import os
import json
import time
import requests
import pandas as pd
from lxml import etree
from datetime import datetime

# --- 基础配置 ---
DATA_ROOT = "data"
FRPS_DIR = os.path.join(DATA_ROOT, "frps_106")
WEATHER_DIR = os.path.join(DATA_ROOT, "weather_82")
EXCEL_PATH = os.path.join(DATA_ROOT, "error_report.xlsx")

for d in [FRPS_DIR, WEATHER_DIR]:
    os.makedirs(d, exist_ok=True)

CONFIG = {
    "106": {
        "base_url": "http://106.75.72.40:7500/api/proxy/tcp",
        "primary_auth": "Basic YWRtaW46bGljYWhr",
        "login_payload": {"username": "admin", "password": "licahk", "recaptcha": ""}
    },
    "82": {
        "base_url": "http://82.156.1.111/weather/php",
        "login": {'username': 'renlixin', 'password': 'licahk', 'login': '123'}
    }
}

error_logs = []


# --- 通用工具函数 ---

def add_error(source, name, reason, latest_time="N/A"):
    """记录错误并计算日期差"""
    days_diff = "N/A"
    if latest_time and latest_time != "N/A":
        try:
            clean_date_str = str(latest_time).split()[0].replace('_', '-')
            target_date = datetime.strptime(clean_date_str, "%Y-%m-%d").date()
            today_date = datetime.now().date()
            diff = (today_date - target_date).days
            days_diff = f"滞后 {diff} 天" if diff > 0 else "当天已同步"
        except:
            days_diff = "解析失败"

    error_logs.append({
        "数据来源": source,
        "站点/代理名称": name,
        "错误原因": reason,
        "日期偏移量": days_diff,
        "最新数据时间": latest_time,
        "检查时间": datetime.now().strftime("%Y-%m-%d %H:%M:%S")
    })


def find_closest_item(items, is_date_level=True):
    """寻找日期最接近今天的那一项"""
    if not items or not isinstance(items, list): return None
    today = datetime.now()
    scored_items = []
    for item in items:
        if not isinstance(item, dict): continue
        name_val = item.get('name', '')
        path_val = item.get('path', '')
        target_str = name_val if name_val else path_val.split('/')[-1]
        try:
            if is_date_level:
                current_date = datetime.strptime(target_str, "%Y_%m_%d")
            else:
                mod_str = item.get('modified', '')
                if mod_str:
                    current_date = datetime.fromisoformat(mod_str.replace('Z', '+00:00'))
                else:
                    continue
            diff = abs((today - current_date.replace(tzinfo=None)).total_seconds())
            scored_items.append((diff, item, target_str))
        except:
            continue
    if not scored_items: return None
    scored_items.sort(key=lambda x: x[0])
    return scored_items[0]


def process_text_content(raw_content):
    if not raw_content: return ""
    lines = str(raw_content).split('\n')
    result, current = [], ""
    for line in lines:
        if "    " in line:
            current += line.strip()
        else:
            if current: result.append(current)
            current = line.strip()
    if current: result.append(current)
    return "\n".join(result)


def save_json(folder, name, data):
    path = os.path.join(folder, f"{name}.json")
    with open(path, 'w', encoding='utf-8') as f:
        json.dump(data, f, indent=4, ensure_ascii=False)


# --- 106 网站逻辑 ---

def get_106_dynamic_token(port):
    """为106特定端口站点执行登录获取 Token"""
    url = f"http://106.75.72.40:{port}/api/login"
    try:
        resp = requests.post(url, json=CONFIG["106"]["login_payload"], timeout=10)
        if resp.status_code == 200:
            return resp.text.strip().replace('"', '')
    except:
        pass
    return None


def run_106_logic():
    print("\n>>> 开始处理 106 网站 (FRPS - 修正逻辑)...")
    c = CONFIG["106"]
    today_str = datetime.now().strftime("%Y_%m_%d")

    try:
        main_headers = {"Authorization": c["primary_auth"], "User-Agent": "Mozilla/5.0"}
        resp = requests.get(c["base_url"], headers=main_headers, timeout=15)
        if resp.status_code != 200:
            add_error("106网站", "主入口API", f"访问失败: HTTP {resp.status_code}")
            return

        proxies = resp.json().get('proxies', [])
        for item in proxies:
            if not isinstance(item, dict): continue
            name = item.get('name', 'Unknown')
            if not name.lower().endswith('_data'): continue

            # 1. 状态预检 - 离线拦截逻辑
            status_raw = item.get('status', '')
            status = str(status_raw).lower().strip() if status_raw else "unknown"

            # 如果离线，直接保存并记录错误，不再进行后续 API 访问
            if status != 'online':
                add_error("106网站", name, f"设备离线 (当前状态: {status})")
                save_json(FRPS_DIR, name, item)
                continue

            # 2. 识别站点类型并进行在线处理
            name_up = name.upper()
            is_tower_underscore = "TOWER_" in name_up
            is_tower_i = "TOWER" in name_up and not is_tower_underscore
            if not (is_tower_underscore or is_tower_i): continue

            try:
                conf = item.get('conf') or {}
                port = conf.get('remote_port')
                if not port:
                    add_error("106网站", name, "配置错误: 缺少 remote_port")
                    continue

                # 只有 Online 才获取子站点 Token
                token = get_106_dynamic_token(port)
                if not token:
                    add_error("106网站", name, "Token获取失败(登录异常)")
                    continue

                headers = {"Authorization": c["primary_auth"], "x-auth": token, "User-Agent": "Mozilla/5.0"}

                # 查找 Data 根目录 (根据类型区分大小写)
                api_root = "/api/resources/Data/" if is_tower_underscore else "/api/resources/data/"
                res2 = requests.get(f"http://106.75.72.40:{port}{api_root}", headers=headers, timeout=10)
                if res2.status_code != 200:
                    add_error("106网站", name, f"无法打开Data目录 (HTTP {res2.status_code})")
                    continue

                it2 = res2.json().get('items', [])
                best_date = find_closest_item(it2, is_date_level=True)
                if not best_date or best_date[2] != today_str:
                    add_error("106网站", name, "未找到今日文件夹", best_date[2] if best_date else "N/A")
                    if not best_date: continue

                date_path = f"{api_root}{best_date[2]}/"

                # 查找文件夹内最新文件
                res3 = requests.get(f"http://106.75.72.40:{port}{date_path}", headers=headers, timeout=10)
                it3 = res3.json().get('items', [])
                best_file = find_closest_item(it3, is_date_level=False)
                if not best_file:
                    add_error("106网站", name, "文件夹内无文件", best_date[2])
                    continue

                file_item = best_file[1]
                full_path = file_item.get('path') or f"{date_path}{file_item.get('name')}"

                # 3. 根据类型下载内容
                if is_tower_i:
                    # TowerI 模式：使用 raw 接口获取二进制
                    raw_url = f"http://106.75.72.40:{port}/api/raw{full_path}"
                    res4 = requests.get(raw_url, headers=headers, timeout=20)
                    if res4.status_code == 200:
                        save_path = os.path.join(FRPS_DIR, f"{name}_{today_str}.bin")
                        with open(save_path, 'wb') as f:
                            f.write(res4.content)
                        print(f"   ✅ {name} 二进制数据保存成功")
                else:
                    # Tower_ 模式：使用 resources 接口获取 JSON
                    file_api_url = f"http://106.75.72.40:{port}/api/resources{full_path}"
                    res4 = requests.get(file_api_url, headers=headers, timeout=20)
                    file_json = res4.json()
                    raw_content = file_json.get('content', '') if file_json else None
                    if raw_content:
                        save_path = os.path.join(FRPS_DIR, f"{name}_{today_str}.json")
                        with open(save_path, 'w', encoding='utf-8') as f:
                            f.write(process_text_content(raw_content))
                        print(f"   ✅ {name} JSON数据保存成功")
                    else:
                        add_error("106网站", name, "文件内容为空", best_date[2])

            except Exception as e:
                add_error("106网站", name, f"站点处理崩溃: {str(e)}")

    except Exception as e:
        add_error("106网站", "全局逻辑", f"主进程崩溃: {str(e)}")


# --- 82 网站逻辑 (保持原样) ---

def run_82_logic():
    print("\n>>> 开始处理 82 网站 (Weather)...")
    c = CONFIG["82"]
    session = requests.Session()
    today_fmt = datetime.now().strftime("%Y-%m-%d")

    try:
        session.post(f"{c['base_url']}/login.php", data=c["login"], timeout=10)
        resp = session.post(f"{c['base_url']}/GetStationList.php", timeout=10)
        if resp.status_code != 200:
            add_error("82网站", "登录模块", f"无法获取列表: HTTP {resp.status_code}")
            return

        stations = etree.HTML(resp.content).xpath('//option/@value')
        stations = [s for s in stations if s and str(s).strip()]

        for sid in stations:
            try:
                r = session.post(f"{c['base_url']}/getLastWeatherData.php", data=str(sid),
                                 headers={'Content-Type': 'text/plain'}, timeout=10)
                if r.status_code != 200:
                    add_error("82网站", sid, f"请求失败: HTTP {r.status_code}")
                    continue

                data = r.json()
                if data is None:
                    add_error("82网站", sid, "返回 Null 数据")
                    continue

                d_list = data.get('date', [])
                if not d_list:
                    add_error("82网站", sid, "返回结果中无日期列表")
                else:
                    latest = str(d_list[-1])
                    if latest.split()[0] != today_fmt:
                        add_error("82网站", sid, "数据非当天更新", latest)

                save_json(WEATHER_DIR, sid, data)
                time.sleep(0.1)
            except Exception as e:
                add_error("82网站", sid, f"数据解析异常: {str(e)}")
    except Exception as e:
        add_error("82网站", "初始化模块", str(e))


# --- 汇总导出 ---

def export_to_excel():
    if not error_logs:
        print("\n[✔] 未发现错误记录。")
        return

    df = pd.DataFrame(error_logs)
    cols = ["数据来源", "站点/代理名称", "错误原因", "日期偏移量", "最新数据时间", "检查时间"]
    df = df[[c for c in cols if c in df.columns]]
    df.to_excel(EXCEL_PATH, index=False)
    print(f"\n[!] 错误报表已生成至: {EXCEL_PATH} (共 {len(error_logs)} 条)")


if __name__ == "__main__":
    run_106_logic()
    run_82_logic()
    export_to_excel()
    print("\n任务全部完成。")