Exemplo n.º 1
0
def get_today_ticks(code=None, retry_count=3, pause=0.001):
    """
        获取当日分笔明细数据
    Parameters
    ------
        code:string
                  股票代码 e.g. 600848
        retry_count : int, 默认 3
                  如遇网络等问题重复执行的次数
        pause : int, 默认 0
                 重复请求数据过程中暂停的秒数,防止请求间隔时间太短出现的问题
     return
     -------
        DataFrame 当日所有股票交易数据(DataFrame)
              属性:成交时间、成交价格、价格变动,成交手、成交金额(元),买卖类型
    """
    if code is None or len(code) != 6:
        return None
    symbol = ct._code_to_symbol(code)
    date = du.today()
    for _ in range(retry_count):
        time.sleep(pause)
        try:
            request = Request(ct.TODAY_TICKS_PAGE_URL %
                              (ct.P_TYPE['http'], ct.DOMAINS['vsf'],
                               ct.PAGES['jv'], date, symbol))
            data_str = urlopen(request, timeout=10).read()
            data_str = data_str.decode('GBK')
            data_str = data_str[1:-1]
            data_str = eval(
                data_str,
                type('Dummy', (dict, ), dict(__getitem__=lambda s, n: n))())
            data_str = json.dumps(data_str)
            data_str = json.loads(data_str)
            pages = len(data_str['detailPages'])
            data = pd.DataFrame()
            ct._write_head()
            for pNo in range(1, pages + 1):
                data = data.append(_today_ticks(symbol, date, pNo, retry_count,
                                                pause),
                                   ignore_index=True)
        except Exception as er:
            print(str(er))
        else:
            return data
    raise IOError(ct.NETWORK_URL_ERROR_MSG)
Exemplo n.º 2
0
def get_k_data(code=None, start='', end='',
                  ktype='D', autype='qfq', 
                  index=False,
                  retry_count=3,
                  pause=0.001):
    """
    获取k线数据
    ---------
    Parameters:
      code:string
                  股票代码 e.g. 600848
      start:string
                  开始日期 format:YYYY-MM-DD 为空时取上市首日
      end:string
                  结束日期 format:YYYY-MM-DD 为空时取最近一个交易日
      autype:string
                  复权类型,qfq-前复权 hfq-后复权 None-不复权,默认为qfq
      ktype:string
                  数据类型,D=日k线 W=周 M=月 5=5分钟 15=15分钟 30=30分钟 60=60分钟,默认为D
      retry_count : int, 默认 3
                 如遇网络等问题重复执行的次数 
      pause : int, 默认 0
                重复请求数据过程中暂停的秒数,防止请求间隔时间太短出现的问题
    return
    -------
      DataFrame
          date 交易日期 (index)
          open 开盘价
          high  最高价
          close 收盘价
          low 最低价
          volume 成交量
          amount 成交额
          turnoverratio 换手率
          code 股票代码
    """
    symbol = ct.INDEX_SYMBOL[code] if index else ct._code_to_symbol(code)
    url = ''
    dataflag = ''
    autype = '' if autype is None else autype
    if (start is not None) & (start != ''):
        end = du.today() if end is None or end == '' else end
    if ktype.upper() in ct.K_LABELS:
        fq = autype if autype is not None else ''
        if code[:1] in ('1', '5') or index:
            fq = ''
        kline = '' if autype is None else 'fq'
        if (start is None or start == '') & (end is None or end == ''):
            urls = [ct.KLINE_TT_URL%(ct.P_TYPE['http'], ct.DOMAINS['tt'],
                                    kline, fq, symbol, 
                                    ct.TT_K_TYPE[ktype.upper()], start, end,
                                    fq, _random(17))]
        else:
            years = du.tt_dates(start, end)
            urls = []
            for year in years:
                startdate = str(year) + '-01-01'
                enddate = str(year+1) + '-12-31'
                url = ct.KLINE_TT_URL%(ct.P_TYPE['http'], ct.DOMAINS['tt'],
                                    kline, fq+str(year), symbol, 
                                    ct.TT_K_TYPE[ktype.upper()], startdate, enddate,
                                    fq, _random(17))
                urls.append(url)
        dataflag = '%s%s'%(fq, ct.TT_K_TYPE[ktype.upper()])
    elif ktype in ct.K_MIN_LABELS:
        urls = [ct.KLINE_TT_MIN_URL%(ct.P_TYPE['http'], ct.DOMAINS['tt'],
                                    symbol, ktype, ktype,
                                    _random(16))]
        dataflag = 'm%s'%ktype
    else:
        raise TypeError('ktype input error.')
    data = pd.DataFrame()
    for url in urls:
        data = data.append(_get_k_data(url, dataflag, 
                                       symbol, code,
                                       index, ktype,
                                       retry_count, pause), 
                           ignore_index=True)
    if ktype not in ct.K_MIN_LABELS:
        if ((start is not None) & (start != '')) & ((end is not None) & (end != '')):
            if data.empty==False:       
                data = data[(data.date >= start) & (data.date <= end)]
    return data
    raise IOError(ct.NETWORK_URL_ERROR_MSG)
Exemplo n.º 3
0
def get_h_data(code, start=None, end=None, autype='qfq',
               index=False, retry_count=3, pause=0.001, drop_factor=True):
    '''
    获取历史复权数据
    Parameters
    ------
      code:string
                  股票代码 e.g. 600848
      start:string
                  开始日期 format:YYYY-MM-DD 为空时取当前日期
      end:string
                  结束日期 format:YYYY-MM-DD 为空时取去年今日
      autype:string
                  复权类型,qfq-前复权 hfq-后复权 None-不复权,默认为qfq
      retry_count : int, 默认 3
                 如遇网络等问题重复执行的次数 
      pause : int, 默认 0
                重复请求数据过程中暂停的秒数,防止请求间隔时间太短出现的问题
      drop_factor : bool, 默认 True
                是否移除复权因子,在分析过程中可能复权因子意义不大,但是如需要先储存到数据库之后再分析的话,有该项目会更加灵活
    return
    -------
      DataFrame
          date 交易日期 (index)
          open 开盘价
          high  最高价
          close 收盘价
          low 最低价
          volume 成交量
          amount 成交金额
    '''
    
    start = du.today_last_year() if start is None else start
    end = du.today() if end is None else end
    qs = du.get_quarts(start, end)
    qt = qs[0]
    ct._write_head()
    data = _parse_fq_data(_get_index_url(index, code, qt), index,
                          retry_count, pause)
    if data is None:
        data = pd.DataFrame()
    if len(qs)>1:
        for d in range(1, len(qs)):
            qt = qs[d]
            ct._write_console()
            df = _parse_fq_data(_get_index_url(index, code, qt), index,
                                retry_count, pause)
            if df is None:  # 可能df为空,退出循环
                break
            else:
                data = data.append(df, ignore_index = True)
    if len(data) == 0 or len(data[(data.date >= start) & (data.date <= end)]) == 0:
        return pd.DataFrame()
    data = data.drop_duplicates('date')
    if index:
        data = data[(data.date >= start) & (data.date <= end)]
        data = data.set_index('date')
        data = data.sort_index(ascending = False)
        return data
    if autype == 'hfq':
        if drop_factor:
            data = data.drop('factor', axis=1)
        data = data[(data.date >= start) & (data.date <= end)]
        for label in ['open', 'high', 'close', 'low']:
            data[label] = data[label].map(ct.FORMAT)
            data[label] = data[label].astype(float)
        data = data.set_index('date')
        data = data.sort_index(ascending = False)
        return data
    else:
        if autype == 'qfq':
            if drop_factor:
                data = data.drop('factor', axis = 1)
            df = _parase_fq_factor(code, start, end)
            df = df.drop_duplicates('date')
            df = df.sort_values('date', ascending = False)
            firstDate = data.head(1)['date']
            frow = df[df.date == firstDate[0]]
            rt = get_realtime_quotes(code)
            if rt is None:
                return pd.DataFrame()
            if ((float(rt['high']) == 0) & (float(rt['low']) == 0)):
                preClose = float(rt['pre_close'])
            
            rate = float(frow['factor']) / preClose
            data = data[(data.date >= start) & (data.date <= end)]
            for label in ['open', 'high', 'low', 'close']:
                data[label] = data[label] / rate
                data[label] = data[label].map(ct.FORMAT)
                data[label] = data[label].astype(float)
            data = data.set_index('date')
            data = data.sort_index(ascending = False)
            return data
        else:
            for label in ['open', 'high', 'close', 'low']:
                data[label] = data[label] / data['factor']
            if drop_factor:
                data = data.drop('factor', axis=1)
            data = data[(data.date >= start) & (data.date <= end)]
            for label in ['open', 'high', 'close', 'low']:
                data[label] = data[label].map(ct.FORMAT)
            data = data.set_index('date')
            data = data.sort_index(ascending = False)
            data = data.astype(float)
            return data