Exemple #1
0
def getprofitreport():
    profitreport = "./output/profit_%d_%d.csv"
    outreport = "./output/profitcontinuous.csv"
    year = 2015
    quarter = 4

    dflist = []
    for i in range(3):
        report = profitreport % (year, quarter)
        if os.path.exists(report):
            print("Get data from csv")
            dfsingle = pd.read_csv(report, encoding="ANSI")
        else:
            print("Get data from internet")
            try:
                dfsingle = ts.get_report_data(year=year, quarter=quarter)
                dfsingle["period"] = "%d_%d" % (year, quarter)
            except Exception as e:
                dfsingle = ts.get_report_data(year=year, quarter=quarter - 1)
                dfsingle["period"] = "%d_%d" % (year, quarter - 1)
            dfsingle.to_csv(report)
        dflist.append(dfsingle)
        year += 1

    dfall = pd.concat(dflist)
    dfposprofit = pd.DataFrame(dfall[(dfall["net_profits"] > 0)
                                     & (dfall["profits_yoy"] > 0)])
    result = dfposprofit.groupby(["code", "name"]).size().reset_index()
    print(result.columns)
    result.columns = ["code", "name", "poscontinue"]
    print(result)
    result = pd.DataFrame(result[(result["poscontinue"] >= 3)],
                          columns=["code", "name"]).drop_duplicates()
    print(result)
    result.to_csv(outreport, index=False)
def main():
    year=2004
    season=1
    while year<2017:
        while season<5:
            try:#本地有记录
                with open('report_'+str(year)+'_'+str(season)+'.pkl', "rb") as f:
                    report = pickle.load(f)
            except:#本地没有存过
                with open('report_' + str(year) + '_' + str(season) + '.pkl', "wb") as f:
                    report_1st = ts.get_report_data(year, season)  # 获取业绩报表
                    report_2nd = ts.get_report_data(year, season)  # 获取业绩报表
                    report_3rd = ts.get_report_data(year, season)  # 获取业绩报表
                    #把三次合并:
                    report_all3 = pd.concat((report_1st,report_2nd,report_3rd),axis=0,join='outer')
                    #report_all3.to_excel(r'D:\work_python\DeepData\tushare_download\201701_01_all3.xlsx')
                    report =report_all3.drop_duplicates()
                    #report.to_excel(r'D:\work_python\DeepData\tushare_download\201701_01_quchong.xlsx')
                    pickle.dump(report, f)
            #report =report.drop_duplicates()
            #report.to_excel(r'D:\work_python\DeepData\tushare_download\201701_01_quchong.xlsx')
            engine = create_engine(
                        'mysql+pymysql://' + DATABASS_USER_NAME + ':' + DATABASS_PASSWORD + '@127.0.0.1/'+DATABASS_NAME+'?charset=utf8')
            #存入数据库
            report.to_sql('report_'+str(year)+'_'+str(season),engine,if_exists='replace')
            season=season+1
        season =1
        year += 1
Exemple #3
0
def save_report_data():
    """
    tushare: save_report_data
    """
    report = ts.get_report_data(2017,2)
    report.to_csv(CSV_REPORT_DATA_S2)

    report = ts.get_report_data(2017,1)
    report.to_csv(CSV_REPORT_DATA_S1)
Exemple #4
0
 def get_report_data(self, year, loops):
     for i in range(1,loops):
         print(i)
         for j in year:
             print(year)
             try:
                 ts.get_report_data(j, 4).to_csv('report_data_%d.csv'%j, encoding='utf-8')
                 print(j)
                 year.remove(j)
             except: pass
def find2017():
    r2017 = ts.get_report_data(2017,3).set_index("code")
    r2016 = ts.get_report_data(2016,4).set_index("code")
    l = []
    l1= []
    for i in r2017.index:
        if i not in r2016.index:
            continue
        rate = (r2017.net_profits[i] - r2016.net_profits[i]) / r2016.net_profits[i]

        try:
            rate = float(rate)
        except:
            rate = rate.values[0]
        if rate > 0.25:
          # l.append((rate,i))
          l1.append(i)
    print(l1)

    r2015 = ts.get_report_data(2015, 4).set_index("code")

    l2 = []
    for i in l1:
        if i not in r2015.index:
            continue
        rate = (r2016.net_profits[i] - r2015.net_profits[i]) / r2015.net_profits[i]

        try:
            rate = float(rate)
        except:
            rate = rate.values[0]
        if rate > 0.3:
            # ll.append((l1[i],rate, i))
            l2.append(i)
    r2014 = ts.get_report_data(2014, 4).set_index("code")
    print(l2)
    l3 = []
    for i in l2:
        if i not in r2014.index:
            continue
        rate = (r2015.net_profits[i] - r2014.net_profits[i]) / r2014.net_profits[i]

        try:
            rate = float(rate)
        except:
            rate = rate.values[0]
        if rate > 0.3:
            # ll.append((l1[i],rate, i))
            l3.append(i)
    print(l3)
    return l3
Exemple #6
0
def basic_information():
    ts.get_cashflow_data(2017, 1).to_sql('cash_flow',
                                         engine,
                                         if_exists='append')
    ts.get_debtpaying_data(2017, 1).to_sql('debtpaying',
                                           engine,
                                           if_exists='append')
    ts.get_growth_data(2017, 1).to_sql('growth', engine, if_exists='append')
    ts.get_operation_data(2017, 1).to_sql('operation',
                                          engine,
                                          if_exists='append')
    ts.get_profit_data(2017, 1).to_sql('profit', engine, if_exists='append')
    ts.get_report_data(2017, 2).to_sql('report', engine, if_exists='append')
    print('basic information over ....')
Exemple #7
0
def _fetch_finance():
    for year in range(2004, 2018):
        set_year = lambda x: str(year) + '-' + x
        for quarter in range(1, 5):
            print(year, ' year ', 'quarter ', quarter)
            rep = ts.get_report_data(
                year, quarter)[['code', 'eps', 'bvps', 'epcf', 'report_date']]
            pro = ts.get_profit_data(year, quarter)[[
                'code', 'roe', 'net_profit_ratio', 'gross_profit_rate',
                'net_profits', 'business_income', 'bips'
            ]]
            ope = ts.get_operation_data(year, quarter)[[
                'code', 'arturnover', 'arturndays', 'inventory_turnover',
                'currentasset_turnover', 'currentasset_days'
            ]]
            gro = ts.get_growth_data(
                year, quarter)[['code', 'mbrg', 'nprg', 'nav', 'epsg', 'seg']]
            deb = ts.get_debtpaying_data(year, quarter)[[
                'code', 'currentratio', 'quickratio', 'cashratio', 'icratio',
                'sheqratio', 'adratio'
            ]]
            cas = ts.get_cashflow_data(year, quarter)[[
                'code', 'cf_sales', 'rateofreturn', 'cf_nm', 'cf_liabilities',
                'cashflowratio'
            ]]

            rep.rename(columns={'report_date': 'date'}, inplace=True)
            rep['date'] = rep['date'].apply(set_year)
            rep = rep.merge(pro, on='code', how='left')
            rep = rep.merge(ope, on='code', how='left')
            rep = rep.merge(gro, on='code', how='left')
            rep = rep.merge(deb, on='code', how='left')
            rep = rep.merge(cas, on='code', how='left')
            finance.insert(rep.to_dict('record'))
            print(year, quarter)
Exemple #8
0
def valuation_factor(year):
    report = ts.get_report_data(year,4)
    report = report.sort_values(by = 'code',axis = 0,ascending = True)
    report = report.reset_index(drop = True)
    report.to_csv("/home/yirui/Desktop/Quant/Report/%s.csv"%year, mode="w")

    profit = ts.get_profit_data(year, 4)
    profit = profit.sort_values(by='code', axis=0, ascending=True)
    profit = profit.reset_index(drop=True)
    profit.to_csv("/home/yirui/Desktop/Quant/Profit/%s.csv"%year,mode= "w")

    operation = ts.get_operation_data(year,4)
    operation = operation.sort_values(by='code', axis=0, ascending=True)
    operation = operation.reset_index(drop=True)
    operation.to_csv("/home/yirui/Desktop/Quant/Operation/%s.csv" % year, mode="w")

    growth = ts.get_growth_data(year,4)
    growth = growth.sort_values(by='code', axis=0, ascending=True)
    growth = growth.reset_index(drop=True)
    growth.to_csv("/home/yirui/Desktop/Quant/Growth/%s.csv" % year, mode="w")

    debtpaying = ts.get_debtpaying_data(year,4)
    debtpaying = debtpaying.sort_values(by='code', axis=0, ascending=True)
    debtpaying = debtpaying.reset_index(drop=True)
    debtpaying.to_csv("/home/yirui/Desktop/Quant/Debtpaying/%s.csv" % year, mode="w")

    cashflow = ts.get_cashflow_data(year,4)
    cashflow = cashflow.sort_values(by='code', axis=0, ascending=True)
    cashflow = cashflow.reset_index(drop=True)
    cashflow.to_csv("/home/yirui/Desktop/Quant/Cashflow/%s.csv" % year, mode="w")
Exemple #9
0
def getData():
    #获取财务数据保存到本地,修改tushare源码使其仅包括金融类股票
    for year in range(2010,2011):
        for season in range(3,5):
            df = ts.get_report_data(year,season)
            filename = str(year) + '-' + str(season)
            df.to_csv(filename,sep=',', encoding='utf-8')
def get_stock_report_manual(year, season):
    frame = ts.get_report_data(year, season)
    table_name = 'stock_report_' + str(year) + 's' + str(season)
    db = MySQLdb.connect(host='localhost',
                         port=3306,
                         user='******',
                         passwd='123456',
                         db='stock_ts',
                         charset='utf8')
    cursor = db.cursor()
    createDBSql = 'create table if not exists ' + table_name + '(code varchar(10), name varchar(16), eps text, eps_yoy text, bvps text, roe text, epcf text, net_profits text, profits_yoy text, distrib text, report_date text)'
    cursor.execute(createDBSql)
    for i in range(0, len(frame)):
        prefix = 'insert into ' + table_name + '(code, name, eps, eps_yoy, bvps, roe, epcf, net_profits, profits_yoy, distrib, report_date) values(\'%s\', \'%s\', \'%s\', \'%s\', \'%s\', \'%s\', \'%s\', \'%s\', \'%s\', \'%s\', \'%s\')'
        sql = prefix % (
            frame['code'][i], str(frame['name'][i]), float(frame['eps'][i]),
            float(frame['eps_yoy'][i]), float(frame['bvps'][i]),
            float(frame['roe'][i]), float(frame['epcf'][i]),
            float(frame['net_profits'][i]), float(frame['profits_yoy'][i]),
            str(frame['distrib'][i]), str(frame['report_date'][i]))
        #print(sql)
        cursor.execute(sql)
        db.commit()
        print(frame['name'][i])

    db.close()
Exemple #11
0
 def get_report_data(self, year, quarter):
     tsdata = ts.get_report_data(
         year=year,
         quarter=quarter,
     )
     jsdata = To_Json(tsdata)
     return jsdata
Exemple #12
0
def get_stock_report(year, season, engine):
	frame = ts.get_report_data(year, season)
	table_name = 'stock_report_' + str(year) + 's' + str(season)
	if useDB == True:
		frame.to_sql(table_name, engine)
	else:
		frame.to_csv(table_name + '.csv')
Exemple #13
0
def fImportReportData(year="", quarter=""):
    df = ts.get_report_data(year=year, quarter=quarter)
    df['YEAR'] = year
    df['QUARTER'] = quarter
    #print(df)
    engine = create_engine('oracle://c##tushare:didierg160@myoracle')
    df.to_sql('tb_stock_report_data', engine, if_exists='append')
Exemple #14
0
def export(exportType, datePicker):
    sb = None
    [year, quarter] = getYearQuarter(datePicker)
    if exportType == "report":
        name = "业绩报表"
        sb = ts.get_report_data(year, quarter)
    elif exportType == "profit":
        name = "盈利能力报表"
        sb = ts.get_profit_data(year, quarter)
    elif exportType == "operation":
        name = "营运能力报表"
        sb = ts.get_operation_data(year, quarter)
    elif exportType == "growth":
        name = "成长能力报表"
        sb = ts.get_growth_data(year, quarter)
    elif exportType == "debtpaying":
        name = "偿债能力报表"
        sb = ts.get_debtpaying_data(year, quarter)
    elif exportType == "cashflow":
        name = "现金流量报表"
        sb = ts.get_cashflow_data(year, quarter)

    filename = quote(name + str(year) + "Q" + str(quarter) + ".xlsx")
    filepath = os.path.join(basedir, app.config['UPLOAD_FOLDER'], filename)

    sb.to_excel(filepath)

    rtn = send_file(filepath, as_attachment=True)
    rtn.headers['Content-Disposition'] += "; filename*=utf-8''%s" % (filename)
    return rtn
Exemple #15
0
def get_basic_datas(
        data_kind):  #datakind 为debtpaying,growth,operation,profit,report
    client = pymongo.MongoClient('localhost', 27017)
    table_stock = client['stock']
    sheet = table_stock[data_kind]
    for year in range(STARTYEAR, ENDYEAR + 1):
        try:
            for season in range(1, 5):
                print('getting ' + datakind + ' data at year:' + str(year) +
                      " season:" + str(season))
                if data_kind == 'debtpaying':
                    tf = ts.get_debtpaying_data(year, season)
                elif data_kind == 'growth':
                    tf = ts.get_growth_data(year, season)
                elif data_kind == 'operation':
                    tf = ts.get_operation_data(year, season)
                elif data_kind == 'profit':
                    tf = ts.get_profit_data(year, season)
                elif data_kind == 'report':
                    tf = ts.get_report_data(year, season)
                else:
                    print('Not available data type of data_kind!')
                    return
                jsonres = json.loads(tf.to_json(orient='records'))
                for j in jsonres:
                    sheet.insert_one(j)
        except:  #数据缺失,tushare接口会报网络错误
            print('the year: ' + str(year) + ' lost data will begin next year')
            continue
Exemple #16
0
def update_basics():
    basics = ts.get_stock_basics()
    f = os.path.join(DATA_DIR, 'basics.h5')
    basics.to_hdf(f, 'basics')

    length = 4 * 5
    year, season = last_report_season()
    for i in range(length):
        f = os.path.join(DATA_DIR, 'basics-{0}-{1}.h5'.format(year, season))
        if os.path.exists(f):
            continue
        report = ts.get_report_data(year, season)
        report.to_hdf(f, 'report')

        profit = ts.get_profit_data(year, season)
        profit.to_hdf(f, 'profit')

        operation = ts.get_operation_data(year, season)
        operation.to_hdf(f, 'operation')

        growth = ts.get_growth_data(year, season)
        growth.to_hdf(f, 'growth')

        debtpaying = ts.get_debtpaying_data(year, season)
        debtpaying.to_hdf(f, 'debtpaying')

        cashflow = ts.get_cashflow_data(year, season)
        cashflow.to_hdf(f, 'cashflow')

        season -= 1
        if season == 0:
            season = 4
            year -= 1
Exemple #17
0
def report_data():

    #获取2014年第3季度的业绩报表数据
    #季度 :1、2、3、4,只能输入这4个季度
    quarter = 1
    # 获取当前的年和月
    nowYear = int(time.strftime('%Y', time.localtime(time.time())))
    nowMonth = int(time.strftime('%m', time.localtime(time.time())))
    if nowMonth <= 3:
        # 上一年的第四季度
        nowYear -= 1
        quarter = 4

    elif nowMonth <= 6:
        # 第一季度
        quarter = 1

    elif nowMonth <= 9:
        # 第一季度
        quarter = 2

    elif nowMonth <= 12:
        # 第一季度
        quarter = 3

    rd = ts.get_report_data(nowYear, quarter)
    rd.to_csv('report_data.csv')
Exemple #18
0
def stat_stock_report(tmp_datetime, max_year=11):
    """
    以后每年7月份取一下上年的年报即可,历史数据不必再取
    经验: 19年4月份取18年的年报是不全的,所以延到7月取
    """
    cur_year = int((tmp_datetime).strftime("%Y"))
    # cur_year = 2005
    i = cur_year - max_year
    # i = 2001
    MAX_RETRY_TIME = 3
    retry_time = 0
    while i < cur_year:
        try:
            data = ts.get_report_data(i, 4)
        except IOError:
            data = None
        if not data is None and len(data) > 0:
            print("\nyear done", i)
            # data = data.drop_duplicates(subset="code", keep="last")
            data.insert(0, "year", [i] * len(data))
            data.head(n=1)
            common.insert_db(data, "ts_stock_report", False, "`year`,`code`")
            i += 1
            retry_time = 0
        else:
            print("\nno data . stock_report year", i)
            retry_time += 1
            if retry_time > MAX_RETRY_TIME:
                i += 1
                retry_time = 0

        time.sleep(5)  # 停止5秒
Exemple #19
0
 def __init__(self,year=None,quarter=None,save=True,updateAll=False,beginyear=1990,beginquarter=1):
     """
     获取季度的公司业绩报告
     @@parm:year:年份
     @@parm:quarter:季度
     @@parm:updateAll是否全量更新历史所有数据
     """
     now = dt.datetime.now()
     if year is None or quarter is None:
             year = now.year
             quarter = math.ceil(now.month/3.0)
             
     if not updateAll:
         
         try:
            profitData  = ts.get_report_data(year,quarter)
            profitData['datatime']  =  now.strftime('%Y-%m-%d')
            profitData['datatimestramp']  =  now.strftime('%H:%M:%S')
            profitData['year'] = year    
            profitData['quarter'] = quarter
            indexlist = ['code','year','quarter']##数据库索引
            tableName = 'companyRepor'  
            database(profitData,indexlist,tableName,save)
            
         except :
             traceback.print_exc()
             
     else:
         ##生成季度时间序列
         quarters = get_quarters((beginyear,beginquarter),(year,quarter))
         for y,q in quarters:
             companyRepor(y,q)##递归获取所有历史数据
Exemple #20
0
 def get_data(self, year, quarter):
     data = ts.get_stock_basics()
     data_roe = pd.DataFrame()
     for i in range(4):
         quarter += 1
         data_roe = pd.concat((data_roe, ts.get_report_data(year, quarter)))
     self.save_data(data, data_roe)
Exemple #21
0
def stock_report(year, quarter, k_index):
    """
        :param k_index: the k_index can be:
        1: name
        2: eps
        3: eps_yoy
        4: bvps
        5: roe
        6: epcf
        7: net_profits
        8: profits_yoy
        9: distrib
        10: report_data
        :return: the data to be captured for k_index and for all code, code is the index of return result
    """

    index_list = [
        'name', 'eps', 'eps_yoy', 'bvps', 'roe', 'epcf', 'net_profits',
        'profits_yoy', 'distrib', 'report_date'
    ]
    if k_index not in index_list:
        raise Exception('invalid k_index - the setting is not in the scope')
    if year <= 0:
        raise Exception('invalid year that should be larger than 0')
    if quarter <= 0 and quarter > 4:
        raise Exception('invalid quarter that we just 4 quarter in market')
    data = ts.get_report_data(year, quarter)
    new_data = data.set_index(['code'])
    return new_data[k_index]
Exemple #22
0
def call_report_v1(year, quarter):
    '''
    code,代码
    name,名称
    esp,每股收益
    eps_yoy,每股收益同比(%)
    bvps,每股净资产
    roe,净资产收益率(%)
    epcf,每股现金流量(元)
    net_profits,净利润(万元)
    profits_yoy,净利润同比(%)
    distrib,分配方案
    report_date,发布日期
    '''

    key = f'v1_report_{year}_{quarter}'
    stores = pd.HDFStore(COMMEN_FILE_PATH)
    if key not in stores:
        df = ts.get_report_data(year, quarter)
        if df.empty:
            return df
        stores[key] = df
        if DEBUG:
            print('STORE:', filePath)
    else:
        df = stores[key]
    stores.close()
    return df
def store_fund_data(quarter_list):
    stock2year_path = os.path.join(LastFilePath, "stock_fundm_info")
    for fun_year, fun_quarter in quarter_list:

        #every dataframe you craw down all needs remove the duplicated row. Only need keep the first row of duplicates.

        # stock2year_report is tushare:get_report_data  (fundamental data).
        stock2year_report = ts.get_report_data(
            fun_year, fun_quarter).drop_duplicates(keep='first')
        # stock2year_prof is tushare.get_profit_data  (fundamental data).
        stock2year_prof = ts.get_profit_data(
            fun_year, fun_quarter).drop_duplicates(keep='first')
        # stock2year_opera is tushare.get_operation_data (fundamental data).
        stock2year_opera = ts.get_operation_data(
            fun_year, fun_quarter).drop_duplicates(keep='first')
        #stock2year_grow is tushare.get_growth_data (fundamental data).
        stock2year_grow = ts.get_growth_data(
            fun_year, fun_quarter).drop_duplicates(keep='first')
        #stock2year_debt is tushare.get_debtpaying_data (fundamental data).
        stock2year_debt = ts.get_debtpaying_data(
            fun_year, fun_quarter).drop_duplicates(keep='first')
        #stock2year_cash is tushare.get_cashflow_data (fundamental data).
        stock2year_cash = ts.get_cashflow_data(
            fun_year, fun_quarter).drop_duplicates(keep='first')
        #stock2year_comb is to combine all the stock2year data of same year and quarter in a same stock code.
        stock2year_list = [stock2year_report,stock2year_prof,stock2year_opera,stock2year_grow, \
                           stock2year_debt,stock2year_cash]
        for every_fund_element in stock2year_list:
            every_fund_element = every_fund_element.set_index('code')
        #use pandas concat to combine all the dataframe along columns.
        total_fund = pd.concat(stock2year_list, axis=1)
        HeadName = fun_year + "/" + fun_quarter + "_" + "fundamt_info"
        CsvName = os.path.join(stock2year_path, "{}.csv".format(HeadName))
        total_fund.to_csv(CsvName)
Exemple #24
0
    def getThingsEveryday(self):
        yearEnd = datetime.now().year - 1

        if self.__flagUpdateReport:
            pdGrowthLastYear = ts.get_growth_data(yearEnd, 4)
            pdGrowthLastYear.to_excel('./' + \
                        str(yearEnd) + 'Growth.xls', sheet_name='Growth')
            pdProfitLastYear = ts.get_profit_data(yearEnd, 4)
            pdProfitLastYear.to_excel('./' + \
                        str(yearEnd) + 'Profit.xls', sheet_name='Profit')
            pdReportLastYear = ts.get_report_data(yearEnd, 4)
            pdReportLastYear.to_excel('./' + \
                        str(yearEnd) + 'y.xls', sheet_name='Report')

        self.__stockBasics = ts.get_stock_basics()  #获得昨天pe
        self.__stockBasics['code'] = self.__stockBasics.index.astype(int)
        self.__stockBasics.sort_index(inplace=True)
        self.__stockTodayAll = ts.get_today_all()  #获得昨收
        self.__stockTodayAll['code'] = self.__stockTodayAll['code'].astype(int)
        self.__pdForwardEps = pd.merge(self.__stockBasics,
                                       self.__stockTodayAll,
                                       on='code')
        self.__pdForwardEps['feps'] = self.__pdForwardEps[
            'settlement'] / self.__pdForwardEps['pe']
        print('\n')
Exemple #25
0
def stat_all(tmp_datetime):
    # 返回 31 天前的数据,做上个季度数据统计。
    tmp_datetime_1month = tmp_datetime + datetime.timedelta(days=-31)
    year = int((tmp_datetime_1month).strftime("%Y"))
    quarter = int(pd.Timestamp(tmp_datetime_1month).quarter)  # 获得上个季度的数据。
    print("############ year %d, quarter %d", year, quarter)
    # 业绩报告(主表)
    data = ts.get_report_data(year, quarter)
    # 增加季度字段。
    data = concat_quarter(year, quarter, data)
    # 处理重复数据,保存最新一条数据。最后一步处理,否则concat有问题。
    data = data.drop_duplicates(subset="code", keep="last")
    global db
    # 插入数据库。
    db.insert_db(data, "ts_report_data", True, "`quarter`,`code`")

    # 盈利能力
    data = ts.get_profit_data(year, quarter)
    # 增加季度字段。
    data = concat_quarter(year, quarter, data)
    # 处理重复数据,保存最新一条数据。
    data = data.drop_duplicates(subset="code", keep="last")
    # 插入数据库。
    db.insert_db(data, "ts_profit_data", True, "`quarter`,`code`")

    # 营运能力
    data = ts.get_operation_data(year, quarter)
    # 增加季度字段。
    data = concat_quarter(year, quarter, data)
    # 处理重复数据,保存最新一条数据。最后一步处理,否则concat有问题。
    data = data.drop_duplicates(subset="code", keep="last")
    # 插入数据库。
    db.insert_db(data, "ts_operation_data", True, "`quarter`,`code`")

    # 成长能力
    data = ts.get_growth_data(year, quarter)
    # 增加季度字段。
    data = concat_quarter(year, quarter, data)
    # 处理重复数据,保存最新一条数据。最后一步处理,否则concat有问题。
    data = data.drop_duplicates(subset="code", keep="last")
    # 插入数据库。
    db.insert_db(data, "ts_growth_data", True, "`quarter`,`code`")

    # 偿债能力
    data = ts.get_debtpaying_data(year, quarter)
    # 增加季度字段。
    data = concat_quarter(year, quarter, data)
    # 处理重复数据,保存最新一条数据。最后一步处理,否则concat有问题。
    data = data.drop_duplicates(subset="code", keep="last")
    # 插入数据库。
    db.insert_db(data, "ts_debtpaying_data", True, "`quarter`,`code`")

    # 现金流量
    data = ts.get_cashflow_data(year, quarter)
    # 增加季度字段。
    data = concat_quarter(year, quarter, data)
    # 处理重复数据,保存最新一条数据。最后一步处理,否则concat有问题。
    data = data.drop_duplicates(subset="code", keep="last")
    # 插入数据库。
    db.insert_db(data, "ts_cashflow_data", True, "`quarter`,`code`")
Exemple #26
0
    def _load_report(self,year,season):
        #从本地或者网上取得report
        #year: int
        #season: int,1,2,3,4
        report_date=None
        eps_yoy,roe,profits_yoy=0.0,0.0,0.0
        try:#本地有记录
            with open('report_'+str(year)+'_'+str(season)+'.pkl', "rb") as f:
                report = pickle.load(f)
        except:#本地没有存过
            with open('report_' + str(year) + '_' + str(season) + '.pkl', "wb") as f:
                report = ts.get_report_data(year, season)  # 获取业绩报表
                pickle.dump(report, f)
        for i in range(report.index.size):
            if(report.iloc[i,0]==self.ticker):
                #公布报告的日期:
                if(season==4):#如果是4季报,发布年份是下一年
                    report_year=year+1
                else:
                    report_year=year
                report_date = date(report_year,int(report.iloc[i,10][0:2]),int(report.iloc[i,10][3:5]))#先看看报告日期.形如'06-16'

                eps_yoy = report.iloc[i,3]/100
                roe = report.iloc[i,5]/100
                profits_yoy = report.iloc[i,8]/100
                break
        return report_date,[eps_yoy,roe,profits_yoy]
Exemple #27
0
def main():
    print("pd version:%s" %pd.__version__)
    print("tushare version:%s" %ts.__version__)
    '''
    data = ts.get_today_all()
    data['Profit yield'] = None

    for i in range(0, len(data.index)):
        if data.at[i, 'per'] == 0:  #可能停牌的股票
            continue

        #if 'ST' in data.at[i, 'name']:
        #    continue

        data.at[i, 'Profit yield'] = 1 / data.at[i, 'per']

    data.to_csv("./get_today_all.csv")
    '''
    databasic = ts.get_stock_basics()
    #databasic.sort_values(by = 'pb').to_csv("./get_today_all.csv")

    dta8_1 = ts.get_report_data(2018, 1)
    
    dtaROE = databasic['pe'] / dta8_1['roe']

    print(dtaROE)
Exemple #28
0
def call_report_v1(year, quarter):
    '''
    code,代码
    name,名称
    esp,每股收益
    eps_yoy,每股收益同比(%)
    bvps,每股净资产
    roe,净资产收益率(%)
    epcf,每股现金流量(元)
    net_profits,净利润(万元)
    profits_yoy,净利润同比(%)
    distrib,分配方案
    report_date,发布日期
    '''

    filePath = COMMEN_FILE_PATH + f'v1_report_{year}_{quarter}.csv'
    if not os.path.exists(filePath):
        df = ts.get_report_data(year, quarter)
        if df.empty:
            return df
        df.to_csv(filePath)
        if DEBUG:
            print('STORE:', filePath)
    else:
        df = pd.read_csv(filePath)
    return df
Exemple #29
0
 def get_temp_data(year, quarter):
     df1 = ts.get_report_data(year, quarter)
     #print (1)
     df1 = df1.merge(ts.get_profit_data(year, quarter),
                     how='inner',
                     on=['code', 'name'])
     #print (2)
     df1 = df1.merge(ts.get_operation_data(year, quarter),
                     how='inner',
                     on=['code', 'name'])
     #print (3, "n", df1)
     df1 = df1.merge(ts.get_growth_data(year, quarter),
                     how='inner',
                     on=['code', 'name'])
     #print (4)
     print(df1)
     df1 = df1.merge(ts.get_debtpaying_data(year, quarter),
                     how='inner',
                     on=['code', 'name'])
     #print (5)
     print(df1)
     df1 = df1.merge(ts.get_cashflow_data(year, quarter),
                     how='inner',
                     on=['code', 'name'])
     #print (6)
     (row, col) = df1.shape
     for i in range(0, row):
         df1.iloc[i, 0] = str(df1.iloc[i, 0])
     return df1
def download_ACH_Q(year, quarter, df):  #按照季度获取信息
    Data = df
    try:

        achievement = ts.get_report_data(year, quarter)
        achievement = pd.DataFrame(achievement)
    except:
        pass
    achievement = achievement.set_index('code')
    achievement = achievement.sort_index()
    for title_name in achievement.columns:
        print(title_name)
        for code_ in achievement.index[1:]:
            if achievement.at['%s' % code_, '%s' % title_name] != NaN:

                try:
                    #print(achievement.at['%s'%code_,'%s'title_name])
                    Data.ix['%s' % code_,
                            '%s-%s-%s' %
                            (title_name, year,
                             quarter)] = achievement.at['%s' % code_,
                                                        '%s' % title_name]
                except:
                    buf = achievement.at['%s' % code_, '%s' % title_name]
                    Data.ix['%s' % code_,
                            '%s-%s-%s' % (title_name, year, quarter)] = buf[0]
            else:
                pass
                #print (code_,title_name)
        #print (achievement.head(9))
    return Data
Exemple #31
0
def value_factor(end_year, season):
    '''
    计算价值因子
    '''
    #价值因子:每股收益与价格比率、每股经营现金流与价格比率、每股净资产与价格比率、股息收益率
    df_report_new = ts.get_report_data(
        end_year, season)[["name", "code", "eps", "epcf", "bvps"]]
    df_new_price = ts.get_today_all()[["name", "code", "settlement"]]
    file = "2005_2011.csv" if end_year in range(2005,
                                                2012) else "2012_2018.csv"
    df_interest = pd.read_csv(file, dtype={'code':
                                           str})[["code",
                                                  str(end_year)]]
    df_interest.columns = ["code", "interest_rate"]
    data = pd.merge(df_report_new, df_new_price, how='inner')
    data = pd.merge(data, df_interest, how='inner')

    # 去掉ST股
    data = data[data.name.map(lambda x: "ST" not in x)]

    data["eps_rate"] = data["eps"] / data["settlement"]
    data["epcf_rate"] = data["epcf"] / data["settlement"]
    data["bvps_rate"] = data["bvps"] / data["settlement"]

    #print("value_factor dataframe length: " + str(len(data)))
    return data[[
        "name", "code", "eps_rate", "epcf_rate", "bvps_rate", "interest_rate"
    ]].drop_duplicates().fillna(0.0)
Exemple #32
0
def get_report_data(year, season):
    if not available(year, season):
        return None

    print("get_report_data")
    save(ts.get_report_data(year, season), "basics/report_data", year, season)

    print("get_profit_data")
    save(ts.get_profit_data(year, season), "basics/profit_data", year, season)

    filename = "operation_data"
    print("get_operation_data")
    save(ts.get_operation_data(year, season), "basics/operation_data", year,
         season)

    filename = "growth_data"
    print("get_growth_data")
    save(ts.get_growth_data(year, season), "basics/growth_data", year, season)

    filename = "get_debtpaying_data"
    print("get_debtpaying_data")
    save(ts.get_debtpaying_data(year, season), "basics/debtpaying_data", year,
         season)

    filename = "get_debtpaying_data"
    print("get_cashflow_data")
    save(ts.get_cashflow_data(year, season), "basics/cashflow_data", year,
         season)
Exemple #33
0
    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
Exemple #34
0
def get_report_data(year, quarter):
    try:
        df = ts.get_report_data(year, quarter)
        engine = create_engine('mysql://*****:*****@127.0.0.1/stock?charset=utf8')
        df.to_sql('report_data', engine, if_exists='append')
        print "message"
    except Exception, e:
        e.message
Exemple #35
0
def sync_report_data():
	'''
	sync report data
	'''
	year = datetime.datetime.now().year
	month = datetime.datetime.now().month
	seaon = month/3
	if month<3:
		year = year - 1
		seaon = 4
	monthstr = '%s%s'%(year,seaon)
	DataFrameToMongo(ts.get_report_data(year, seaon), MongoClient(mongourl)['stoinfo']['report_data'], ['code'], monthstr)
Exemple #36
0
def getQuerterReport(year):
    for q in [1, 2, 3, 4]:
        getDownloaded()
        if not isDownloaded(reportByYear(year, q)):
            try:
                print('Report of ' + str(year) + ' and quarter is ' + str(q))
                report_df = ts.get_report_data(year, q)
                report_df.to_csv(path + reportByYear(year, q))
                time.sleep(5)
            except Exception:
                logError('Report: ' + str(year) + '-' + str(q) +
                         ' there is problem, will skip it. ')
Exemple #37
0
def collect_report_data(year, term):
    try:
        report_data_path = os.path.join(BASE_FOLDER, 'report')
        if not os.path.isdir(report_data_path):
            os.makedirs(report_data_path)
        path = os.path.join(report_data_path, '{}-{}.csv'.format(year, term))
        if not os.path.exists(path):
            df = ts.get_report_data(year, term)
            df.to_csv(path)
        return report_data_path
    except Exception as ex:
        print("error occurred in retrieving report data: ", ex)
        return None
    def profit(self):
        df_2016=ts.get_report_data(2016,4)

        #第四季度就是年报
        #df= df.sort_values('profits_yoy',ascending=False)
        #df.to_excel('profit.xls')
        df_2015=ts.get_report_data(2015,4)
        df_2016.to_excel('2016_report.xls')
        df_2015.to_excel('2015_report.xls')
        code_2015_lost=df_2015[df_2015['net_profits']<0]['code'].values
        code_2016_lost=df_2016[df_2016['net_profits']<0]['code'].values

        print code_2015_lost
        print code_2016_lost
        two_year_lost=[]
        #two_year_lost_name=[]
        for i in code_2015_lost:
            if i in code_2016_lost:
                print i,
                #name=self.base[self.base['code']==i].values[0]
                two_year_lost.append(i)

        self.saveList(two_year_lost,'st_dangours.csv')
def get_basic():
    hsdq = stock_info.ix['300141']
    print(hsdq)
    report = ts.get_report_data(2014, 1)
    print(report)

    # hsdq=stock_info.ix['300141']
    # print(hsdq)
    # report=ts.get_report_data(2014,1)
    # print(report)
    print('*' * 20)
    df = ts.get_today_all()
    zrkj = df[df['code'] == '300333']
    print(type(zrkj))
    print(type(zrkj['code']))
    print(zrkj['name'].values[0])
Exemple #40
0
def updatereport():
    reportdatalist=ts.get_report_data(2014,4)
    reportdata=pd.DataFrame(reportdatalist)
    conn= ms.connect(host='localhost',port = 3306,user='******', passwd='123456',db ='investment',charset="utf8")
    cur = conn.cursor()
    values=[]
    for index,row in reportdata.iterrows():
        if math.isnan(row['eps']):
            eps=0
        else:
            eps=row['eps']
         
        if math.isnan(row['eps_yoy']):
            eps_yoy=0
        else:
            eps_yoy=row['eps_yoy']
             
        if math.isnan(row['bvps']):
            bvps=0
        else:
            bvps=row['bvps']
             
        if math.isnan(row['epcf']):
            epcf=0
        else:
            epcf=row['epcf']
         
        if math.isnan(row['roe']):
            roe=0
        else:
            roe=row['roe'] 
               
        if math.isnan(row['net_profits']):
            net_profits=0
        else:
            net_profits=row['net_profits']  
             
        if math.isnan(row['profits_yoy']):
            profits_yoy=0
        else:
            profits_yoy=row['profits_yoy']  
        values.append((row['code'],row['name'],eps,eps_yoy,bvps,roe,epcf,net_profits,profits_yoy,'2015-'+row['report_date']))
    cur.executemany('insert into report20144 (code,name,eps,eps_yoy,bvps,roe,epcf,net_profis,profis_yoy,report_date) values(%s,%s,%s,%s,%s,%s,%s,%s,%s,%s)',values)
    conn.commit()
    cur.close()
    conn.close()
Exemple #41
0
def update_basics():
    basics = ts.get_stock_basics()
    f = os.path.join(base_dir, 'basics.h5')
    basics.to_hdf(f, 'basics')

    today = datetime.date.today()
    current_year = today.year
    current_season = today.month / 3
    if current_season == 0:
        current_year -= 1
        current_season = 4
    length = 4 * 5

    year = current_year
    season = current_season
    for i in range(length):
        f = os.path.join(base_dir, 'basics-{0}-{1}.h5'.format(year, season))
        if os.path.exists(f):
            continue
        print(f)
        report = ts.get_report_data(year, season)
        report.to_hdf(f, 'report')

        profit = ts.get_profit_data(year, season)
        profit.to_hdf(f, 'profit')

        operation = ts.get_operation_data(year, season)
        operation.to_hdf(f, 'operation')

        growth = ts.get_growth_data(year, season)
        growth.to_hdf(f, 'growth')

        debtpaying = ts.get_debtpaying_data(year, season)
        debtpaying.to_hdf(f, 'debtpaying')

        cashflow = ts.get_cashflow_data(year, season)
        cashflow.to_hdf(f, 'cashflow')

        season -= 1
        if season == 0:
            season = 4
            year -= 1
Exemple #42
0
def finance_report(year=2018, quarter=2):
    latest_equity = Equity.objects().order_by('-date').first()
    # print(latest_equity)
    date = latest_equity.date

    df = ts.get_report_data(year, quarter)
    print(df)
    data = df.to_dict('index')
    print(data)
    print(len(data.items()))
    from mongoengine.queryset.visitor import Q
    for index, value in sorted(data.items()):
        code = value['code']
        name = value['name']
        roe = value['roe']
        eps = value['eps']
        report_date = value['report_date']
        # print('code:{} roe:{}'.format(code, roe))
        FinanceReport.objects(code=code, year=year, quarter=quarter).update_one(code=code, name=name,
                                                                                year=year, quarter=quarter,
                                                                                report_date=report_date,
                                                                                roe=roe, eps=eps, upsert=True)
    def store_data(self):
        # 预测
        # year_2016=ts.forecast_data(2016, 4)
        # self.save_to_excel(year_2016,'2016-profit.xls')

        # year_2017=ts.forecast_data(2017, 4)
        # self.save_to_excel(year_2017,'2017-profit.xls')
        # 盈利能力
        # profit_2016=ts.get_profit_data(2016,4)
        # profit_2017=ts.get_profit_data(2017,3)
       # self.save_to_excel(profit_2016, '2016-profit.xls')
        # self.save_to_excel(profit_2017, '2017-3rdprofit.xls')
        # 股票基本信息
        # basic=ts.get_stock_basics()
        # basic.to_csv('temp.xls',encoding='gbk')
        # df=pd.read_csv('temp.xls',encoding='gbk',dtype={'code':str})
        # # print df
        # self.save_to_excel(df,'Markets.xls')

        # 基本面 每股净资产<1
        df=ts.get_report_data(2017, 3)
        self.save_to_excel(df,'2017-3rd-report.xls')
"""
Created on Wed Oct 07 09:21:05 2015

@author: Fuqian
"""

from sqlalchemy import create_engine
import tushare as ts
#import pymongo
import pandas as pd



df_base = ts.get_stock_basics()

df_report_1503 = ts.get_report_data(2015,3)

df_profit_1503 = ts.get_profit_data(2015,3)

df_growth_1503 = ts.get_growth_data(2015,3)

#detail_daily={}

engine = create_engine('mysql://*****:*****@127.0.0.1/stock?charset=utf8')


for row_index, row in df_base.iterrows():
    try:
        f = open('qfq_err', 'a')
        f_d = open('detailDay_err','a')
        
#coding:utf-8
from sqlalchemy import create_engine
import tushare as ts
# define engine
engine = create_engine('mysql://*****:*****@127.0.0.1/tushare?charset=utf8')

df = ts.get_report_data(2014,4)
df = df.assign(quater=20144)
df.to_sql('report_data',engine, if_exists='append')
df = ts.get_report_data(2015,4)
df = df.assign(quater=20154)
df.to_sql('report_data',engine, if_exists='append')
df = ts.get_report_data(2016,4)
df = df.assign(quater=20164)
df.to_sql('report_data',engine, if_exists='append')

Exemple #46
0
    def pick_data(self, max_num_threads = 20, pause = 0):
        """
        pick all necessary data from local database and from internet for loaded stocks. This function will take a while.
        """
        logging.info('getting basics from tushare')
        self._init_stock_objs()

        # self.data_manager.drop_stock()
        # self.stocks = {key: self.stocks[key] for key in ['600233', '600130']}
        logging.info('totally there are %d listed companies' % len(self.stocks))

        logging.info('get indexes from tushare')
        self._get_indexes()

        # self._pick_hist_data_and_save(self.stocks, False, self.indexes['000001'].hist_start_date, max_num_threads)

        logging.info('getting last stock trading data')
        df = ts.get_today_all()
        self._extract_from_dataframe(df,
                    ignore=('changepercent', 'open', 'high', 'low', 'settlement', 'volume', 'turnoverratio', 'amount'),
                    remap={'trade': 'price', 'per': 'pe'})

        # calculate the report quarter
        report_year, report_quarter = ts.get_last_report_period()

        logging.info('getting last report (%d quarter %d) from tushare' % (report_year, report_quarter))
        df = ts.get_report_data(report_year, report_quarter)
        self._extract_from_dataframe(df)

        logging.info('getting last profit data from tushare')
        df = ts.get_profit_data(report_year, report_quarter)
        self._extract_from_dataframe(df, ignore=('net_profits', 'roe', 'eps'))

        logging.info('getting last operation data from tushare')
        df = ts.get_operation_data(report_year, report_quarter)
        self._extract_from_dataframe(df)

        logging.info('getting last growth data from tushare')
        df = ts.get_growth_data(report_year, report_quarter)
        self._extract_from_dataframe(df)

        logging.info('getting last debtpaying data from tushare')
        df = ts.get_debtpaying_data(report_year, report_quarter)
        self._extract_from_dataframe(df)

        logging.info('getting last cashflow data from tushare')
        df = ts.get_cashflow_data(report_year, report_quarter)
        self._extract_from_dataframe(df)

        logging.info('getting history trading data from tushare')
        start_from = self.indexes['000001'].hist_start_date
        data_full = self._pick_hist_data_and_save(self.stocks, False, start_from, max_num_threads, pause)  # anything that pulling data must before here

        self._remove_unavailable_stocks()

        '''
        # calculate qianfuquan data
        # deprecated due to precision issue

        for code, stock in self.stocks.items():
            for i in range(1, len(stock.hist_data.index)-1):
                b = stock.hist_data.at[stock.hist_data.index[i], 'close']
                a = stock.hist_data.at[stock.hist_data.index[i+1], 'close']
                p = stock.hist_data.at[stock.hist_data.index[i+1], 'p_change'] / 100.0

                q = (p*a+a)/b
                if q > 1.1:
                    print('%s chuq-uan %s: %s %s %s, 1/%s' % (stock, stock.hist_data.index[i], b, a, p, q))
        '''

        return data_full
import tushare as ts
import marshal, pickle

year = 2015

print(report_data)

report_data = ts.get_report_data(year, 4)
f = file('report_data', 'w')
pickle.dump(report_data, f)
f.close()

profit_data = ts.get_profit_data(year, 4)
f = file('profit_data', 'w')
pickle.dump(profit_data, f)
f.close()

growth_data = ts.get_growth_data(year, 4)
f = file('growth_data', 'w')
pickle.dump(growth_data, f)
f.close()
Exemple #48
0
# -*- coding: utf8 -*-

import tushare as ts

print(ts.__version__)
report = ts.get_report_data(2016, 1)
report['year'] = 2016
report['quarter'] = 1
for y in range(2005, 2016):
    for q in range(1, 5):
        print(y, q)
        r = ts.get_report_data(y, q)
        r['year'] = y
        r['quarter'] = q
        report = report.append(r, ignore_index=True)

report.to_csv('report05Q4-16Q1.csv')
Exemple #49
0
#coding=utf-8
import tushare as ts

# 获取沪深上市公司基本情况
df = ts.get_stock_basics()
date = df.ix['600848']['timeToMarket']#上市日期YYYYMMDD

#获取2014年第3季度的业绩报表数据
ts.get_report_data(2014,3)

#获取2014年第3季度的盈利能力数据
ts.get_profit_data(2014,3)


#获取2014年第3季度的营运能力数据
ts.get_operation_data(2014,3)


#获取2014年第3季度的成长能力数据
ts.get_growth_data(2014,3)

#获取2014年第3季度的偿债能力数据
ts.get_debtpaying_data(2014,3)

#获取2014年第3季度的现金流量数据
ts.get_cashflow_data(2014,3)
def download_report_info(file_path, year, quarter):
    report_data = ts.get_report_data(year, quarter)
    if report_data is not None:
        report_data.to_csv(file_path + 'report_' + str(year) + '_' + str(quarter) + '.csv', encoding='utf-8')
Exemple #51
0
   `eps` double NOT NULL DEFAULT '0' COMMENT '每股收益',\
   `eps_yoy` double NOT NULL DEFAULT '0' COMMENT '每股收益同比(%)',\
   `bvps` double NOT NULL DEFAULT '0' COMMENT '每股净资产',\
   `roe` double NOT NULL DEFAULT '0' COMMENT '净资产收益率(%)',\
   `epcf` double NOT NULL DEFAULT '0' COMMENT '每股现金流量(元)',\
   `net_profits` double NOT NULL DEFAULT '0' COMMENT '净利润(万元)',\
   `profits_yoy` double NOT NULL DEFAULT '0' COMMENT '净利润同比(%)',\
   `distrib` char(30) NOT NULL DEFAULT '' COMMENT '分配方案',\
   `report_date` char(10) NOT NULL DEFAULT '' COMMENT '发布日期',\
   PRIMARY KEY (`id`),\
   UNIQUE KEY `code` (`code`)\
) ENGINE=MyISAM DEFAULT CHARSET=utf8 COMMENT='业绩报告(主表)';"
cursor.execute(sql)

try:
    df = ts.get_report_data(year, season)
    if df is not None:
        for id in df.index:
            temp = df.ix[id]
            
            code = temp['code']
            
            # 以下先判断temp['']是否为NAN
            if temp['eps'] != temp['eps']:
                eps = 0.00
            else:
                eps = temp['eps']
            if temp['eps_yoy'] != temp['eps_yoy']:
                eps_yoy = 0.00
            else:
                eps_yoy = temp['eps_yoy']
# v_ma10:10日均量
# v_ma20:20日均量
# turnover:换手率[注:指数无此项] 4/4

raw_data_profit = ts.get_profit_data()
# code,代码
# name,名称
# roe,净资产收益率(%) 7
# net_profit_ratio,净利率(%)
# gross_profit_rate,毛利率(%)
# net_profits,净利润(万元)
# eps,每股收益
# business_income,营业收入(百万元)
# bips,每股主营业务收入(元)

raw_data_report = ts.get_report_data()

# code,代码
# name,名称
# eps,每股收益
# eps_yoy,每股收益同比(%)
# bvps,每股净资产
# roe,净资产收益率(%)
# epcf,每股现金流量(元)
# net_profits,净利润(万元)
# profits_yoy,净利润同比(%)
# distrib,分配方案
# report_date,发布日期

raw_data_growth = ts.get_growth_data()
Exemple #53
0
import tushare as ts
import sys
df = ts.get_report_data(int(sys.argv[1]), int(sys.argv[2]))
df.to_csv(sys.argv[3], encoding="utf8")
Exemple #54
0
####################
# Financial Report #
####################

# FinancialReport: EPS, EPS_YOY, ROE, net_profits, profits_yoy
# ProfitData: ROE, net_profit_ratio, gross_profit_rate, EPS, bips (business income per share)
# GrowthData: mbrg (main business rate growth), nprg (net profit), 
#             nav, targ (total asset), epsg, seg (shareholder's eqty)
# DebtPayingData: currentratio, quickratio, cashratio, icratio (interest coverage)


# TODO Data is available quarterly
# TODO Compare data for FinancialReport and ProfitData

FinancialData = ts.get_report_data(CURRENT.year, np.floor((CURRENT.month+2)/3)-1)
FinancialData = FinancialData.set_index('code')
FinancialData = FinancialData.drop(['name', 'bvps', 'distrib', 'epcf', 'report_date'], axis = 1)
FinancialData.to_csv('./ASHR/DATA/FinancialData_2015_1.csv', index = True)

ProfitData = ts.get_profit_data(CURRENT.year, np.floor((CURRENT.month+2)/3)-1)
ProfitData = ProfitData.set_index('code')
ProfitData = ProfitData.drop(['name', 'business_income', 'net_profits'], axis = 1)
ProfitData.to_csv('./ASHR/DATA/ProfitData_2015_1.csv', index = True)

GrowthData = ts.get_growth_data(CURRENT.year, np.floor((CURRENT.month+2)/3)-1)
GrowthData = GrowthData.set_index('code')
GrowthData = GrowthData.drop(['name'], axis = 1)
GrowthData.to_csv('./ASHR/DATA/GrowthData_2015_1.csv', index = True)

DebtPayingData = ts.get_debtpaying_data(CURRENT.year, np.floor((CURRENT.month+2)/3)-1)
Exemple #55
0
df = ts.get_stock_basics()
df.to_sql('basic_info',engine, if_exists='replace')

# import report data
#code,代码
#name,名称
#esp,每股收益
#eps_yoy,每股收益同比(%)
#bvps,每股净资产
#roe,净资产收益率(%)
#epcf,每股现金流量(元)
#net_profits,净利润(万元)
#profits_yoy,净利润同比(%)
#distrib,分配方案
#report_date,发布日期
df = ts.get_report_data(lastYear,lastSeason)
df = df.assign(quater=lastQuater)
df.to_sql('report_data',engine, if_exists='replace')
df = ts.get_report_data(currentYear,currentSeason)
df = df.assign(quater=currentQuater)
df.to_sql('report_data',engine, if_exists='append')

# import profit data
#code,代码
#name,名称
#roe,净资产收益率(%)
#net_profit_ratio,净利率(%)
#gross_profit_rate,毛利率(%)
#net_profits,净利润(万元)
#esp,每股收益
#business_income,营业收入(百万元)