# api/courseware/qa_optimizer.py
"""
Course QA Optimizer：基于学生答题数据（Smart Quiz）分析弱点，自动优化后续教学建议。
"""
from typing import Optional, List, Tuple

from api.config import client, DEFAULT_MODEL
from api.courseware.rag import get_rag_context_with_refs, inject_refs_instruction
from api.courseware.references import append_references_to_content
from api.courseware.prompts import QA_OPTIMIZER_SYSTEM, QA_OPTIMIZER_USER_TEMPLATE


def optimize_from_quiz_data(
    quiz_summary: str,
    course_topic: Optional[str] = None,
    max_tokens: int = 1500,
    history: Optional[list] = None,
) -> str:
    """
    基于 Smart Quiz 答题数据摘要，分析薄弱点并给出后续教学优化建议。
    """
    query = f"{quiz_summary}\n{course_topic or ''}"[:2000]
    rag_context, refs = get_rag_context_with_refs(query, top_k=6, max_context_chars=4000)
    ref_instruction = inject_refs_instruction(refs)
    user_content = QA_OPTIMIZER_USER_TEMPLATE.format(
        quiz_summary=quiz_summary.strip() or "（未提供答题数据）",
        course_topic=(course_topic or "（未指定）").strip(),
        rag_context=rag_context or "（无检索到知识库摘录。）",
        ref_instruction=ref_instruction,
    )
    messages = [{"role": "system", "content": QA_OPTIMIZER_SYSTEM}]
    if history:
        for user_msg, assistant_msg in history[-10:]:
            if user_msg:
                messages.append({"role": "user", "content": user_msg})
            if assistant_msg:
                messages.append({"role": "assistant", "content": assistant_msg})
    messages.append({"role": "user", "content": user_content})
    
    try:
        resp = client.chat.completions.create(
            model=DEFAULT_MODEL,
            messages=messages,
            temperature=0.4,
            max_tokens=max_tokens,
            timeout=90,
        )
        out = (resp.choices[0].message.content or "").strip()
    except Exception as e:
        out = f"生成失败：{e}。请稍后重试。"
    if refs and "## References" not in out and "[Source:" not in out:
        out = append_references_to_content(out, refs)
    return out