Cn Last30days

interface:
  display_name: "Cn Last 30 Days"
  short_description: "研究过去30天中国社媒平台上任何话题的真实讨论。覆盖小红书、抖音、公众号三大平台，返回跨平台舆情综合分析。"
  default_prompt: "研究过去30天这个话题在小红书、抖音、公众号上的讨论。综合分析用户在说什么、什么内容被点赞收藏最多、各平台态度有什么差异。"
  brand_color: "#E63946"

policy:
  allow_implicit_invocation: true

Bud1�

cache_
__pycache__dsclbool

 
@
�





 
@
�






 
@
�






 
@E�
DSDB
 `
�




 
@
�






 
@
�






 
@

#!/usr/bin/env python3
"""
cn-last30days: 中国社媒平台话题研究工具
==========================================
从小红书、抖音、公众号三大平台搜索过去30天内人们关于某话题的真实讨论。

Usage:
    python cn_last30days.py "AI视频工具"
    python cn_last30days.py "大模型" --output-format html
    python cn_last30days.py "小红书运营" --platforms xhs,gzh
"""

from __future__ import annotations

import argparse
import json
import os
import sys
import time
from datetime import datetime, timedelta, timezone
from pathlib import Path
from typing import Any
from urllib.parse import quote

# Windows stdout UTF-8
if os.name == "nt":
    for stream in (sys.stdout, sys.stderr):
        if hasattr(stream, "reconfigure"):
            stream.reconfigure(encoding="utf-8", errors="replace")

# ─── 常量 ──────────────────────────────────────────────────────────────────────────
API_BASE = "https://redfox.hk/story/api/multiPlatform/workSearch"
PLATFORMS = {
    "xhs": {
        "label": "小红书",
        "result_key": "xhsResult",
    },
    "dy": {
        "label": "抖音",
        "result_key": "dyResult",
    },
    "gzh": {
        "label": "公众号",
        "result_key": "gzhResult",
    },
}
DEFAULT_COUNT = 50
SOURCE_LABEL = "多平台话题研究-GitHub"

# ─── API Key ────────────────────────────────────────────────────────────────────────


class InsufficientCreditsError(Exception):
    """API 积分不足错误"""
    pass


def get_api_key(cli_key: str | None = None) -> str:
    """按优先级获取 API Key: 命令行 > 环境变量 > 配置文件"""
    if cli_key:
        return cli_key
    # 环境变量
    for env_name in ("REDFOX_API_KEY", "X_API_KEY"):
        val = os.environ.get(env_name, "").strip()
        if val:
            return val
    # 配置文件
    config_path = os.path.expanduser("~/.qoder/apis/redfox.json")
    if os.path.isfile(config_path):
        try:
            with open(config_path) as f:
                cfg = json.load(f)
            val = (cfg.get("api_key") or "").strip()
            if val:
                return val
        except Exception:
            pass
    return ""


# ─── 数量解析 ───────────────────────────────────────────────────────────────────────
def parse_count(value: Any) -> int:
    """解析数量字段，支持 '1.2w'、'5000+' 等中文格式"""
    if value is None:
        return 0
    if isinstance(value, (int, float)):
        return int(value)
    text = str(value).replace("+", "").replace(",", "").strip()
    if not text:
        return 0
    try:
        if "w" in text.lower():
            return int(float(text.lower().replace("w", "")) * 10000)
        if text.endswith("万"):
            return int(float(text[:-1]) * 10000)
        if text.endswith("亿"):
            return int(float(text[:-1]) * 100000000)
        return int(float(text))
    except (TypeError, ValueError):
        return 0


def fuzzy_count(value: Any) -> str:
    """模糊化互动数，5000以下保留原始值"""
    num = parse_count(value)
    if num <= 0:
        return "--"
    if num < 5000:
        return str(num)
    if num < 10000:
        return "5000+"
    wan = num // 10000
    return f"{wan}w+"


# ─── HTTP 请求 ──────────────────────────────────────────────────────────────────────
def _http_post(url: str, payload: dict, api_key: str, max_retries: int = 3) -> dict:
    """带重试的 HTTP POST 请求"""
    import urllib.request
    import urllib.error

    headers = {
        "Content-Type": "application/json",
        "X-API-KEY": api_key,
        "User-Agent": "cn-last30days/1.0",
    }
    body = json.dumps(payload, ensure_ascii=False).encode("utf-8")
    last_error = None

    for attempt in range(max_retries):
        try:
            req = urllib.request.Request(url, data=body, headers=headers, method="POST")
            with urllib.request.urlopen(req, timeout=30) as resp:
                raw = resp.read().decode("utf-8")
            result = json.loads(raw)
            code = result.get("code")
            if code == 3108:
                # 限频，等待重试
                time.sleep(5 * (attempt + 1))
                continue
            if code == 3201:
                # 积分不足，不可重试
                raise InsufficientCreditsError(result.get("msg", "积分不足"))
            if code not in (200, 2000):
                raise Exception(f"API 错误 code={code}: {result.get('msg', '未知')}")
            return result
        except urllib.error.HTTPError as e:
            last_error = f"HTTP {e.code}"
            if attempt < max_retries - 1:
                time.sleep(2 ** attempt)
        except urllib.error.URLError as e:
            last_error = f"网络错误: {e.reason}"
            if attempt < max_retries - 1:
                time.sleep(2 ** attempt)
        except Exception as e:
            last_error = str(e)
            if attempt < max_retries - 1:
                time.sleep(2 ** attempt)

    raise Exception(f"请求失败: {last_error}（已尝试 {max_retries} 次）")


def _first_of(art: dict, *keys: str, default: Any = None) -> Any:
    """从文章字典中按优先级取第一个非空值"""
    for k in keys:
        v = art.get(k)
        if v is not None and v != "" and v != 0:
            return v
    return default


def _normalize_article(art: dict, platform: str, idx: int) -> dict:
    """将不同平台的数据归一化为统一格式"""
    if platform == "xhs":
        return _normalize_xhs(art, idx)
    elif platform == "dy":
        return _normalize_dy(art, idx)
    elif platform == "gzh":
        return _normalize_gzh(art, idx)
    return art


def _normalize_xhs(art: dict, idx: int) -> dict:
    """归一化小红书数据 - 兼容 xhsUser/searchArticle (work*前缀) 和 xhs/search/search 两种格式"""
    note_id = str(_first_of(art, "workId", "id", "noteId", "workUuid", "uuid", default=""))
    author_id = str(_first_of(art, "accountUserid", "authorId", "accountId", default=""))
    title_raw = _first_of(art, "workTitle", "title", "displayTitle", default="")
    desc_raw = _first_of(art, "workDesc", "desc", "displayDesc", "summary", default="")
    title = (title_raw or desc_raw or "无标题")[:200]
    desc = (desc_raw or "")[:500]
    # 链接
    note_link = _first_of(art, "workUrl", "shareInfoLink", "url", default="")
    if not note_link and note_id:
        xsec_token = art.get("xsecToken", "")
        if xsec_token:
            note_link = f"https://www.xiaohongshu.com/explore/{note_id}?xsec_token={xsec_token}"
        else:
            note_link = f"https://www.xiaohongshu.com/explore/{note_id}"
    author_link = f"https://www.xiaohongshu.com/user/profile/{author_id}" if author_id else ""
    # 作者
    author_name = _first_of(art, "accountNickname", "authorNickname", "author", "accountName", "nickname", default="未知")
    # 时间
    pub_time = _first_of(art, "workPublishTime", "createTime", "publishTime", "time", default="")
    if isinstance(pub_time, (int, float)) and pub_time > 1000000000000:
        from datetime import datetime as _dt
        try:
            pub_time = _dt.fromtimestamp(pub_time / 1000.0).strftime("%Y-%m-%d %H:%M:%S")
        except (OSError, ValueError):
            pub_time = str(pub_time)
    # 封面
    cover = _first_of(art, "coverUrl", "cover", default="")
    # 账号类型
    account_type = _first_of(art, "accountType", default="")
    # 笔记类型
    work_type = _first_of(art, "workType", "noteType", default="")
    return {
        "id": f"XHS{idx}",
        "platform": "小红书",
        "platform_key": "xhs",
        "title": title,
        "desc": desc,
        "url": note_link,
        "author": author_name,
        "author_id": author_id,
        "author_link": author_link,
        "author_fans": fuzzy_count(_first_of(art, "authorFans", "followerCount", default=0)),
        "published_at": str(pub_time),
        "engagement": {
            "likes": parse_count(_first_of(art, "workLikedCount", "likedCount", "likeCount", default=0)),
            "comments": parse_count(_first_of(art, "workCommentsCount", "commentsCount", "commentCount", default=0)),
            "collects": parse_count(_first_of(art, "workCollectedCount", "collectedCount", "collectCount", default=0)),
            "shares": parse_count(_first_of(art, "workSharedCount", "sharedCount", "shareCount", default=0)),
            "interactions": parse_count(_first_of(art, "interactiveCount", default=0)),
        },
        "engagement_display": _engagement_display(art, "xhs"),
        "cover": cover,
        "scores": _extract_scores(art),
        "account_type": account_type,
        "work_type": work_type,
    }


def _normalize_dy(art: dict, idx: int) -> dict:
    """归一化抖音数据 - 兼容 dyData/searchArticle 和 dy/search/search 两种格式"""
    work_url = _first_of(art, "workUrl", "url", default="")
    title_raw = _first_of(art, "title", "desc", default="")
    desc_raw = _first_of(art, "desc", "summary", default="")
    title = (title_raw or "无标题")[:200]
    desc = (desc_raw or "")[:500]
    author_name = _first_of(art, "accountName", "author", "authorNickname", default="未知")
    author_id = str(_first_of(art, "accountId", "authorId", default=""))
    pub_time = _first_of(art, "publishTime", "createTime", default="")
    cover = _first_of(art, "cover", "coverUrl", default="")
    return {
        "id": f"DY{idx}",
        "platform": "抖音",
        "platform_key": "dy",
        "title": title,
        "desc": desc,
        "url": work_url,
        "author": author_name,
        "author_id": author_id,
        "author_link": f"https://www.douyin.com/user/{author_id}" if author_id else "",
        "author_fans": fuzzy_count(_first_of(art, "followerCount", "authorFans", default=0)),
        "published_at": str(pub_time),
        "engagement": {
            "likes": parse_count(_first_of(art, "likeCount", "likedCount", default=0)),
            "comments": parse_count(_first_of(art, "commentCount", "commentsCount", default=0)),
            "collects": parse_count(_first_of(art, "collectCount", "collectedCount", default=0)),
            "shares": parse_count(_first_of(art, "shareCount", "sharedCount", default=0)),
        },
        "engagement_display": _engagement_display(art, "dy"),
        "cover": cover,
        "scores": _extract_scores(art),
    }


def _normalize_gzh(art: dict, idx: int) -> dict:
    """归一化公众号数据 - 适配 gzh/search/hotArticle 格式"""
    url = _first_of(art, "url", "workUrl", default="")
    title = (art.get("title") or "无标题")[:200]
    summary = _first_of(art, "summary", "desc", default="")
    author_name = _first_of(art, "author", "accountName", default="-")
    author_id = str(_first_of(art, "accountId", "authorId", default=""))
    pub_time = _first_of(art, "publicTime", "publishTime", "createTime", default="")
    cover = _first_of(art, "imageUrl", "coverUrl", "cover", default="")
    return {
        "id": f"GZH{idx}",
        "platform": "公众号",
        "platform_key": "gzh",
        "title": title,
        "desc": (summary or "")[:500],
        "url": url,
        "author": author_name,
        "author_id": author_id,
        "author_link": "",
        "author_fans": fuzzy_count(_first_of(art, "followerCount", "authorFans", default=0)),
        "published_at": str(pub_time),
        "engagement": {
            "reads": parse_count(_first_of(art, "clicksCount", "readCount", default=0)),
            "likes": parse_count(_first_of(art, "likeCount", "likedCount", default=0)),
            "watches": parse_count(_first_of(art, "watchCount", default=0)),
            "collects": parse_count(_first_of(art, "collectCount", "collectedCount", default=0)),
            "shares": parse_count(_first_of(art, "shareCount", "sharedCount", default=0)),
            "comments": parse_count(_first_of(art, "commentsCount", "commentCount", default=0)),
        },
        "engagement_display": _engagement_display(art, "gzh"),
        "cover": cover,
        "scores": _extract_scores(art),
    }


def _engagement_display(art: dict, platform: str) -> str:
    """生成可读的互动数据字符串"""
    if platform == "xhs":
        likes = fuzzy_count(_first_of(art, "workLikedCount", "likedCount", "likeCount", default=0))
        comments = fuzzy_count(_first_of(art, "workCommentsCount", "commentsCount", "commentCount", default=0))
        collects = fuzzy_count(_first_of(art, "workCollectedCount", "collectedCount", "collectCount", default=0))
        interactions = fuzzy_count(_first_of(art, "interactiveCount", default=0))
        return f"🔥{interactions}互动 👍{likes} ⭐{collects} 💬{comments}"
    elif platform == "dy":
        likes = fuzzy_count(_first_of(art, "workLikedCount", "likeCount", "likedCount", default=0))
        comments = fuzzy_count(_first_of(art, "workCommentsCount", "commentCount", "commentsCount", default=0))
        shares = fuzzy_count(_first_of(art, "workSharedCount", "shareCount", "sharedCount", default=0))
        collects = fuzzy_count(_first_of(art, "workCollectedCount", "collectCount", "collectedCount", default=0))
        return f"👍{likes} 💬{comments} ⭐{collects} 🔄{shares}"
    elif platform == "gzh":
        reads = fuzzy_count(_first_of(art, "clicksCount", "readCount", default=0))
        likes = fuzzy_count(_first_of(art, "likeCount", "likedCount", default=0))
        watches = fuzzy_count(_first_of(art, "watchCount", default=0))
        comments = fuzzy_count(_first_of(art, "commentsCount", "commentCount", default=0))
        shares = fuzzy_count(_first_of(art, "shareCount", "sharedCount", default=0))
        return f"📖{reads} 👍{likes} 👁{watches} 💬{comments} 🔄{shares}"
    return ""


def _extract_scores(art: dict) -> dict:
    """提取评分字段（如有关键词搜索评分）"""
    return {
        "total": art.get("totalScore", 0),
        "relevance": art.get("relevanceScore", 0),
        "popularity": art.get("popularityScore", 0),
        "recency": art.get("recencyScore", 0),
    }


# ─── 主搜索函数 ─────────────────────────────────────────────────────────────────────
def search(
    keyword: str,
    platforms: list[str] | None = None,
    count: int = DEFAULT_COUNT,
    api_key: str | None = None,
    days: int = 30,
) -> dict:
    """通过统一接口搜索多平台话题数据"""
    if not platforms:
        platforms = list(PLATFORMS.keys())

    key = get_api_key(api_key)
    if not key:
        sys.stderr.write("\u274c 未找到 API Key，请先配置：\n")
        sys.stderr.write("   export REDFOX_API_KEY=ak_你的密钥\n")
        sys.stderr.write("   或使用 --api-key 参数传入\n")
        sys.stderr.write("   注册地址: https://www.redfox.hk/login\n")
        sys.stderr.flush()
        sys.exit(1)

    # 构建统一请求参数
    today = datetime.now()
    start_date = (today - timedelta(days=days)).strftime("%Y-%m-%d")
    end_date = today.strftime("%Y-%m-%d")
    payload = {
        "keyword": keyword,
        "source": SOURCE_LABEL,
        "startDate": start_date,
        "endDate": end_date,
    }

    sys.stderr.write(f"[\u2699\ufe0f] 搜索中: {keyword} ...\n")
    sys.stderr.flush()

    results = {}
    credit_error = False

    try:
        result = _http_post(API_BASE, payload, key)
        data = result.get("data") or {}

        for p in platforms:
            if p not in PLATFORMS:
                results[p] = {"platform": p, "label": p, "items": [], "total": 0, "error": "未知平台"}
                continue

            plat = PLATFORMS[p]
            label = plat["label"]
            result_key = plat["result_key"]
            articles = data.get(result_key, [])
            if isinstance(articles, dict):
                articles = articles.get("articles", [])
            if not isinstance(articles, list):
                articles = []

            # 去重并归一化
            all_articles = []
            seen_ids = set()
            for art in articles:
                uid = (
                    art.get("workUuid") or art.get("uuid")
                    or art.get("id") or art.get("noteId")
                    or ""
                )
                if uid and uid in seen_ids:
                    continue
                if uid:
                    seen_ids.add(uid)
                item = _normalize_article(art, p, len(all_articles) + 1)
                all_articles.append(item)
                if len(all_articles) >= count:
                    break

            sys.stderr.write(f"[{label}] 获取 {len(all_articles)} 条\n")
            sys.stderr.flush()
            results[p] = {
                "platform": p,
                "label": label,
                "items": all_articles[:count],
                "total": len(all_articles[:count]),
            }

    except InsufficientCreditsError as e:
        sys.stderr.write(f"⚠️ {e}\n")
        sys.stderr.write(f"请配置个人 API Key: export REDFOX_API_KEY=你的密钥\n")
        sys.stderr.write(f"注册地址: https://www.redfox.hk/login\n")
        sys.stderr.flush()
        credit_error = True
    except Exception as e:
        sys.stderr.write(f"请求失败: {e}\n")
        sys.stderr.flush()

    # 为未处理的平台填充空结果
    for p in platforms:
        if p not in results:
            results[p] = {
                "platform": p,
                "label": PLATFORMS[p]["label"],
                "items": [],
                "total": 0,
            }
            if credit_error:
                results[p]["error"] = "积分不足，请配置个人 API Key"

    # 汇总统计
    total_items = sum(r["total"] for r in results.values())
    today_utc = datetime.now(timezone.utc)
    return {
        "keyword": keyword,
        "searched_at": today_utc.isoformat(),
        "date_range": {
            "from": (today_utc - timedelta(days=days)).strftime("%Y-%m-%d"),
            "to": today_utc.strftime("%Y-%m-%d"),
        },
        "platforms": results,
        "total_items": total_items,
    }


# ─── JSON 输出 ──────────────────────────────────────────────────────────────────────
def format_as_json(data: dict, max_items: int = 50) -> dict:
    """精简 JSON 格式（供 AI 智能体分析使用）"""
    output = {
        "keyword": data["keyword"],
        "searched_at": data["searched_at"],
        "date_range": data["date_range"],
        "total_items": data["total_items"],
        "platforms": {},
    }

    for pkey, pdata in data["platforms"].items():
        items = []
        for item in pdata.get("items", [])[:max_items]:
            items.append({
                "id": item["id"],
                "platform": item["platform"],
                "title": item["title"],
                "author": item["author"],
                "author_fans": item["author_fans"],
                "published_at": item["published_at"],
                "engagement_display": item["engagement_display"],
                "engagement": item["engagement"],
                "url": item["url"],
                "desc": item["desc"][:200],
                "scores": item.get("scores", {}),
            })
        output["platforms"][pkey] = {
            "label": pdata["label"],
            "total": pdata["total"],
            "items": items,
        }
        if pdata.get("error"):
            output["platforms"][pkey]["error"] = pdata["error"]

    return output


# ─── HTML 报告 ──────────────────────────────────────────────────────────────────────

def _md_to_html(text: str) -> str:
    """简易 Markdown → HTML 转换（无第三方依赖）"""
    import re
    lines = text.split("\n")
    out = []
    in_list = False
    for line in lines:
        stripped = line.strip()
        # 标题
        if stripped.startswith("### "):
            if in_list:
                out.append("</ul>"); in_list = False
            out.append(f'<h4>{stripped[4:]}</h4>')
        elif stripped.startswith("## "):
            if in_list:
                out.append("</ul>"); in_list = False
            out.append(f'<h3>{stripped[2:]}</h3>')
        elif stripped.startswith("# "):
            if in_list:
                out.append("</ul>"); in_list = False
            out.append(f'<h2>{stripped[2:]}</h2>')
        # 分隔线
        elif stripped == "---":
            if in_list:
                out.append("</ul>"); in_list = False
            out.append('<hr>')
        # 无序列表
        elif stripped.startswith("- "):
            if not in_list:
                out.append('<ul>'); in_list = True
            content = _md_inline(stripped[2:])
            out.append(f'<li>{content}</li>')
        # 有序列表
        elif re.match(r'^\d+\.\s', stripped):
            if not in_list:
                out.append('<ol>'); in_list = True
            content = _md_inline(re.sub(r'^\d+\.\s', '', stripped))
            out.append(f'<li>{content}</li>')
        # 空行
        elif not stripped:
            if in_list:
                out.append("</ul>"); in_list = False
            out.append('')
        # 普通段落
        else:
            if in_list:
                out.append("</ul>"); in_list = False
            out.append(f'<p>{_md_inline(stripped)}</p>')
    if in_list:
        out.append("</ul>")
    return "\n".join(out)


def _md_inline(text: str) -> str:
    """行内 Markdown 转换：粗体、链接"""
    import re
    text = text.replace("&", "&amp;").replace("<", "&lt;").replace(">", "&gt;")
    # [text](url)
    text = re.sub(r'\[([^\]]+)\]\(([^)]+)\)', r'<a href="\2" target="_blank">\1</a>', text)
    # **bold**
    text = re.sub(r'\*\*(.+?)\*\*', r'<strong>\1</strong>', text)
    return text


def format_as_html(data: dict, max_items: int = 50, report_html: str = "") -> str:
    """生成网站风格 HTML 报告"""
    keyword = data["keyword"]
    total = data["total_items"]
    date_range = data["date_range"]

    # 平台配色和图标
    platform_meta = {
        "xhs": {"primary": "#ff2442", "bg": "#fff1f0", "icon": "📕", "name": "小红书"},
        "dy": {"primary": "#161823", "bg": "#f5f5f5", "icon": "🎵", "name": "抖音"},
        "gzh": {"primary": "#07c160", "bg": "#f0fff4", "icon": "📖", "name": "公众号"},
    }

    # 统计卡片
    stats_html = ""
    for pkey, pdata in data["platforms"].items():
        meta = platform_meta.get(pkey, platform_meta["xhs"])
        ptotal = pdata["total"]
        # 统计总互动
        total_likes = sum(it.get("engagement", {}).get("likes", 0) for it in pdata.get("items", [])[:max_items])
        total_reads = sum(
            it.get("engagement", {}).get("reads", 0) + it.get("engagement", {}).get("likes", 0) + it.get("engagement", {}).get("collects", 0) + it.get("engagement", {}).get("shares", 0) + it.get("engagement", {}).get("comments", 0)
            for it in pdata.get("items", [])[:max_items]
        )
        m_primary = meta["primary"]
        m_bg = meta["bg"]
        m_icon = meta["icon"]
        m_name = meta["name"]
        stats_html += f'''
            <div class="stat-card" style="--p-color: {m_primary}; --p-bg: {m_bg}">
                <div class="stat-icon">{m_icon}</div>
                <div class="stat-body">
                    <div class="stat-label">{m_name}</div>
                    <div class="stat-num">{ptotal} <small>条</small></div>
                </div>
            </div>'''

    # 构建 Tab 和内容
    tabs_html = ""
    panels_html = ""
    for pkey, pdata in data["platforms"].items():
        meta = platform_meta.get(pkey, platform_meta["xhs"])
        label = pdata["label"]
        ptotal = pdata["total"]
        is_first = pkey == list(data["platforms"].keys())[0]

        active = " active" if is_first else ""
        m_primary = meta["primary"]
        m_icon = meta["icon"]
        tabs_html += (
            '<button class="tab-btn' + active + '" data-platform="' + pkey + '" '
            'style="--tab-color: ' + m_primary + '">' + m_icon + ' ' + label
            + ' <span class="tab-count">' + str(ptotal) + '</span></button>\n'
        )

        display = "block" if is_first else "none"
        items = pdata.get("items", [])[:max_items]
        error_html = ""
        if pdata.get("error"):
            error_html = '<div class="error-banner">⚠️ ' + pdata["error"] + '</div>'

        cards = ""
        for idx, item in enumerate(items):
            title_escaped = item["title"].replace("&", "&amp;").replace("<", "&lt;").replace(">", "&gt;").replace('"', "&quot;")
            desc_escaped = item["desc"][:200].replace("&", "&amp;").replace("<", "&lt;").replace(">", "&gt;") if item.get("desc") else ""
            author_escaped = item["author"].replace("&", "&amp;").replace("<", "&lt;").replace(">", "&gt;")

            item_url = item.get("url", "")
            url_attr = 'href="' + item_url + '"' if item_url else 'href="#"'
            author_link = item.get("author_link", "")
            author_html = ('<a href="' + author_link + '" class="author-link" target="_blank">' + author_escaped + '</a>') if author_link else ('<span class="author-name">' + author_escaped + '</span>')

            # 互动数据标签 — 按平台展示对应字段，始终显示（包括0值），使用fuzzy_count格式化
            eng = item.get("engagement", {})
            eng_tags = ""
            if pkey == "gzh":
                reads = fuzzy_count(eng.get("reads", 0))
                likes = fuzzy_count(eng.get("likes", 0))
                shares = fuzzy_count(eng.get("shares", 0))
                eng_tags = ('<span class="eng-tag reads">📖 ' + reads + '</span>'
                            + '<span class="eng-tag">👍 ' + likes + '</span>'
                            + '<span class="eng-tag">🔄 ' + shares + '</span>')
            elif pkey == "xhs":
                likes = fuzzy_count(eng.get("likes", 0))
                collects = fuzzy_count(eng.get("collects", 0))
                comments = fuzzy_count(eng.get("comments", 0))
                eng_tags = ('<span class="eng-tag">👍 ' + likes + '</span>'
                            + '<span class="eng-tag">⭐ ' + collects + '</span>'
                            + '<span class="eng-tag">💬 ' + comments + '</span>')
            elif pkey == "dy":
                likes = fuzzy_count(eng.get("likes", 0))
                comments = fuzzy_count(eng.get("comments", 0))
                shares = fuzzy_count(eng.get("shares", 0))
                eng_tags = ('<span class="eng-tag">👍 ' + likes + '</span>'
                            + '<span class="eng-tag">💬 ' + comments + '</span>'
                            + '<span class="eng-tag">🔄 ' + shares + '</span>')

            author_fans = item.get("author_fans", "--")
            pub_date = item.get("published_at", "")[:10] if item.get("published_at") else "--"
            desc_html = ('<p class="card-desc">' + desc_escaped + '</p>') if desc_escaped else ''
            rank_num = idx + 1

            # 公众号没有粉丝数，不展示粉丝字段
            if pkey == "gzh":
                fans_html = ''
            else:
                fans_html = (
                    '                        <span class="dot">·</span>\n'
                    '                        <span class="fans">' + str(author_fans) + '粉</span>\n'
                )

            cards += (
                '\n            <div class="card" style="--card-accent: ' + m_primary + '">\n'
                '                <div class="card-rank">' + str(rank_num) + '</div>\n'
                '                <div class="card-body">\n'
                '                    <a ' + url_attr + ' class="card-title" target="_blank">' + title_escaped + '</a>\n'
                '                    <div class="card-meta">\n'
                '                        ' + author_html + '\n'
                + fans_html +
                '                        <span class="dot">·</span>\n'
                '                        <span class="time">' + pub_date + '</span>\n'
                '                    </div>\n'
                '                    ' + desc_html + '\n'
                '                    <div class="card-footer">\n'
                '                        <div class="engagement">' + eng_tags + '</div>\n'
                '                        <a ' + url_attr + ' class="view-btn" target="_blank">查看原文 ↗</a>\n'
                '                    </div>\n'
                '                </div>\n'
                '            </div>'
            )

        xhs_notice = (
            '<div style="background:#fff3cd;border:1px solid #ffc107;border-radius:8px;padding:10px 14px;margin-bottom:14px;color:#664d03;font-size:13px;line-height:1.6">'
            '⚠️ 受小红书风控规则限制，部分作品链接可能无法正常跳转，您可复制对应作品标题前往小红书搜索查看，感谢理解🙇‍♀️🙇‍♀️'
            '</div>'
        ) if pkey == "xhs" else ""
        no_data_html = "<div class='no-data-hint'><p>未查询到相关内容，建议更换关键词重试。</p></div>" if not items else ""
        panels_html += (
            '\n        <div class="tab-panel" id="panel-' + pkey + '" style="display: ' + display + '">\n'
            '            ' + error_html + '\n'
            '            ' + xhs_notice + '\n'
            '            <div class="card-list">\n'
            '                ' + cards + '\n'
            '            </div>\n'
            '            ' + no_data_html + '\n'
            '        </div>'
        )

    # ── 研究报告区域 ──
    report_section = ""
    if report_html:
        report_section = f'''
    <div class="report-section">
        <h2 class="section-title">📝 研究报告</h2>
        <div class="report-content">{report_html}</div>
    </div>'''

    html = f'''<!DOCTYPE html>
<html lang="zh-CN">
<head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>cn-last30days · {keyword}</title>
    <style>
        * {{ margin: 0; padding: 0; box-sizing: border-box; }}
        :root {{
            --primary: #4f46e5;
            --primary-light: #e0e7ff;
            --text: #1f2937;
            --text-secondary: #6b7280;
            --border: #e5e7eb;
            --bg: #f9fafb;
            --card-bg: #ffffff;
            --radius: 12px;
        }}
        body {{
            font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', 'PingFang SC', 'Noto Sans SC', Roboto, sans-serif;
            background: var(--bg); color: var(--text); line-height: 1.6;
        }}
        /* ── 导航栏 ── */
        .navbar {{
            background: white; border-bottom: 1px solid var(--border);
            padding: 12px 24px; position: sticky; top: 0; z-index: 100;
            display: flex; align-items: center; gap: 12px;
        }}
        .navbar .logo {{ font-size: 18px; font-weight: 800; color: var(--primary); letter-spacing: -0.5px; }}
        .navbar .logo span {{ color: #ef4444; }}
        .navbar .badge {{
            background: var(--primary-light); color: var(--primary);
            font-size: 11px; font-weight: 600; padding: 2px 8px; border-radius: 20px;
        }}
        /* ── Hero ── */
        .hero {{
            background: linear-gradient(135deg, #1e1b4b 0%, #312e81 50%, #4338ca 100%);
            color: white; padding: 48px 24px; text-align: center;
        }}
        .hero h1 {{ font-size: 28px; font-weight: 800; margin-bottom: 8px; }}
        .hero .keyword {{
            display: inline-block; background: rgba(255,255,255,0.15); border-radius: 8px;
            padding: 4px 16px; font-size: 16px; margin: 8px 0;
        }}
        .hero .date-range {{ font-size: 14px; opacity: 0.8; }}
        /* ── 统计卡片 ── */
        .stats-row {{
            display: flex; gap: 16px; padding: 24px; max-width: 960px; margin: -28px auto 0;
            position: relative; z-index: 10; flex-wrap: wrap; justify-content: center;
        }}
        .stat-card {{
            background: white; border-radius: var(--radius); padding: 20px 24px;
            box-shadow: 0 4px 12px rgba(0,0,0,0.08); flex: 1; min-width: 180px;
            display: flex; align-items: center; gap: 16px; border-left: 4px solid var(--p-color);
        }}
        .stat-icon {{ font-size: 32px; }}
        .stat-label {{ font-size: 13px; color: var(--text-secondary); }}
        .stat-num {{ font-size: 24px; font-weight: 800; color: var(--text); }}
        .stat-num small {{ font-size: 13px; font-weight: 400; color: var(--text-secondary); }}
        /* ── 主内容区 ── */
        .main {{ max-width: 960px; margin: 24px auto; padding: 0 24px; }}
        /* ── Tab ── */
        .tab-bar {{
            display: flex; gap: 4px; background: white; border-radius: var(--radius) var(--radius) 0 0;
            padding: 8px 8px 0; border: 1px solid var(--border); border-bottom: none;
        }}
        .tab-btn {{
            padding: 12px 24px; border: none; background: transparent;
            font-size: 15px; font-weight: 600; cursor: pointer; color: var(--text-secondary);
            border-radius: 8px 8px 0 0; transition: all 0.2s; position: relative;
        }}
        .tab-btn:hover {{ background: var(--bg); color: var(--tab-color); }}
        .tab-btn.active {{
            background: var(--bg); color: var(--tab-color);
        }}
        .tab-btn.active::after {{
            content: ''; position: absolute; bottom: 0; left: 20%; right: 20%;
            height: 3px; background: var(--tab-color); border-radius: 3px 3px 0 0;
        }}
        .tab-count {{
            font-size: 12px; background: var(--bg); padding: 2px 8px;
            border-radius: 10px; font-weight: 500; margin-left: 4px;
        }}
        .tab-btn.active .tab-count {{ background: var(--primary-light); color: var(--tab-color); }}
        /* ── 卡片列表 ── */
        .card-list {{ background: white; border: 1px solid var(--border); border-radius: 0 0 var(--radius) var(--radius); }}
        .card {{
            display: flex; gap: 16px; padding: 20px 24px;
            border-bottom: 1px solid #f3f4f6; transition: background 0.15s;
        }}
        .card:last-child {{ border-bottom: none; }}
        .card:hover {{ background: #fafbfc; }}
        .card-rank {{
            font-size: 20px; font-weight: 800; color: var(--card-accent);
            min-width: 32px; text-align: center; padding-top: 2px; opacity: 0.8;
        }}
        .card-body {{ flex: 1; min-width: 0; }}
        .card-title {{
            font-size: 16px; font-weight: 600; color: var(--text); text-decoration: none;
            line-height: 1.5; display: block; transition: color 0.15s;
        }}
        .card-title:hover {{ color: var(--card-accent); }}
        .card-meta {{ font-size: 13px; color: var(--text-secondary); margin-top: 6px; display: flex; align-items: center; gap: 0; flex-wrap: wrap; }}
        .author-link {{ color: var(--primary); text-decoration: none; font-weight: 500; transition: color 0.15s; }}
        .author-link:hover {{ color: var(--card-accent); text-decoration: underline; }}
        .author-name {{ color: var(--primary); font-weight: 500; }}
        .dot {{ margin: 0 6px; }}
        .card-desc {{ font-size: 14px; color: var(--text-secondary); line-height: 1.6; margin-top: 8px; display: -webkit-box; -webkit-line-clamp: 2; -webkit-box-orient: vertical; overflow: hidden; }}
        .card-footer {{ display: flex; justify-content: space-between; align-items: center; margin-top: 10px; }}
        .engagement {{ display: flex; gap: 8px; flex-wrap: wrap; }}
        .eng-tag {{ font-size: 12px; color: var(--text-secondary); background: #f3f4f6; padding: 2px 8px; border-radius: 4px; }}
        .eng-tag.reads {{ background: #ecfdf5; color: #065f46; }}
        .view-btn {{
            font-size: 13px; color: var(--card-accent); text-decoration: none;
            font-weight: 500; white-space: nowrap; transition: opacity 0.15s;
        }}
        .view-btn:hover {{ opacity: 0.7; }}
        /* ── 其他 ── */
        .error-banner {{
            background: #fef3c7; color: #92400e; padding: 12px 16px;
            border-radius: 8px; margin: 16px 24px; font-size: 14px;
        }}
        .no-data-hint {{ text-align: center; color: var(--text-secondary); padding: 48px; }}
        /* ── 研究报告 ── */
        .report-section {{
            max-width: 960px; margin: 0 auto; padding: 0 24px;
        }}
        .section-title {{
            font-size: 20px; font-weight: 800; color: var(--text); margin-bottom: 16px;
            padding-bottom: 8px; border-bottom: 2px solid var(--primary);
        }}
        .report-content {{
            background: white; border: 1px solid var(--border); border-radius: var(--radius);
            padding: 32px; line-height: 1.8; color: var(--text); font-size: 15px;
        }}
        .report-content h2 {{ font-size: 20px; font-weight: 800; color: var(--primary); margin: 24px 0 12px; }}
        .report-content h3 {{ font-size: 18px; font-weight: 700; color: var(--text); margin: 20px 0 10px; }}
        .report-content h4 {{ font-size: 16px; font-weight: 700; color: var(--text); margin: 16px 0 8px; }}
        .report-content p {{ margin-bottom: 12px; }}
        .report-content strong {{ color: #1e1b4b; }}
        .report-content a {{ color: var(--primary); text-decoration: none; }}
        .report-content a:hover {{ text-decoration: underline; }}
        .report-content ul, .report-content ol {{ margin: 8px 0 12px 20px; }}
        .report-content li {{ margin-bottom: 4px; }}
        .report-content hr {{ border: none; border-top: 1px solid var(--border); margin: 20px 0; }}
        .footer {{
            text-align: center; color: var(--text-secondary); font-size: 12px;
            padding: 32px 24px; margin-top: 16px;
        }}
        .footer a {{ color: var(--primary); text-decoration: none; }}
        /* ── 响应式 ── */
        @media (max-width: 640px) {{
            .hero {{ padding: 32px 16px; }}
            .hero h1 {{ font-size: 20px; }}
            .stats-row {{ padding: 16px; margin-top: -20px; }}
            .stat-card {{ min-width: 140px; padding: 14px 16px; }}
            .stat-num {{ font-size: 20px; }}
            .main {{ padding: 0 12px; }}
            .card {{ padding: 14px 16px; gap: 10px; }}
            .card-rank {{ font-size: 16px; min-width: 24px; }}
            .card-title {{ font-size: 15px; }}
        }}
    </style>
</head>
<body>
    <nav class="navbar">
        <div class="logo">🇨🇳 cn<span>-last30days</span></div>
        <div class="badge">v2.0</div>
    </nav>
    <div class="hero">
        <h1>中国社媒话题研究</h1>
        <div class="keyword">{keyword}</div>
        <div class="date-range">{date_range["from"]} ~ {date_range["to"]}</div>
    </div>
    <div class="stats-row">
        {stats_html}
        <div class="stat-card" style="--p-color: var(--primary); --p-bg: var(--primary-light)">
            <div class="stat-icon">📊</div>
            <div class="stat-body">
                <div class="stat-label">合计</div>
                <div class="stat-num">{total} <small>条</small></div>
            </div>
        </div>
    </div>
    {report_section}
    <div class="main">
        <div class="tab-bar">
            {tabs_html}
        </div>
        {panels_html}
    </div>
    <div class="footer">
        数据来源：<a href="https://redfox.hk" target="_blank">redfox.hk</a> API · 小红书 / 抖音 / 公众号 · 近30天热门内容<br>
        互动数据为入库快照，实时数据可能持续增长
    </div>
    <script>
        document.querySelectorAll('.tab-btn').forEach(btn => {{
            btn.addEventListener('click', () => {{
                document.querySelectorAll('.tab-btn').forEach(b => b.classList.remove('active'));
                document.querySelectorAll('.tab-panel').forEach(p => p.style.display = 'none');
                btn.classList.add('active');
                document.getElementById('panel-' + btn.dataset.platform).style.display = 'block';
            }});
        }});
    </script>
</body>
</html>'''

    return html


# ─── CLI ─────────────────────────────────────────────────────────────────────────────
def main():
    parser = argparse.ArgumentParser(
        description="cn-last30days: 中国社媒平台话题研究工具"
    )
    parser.add_argument("keyword", nargs="?", default="dummy", help="搜索关键词（--from-json 模式下可省略）")
    parser.add_argument(
        "--platforms", "-p",
        default="xhs,dy,gzh",
        help="平台列表，逗号分隔（默认: xhs,dy,gzh）"
    )
    parser.add_argument(
        "--count", "-n",
        type=int,
        default=DEFAULT_COUNT,
        help=f"每个平台获取条数（默认: {DEFAULT_COUNT}）"
    )
    parser.add_argument(
        "--days", "-d",
        type=int,
        default=30,
        help="搜索时间范围，最近多少天（默认: 30，最大: 30）"
    )
    parser.add_argument(
        "--output-format", "-f",
        choices=["json", "html", "both"],
        default="json",
        help="输出格式（默认: json，综合报告后再按需生成HTML）"
    )
    parser.add_argument(
        "--output-dir",
        default=str(Path.home() / "Downloads" / "CnLast30Days"),
        help="HTML 输出目录"
    )
    parser.add_argument(
        "--api-key",
        default=None,
        help="API Key（覆盖环境变量和配置文件）"
    )
    parser.add_argument(
        "--max-items",
        type=int,
        default=50,
        help="输出中最多展示条数（默认: 50）"
    )
    parser.add_argument(
        "--from-json",
        default=None,
        help="从已有 JSON 文件生成 HTML，不调用 API（值: JSON 文件路径）"
    )
    parser.add_argument(
        "--report-file",
        default=None,
        help="研究报告 Markdown 文件路径（嵌入到 HTML 报告顶部）"
    )
    parser.add_argument(
        "--debug",
        action="store_true",
        help="调试模式，打印原始 API 响应"
    )

    args = parser.parse_args()

    # ── 从 JSON 生成 HTML 模式 ──
    if args.from_json:
        json_path = Path(args.from_json)
        if not json_path.exists():
            sys.stderr.write(f"错误: JSON 文件不存在: {json_path}\n")
            sys.exit(1)
        raw = json.loads(json_path.read_text(encoding="utf-8"))
        # 从精简 JSON 反向构造 data 结构
        data = {
            "keyword": raw.get("keyword", ""),
            "total_items": raw.get("total_items", 0),
            "date_range": raw.get("date_range", {}),
            "platforms": {},
        }
        for pkey, pdata in raw.get("platforms", {}).items():
            items = []
            for it in pdata.get("items", []):
                item = dict(it)
                item["source"] = pkey
                items.append(item)
            data["platforms"][pkey] = {
                "label": pdata.get("label", pkey),
                "total": pdata.get("total", len(items)),
                "items": items,
            }
        # 读取研究报告
        report_html = ""
        if args.report_file:
            report_path = Path(args.report_file)
            if report_path.exists():
                report_md = report_path.read_text(encoding="utf-8")
                report_html = _md_to_html(report_md)
            else:
                sys.stderr.write(f"⚠️ 报告文件不存在: {report_path}，跳过\n")
        html_content = format_as_html(data, max_items=args.max_items, report_html=report_html)
        output_dir = Path(args.output_dir)
        output_dir.mkdir(parents=True, exist_ok=True)
        base_name = json_path.stem  # 用原 JSON 文件名
        html_file = output_dir / f"{base_name}.html"
        html_file.write_text(html_content, encoding="utf-8")
        sys.stderr.write(f"✅ HTML 已保存: {html_file}\n")
        print(json.dumps({"files": {"html": str(html_file)}}, ensure_ascii=False))
        return

    # ── 正常搜索模式 ──
    # 解析平台列表
    platforms = []
    for p in args.platforms.split(","):
        p = p.strip().lower()
        if p in PLATFORMS:
            platforms.append(p)
        else:
            sys.stderr.write(f"未知平台: {p}，可用: {', '.join(PLATFORMS.keys())}\n")
    if not platforms:
        sys.stderr.write("错误: 未指定有效平台\n")
        sys.exit(1)

    # 执行搜索
    keyword = args.keyword.strip()
    if not keyword:
        sys.stderr.write("错误: 关键词不能为空\n")
        sys.exit(1)

    sys.stderr.write(f"\n{'='*60}\n")
    sys.stderr.write(f"cn-last30days · 搜索: {keyword}\n")
    sys.stderr.write(f"平台: {', '.join(PLATFORMS[p]['label'] for p in platforms)}\n")
    sys.stderr.write(f"每平台: {args.count} 条 | 时间: 近{args.days}天\n")
    sys.stderr.write(f"{'='*60}\n\n")
    sys.stderr.flush()

    try:
        data = search(
            keyword=keyword,
            platforms=platforms,
            count=args.count,
            api_key=args.api_key,
            days=args.days,
        )
    except Exception as e:
        sys.stderr.write(f"\n❌ 搜索失败: {e}\n")
        sys.exit(1)

    # 准备输出目录和文件名
    output_dir = Path(args.output_dir)
    output_dir.mkdir(parents=True, exist_ok=True)
    keyword_safe = keyword.replace('"', '').replace(' ', '_')[:30]
    timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
    base_name = f"cn30days_{keyword_safe}_{timestamp}"

    # 保存 JSON
    json_file = None
    if args.output_format in ("json", "both"):
        json_data = format_as_json(data, max_items=args.max_items)
        json_file = output_dir / f"{base_name}.json"
        json_file.write_text(
            json.dumps(json_data, ensure_ascii=False, indent=2),
            encoding="utf-8",
        )
        sys.stderr.write(f"\n✅ JSON 已保存: {json_file}\n")

    # 保存 HTML
    html_file = None
    if args.output_format in ("html", "both"):
        html_content = format_as_html(data, max_items=args.max_items)
        html_file = output_dir / f"{base_name}.html"
        html_file.write_text(html_content, encoding="utf-8")
        sys.stderr.write(f"✅ HTML 已保存: {html_file}\n")

    # 统计（stderr，不影响 stdout）
    sys.stderr.write(f"\n{'='*60}\n")
    sys.stderr.write(f"搜索完成！共 {data['total_items']} 条结果\n")
    for pkey, pdata in data["platforms"].items():
        status = f"✅ {pdata['total']} 条" if not pdata.get("error") else f"❌ {pdata['error']}"
        sys.stderr.write(f"  {pdata['label']}: {status}\n")
    sys.stderr.write(f"{'='*60}\n")
    sys.stderr.flush()

    # stdout 输出简洁摘要（供 AI 智能体解析，单行 JSON）
    summary = {
        "keyword": keyword,
        "date_range": data["date_range"],
        "total_items": data["total_items"],
        "platforms": {p: v["total"] for p, v in data["platforms"].items()},
        "files": {},
    }
    if json_file:
        summary["files"]["json"] = str(json_file)
    if html_file:
        summary["files"]["html"] = str(html_file)
    print(json.dumps(summary, ensure_ascii=False))


if __name__ == "__main__":
    main()