{
  "update_rule": {
    "update_hour": 9,
    "update_minute": 0,
    "lookback_days": 7,
    "update_note": "每日早上 9:00 更新前一天榜单"
  },
  "genders": [
    {"value": "all", "zh": "全部", "keywords": ["全部", "所有性别", "不分性别", "男女都", "all"]},
    {"value": "female", "zh": "女", "keywords": ["女性", "女生", "女榜", "女", "female"]},
    {"value": "male", "zh": "男", "keywords": ["男性", "男生", "男榜", "男", "male"]}
  ],
  "categories": [
    {"zh": "全部行业", "keywords": ["全部行业", "全品类", "全部", "综合", "所有", "不限"]},
    {"zh": "商业创业", "keywords": ["商业创业", "创业", "商业", "生意", "企业家", "经商"]},
    {"zh": "汽车出行", "keywords": ["汽车出行", "汽车", "车评", "新能源车", "摩托车", "出行", "车"]},
    {"zh": "设计创意", "keywords": ["设计创意", "设计", "创意", "平面设计", "工业设计", "UI设计"]},
    {"zh": "航空领域", "keywords": ["航空领域", "航空", "飞机", "民航", "飞行员", "机场"]},
    {"zh": "电商零售", "keywords": ["电商零售", "电商", "零售", "跨境电商", "直播带货", "网购"]},
    {"zh": "教育培训", "keywords": ["教育培训", "教育", "培训", "考研", "考试", "课程", "老师", "学习"]},
    {"zh": "工程运营", "keywords": ["工程运营", "工程", "运营", "制造", "机械", "运维"]},
    {"zh": "娱乐文化", "keywords": ["娱乐文化", "娱乐", "文化", "明星", "影视", "综艺", "追星", "八卦"]},
    {"zh": "环境能源", "keywords": ["环境能源", "环境", "能源", "气候", "环保", "电力", "光伏"]},
    {"zh": "家庭育儿", "keywords": ["家庭育儿", "育儿", "母婴", "亲子", "宝宝", "早教", "家庭"]},
    {"zh": "时尚美妆", "keywords": ["时尚美妆", "美妆", "时尚", "穿搭", "护肤", "彩妆", "潮流"]},
    {"zh": "金融投资", "keywords": ["金融投资", "金融", "投资", "交易", "股票", "基金", "理财", "加密货币", "财经", "炒股"]},
    {"zh": "美食饮品", "keywords": ["美食饮品", "美食", "饮品", "探店", "烹饪", "咖啡", "烘焙", "吃播", "菜谱"]},
    {"zh": "游戏电竞", "keywords": ["游戏电竞", "游戏", "电竞", "手游", "网游", "主机游戏"]},
    {"zh": "健康医疗", "keywords": ["健康医疗", "医疗", "医学", "医生", "疾病", "药学", "临床"]},
    {"zh": "牧场马术", "keywords": ["牧场马术", "马术", "牧场", "骑马", "马场"]},
    {"zh": "宗教信仰", "keywords": ["宗教信仰", "宗教", "信仰", "灵性", "佛教", "基督教"]},
    {"zh": "科学学术", "keywords": ["科学学术", "科学", "学术", "科研", "物理", "化学", "天文", "学者"]},
    {"zh": "体育健身", "keywords": ["体育健身", "体育", "健身", "运动", "篮球", "足球", "跑步", "瑜伽", "球赛"]},
    {"zh": "科技软件", "keywords": ["科技软件", "科技", "软件", "编程", "互联网", "人工智能", "AI", "程序员", "数码", "SaaS"]},
    {"zh": "旅行户外", "keywords": ["旅行户外", "旅行", "旅游", "户外", "露营", "徒步", "景点", "攻略"]},
    {"zh": "视觉艺术", "keywords": ["视觉艺术", "绘画", "摄影", "插画", "雕塑", "艺术"]},
    {"zh": "健康生活", "keywords": ["健康生活", "养生", "生活方式", "冥想", "心理健康"]},
    {"zh": "居家手工", "keywords": ["居家手工", "居家", "手工", "手作", "DIY", "家居", "收纳"]},
    {"zh": "人力职场", "keywords": ["人力职场", "人力", "职场", "招聘", "简历", "职业发展", "HR"]},
    {"zh": "公益平权", "keywords": ["公益平权", "公益", "平权", "慈善", "社会责任"]},
    {"zh": "法律法务", "keywords": ["法律法务", "法律", "法务", "律师", "司法", "法规"]},
    {"zh": "市场营销", "keywords": ["市场营销", "营销", "销售", "广告", "品牌", "增长黑客"]},
    {"zh": "音乐领域", "keywords": ["音乐领域", "音乐", "歌手", "乐队", "说唱", "钢琴", "吉他"]},
    {"zh": "萌宠动物", "keywords": ["萌宠动物", "宠物", "萌宠", "猫咪", "狗狗", "猫", "狗", "动物", "兔子"]},
    {"zh": "政治新闻", "keywords": ["政治新闻", "政治", "新闻", "时政", "政策", "记者", "媒体"]}
  ]
}

# X 热门账号榜 API 接口文档

## 接口基础信息

| 字段 | 值 |
|------|-----|
| 接口地址 | `https://redfox.hk/story/api/x/hotAccount/rankList` |
| 请求方式 | POST |
| Content-Type | `application/json` |
| 认证方式 | `X-API-KEY`（从环境变量 `REDFOX_API_KEY` 获取，格式 `ak_xxxxxxxx`） |

---

## 请求参数

```json
{
  "pageNum": 1,                  // 页码，从 1 开始（每页固定 20 条）
  "rankDate": "2026-09-20",     // 榜单日期 YYYY-MM-DD
  "gender": "all",              // 性别：all / male / female
  "category": "all",            // 行业分类：all 或行业中文名（如 "萌宠动物"）
  "source": "X账号榜-GitHub"     // 固定来源标识，每次请求必传
}
```

> ⚠️ **重要**：`category` 传**行业中文名**（如用户问宠物 → 传 `"萌宠动物"`），全部行业传 `"all"`；skill 内部不暴露英文分类值。

---

## 响应结构

```json
{
  "code": 2000,
  "msg": null,
  "data": {
    "pageNum": 1,
    "pageSize": 20,
    "pages": 10,
    "total": 200,
    "list": [
      {
        "rankNo": 1,
        "fullName": "Donald J. Trump",
        "profileUrl": "https://www.twitter.com/realdonaldtrump/",
        "pictureUrl": "https://favikon-medias.s3.../25073877.jpg",
        "title": "Former U.S. President and Political Leader",
        "biography": "Donald J. Trump is the 45th and 47th President...",
        "countryCode": "us",
        "countryName": "United States",
        "gender": "male",
        "xScore": 98.6,
        "growthPercentage": "+0.11 %",
        "growthAbsolute": "+186.1K",
        "xFollowerCount": 111800000,
        "statDate": "2026-09-20",
        "typologie": null,
        "categoriesJson": "[{\"root_name\":\"⚖️ Law, Medias & Politics\",\"name\":\"Politicians\",\"primary\":true,...}]",
        "networksJson": "{\"youtube\":{\"follower_count\":\"4M\",\"profile_url\":\"...\"},\"twitter\":{...},\"tiktok\":{...}}",
        "badgesJson": "{\"hasTopics\":\"US Elections 2024, Trump Government\",\"hasCause\":\"Conservatives,...\",\"verifiedCreator\":false,...}"
      }
    ]
  }
}
```

---

## 错误码

| code | 含义 |
|------|------|
| 2000 | **成功**（注意：成功码是 2000 不是 0！） |
| 其他 | 错误；目标日期/筛选无数据时 `data.list` 为空数组、`total=0` |

---

## 字段映射（fetch_rank.py normalized JSON）

| 接口原始字段 | normalized 字段 | 含义 |
|------------|----------------|------|
| `rankNo` | `rank` | 排名 |
| `fullName` | `fullName` | 账号名 |
| `profileUrl` | `profileUrl` | X 主页链接 |
| `pictureUrl` | `pictureUrl` | 头像 |
| `title` | `title` | 头街（一句话介绍） |
| `biography` | `biography` | 账号简介 |
| `countryName` / `countryCode` | `country` / `countryCode` | 国家/地区 |
| `categoriesJson`（primary=true 的 root_name，去 emoji） | `industries` | 主行业 |
| `xScore` | `score` | 影响力评分（满分 100） |
| `growthPercentage` | `growthPercentage` | 日涨幅（去空格，如 `+0.11%`） |
| `growthAbsolute` | `growthAbsolute` | 日涨粉绝对值 |
| `xFollowerCount` | `xFollowers` / `xFollowersFmt` | X 粉丝数（原始值 / 格式化） |
| `networksJson`（除 twitter 外） | `otherNetworks` | 其他平台粉丝数列表 |
| `badgesJson.hasTopics` / `hasCause` | `topics` / `cause` | 专题标签 / 公益标签 |

**JSON 字符串字段**：`categoriesJson`、`networksJson`、`badgesJson` 均为 JSON 字符串，需二次 `json.loads` 解析；解析失败时按空值容错处理。

**头像时效**：`pictureUrl` 为 S3 签名 URL（有效期约 15 分钟），HTML 报告内置 `onerror` 容错，头像失效时自动隐藏、不影响排版。

---

## 更新规则与日期计算

| 规则 | 值 |
|------|-----|
| 更新时间 | 每日早上 9:00 更新前一天榜单 |
| 最新可用日期 | 当前时间 ≥ 9:00 → 昨日；< 9:00 → 前日 |
| 回溯范围 | 过去 7 天 |
| 空数据回退 | 目标日期 `total=0` 时，自最新可用日期向前逐日扫描（最多 7 天） |

---

## 分页规则

| 字段 | 值 |
|------|-----|
| pageSize | 固定 20 |
| pages | 10（全量 200 条） |
| 翻页方式 | 请求参数 `pageNum` 递增；对话中逐页展示并询问是否继续 |

# 行业分类映射表与匹配规则

## 传参规则

- 接口 `category` 参数传**行业中文名**（如 `"萌宠动物"`），全部行业传 `"all"`
- 用户自然语言描述 → 按关键词模糊匹配到行业 → 传对应中文名
- 例：用户问"宠物" → 传 `"萌宠动物"`；未提及行业 → 传 `"all"`
- 无法识别时降级为 `all`（全部行业），并列出完整行业列表供用户选择
- skill 内部（文档、配置、输出）不暴露英文分类值

## 性别映射

| 用户说 | 传参 | 展示 |
|--------|------|------|
| 男、男性、男生、男榜 | `male` | 男 |
| 女、女性、女生、女榜 | `female` | 女 |
| 全部、不分性别、未提及 | `all` | 全部 |

## 行业分类表（32 类）

| 传参值（中文） | 常见用户说法（模糊匹配关键词） |
|--------------|------------------------------|
| `all`（全部行业） | 全部、全品类、综合、所有、不限 |
| `商业创业` | 创业、商业、生意、企业家、经商 |
| `汽车出行` | 汽车、车评、新能源车、摩托车、出行 |
| `设计创意` | 设计、创意、平面设计、工业设计、UI设计 |
| `航空领域` | 航空、飞机、民航、飞行员、机场 |
| `电商零售` | 电商、零售、跨境电商、直播带货、网购 |
| `教育培训` | 教育、培训、考研、考试、课程、老师、学习 |
| `工程运营` | 工程、运营、制造、机械、运维 |
| `娱乐文化` | 娱乐、文化、明星、影视、综艺、追星、八卦 |
| `环境能源` | 环境、能源、气候、环保、电力、光伏 |
| `家庭育儿` | 育儿、母婴、亲子、宝宝、早教、家庭 |
| `时尚美妆` | 美妆、时尚、穿搭、护肤、彩妆、潮流 |
| `金融投资` | 金融、投资、交易、股票、基金、理财、加密货币、财经、炒股 |
| `美食饮品` | 美食、饮品、探店、烹饪、咖啡、烘焙、吃播、菜谱 |
| `游戏电竞` | 游戏、电竞、手游、网游、主机游戏 |
| `健康医疗` | 医疗、医学、医生、疾病、药学、临床 |
| `牧场马术` | 马术、牧场、骑马、马场 |
| `宗教信仰` | 宗教、信仰、灵性、佛教、基督教 |
| `科学学术` | 科学、学术、科研、物理、化学、天文、学者 |
| `体育健身` | 体育、健身、运动、篮球、足球、跑步、瑜伽、球赛 |
| `科技软件` | 科技、软件、编程、互联网、人工智能、AI、程序员、数码、SaaS |
| `旅行户外` | 旅行、旅游、户外、露营、徒步、景点、攻略 |
| `视觉艺术` | 绘画、摄影、插画、雕塑、艺术 |
| `健康生活` | 养生、生活方式、冥想、心理健康 |
| `居家手工` | 居家、手工、手作、DIY、家居、收纳 |
| `人力职场` | 人力、职场、招聘、简历、职业发展、HR |
| `公益平权` | 公益、平权、慈善、社会责任 |
| `法律法务` | 法律、法务、律师、司法、法规 |
| `市场营销` | 营销、销售、广告、品牌、增长黑客 |
| `音乐领域` | 音乐、歌手、乐队、说唱、钢琴、吉他 |
| `萌宠动物` | 宠物、萌宠、猫咪、狗狗、猫、狗、动物、兔子 |
| `政治新闻` | 政治、新闻、时政、政策、记者、媒体 |

## 匹配策略

- **长词优先**：所有关键词按长度降序扫描，先命中长者（如 "健康生活" 优先于 "健康"）
- **全文扫描**：关键词出现在用户文本任意位置即命中
- **显式优先**：`--category` 参数支持直接传行业中文名或匹配关键词，优先级高于文本解析
- **降级兜底**：无任何命中时传 `all`，并在对话中告知用户当前为全部行业榜单

#!/usr/bin/env python3
"""
X (Twitter) 热门账号榜数据抓取脚本
用法：
    python fetch_rank.py --query "给我2026年4月25日的Twitter全品类日榜"
    python fetch_rank.py --query "宠物类女博主榜单" --page 2 --html
    python fetch_rank.py --date 2026-09-20 --gender female --category "萌宠动物"
"""
import argparse
import json
import os
import re
import subprocess
import sys
from datetime import date, datetime, timedelta

import urllib.request
import urllib.error

API_URL = "https://redfox.hk/story/api/x/hotAccount/rankList"

# ─────────────────────────────────────────────────────────
# 加载外部静态配置（assets/）
# ─────────────────────────────────────────────────────────
_ASSETS_DIR = os.path.join(os.path.dirname(os.path.abspath(__file__)), "..", "assets")
_SKILL_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
OUTPUT_DIR = os.path.join(_SKILL_DIR, "output")  # 数据与报告统一落盘 output/，不污染 skill 根目录


def _load_json(filename):
    with open(os.path.join(_ASSETS_DIR, filename), encoding="utf-8") as f:
        return json.load(f)


_cfg = _load_json("category_config.json")
UPDATE_HOUR = _cfg["update_rule"]["update_hour"]
UPDATE_MINUTE = _cfg["update_rule"]["update_minute"]
LOOKBACK_DAYS = _cfg["update_rule"]["lookback_days"]
UPDATE_NOTE = _cfg["update_rule"]["update_note"]
GENDERS = _cfg["genders"]
CATEGORIES = _cfg["categories"]

GENDER_LABELS = {g["value"]: g["zh"] for g in GENDERS}
ALL_CATEGORY_ZH = "全部行业"
CATEGORY_BY_ZH = {c["zh"]: c for c in CATEGORIES}

CN_NUM = {"一": 1, "二": 2, "三": 3, "四": 4, "五": 5,
          "六": 6, "七": 7, "八": 8, "九": 9, "十": 10}


def _get_api_key():
    """从环境变量 REDFOX_API_KEY 获取 API Key。"""
    api_key = os.environ.get("REDFOX_API_KEY", "").strip()
    if api_key:
        return api_key
    print("[ERROR] 未设置 REDFOX_API_KEY 环境变量", file=sys.stderr)
    print("[HINT] 请运行: export REDFOX_API_KEY=<你的apikey>", file=sys.stderr)
    print("[HINT] 访问 https://redfox.hk/settings/api-keys?source=github 获取 API Key", file=sys.stderr)
    sys.exit(4)


def _api_post(payload, timeout=20):
    """使用原生 urllib 发送 HTTPS POST 请求，携带 X-API-KEY 鉴权。"""
    headers = {
        "Content-Type": "application/json",
        "Accept": "application/json",
        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36",
        "X-API-KEY": _get_api_key(),
    }
    data = json.dumps(payload).encode("utf-8")
    req = urllib.request.Request(API_URL, data=data, headers=headers, method="POST")
    try:
        with urllib.request.urlopen(req, timeout=timeout) as resp:
            return json.loads(resp.read().decode("utf-8"))
    except urllib.error.HTTPError as e:
        body = e.read().decode("utf-8", errors="replace")
        print(f"[ERROR] HTTP {e.code}：{body[:200]}", file=sys.stderr)
        sys.exit(2)


# ─────────────────────────────────────────────────────────
# 自然语言解析：性别 / 行业 / 日期 / 页码
# ─────────────────────────────────────────────────────────
def match_gender(text, explicit=None):
    if explicit:
        return explicit
    low = (text or "").lower()
    for g in GENDERS:  # 配置中 female 先于 male，避免"女"被"男"误吞
        if g["value"] == "all":
            continue
        for kw in g["keywords"]:
            if kw.lower() in low:
                return g["value"]
    return "all"


def match_category(text, explicit=None):
    """返回接口 category 传参值："all" 或行业中文名。"""
    if explicit:
        key = explicit.strip()
        if key in ("all", ALL_CATEGORY_ZH):
            return "all"
        if key in CATEGORY_BY_ZH:
            return key
        for c in CATEGORIES:  # 支持传匹配关键词
            if key in c["keywords"]:
                return "all" if c["zh"] == ALL_CATEGORY_ZH else c["zh"]
        print(f"[WARN] 无法识别的行业分类「{explicit}」，已降级为全部行业", file=sys.stderr)
        return "all"
    low = (text or "").lower()
    pairs = []
    for c in CATEGORIES:
        if c["zh"] == ALL_CATEGORY_ZH:
            continue
        for kw in c["keywords"]:
            pairs.append((kw.lower(), c["zh"]))
    pairs.sort(key=lambda p: len(p[0]), reverse=True)  # 长词优先
    for kw, zh in pairs:
        if kw in low:
            return zh
    return "all"


def parse_page(text, explicit=None):
    if explicit:
        return max(1, int(explicit))
    m = re.search(r"第\s*([0-9]+|[一二三四五六七八九十])\s*页", text or "")
    if m:
        v = m.group(1)
        return CN_NUM.get(v, int(v) if v.isdigit() else 1)
    return 1


def parse_date(text, explicit=None):
    if explicit:
        return date.fromisoformat(explicit)
    t = text or ""
    if "前天" in t:
        return date.today() - timedelta(days=2)
    if "昨天" in t or "昨日" in t:
        return date.today() - timedelta(days=1)
    if "今天" in t or "今日" in t:
        return date.today()
    m = re.search(r"(20\d{2})\s*[-年/.]\s*(\d{1,2})\s*[-月/.]\s*(\d{1,2})", t)
    if m:
        return date(int(m.group(1)), int(m.group(2)), int(m.group(3)))
    m = re.search(r"(\d{1,2})\s*[-月/.]\s*(\d{1,2})\s*日?", t)
    if m:
        return date(date.today().year, int(m.group(1)), int(m.group(2)))
    return None


# ─────────────────────────────────────────────────────────
# 日期计算与容错回退
# ─────────────────────────────────────────────────────────
def latest_rank_date():
    """每日早上 9:00 更新前一天榜单：9 点后最新为昨日，9 点前为前日。"""
    now = datetime.now()
    updated = (now.hour, now.minute) >= (UPDATE_HOUR, UPDATE_MINUTE)
    offset = 1 if updated else 2
    return date.today() - timedelta(days=offset)


def fetch_page(rank_date, gender, category, page):
    """请求单页榜单，返回 data dict；无数据或失败返回 None。"""
    payload = {
        "pageNum": page,
        "rankDate": str(rank_date),
        "gender": gender,
        "category": category,
        "source": "X账号榜-GitHub",
    }
    resp = _api_post(payload)
    if resp.get("code") != 2000:
        print(f"[ERROR] 接口返回异常 code={resp.get('code')} msg={resp.get('msg')}", file=sys.stderr)
        return None
    data = resp.get("data") or {}
    if not data.get("list"):
        return None
    return data


def resolve_date(target, gender, category, page, warnings):
    """校验目标日期并容错回退到最近可用日期。"""
    today = date.today()
    if target is None:
        target = latest_rank_date()
    else:
        diff = (today - target).days
        if diff < 1:
            warnings.append("非常抱歉🙏，我们最新的是昨日的数据，将为您提供最接近您需求的昨日热榜。")
            target = latest_rank_date()
        elif diff > LOOKBACK_DAYS:
            warnings.append(f"非常抱歉🙏，目前榜单最多支持回溯「过去{LOOKBACK_DAYS}天」，我将为您查询最接近您需求的时间范围~")
            target = latest_rank_date()

    data = fetch_page(target, gender, category, page)
    if data:
        return target, data

    # 目标日期暂无数据：自最新可用日期起向前扫描回退
    start = latest_rank_date()
    for i in range(LOOKBACK_DAYS):
        cand = start - timedelta(days=i)
        data = fetch_page(cand, gender, category, page)
        if data:
            if cand != target:
                warnings.append(f"「{target}」暂无榜单数据，已自动切换至最近可用日期 {cand}。")
            return cand, data
    return target, None


# ─────────────────────────────────────────────────────────
# 字段归一化
# ─────────────────────────────────────────────────────────
def _strip_emoji(s):
    return re.sub(r"[^\x20-\x7E\u4e00-\u9fff]", "", s or "").strip()


def _safe_json(s, default):
    """解析 JSON 字符串字段；兼容双重编码，类型不符时回退默认值。"""
    cur = s
    for _ in range(3):
        if not isinstance(cur, str):
            break
        try:
            cur = json.loads(cur)
        except (ValueError, TypeError):
            return default
    if isinstance(default, dict) and not isinstance(cur, dict):
        return default
    if isinstance(default, list) and not isinstance(cur, list):
        return default
    return cur


def fmt_followers(n):
    """111800000 -> 1.1亿+；31000000 -> 3,100w+；1800000 -> 180w+"""
    try:
        n = int(n)
    except (ValueError, TypeError):
        return "-"
    if n >= 100000000:
        return f"{n / 100000000:.1f}亿+"
    if n >= 10000:
        return f"{n // 10000:,}w+"
    return str(n)


def parse_count(s):
    """解析接口粉丝数字符串：4M -> 4000000；290K -> 290000；4.4M -> 4400000"""
    if not s or s == "-":
        return None
    m = re.match(r"^([\d.]+)\s*([KMB]?)$", str(s).strip(), re.IGNORECASE)
    if not m:
        return None
    mult = {"": 1, "K": 1000, "M": 1000000, "B": 1000000000}[m.group(2).upper()]
    return int(float(m.group(1)) * mult)


_NETWORK_ORDER = ["youtube", "tiktok", "instagram", "twitch", "linkedin", "facebook"]
_NETWORK_LABELS = {"youtube": "YouTube", "tiktok": "TikTok", "instagram": "Instagram",
                   "twitch": "Twitch", "linkedin": "LinkedIn", "facebook": "Facebook"}


def normalize_item(raw):
    cats = _safe_json(raw.get("categoriesJson"), [])
    industries, seen = [], set()
    for c in cats:
        if c.get("primary"):
            name = _strip_emoji(c.get("root_name"))
            if name and name not in seen:
                seen.add(name)
                industries.append(name)
    if not industries and cats:
        industries = [_strip_emoji(cats[0].get("root_name"))]

    networks = _safe_json(raw.get("networksJson"), {}) or {}
    others = []
    for key in _NETWORK_ORDER + [k for k in networks if k not in _NETWORK_ORDER]:
        v = networks.get(key)
        if not v or key == "twitter":
            continue
        fc = v.get("follower_count") or "-"
        cnt = parse_count(fc)
        others.append({
            "network": _NETWORK_LABELS.get(key, key.capitalize()),
            "followers": fc,
            "followersFmt": fmt_followers(cnt) if cnt is not None else fc,
            "profileUrl": v.get("profile_url") or "",
        })

    badges = _safe_json(raw.get("badgesJson"), {}) or {}
    growth = (raw.get("growthPercentage") or "").replace(" ", "")

    return {
        "rank": raw.get("rankNo"),
        "fullName": raw.get("fullName") or "-",
        "profileUrl": raw.get("profileUrl") or "",
        "pictureUrl": raw.get("pictureUrl") or "",
        "title": raw.get("title") or "",
        "biography": raw.get("biography") or "",
        "country": raw.get("countryName") or "-",
        "countryCode": (raw.get("countryCode") or "").lower(),
        "gender": raw.get("gender") or "",
        "industries": industries,
        "score": raw.get("xScore"),
        "growthPercentage": growth,
        "growthAbsolute": raw.get("growthAbsolute") or "",
        "xFollowers": raw.get("xFollowerCount"),
        "xFollowersFmt": fmt_followers(raw.get("xFollowerCount")),
        "otherNetworks": others,
        "topics": badges.get("hasTopics") or "",
        "cause": badges.get("hasCause") or "",
    }


def main():
    parser = argparse.ArgumentParser(description="X (Twitter) 热门账号榜查询")
    parser.add_argument("--query", default="", help="用户原始问题（自动解析日期/性别/行业/页码）")
    parser.add_argument("--date", default=None, help="榜单日期 YYYY-MM-DD")
    parser.add_argument("--gender", default=None, choices=["all", "male", "female"])
    parser.add_argument("--category", default=None, help="行业分类（中文名，如 萌宠动物；默认全部行业）")
    parser.add_argument("--page", type=int, default=None, help="页码（每页 20 条）")
    parser.add_argument("--output", default=None, help="JSON 输出路径")
    parser.add_argument("--html", action="store_true", help="同时生成 HTML 报告")
    args = parser.parse_args()

    gender = match_gender(args.query, args.gender)
    category = match_category(args.query, args.category)
    page = parse_page(args.query, args.page)
    target = parse_date(args.query, args.date)

    warnings = []
    rank_date, data = resolve_date(target, gender, category, page, warnings)
    if not data:
        print("[ERROR] 该筛选条件下暂无榜单数据，请更换日期/性别/行业后重试", file=sys.stderr)
        sys.exit(3)

    items = [normalize_item(x) for x in data.get("list", [])]
    meta = {
        "rankDate": str(rank_date),
        "gender": gender,
        "genderLabel": GENDER_LABELS.get(gender, "全部"),
        "category": category,
        "categoryLabel": ALL_CATEGORY_ZH if category == "all" else category,
        "pageNum": data.get("pageNum", page),
        "pageSize": data.get("pageSize", 20),
        "pages": data.get("pages", 1),
        "total": data.get("total", len(items)),
        "updateNote": UPDATE_NOTE,
        "warnings": warnings,
        "generatedAt": datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
    }
    result = {"meta": meta, "items": items}

    out_path = args.output or os.path.join(OUTPUT_DIR, "x_top_account_data.json")
    out_dir = os.path.dirname(out_path)
    if out_dir:
        os.makedirs(out_dir, exist_ok=True)
    with open(out_path, "w", encoding="utf-8") as f:
        json.dump(result, f, ensure_ascii=False, indent=2)

    for w in warnings:
        print(f"[WARN] {w}")
    print(f"[INFO] 数据日期：{meta['rankDate']}｜性别：{meta['genderLabel']}｜行业：{meta['categoryLabel']}"
          f"｜第 {meta['pageNum']}/{meta['pages']} 页｜共 {meta['total']} 条")
    print(f"[INFO] 数据文件：{out_path}")

    if args.html:
        script = os.path.join(os.path.dirname(os.path.abspath(__file__)), "generate_report.py")
        subprocess.run([sys.executable, script, "--data", out_path], check=False)


if __name__ == "__main__":
    main()

#!/usr/bin/env python3
"""
X (Twitter) 热门账号榜 HTML 报告生成脚本
用法：
    python generate_report.py --data /path/to/x_top_account_data.json
    python generate_report.py --data data.json --output /path/to/report.html
X 平台风格（纯白底 + X 蓝 #1d9bf0），支持一键导出 PDF / 高清图片。
"""
import argparse
import html
import json
import os
import subprocess
import sys
from datetime import datetime

# X 平台视觉规范
X_BLUE = "#1d9bf0"
X_TEXT = "#0f1419"
X_SUB = "#536471"
X_BORDER = "#eff3f4"
X_HOVER = "#f7f9f9"
X_GREEN = "#00ba7c"
X_RED = "#f4212e"

# HTML 报告默认输出目录：skill 目录下的 output/（运行时自动创建）
_SKILL_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
OUTPUT_DIR = os.path.join(_SKILL_DIR, "output")

MEDALS = {1: ("🥇", "#ffd700"), 2: ("🥈", "#c0c0c0"), 3: ("🥉", "#cd7f32")}


def esc(s):
    return html.escape(str(s if s is not None else ""), quote=True)


def flag_emoji(cc):
    cc = (cc or "").upper()
    if len(cc) != 2 or not cc.isalpha():
        return ""
    return "".join(chr(0x1F1E6 + ord(ch) - 65) for ch in cc)


def score_fmt(score):
    try:
        return f"{round(float(score))}/100"
    except (TypeError, ValueError):
        return "-"


def growth_html(g):
    g = (g or "").strip()
    if not g:
        return "<span class='dim'>—</span>"
    cls = "up" if g.startswith("+") else ("down" if g.startswith("-") else "dim")
    return f"<span class='{cls}'>{esc(g)}</span>"


CSS = f"""
* {{ box-sizing: border-box; margin: 0; padding: 0; }}
body {{
  background: #ffffff; color: {X_TEXT};
  font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto,
    "Helvetica Neue", "PingFang SC", "Microsoft YaHei", sans-serif;
  padding: 32px 16px;
}}
.report {{ max-width: 1440px; margin: 0 auto; }}
.header {{ display: flex; align-items: center; gap: 14px; margin-bottom: 6px; }}
.logo {{
  width: 44px; height: 44px; border-radius: 50%; background: {X_TEXT};
  color: #fff; font-size: 26px; font-weight: 800;
  display: flex; align-items: center; justify-content: center; flex: none;
}}
.header h1 {{ font-size: 24px; font-weight: 800; letter-spacing: .2px; }}
.header .sub {{ color: {X_SUB}; font-size: 13px; margin-top: 4px; }}
.export-bar {{ text-align: right; margin: 10px 0 14px; }}
.btn {{
  border: 1px solid {X_BORDER}; background: #fff; color: {X_BLUE};
  border-radius: 999px; padding: 8px 18px; font-size: 13px; font-weight: 700;
  cursor: pointer; margin-left: 8px;
}}
.btn-primary {{ background: {X_BLUE}; border-color: {X_BLUE}; color: #fff; }}
.btn:hover {{ opacity: .85; }}
.stats {{ display: flex; gap: 12px; flex-wrap: nowrap; margin-bottom: 16px; }}
.stat {{
  flex: 1; min-width: 150px; background: #fff; border: 1px solid {X_BORDER};
  border-radius: 16px; padding: 14px 18px;
}}
.stat .v {{ font-size: 22px; font-weight: 800; color: {X_BLUE}; }}
.stat .k {{ font-size: 12px; color: {X_SUB}; margin-top: 4px; }}
.notice {{
  background: {X_HOVER}; border: 1px solid {X_BORDER}; border-radius: 16px;
  padding: 14px 18px; font-size: 13px; color: {X_SUB}; line-height: 1.8;
  margin-bottom: 16px;
}}
.notice b {{ color: {X_TEXT}; }}
.table-wrap {{
  border: 1px solid {X_BORDER}; border-radius: 16px; overflow: hidden;
  background: #fff;
}}
table {{ width: 100%; border-collapse: collapse; font-size: 13px; table-layout: fixed; }}
thead th:nth-child(1) {{ width: 56px; }}
thead th:nth-child(2) {{ width: 186px; }}
thead th:nth-child(3) {{ width: 194px; }}
thead th:nth-child(4) {{ width: 84px; }}
thead th:nth-child(5) {{ width: 128px; }}
thead th:nth-child(6) {{ width: 84px; }}
thead th:nth-child(7) {{ width: 76px; }}
thead th:nth-child(8) {{ width: 84px; }}
thead th:nth-child(9) {{ width: 150px; }}
thead th:nth-child(10) {{ width: 150px; }}
thead th {{
  background: {X_HOVER}; color: {X_SUB}; font-size: 12px; font-weight: 700;
  text-align: left; padding: 12px 10px; border-bottom: 1px solid {X_BORDER};
}}
tbody td {{ padding: 12px 10px; border-bottom: 1px solid {X_BORDER}; vertical-align: top; }}
tbody tr:last-child td {{ border-bottom: none; }}
tbody tr:hover {{ background: {X_HOVER}; }}
.rank {{ font-weight: 800; font-size: 15px; white-space: nowrap; }}
.acct {{ display: flex; align-items: center; gap: 8px; min-width: 0; max-width: 100%; }}
.acct img {{
  width: 28px; height: 28px; border-radius: 50%; object-fit: cover; flex: none;
  background: {X_BORDER};
}}
.acct a {{
  color: {X_BLUE}; text-decoration: none; font-weight: 700;
  white-space: nowrap; overflow: hidden; text-overflow: ellipsis; display: block;
}}
.acct a:hover {{ text-decoration: underline; }}
.bio {{ max-width: 100%; color: {X_SUB}; font-size: 12px; line-height: 1.6; }}
.acct-body {{ min-width: 0; max-width: 100%; }}
.acct .t {{
  color: {X_SUB}; font-size: 12px; margin-top: 2px;
  white-space: nowrap; overflow: hidden; text-overflow: ellipsis; max-width: 100%;
}}
.bio .d {{
  display: -webkit-box; -webkit-line-clamp: 2; -webkit-box-orient: vertical;
  overflow: hidden;
}}
.ind {{ color: {X_SUB}; font-size: 12px; line-height: 1.6; word-break: break-word; }}
.score {{ font-weight: 800; white-space: nowrap; }}
.up {{ color: {X_GREEN}; font-weight: 700; white-space: nowrap; }}
.down {{ color: {X_RED}; font-weight: 700; white-space: nowrap; }}
.dim {{ color: {X_SUB}; }}
.fans {{ font-weight: 700; white-space: nowrap; }}
.nets {{ margin: 0; padding-left: 16px; color: {X_SUB}; font-size: 12px; line-height: 1.7; word-break: break-word; }}
.nets a {{ color: {X_SUB}; text-decoration: none; }}
.nets a:hover {{ color: {X_BLUE}; }}
.tags {{ color: {X_SUB}; font-size: 12px; max-width: 100%; word-break: break-word; }}
.footer {{
  margin-top: 18px; color: {X_SUB}; font-size: 12px; text-align: center; line-height: 1.8;
}}
@media (max-width: 768px) {{
  .stats {{ flex-wrap: wrap; }}
}}
@media print {{
  .export-bar {{ display: none; }}
  body {{ padding: 0; }}
  .table-wrap {{ box-shadow: none; }}
}}
"""


def build_rows(items):
    rows = []
    for it in items:
        rank = it.get("rank")
        medal = MEDALS.get(rank)
        rank_cell = f"<span>{medal[0]} {rank}</span>" if medal else f"<span class='dim'>{rank}</span>"

        name = esc(it.get("fullName"))
        url = esc(it.get("profileUrl"))
        pic = esc(it.get("pictureUrl"))
        avatar = f"<img src='{pic}' alt='' onerror=\"this.style.visibility='hidden'\">" if pic else ""
        title = esc(it.get("title"))
        acct = (f"<div class='acct'>{avatar}<div class='acct-body'>"
                f"<a href='{url}' target='_blank' rel='noopener' title='{name}'>{name}</a>"
                f"<div class='t' title='{title}'>{title or '—'}</div></div></div>")

        bio_full = esc(it.get("biography"))
        bio = f"<div class='bio'><div class='d' title='{bio_full}'>{bio_full}</div></div>"

        flag = flag_emoji(it.get("countryCode"))
        country = f"{flag} {esc(it.get('country'))}".strip()

        inds = it.get("industries") or []
        ind = esc(" / ".join(inds[:2])) if inds else "<span class='dim'>—</span>"

        nets = it.get("otherNetworks") or []
        if nets:
            lis = "".join(
                f"<li><a href='{esc(n.get('profileUrl'))}' target='_blank' rel='noopener'>"
                f"{esc(n.get('network'))} {esc(n.get('followersFmt') or n.get('followers'))}</a></li>"
                for n in nets)
            nets_cell = f"<ul class='nets'>{lis}</ul>"
        else:
            nets_cell = "<span class='dim'>—</span>"

        topics = it.get("topics") or it.get("cause") or ""
        tags = esc(topics) if topics else "<span class='dim'>—</span>"

        rows.append(
            "<tr>"
            f"<td class='rank'>{rank_cell}</td>"
            f"<td>{acct}</td>"
            f"<td>{bio}</td>"
            f"<td>{country}</td>"
            f"<td class='ind'>{ind}</td>"
            f"<td class='score'>{score_fmt(it.get('score'))}</td>"
            f"<td>{growth_html(it.get('growthPercentage'))}</td>"
            f"<td class='fans'>{esc(it.get('xFollowersFmt'))}</td>"
            f"<td>{nets_cell}</td>"
            f"<td class='tags'>{tags}</td>"
            "</tr>")
    return "\n".join(rows)


def build_html(payload):
    meta = payload.get("meta", {})
    items = payload.get("items", [])

    scores = [float(i["score"]) for i in items if i.get("score") is not None]
    avg_score = f"{sum(scores) / len(scores):.1f}" if scores else "-"
    growths = []
    for i in items:
        g = (i.get("growthPercentage") or "").replace("%", "").replace("+", "")
        try:
            growths.append((float(g), i.get("fullName")))
        except ValueError:
            pass
    top_growth = f"+{max(growths)[0]}% · {max(growths)[1]}" if growths else "-"

    rows = build_rows(items)
    now = datetime.now().strftime("%Y-%m-%d %H:%M")
    img_name = esc(report_filename(meta)[:-5] + ".png")

    # 标题：X(Twitter)·{行业}·热门{性别}账号榜；行业/性别为 all 时省略对应字段
    gender_zh = {"male": "男", "female": "女"}.get(meta.get("gender", "all"), "")
    cat_part = f"·{meta.get('categoryLabel')}" if meta.get("category") != "all" else ""
    rank_part = f"热门{gender_zh}性账号榜" if gender_zh else "热门账号榜"
    h1 = f"X(Twitter){cat_part}·{rank_part}"

    return f"""<!DOCTYPE html>
<html lang="zh-CN">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>{esc(h1)} · {esc(meta.get('rankDate'))}</title>
<style>{CSS}</style>
</head>
<body>
<div class="report" id="report">
  <div class="header">
    <div class="logo">𝕏</div>
    <div>
      <h1>{esc(h1)}</h1>
      <div class="sub">数据日期 {esc(meta.get('rankDate'))} · 第 {esc(meta.get('pageNum'))}/{esc(meta.get('pages'))} 页 · 每页 {esc(meta.get('pageSize'))} 条 · 影响力评分：结合粉丝数、互动量、曝光数等加权综合评估</div>
    </div>
  </div>

  <div class="export-bar">
    <button class="btn btn-primary" onclick="window.print()">🖨️ 导出 PDF</button>
    <button class="btn" id="downloadImgBtn" onclick="downloadAsImage()">📷 导出图片</button>
  </div>

  <div class="stats">
    <div class="stat"><div class="v">{esc(meta.get('total'))}</div><div class="k">上榜账号（总）</div></div>
    <div class="stat"><div class="v">{esc(meta.get('pageNum'))}/{esc(meta.get('pages'))}</div><div class="k">当前页 / 总页数</div></div>
    <div class="stat"><div class="v">{avg_score}</div><div class="k">本页平均影响力评分</div></div>
    <div class="stat"><div class="v" style="font-size:15px;line-height:30px;">{esc(top_growth)}</div><div class="k">本页最高日涨幅</div></div>
  </div>

  <div class="notice">
    <div>💡 <b>榜单说明</b>：{esc(meta.get('updateNote'))}，与实时数据存在差异。</div>
    <div>📐 <b>影响力评分</b>：结合粉丝数、互动量、曝光数等加权综合评估，满分 100。</div>
  </div>

  <div class="table-wrap">
    <table>
      <thead>
        <tr>
          <th>排名</th><th>账号名</th><th>账号简介</th><th>国家/地区</th><th>主行业</th>
          <th>影响力评分</th><th>粉丝数日涨幅</th><th>X粉丝数</th><th>其他平台粉丝数</th><th>专题标签</th>
        </tr>
      </thead>
      <tbody>
{rows}
      </tbody>
    </table>
  </div>

  <div class="footer">
    数据来源：红狐数据 · 生成时间 {now}<br>
    本报告支持浏览器一键导出 PDF / 高清图片
  </div>
</div>

<script src="https://cdn.jsdelivr.net/npm/html2canvas@1.4.1/dist/html2canvas.min.js"></script>
<script>
async function downloadAsImage() {{
  const btn = document.getElementById('downloadImgBtn');
  btn.disabled = true; btn.textContent = '⏳ 生成中...';
  try {{
    const canvas = await html2canvas(document.getElementById('report'), {{
      scale: 2, backgroundColor: '#ffffff', useCORS: true
    }});
    const a = document.createElement('a');
    a.download = '{img_name}';
    a.href = canvas.toDataURL('image/png');
    a.click();
  }} catch (e) {{
    alert('图片导出失败：' + e.message + '\\n可改用「导出 PDF」按钮。');
  }} finally {{
    btn.disabled = false; btn.textContent = '📷 导出图片';
  }}
}}
</script>
</body>
</html>"""


def open_in_browser(path):
    try:
        abs_path = os.path.abspath(path)
        if sys.platform == "darwin":
            subprocess.run(["open", abs_path], check=True)
        elif sys.platform.startswith("win"):
            os.startfile(abs_path)  # noqa
        else:
            subprocess.run(["xdg-open", abs_path], check=True)
        print(f"\n✓ HTML 报告已自动打开: {abs_path}", file=sys.stderr)
    except Exception:
        print(f"\n✓ HTML 报告已生成: {os.path.abspath(path)}", file=sys.stderr)


def report_filename(meta):
    """报告文件名：X账号榜_性别_分类_日期.html；all 展示为全部性别/全部分类。"""
    gender = meta.get("gender", "all")
    gender_part = "全部性别" if gender == "all" else (meta.get("genderLabel") or gender)
    category = meta.get("category", "all")
    category_part = "全部分类" if category == "all" else (meta.get("categoryLabel") or category)
    return f"X账号榜_{gender_part}_{category_part}_{meta.get('rankDate', 'latest')}.html"


def main():
    parser = argparse.ArgumentParser(description="生成 X 热门账号榜 HTML 报告")
    parser.add_argument("--data", required=True, help="fetch_rank.py 输出的 JSON 路径")
    parser.add_argument("--output", default=None, help="HTML 输出路径")
    parser.add_argument("--no-open", action="store_true", help="生成后不自动打开浏览器")
    args = parser.parse_args()

    with open(args.data, encoding="utf-8") as f:
        payload = json.load(f)
    meta = payload.get("meta", {})

    out = args.output or os.path.join(OUTPUT_DIR, report_filename(meta))
    os.makedirs(os.path.dirname(os.path.abspath(out)), exist_ok=True)

    with open(out, "w", encoding="utf-8") as f:
        f.write(build_html(payload))
    print(f"[INFO] HTML 报告已生成：{out}")

    if not args.no_open:
        open_in_browser(out)


if __name__ == "__main__":
    main()