"""解析模型返回为候选草稿（纯函数，无 IO）。

模型可能返回纯 JSON、markdown 代码块或夹杂解释文本；本模块容错提取首个 JSON 数组，
并归一化为候选草稿。evidence 序号由 application 映射为 ``import_items.id``。
"""

import json
import re
from dataclasses import dataclass, field
from typing import Any


@dataclass
class ProfileCandidateDraft:
    field_name: str
    value: str
    confidence: float
    reason: str
    evidence: list[int] = field(default_factory=list)


@dataclass
class MemoryCandidateDraft:
    type: str
    title: str
    content: str
    importance: float
    sensitivity: str
    reason: str
    evidence: list[int] = field(default_factory=list)


@dataclass
class PersonalityProposalDraft:
    field_name: str
    change: str
    reason: str
    evidence: list[int] = field(default_factory=list)


def parse_profile_candidates(raw: str) -> list[ProfileCandidateDraft]:
    drafts: list[ProfileCandidateDraft] = []
    for item in _extract_json_array(raw):
        if not isinstance(item, dict):
            continue
        field_name = str(item.get("field", "")).strip()
        value = str(item.get("value", "")).strip()
        if not field_name or not value:
            continue
        drafts.append(
            ProfileCandidateDraft(
                field_name=field_name,
                value=value,
                confidence=_clamp_float(item.get("confidence"), 0.5),
                reason=str(item.get("reason", "")).strip(),
                evidence=_int_list(item.get("evidence")),
            )
        )
    return drafts


def parse_memory_candidates(raw: str) -> list[MemoryCandidateDraft]:
    drafts: list[MemoryCandidateDraft] = []
    for item in _extract_json_array(raw):
        if not isinstance(item, dict):
            continue
        title = str(item.get("title", "")).strip()
        content = str(item.get("content", "")).strip()
        if not title or not content:
            continue
        drafts.append(
            MemoryCandidateDraft(
                type=str(item.get("type", "other")).strip() or "other",
                title=title,
                content=content,
                importance=_clamp_float(item.get("importance"), 0.5),
                sensitivity=_sensitivity(item.get("sensitivity")),
                reason=str(item.get("reason", "")).strip(),
                evidence=_int_list(item.get("evidence")),
            )
        )
    return drafts


def parse_personality_proposals(raw: str) -> list[PersonalityProposalDraft]:
    drafts: list[PersonalityProposalDraft] = []
    for item in _extract_json_array(raw):
        if not isinstance(item, dict):
            continue
        field_name = str(item.get("field", "")).strip()
        change = str(item.get("change", "")).strip()
        if not field_name or not change:
            continue
        drafts.append(
            PersonalityProposalDraft(
                field_name=field_name,
                change=change,
                reason=str(item.get("reason", "")).strip(),
                evidence=_int_list(item.get("evidence")),
            )
        )
    return drafts


def _extract_json_array(raw: str) -> list[Any]:
    text = raw.strip()
    matched = re.search(r"```(?:json)?\s*(\[.*?\])\s*```", text, re.DOTALL)
    if matched:
        text = matched.group(1)
    else:
        start, end = text.find("["), text.rfind("]")
        if start != -1 and end > start:
            text = text[start : end + 1]
    try:
        data = json.loads(text)
    except json.JSONDecodeError:
        return []
    return data if isinstance(data, list) else []


def _clamp_float(value: Any, default: float) -> float:
    try:
        num = float(value)
    except (TypeError, ValueError):
        return default
    return max(0.0, min(1.0, num))


def _int_list(value: Any) -> list[int]:
    if not isinstance(value, list):
        return []
    result: list[int] = []
    for item in value:
        try:
            result.append(int(item))
        except (TypeError, ValueError):
            continue
    return result


def _sensitivity(value: Any) -> str:
    text = str(value or "").strip().lower()
    return text if text in {"low", "medium", "high"} else "low"