"""创作 query 筛选器：调用 acquisition/query_filter.txt，批量判一组 query。

模型对每条 query 打两项：语义合法性 valid(0-10) + 创作相关性 relevant(bool)。
最终保留 keep = valid ≥ VALID_MIN 且 relevant —— 阈值放代码里(好调，不改提示词)。
  · valid：词组合本身说不说得通(机械正交易产生的废话短语在此淘汰)。
  · relevant：搜回来是否和创作知识相关(三把尺=设计决策vs工艺/可迁移/业务阶段；再排 制作/题材本身/应试政企/作品)。
返回与输入等长的 [{keep, valid, relevant, reason}]。供 build_creation_demo、filter_multiaxis 等共用。
"""
from __future__ import annotations

import json
import os
from pathlib import Path

import httpx

from core.config import Settings, load_env_file

PROMPT = Path(__file__).resolve().parent / "query_filter.txt"
VALID_MIN = 6   # 语义合法性阈值：valid ≥ 此值才算说得通；调这里即可松紧


def filter_queries(queries: list[str], settings: Settings, *, batch: int = 40) -> list[dict]:
    """批量过滤（分批调用，避免一次喂太多）。返回 [{keep, reason}]，与 queries 对齐。"""
    out: list[dict] = []
    for i in range(0, len(queries), batch):
        out.extend(_filter_batch(queries[i:i + batch], settings))
    return out


def _filter_batch(queries: list[str], settings: Settings) -> list[dict]:
    if not queries:
        return []
    user = json.dumps([{"idx": i, "query": q} for i, q in enumerate(queries)], ensure_ascii=False)
    messages = [
        {"role": "system", "content": PROMPT.read_text("utf-8")},
        {"role": "user", "content": user},
    ]
    try:
        txt = _chat_content(settings, messages)
        # query_filter 要求输出数组；有的模型会包一层 {"result":[...]}，都兜住
        data = json.loads(txt)
        arr = data if isinstance(data, list) else next((v for v in data.values() if isinstance(v, list)), [])
        by = {d.get("idx"): d for d in arr if isinstance(d, dict)}
        out = []
        for i in range(len(queries)):
            d = by.get(i, {})
            try:
                valid = int(d.get("valid", 10))
            except (TypeError, ValueError):
                valid = 10
            relevant = bool(d.get("relevant", True))
            out.append({"keep": valid >= VALID_MIN and relevant,   # 阈值在代码里判
                        "valid": valid, "relevant": relevant,
                        "reason": str(d.get("reason", ""))[:50]})
        return out
    except Exception as exc:
        return [{"keep": True, "valid": 10, "relevant": True, "reason": f"筛选失败:{str(exc)[:30]}"} for _ in queries]


def _chat_content(settings: Settings, messages: list[dict]) -> str:
    """Return chat content. Prefer OpenRouter; fall back to Ark when OpenRouter is region-blocked locally."""
    body = {"model": settings.llm_model, "messages": messages, "response_format": {"type": "json_object"}}
    env = load_env_file(os.getenv("CK_ENV_FILE", ".env"))
    prefer_ark = os.getenv("QUERY_FILTER_PROVIDER") == "ark" or bool(os.getenv("ARK_CHAT_MODEL"))
    openrouter_exc: Exception | None = None
    if settings.openrouter_api_key and not prefer_ark:
        try:
            resp = httpx.post(
                settings.openrouter_base_url.rstrip("/") + "/chat/completions",
                headers={"Authorization": f"Bearer {settings.openrouter_api_key}", "Content-Type": "application/json"},
                json=body,
                timeout=120,
            )
            resp.raise_for_status()
            return resp.json()["choices"][0]["message"]["content"]
        except Exception as exc:
            openrouter_exc = exc

    ark_key = os.getenv("ARK_API_KEY") or env.get("ARK_API_KEY")
    if ark_key:
        ark_model = os.getenv("ARK_CHAT_MODEL") or env.get("ARK_CHAT_MODEL") or "doubao-seed-1-6-flash-250615"
        ark_url = os.getenv("ARK_CHAT_URL") or env.get("ARK_CHAT_URL") or "https://ark.cn-beijing.volces.com/api/v3/chat/completions"
        resp = httpx.post(
            ark_url,
            headers={"Authorization": f"Bearer {ark_key}", "Content-Type": "application/json"},
            json={"model": ark_model, "messages": messages, "response_format": {"type": "json_object"}},
            timeout=120,
        )
        resp.raise_for_status()
        return resp.json()["choices"][0]["message"]["content"]

    if openrouter_exc:
        raise openrouter_exc
    raise RuntimeError("missing OpenRouter/Ark chat credentials")