网易大宗交易爬取（unicode

def dzjy(code=None,Page = 0,end_date = None,data =pd.DataFrame()):
    '''
    # 获取大宗交易数据
    :param code:查询特定股票大宗交易数据，若无，则查询全部股票
    :param Page:不需要设置
    :param end_date:最早交易日期，若为空则返回第一页的数据
    :param data:不需要设置
    :return:DataFrame
            NO：序号
            PUBLISHDATE：交易日期
            EXCHANGE：不知道
            SYMBOL：股票代码
            SNAME：股票名称
            STYPE：股票类型
            DZJY2：成交数量（万股）
            DZJY5：成交价格
            DZJY6：成交金额（万元）
            DZJY9：买方名称
            DZJY11：卖方名称
            TCLOSE：收盘价格
            DZJY55：折溢价率
            CODE：股票代码
            NAME：不知道
    '''
    if code is None:url = 'http://quotes.money.163.com/hs/marketdata/service/dzjy.php?host=/hs/marketdata/service/dzjy.php&page={}&fields=NO,SYMBOL,SNAME,PUBLISHDATE,DZJY2,DZJY5,TCLOSE,DZJY55,DZJY6,DZJY9,DZJY11&sort=PUBLISHDATE&order=desc&count=1000&type=query&initData=[object Object]'.format(Page)
    else:url = 'http://quotes.money.163.com/hs/marketdata/service/dzjy.php?host=/hs/marketdata/service/dzjy.php&page={}&query=symbol:{};&fields=NO,SYMBOL,SNAME,PUBLISHDATE,DZJY2,DZJY5,TCLOSE,DZJY55,DZJY6,DZJY9,DZJY11&sort=PUBLISHDATE&order=desc&count=25&type=query&initData=[object Object]'.format(Page,code)
    # url = 'http://quotes.money.163.com/hs/marketdata/service/dzjy.php?host=/hs/marketdata/service/dzjy.php&page=0&fields=NO,SYMBOL,SNAME,PUBLISHDATE,DZJY2,DZJY5,TCLOSE,DZJY55,DZJY6,DZJY9,DZJY11&sort=PUBLISHDATE&order=desc&count=25&type=query&i nitData=[object Object]&callback=callback_1717769636&req=31952'
    # 'http://quotes.money.163.com/hs/marketdata/service/dzjy.php?host=/hs/marketdata/service/dzjy.php&page=0&query=symbol:000568;&fields=NO,SYMBOL,SNAME,PUBLISHDATE,DZJY2,DZJY5,TCLOSE,DZJY55,DZJY6,DZJY9,DZJY11&sort=PUBLISHDATE&order=desc&count=25&type=query&initData=[object Object]&callback=callback_472758631&req=02126'
    req = Request(url=url, headers=header)
    bs = urlopen(req).read()
    bi = io.BytesIO(bs)
    gf = gzip.GzipFile(fileobj=bi, mode="rb")
    mainPage = gf.read().decode("unicode-escape")
    page_list = ast.literal_eval_r(mainPage)['list']
    df = pd.DataFrame.from_records(page_list)
    df =df[['NO','PUBLISHDATE','EXCHANGE','SYMBOL','SNAME','STYPE','DZJY2','DZJY5','DZJY6','DZJY9','DZJY11','TCLOSE','DZJY55','CODE','NAME']]
    df['DZJY2'] = df['DZJY2'].astype(float)
    df['DZJY5'] = df['DZJY5'].astype(float)
    df['DZJY6'] = df['DZJY6'].astype(float)
    df['TCLOSE'] = df['TCLOSE'].astype(float)
    df['DZJY55'] = df['DZJY55'].astype(float)
    data = pd.concat([data, df], ignore_index=True)
    if end_date is None:
        return data
    count = (df["PUBLISHDATE"] < end_date).sum()
    if count == 0:
        Page += 1
        return dzjy(code=code, Page=Page, end_date=end_date, data=data)
    else:
        return data