Python tushare.get_hist_data() Examples

def _init_coll(self):
        """
        Get all the history data when initiate the library.
        1. Connect to arctic and create the library.
        2. Get all the history data from tushare and strip the unused columns.
        3. Store the data to arctic.
        :return: None
        """

        # if collection is not initialized
        if self._coll_name not in self._library.list_symbols():
            self._new_added_colls.append(self._coll_name)
            his_data = ts.get_hist_data(code=self._coll_name, retry_count=5).sort_index()
            if len(his_data) == 0:
                logger.warning(
                    f'data of stock {self._coll_name} when initiation is empty'
                )
                return

            his_data = bdu.Utils.strip_unused_cols(his_data, *self._unused_cols)

            logger.debug(f'write history data for stock: {self._coll_name}.')
            self._library.write(self._coll_name, his_data) 

def get_hist_data_cache(code, date_start, date_end):
    cache_dir = bash_stock_tmp % (date_end[0:7], date_end)
    # 如果没有文件夹创建一个。月文件夹和日文件夹。方便删除。
    # print("cache_dir:", cache_dir)
    if not os.path.exists(cache_dir):
        os.makedirs(cache_dir)
    cache_file = cache_dir + "%s^%s.gzip.pickle" % (date_end, code)
    # 如果缓存存在就直接返回缓存数据。压缩方式。
    if os.path.isfile(cache_file):
        print("######### read from cache #########", cache_file)
        return pd.read_pickle(cache_file, compression="gzip")
    else:
        print("######### get data, write cache #########", code, date_start, date_end)
        stock = ts.get_hist_data(code, start=date_start, end=date_end)
        if stock is None:
            return None
        stock = stock.sort_index(0)  # 将数据按照日期排序下。
        stock.to_pickle(cache_file, compression="gzip")
        return stock 

def _cache_monthly(self, date):
        '''
        缓存date所在月的数据，等缓存成功后返回True
        :param code:
        :param ktype:
        :param date:
        :return:
        '''
        print 'caching...'
        start, end = self._get_date_range_of_month(date, 'str')
        df = ts.get_hist_data(code=self.code, ktype=self.ktype, start=start, end=end, retry_count=6)
        if df is not None:
            df.to_csv(self._get_cache_filename(date))

        waiting_seconds = 0
        while not self._in_cache(date=date):
            sleep(1)
            waiting_seconds += 1
            if waiting_seconds > 30:
                self._cache_monthly(date=date)
        return True 

def home(request):
    stock_his_data = ts.get_hist_data('sh000001')
    stock_name = get_stock_name('sh000001')

    date = stock_his_data.index.tolist()
    open = stock_his_data['open'].tolist()
    close = stock_his_data['close'].tolist()
    high = stock_his_data['high'].tolist()
    low = stock_his_data['low'].tolist()
    volume = stock_his_data['volume'].tolist()
    dataMA5 = stock_his_data['ma5'].tolist()
    dataMA10 = stock_his_data['ma10'].tolist()
    dataMA20 = stock_his_data['ma20'].tolist()

    return render(request, 'home.html', {'date': json.dumps(date), 'open': json.dumps(open), 'close': json.dumps(close),
                                         'high': json.dumps(high), 'low': json.dumps(low), 'volume': json.dumps(volume),
                                         'dataMA5': json.dumps(dataMA5), 'dataMA10': json.dumps(dataMA10),
                                         'dataMA20': json.dumps(dataMA20), 'stock_name': json.dumps(stock_name)}) 

def dash_index(request):
    stock_his_data = ts.get_hist_data('sz')
    stock_name = get_stock_name('sz')

    stock_his_data_dic = json.dumps(stock_his_data.to_json(orient='index'))
    pprint.pprint(stock_his_data_dic)

    print(type(stock_his_data_dic))
    date = stock_his_data.index.tolist()
    open = stock_his_data['open'].tolist()
    close = stock_his_data['close'].tolist()
    high = stock_his_data['high'].tolist()
    low = stock_his_data['low'].tolist()
    volume = stock_his_data['volume'].tolist()
    dataMA5 = stock_his_data['ma5'].tolist()
    dataMA10 = stock_his_data['ma10'].tolist()
    dataMA20 = stock_his_data['ma20'].tolist()

    return render(request, 'base_dash.html',
                  {'stock_his_data_dic': stock_his_data_dic, 'date': json.dumps(date), 'open': json.dumps(open),
                   'close': json.dumps(close), 'high': json.dumps(high), 'low': json.dumps(low),
                   'volume': json.dumps(volume), 'dataMA5': json.dumps(dataMA5), 'dataMA10': json.dumps(dataMA10),
                   'dataMA20': json.dumps(dataMA20), 'stock_name': json.dumps(stock_name)}) 

def dash_index(request):
    stock_his_data = ts.get_hist_data('sz')
    stock_name = get_stock_name('sz')

    stock_his_data_dic = json.dumps(stock_his_data.to_json(orient='index'))
    pprint.pprint(stock_his_data_dic)

    print(type(stock_his_data_dic))
    date = stock_his_data.index.tolist()
    open = stock_his_data['open'].tolist()
    close = stock_his_data['close'].tolist()
    high = stock_his_data['high'].tolist()
    low = stock_his_data['low'].tolist()
    volume = stock_his_data['volume'].tolist()
    dataMA5 = stock_his_data['ma5'].tolist()
    dataMA10 = stock_his_data['ma10'].tolist()
    dataMA20 = stock_his_data['ma20'].tolist()

    return render(request, 'base_dash.html', {'stock_his_data_dic': stock_his_data_dic, 'date': json.dumps(date), 'open': json.dumps(open), 'close': json.dumps(close), 'high': json.dumps(high), 'low': json.dumps(low), 'volume': json.dumps(volume), 'dataMA5': json.dumps(dataMA5), 'dataMA10': json.dumps(dataMA10), 'dataMA20': json.dumps(dataMA20), 'stock_name': json.dumps(stock_name)}) 

def get_stock_price(code, include_realtime_price):
    """
    获取个股股价
    :param code: 股票代码
    :param include_realtime_price: 是否含实时股价
    :return:
    """

    # 获取历史股价
    df = ts.get_hist_data(code)
    df = df[['close']]
    df['date'] = df.index

    if include_realtime_price:
        df_today = ts.get_today_all()
        df_code = df_today[df_today['code']==code]
        df_code = df_code[['trade']]
        df_code['date'] = GetNowDate()
        df_code.rename(columns={'trade': 'close'}, inplace=True)
        df = pd.concat([df, df_code], ignore_index=True)

    df.sort(columns='date', inplace=True)
    df = df.drop_duplicates(['date'])
    df.index = range(len(df))
    print '\n'
    # print df.head()
    print df.tail()
    return df 

def download_delta_data(self):
        """
        Get yesterday's data and append it to collection,
        this method is planned to be executed at each day's 8:30am to update the data.
        1. Connect to arctic and get the library.
        2. Get today's history data from tushare and strip the unused columns.
        3. Store the data to arctic.
        :return: None
        """

        self._init_coll()

        if self._coll_name in self._new_added_colls:
            return

        # 15:00 PM can get today data
        # start = latest_date + 1 day
        latest_date = self.get_data().index[-1]
        start = latest_date + dt.timedelta(days=1)
        start = dt.datetime.strftime(start, '%Y-%m-%d')

        his_data = ts.get_hist_data(
            code=self._coll_name,
            start=start,
            retry_count=5
        )

        # delta data is empty
        if len(his_data) == 0:
            logger.info(
                f'delta data of stock {self._coll_name} is empty, after {start}')
            return

        his_data = bdu.Utils.strip_unused_cols(his_data, *self._unused_cols)

        logger.info(f'got delta data of stock: {self._coll_name}, after {start}')
        self._library.append(self._coll_name, his_data) 

def get(self):
        symbol = self.get_argument("symbol")#股票代码
        period = self.get_argument("period")#时间周期,单位-分钟
        period_allow_list = ["5","15","30","60","1440","M","W"]
        if period not in period_allow_list:
            return 
       
        data = None
        if period =="1440":
            data = ts.get_hist_data(symbol)
        else:
            data = ts.get_k_data(symbol,ktype=period)
        print "=========",symbol,":",period
        resultlist = []
        lens = len(data)
        for unit in data.iterrows():
            obj  = {}
            dates = None
            if period =="1440":
                dates = unit[1].name
            else:
                dates = unit[1]['date']
            print "len",len(dates)
#             长度等于10的是%Y-%m-%d格式,16的是%Y-%m-%d %H:%M 格式
            dataformate = "%Y-%m-%d %H:%M"
            date_len = len(dates)
            if date_len == 10 :
                dataformate = "%Y-%m-%d"
            d=datetime.datetime.strptime(dates,dataformate)
            obj["date"]=int(time.mktime(d.timetuple()))
            obj["open"]=unit[1]['open']
            obj["close"]=unit[1]['close']
            obj["high"]=unit[1]['high']
            obj["low"]=unit[1]['low']
            obj["volume"]=unit[1]['volume']
            resultlist.append(obj)
        
        resultlist.sort(key=lambda obj:obj.get('date'), reverse=False) 
        
        s = json.dumps(resultlist)
        self.write(s) 

def _retrieve_online_historical_data(self, symbol):
        """
        Retrieve historical data from web
        """
        data = ts.get_hist_data(symbol, start=self._start_date.strftime('%Y-%m-%d'), end=self._end_date.strftime('%Y-%m-%d'))
        data = data.sort_index()
        data.index = data.index.to_datetime()
        #data.index = data.index.tz_localize('UTC')

        self._hist_data[symbol] = data
        self._hist_data[symbol]["FullSymbol"] = symbol         # attach symbol to data; it will be merged into _data_stream 

def stock_his_data2json(stock_code):
    stock_name = get_stock_name(stock_code)
    stock_his_data_json = json.dumps(ts.get_hist_data(stock_code).to_json(orient='index'))
    return stock_name,stock_his_data_json

#----------------------------- Request Methos -----------------------------# 

def __init__(self,code=None,start=None,end=None,save=True):
        """
        get symbol history data
        @@param:code:股票代码，如果code is None,或者全部股票
        @@parm:start:开始时间
        @@end:结束时间
        """
        now = dt.datetime.now()
        if end is None:
            end = (now -dt.timedelta(1)).strftime('%Y-%m-%d')
        if start is None:
            start = '1990-01-01'
            
        if code is None:
            mongo =mongodb()##获取股票代码
            collectname = 'SymbolMdm'
            where = {}
            kwargs = {'code':1}##获取code代码字断
            codes = mongo.select(collectname,where,kwargs)
            for code in codes:
                HistData(code.get('code'),HistData,end)##递归请求历史数据
        else:
            try:
                data = ts.get_hist_data(code,start=start,end=end) 
                data['datatime']  =  now.strftime('%Y-%m-%d')
                data['datatimestramp']  =  now.strftime('%H:%M:%S')
                data['date'] =data.index
                data['code'] = code
                indexlist = ['code','date']##数据库索引
                tableName = 'HistData'  
                database(data,indexlist,tableName,save)
                print 'info:{} downloaded is ok,num:{}!       '.format(code,len(data))
            except:
                pass 

def appends():
    filename = 'c:/day/bigfile.csv'
    for code in ['000875', '600848', '000981']:
        df = ts.get_hist_data(code)
        if os.path.exists(filename):
            df.to_csv(filename, mode='a', header=None)
        else:
            df.to_csv(filename) 

def json():
    df = ts.get_hist_data('000875')
    df.to_json('c:/day/000875.json',orient='records')

    #或者直接使用
    print(df.to_json(orient='records')) 

def hdf():
    df = ts.get_hist_data('000875')
#     df.to_hdf('c:/day/store.h5','table')
    
    store = HDFStore('c:/day/store.h5')
    store['000875'] = df
    store.close() 

def xls():
    df = ts.get_hist_data('000875')
    #直接保存
    df.to_excel('c:/day/000875.xlsx', startrow=2,startcol=5) 

def csv():
    df = ts.get_hist_data('000875')
    df.to_csv('c:/day/000875.csv',columns=['open','high','low','close']) 

def home(request):
    stock_his_data = ts.get_hist_data('sh000001')
    stock_name = get_stock_name('sh000001')

    date = stock_his_data.index.tolist()
    open = stock_his_data['open'].tolist()
    close = stock_his_data['close'].tolist()
    high = stock_his_data['high'].tolist()
    low = stock_his_data['low'].tolist()
    volume = stock_his_data['volume'].tolist()
    dataMA5 = stock_his_data['ma5'].tolist()
    dataMA10 = stock_his_data['ma10'].tolist()
    dataMA20 = stock_his_data['ma20'].tolist()

    return render(request, 'home.html', {'date': json.dumps(date), 'open': json.dumps(open), 'close': json.dumps(close), 'high': json.dumps(high), 'low': json.dumps(low), 'volume': json.dumps(volume), 'dataMA5': json.dumps(dataMA5), 'dataMA10': json.dumps(dataMA10), 'dataMA20': json.dumps(dataMA20), 'stock_name': json.dumps(stock_name)}) 

def stock_his_data2json(stock_code):
    stock_name = get_stock_name(stock_code)
    stock_his_data_json = json.dumps(ts.get_hist_data(stock_code).to_json(orient='index'))
    return stock_name,stock_his_data_json

#----------------------------- Request Methos -----------------------------# 

def appends():
    filename = 'c:/day/bigfile.csv'
    for code in ['000875', '600848', '000981']:
        df = ts.get_hist_data(code)
        if os.path.exists(filename):
            df.to_csv(filename, mode='a', header=None)
        else:
            df.to_csv(filename) 

def json():
    df = ts.get_hist_data('000875')
    df.to_json('c:/day/000875.json',orient='records')

    #或者直接使用
    print(df.to_json(orient='records')) 

def hdf():
    df = ts.get_hist_data('000875')
#     df.to_hdf('c:/day/store.h5','table')
    
    store = HDFStore('c:/day/store.h5')
    store['000875'] = df
    store.close() 

def xls():
    df = ts.get_hist_data('000875')
    #直接保存
    df.to_excel('c:/day/000875.xlsx', startrow=2,startcol=5) 

def csv():
    df = ts.get_hist_data('000875')
    df.to_csv('c:/day/000875.csv',columns=['open','high','low','close']) 

def divident_rate(self):
        stock = ts.get_hist_data(self.id)
        df = dividend_rate.get_bonus_table(self)
        df_dividend = df[['年度','派息','登记日']]
#        print(df_dividend)
        stock_close_price = stock["close"]
        sIndex = stock_close_price.index.tolist()
        # 获取登记日
        regis = df_dividend['登记日'].tolist()
#        print(sIndex)
#        print(regis)
        close_price = []
        diVi = []
        aPe = []
        bonus = []
        div_year = []
        for i in regis:
            if i != "--" and i in sIndex:
                cprice = stock_close_price.loc[i]
                close_price.append(cprice)
                aDiv = df_dividend[df_dividend['登记日'] == i]['派息'].tolist()[0]
                year = df_dividend[df_dividend['登记日'] == i]['年度'].values #获得年份
                div_year.append(year[0])
                
                #此处的bonus暂时通过ts获得，以后可以直接搜索本地数据库
                profit_table = ts.get_report_data(year[0],4) #获取年度eps
                print('')
                target_eps = profit_table[profit_table['code'] == self.id]['eps'].values
                eps = target_eps[0].item()  #numpy.float64 -> float
                per_bonus = round(float(aDiv) / 10 / eps * 100, 2)
#                per_bonus = 1   #测试时开启
                
                bonus.append(per_bonus)

                diVi.append(float(aDiv)/10) #10股派息转1股派息
        div_ratio = []
        for i,j in zip(diVi,close_price):
            adivr = float(i) / float(j) * 100
            div_ratio.append(round(adivr,2))
            aPe.append(round(100/adivr,2))

        reDf = pd.DataFrame({"cash_div":diVi,   #每股派现方案
                             "div_ratio(%)":div_ratio, #股息率
                             'ape':aPe, #真实市盈率
                             'bonus_ratio(%)':bonus #分红率
                             },index = div_year)
                
        # 统计输出
        print(self.id + '分红情况统计如下：')
        avg_bonus = round(sum(bonus)/len(bonus),2)
        print('1.平均分红率:',avg_bonus,'%')
        avg_div = round(sum(div_ratio)/len(div_ratio),2)
        print('2.平均股息率:',avg_div,'%')
        print('3.详细列表如下所示')
        return reDf

##############################################