import requests
from xlog03 import *
from rds_model.db_redis import DbRedis
import random
import json
import threading
from log.print_log import PrintLog

redisModel = DbRedis.connect()
redisDyModel = DbRedis.douyin_connect()

def get_user_info(user_id, proxies):
    params = {
        'user_id': user_id,
        'version_code': '9.8.1',
        'js_sdk_version': '1.47.2.2',
        'app_name': 'aweme',
        'app_version': '9.8.1',
        'channel': 'App%20Store',
        'mcc_mnc': 46002,
        'aid': 1128,
        'screen_width': 640,
        'os_api': 18,
        'ac': 'WIFI',
        'os_version': '13.3.1',
        'device_platform': 'iphone',
        'build_number': 98107,
        'device_type': 'iPhone8,4',
        'address_book_access': 1
    }
    url = 'https://api3-normal-c-lf.amemv.com/aweme/v1/user/profile/self/?'

    douyin_url = parse_params(url, params)
    response = http_get(douyin_url, proxies)

    return response

def parse_params(url, params):
    if params is None:
        params = {}

    if not url.endswith('?'):
        url = url + '?'

    common_params = parse.urlencode(params)
    douyin_url = url + common_params

    return douyin_url

def http_get(douyin_url, proxies):
    if proxies:
        resp = requests.get(douyin_url, proxies=proxies, verify=False, timeout=10)
    else:
        resp = requests.get(douyin_url, verify=False, timeout=10)

    return resp

def get_scraper_proxy(proxy_key):
    proxy_dict = redisModel.hgetall(proxy_key)
    if (proxy_dict is None) or (len(proxy_dict) == 0):
        return
    proxy_list = list(proxy_dict)

    now = int(time.time())
    while True:
        proxy = random.choice(proxy_list)
        if proxy is None:
            return
        proxy_info = proxy_dict.get(proxy)
        if proxy_info is None:
            continue
        proxy_info = json.loads(proxy_info)
        expire_at = int(proxy_info.get('expired_at'))
        # 删除过期的代理
        if expire_at <= now:
            redisModel.hdel(proxy_key, proxy)
            proxy_list.remove(proxy)
            continue

        return proxy

def scraper():
    pKey = 'IpProxyHash'
    uKey = 'SL:List:Douyin:BarrageUserScrapeQueue'
    sKey = 'SL:List:Douyin:BarrageUserDataQueue'
    hKey = 'SL:Hash:Douyin:BarrageUserScrapeRecord'

    cur_time = int(time.time())
    while True:
        now_time = int(time.time())
        if (now_time - cur_time) > 270:
            print('thrend_' + threading.current_thread().name + ' finish')
            break

        ullen = redisDyModel.llen(uKey)
        if ullen == 0:
            time.sleep(2)
            continue

        str_time = time.strftime("%H:%M:%S", time.localtime())
        user_id = redisDyModel.rpop(uKey)

        try:
            proxy = get_scraper_proxy(pKey)
            if proxy:
                proxies = {
                    "http": "http://" + proxy,
                    "https": "http://" + proxy
                }
            else:
                time.sleep(2)
                continue

            print(str_time + ' 爬取代理:' + str(proxies))
            PrintLog.print(str_time + ' 爬取代理:' + str(proxies))

            response = get_user_info(user_id, proxies)
            if len(response.text) > 0 and response.json()['status_code'] == 0 and response.json()['user']:
                print(str_time + ' user_id:' + str(user_id) + ' 爬取成功')
                PrintLog.print(str_time + ' user_id:' + str(user_id) + ' 爬取成功')
                redisDyModel.lpush(sKey, response.text)
                redisDyModel.hdel(hKey, user_id)
            else:
                print(str_time + ' user_id:' + str(user_id) + ' 爬取失败')
                PrintLog.print(str_time + ' user_id:' + str(user_id) + ' 爬取失败')

                record_json = redisDyModel.hget(hKey, user_id)
                if record_json:
                    record_dict = json.loads(record_json)
                    if record_dict['times'] < 10:
                        record_dict['times'] += 1
                        redisDyModel.hset(hKey, user_id, json.dumps(record_dict))
                    else:
                        redisDyModel.hdel(hKey, user_id)
                else:
                    record_dict = {'times': 1}
                    redisDyModel.hset(hKey, user_id, json.dumps(record_dict))

            time.sleep(1)
        except Exception as e:
            print(str_time + ' 错误:' + str(e))
            PrintLog.print(str_time + ' 错误:' + str(e))
            redisDyModel.lpush(uKey, user_id)
            continue

if __name__ == '__main__':
    import warnings
    warnings.filterwarnings("ignore")

    print('===========爬取程序===========')
    PrintLog.print('===========爬取程序===========')

    threading_count = 50
    for i in range(0, threading_count):
        task = threading.Thread(target=scraper, name=i)
        task.start()  # 准备就绪,等待cpu执行