feat: add backend application foundation

2026-07-13 11:24:05 +08:00
parent 81b0faa67a
commit 44e01e2f93
10 changed files with 314 additions and 0 deletions
@@ -0,0 +1,66 @@
+from functools import lru_cache
+from pathlib import Path
+
+from pydantic import Field, field_validator
+from pydantic_settings import BaseSettings, SettingsConfigDict
+
+
+class Settings(BaseSettings):
+    model_config = SettingsConfigDict(
+        env_file=".env",
+        env_prefix="KBQA_",
+        extra="ignore",
+    )
+
+    app_name: str = "KBQA"
+    debug: bool = False
+
+    dashscope_api_key: str = ""
+    dashscope_base_url: str = "https://dashscope.aliyuncs.com/compatible-mode/v1"
+    chat_model: str = "qwen3.5-flash"
+    embedding_model: str = "text-embedding-v4"
+    embedding_dim: int = 1024
+
+    database_url: str = "sqlite+aiosqlite:///./data/kbqa.sqlite3"
+    milvus_uri: str = "./data/milvus/kbqa.db"
+    raw_data_dir: Path = Path("./data/raw")
+
+    chunk_size: int = Field(default=500, ge=1)
+    chunk_overlap: int = Field(default=50, ge=0)
+    retrieval_top_k: int = Field(default=5, ge=1, le=100)
+    max_research_searches: int = Field(default=3, ge=1, le=10)
+    max_sources: int = Field(default=8, ge=1, le=50)
+    history_message_limit: int = Field(default=20, ge=1, le=200)
+    max_upload_mib: int = Field(default=20, ge=1, le=1024)
+    max_query_length: int = Field(default=10_000, ge=1, le=100_000)
+    cors_origins: list[str] = [
+        "http://localhost:5173",
+        "http://127.0.0.1:5173",
+    ]
+
+    run_live_tests: bool = False
+    live_test_file: Path | None = None
+
+    @field_validator("chunk_overlap")
+    @classmethod
+    def overlap_must_be_smaller_than_chunk(cls, value: int, info) -> int:
+        chunk_size = info.data.get("chunk_size", 500)
+        if value >= chunk_size:
+            raise ValueError("chunk_overlap must be smaller than chunk_size")
+        return value
+
+
+@lru_cache
+def get_settings() -> Settings:
+    return Settings()
+
+
+def validate_live_settings(settings: Settings) -> None:
+    missing: list[str] = []
+    if not settings.dashscope_api_key:
+        missing.append("KBQA_DASHSCOPE_API_KEY")
+    if not settings.dashscope_base_url:
+        missing.append("KBQA_DASHSCOPE_BASE_URL")
+    if missing:
+        joined = ", ".join(missing)
+        raise RuntimeError(f"Missing live model configuration: {joined}")