"""
  Rerank backends - pluggable implementations of the rerank protocol.
  
  Each backend implements score_with_meta(query, docs, normalize) -> (scores, meta).
  Service loads one backend via get_rerank_backend(name, config) from config.
  """
  
  from __future__ import annotations
  
  from typing import Any, Dict, List, Protocol, Tuple
  
  
  class RerankBackendProtocol(Protocol):
      """Protocol for reranker backends (service-internal)."""
  
      def score_with_meta(
          self,
          query: str,
          docs: List[str],
          normalize: bool = True,
      ) -> Tuple[List[float], Dict[str, Any]]:
          """
          Input:
            query: search query string
            docs: list of documents, scores must align 1:1 with docs
            normalize: whether to normalize scores (e.g. sigmoid)
          Output:
            scores: list same length as docs, same order
            meta: at least input_docs, usable_docs, unique_docs, elapsed_ms
          """
          ...
  
  
  def get_rerank_backend(name: str, config: Dict[str, Any]) -> RerankBackendProtocol:
      """
      Factory: return a reranker backend instance for the given name and config.
      Config is the corresponding block from services.rerank.backends.<name>.
      """
      name = (name or "bge").strip().lower()
      if name == "bge":
          from reranker.backends.bge import BGERerankerBackend
          return BGERerankerBackend(config)
      if name == "qwen3_vllm":
          from reranker.backends.qwen3_vllm import Qwen3VLLMRerankerBackend
          return Qwen3VLLMRerankerBackend(config)

      if name == "qwen3_vllm_score":
          from reranker.backends.qwen3_vllm_score import Qwen3VLLMScoreRerankerBackend
          return Qwen3VLLMScoreRerankerBackend(config)

      if name == "qwen3_transformers":
          from reranker.backends.qwen3_transformers import Qwen3TransformersRerankerBackend
          return Qwen3TransformersRerankerBackend(config)

      if name == "qwen3_transformers_packed":
          from reranker.backends.qwen3_transformers_packed import (
              Qwen3TransformersPackedRerankerBackend,
          )
          return Qwen3TransformersPackedRerankerBackend(config)

      if name == "qwen3_gguf":
          from reranker.backends.qwen3_gguf import Qwen3GGUFRerankerBackend

          gguf_config = dict(config or {})
          gguf_config.setdefault("_backend_name", "qwen3_gguf")
          return Qwen3GGUFRerankerBackend(gguf_config)
      if name == "qwen3_gguf_06b":
          from reranker.backends.qwen3_gguf import Qwen3GGUFRerankerBackend
          gguf_config = dict(config or {})
          gguf_config.setdefault("_backend_name", "qwen3_gguf_06b")
          return Qwen3GGUFRerankerBackend(gguf_config)

      if name == "dashscope_rerank":
          from reranker.backends.dashscope_rerank import DashScopeRerankBackend
          return DashScopeRerankBackend(config)

      raise ValueError(

          f"Unknown rerank backend: {name!r}. Supported: bge, qwen3_vllm, qwen3_vllm_score, qwen3_transformers, qwen3_transformers_packed, qwen3_gguf, qwen3_gguf_06b, dashscope_rerank"

      )

  
  
  __all__ = ["RerankBackendProtocol", "get_rerank_backend"]