+
+    @property
+    def index(self) -> IndexConfig:
+        return IndexConfig(
+            name=self.index_name,
+            endpoint=self.index_endpoint,
+            dimensions=self.index_dimensions,
+            machine_type=self.index_machine_type,
+            origin=self.index_origin,
+            destination=self.index_destination,
+            chunk_limit=self.index_chunk_limit,
+        )
+
+    @property
+    def bigquery(self) -> BigQueryConfig:
+        return BigQueryConfig(
+            dataset_id=self.bigquery_dataset_id,
+            project_id=self.bigquery_project_id,
+            table_ids=self.bigquery_table_ids,
+        )
+
+    @classmethod
+    def settings_customise_sources(
+        cls,
+        settings_cls: type[BaseSettings],
+        init_settings: PydanticBaseSettingsSource,
+        env_settings: PydanticBaseSettingsSource,
+        dotenv_settings: PydanticBaseSettingsSource,
+        file_secret_settings: PydanticBaseSettingsSource,
+    ) -> tuple[PydanticBaseSettingsSource, ...]:
+        return (
+            env_settings,
+            YamlConfigSettingsSource(settings_cls),
+        )
+
+
+settings = Settings()