def get_history_data(date):
df = pd.read_csv(HISTORY_FILE)
data = df.loc[df[STOCK_NAMES[0]] == date_utils.lxrDate2csvDate(date)]
result = data.values
if len(result) > 0:
return result[0]
else:
return None
# result = get_history_data('2012')
# print(result)
# y_axises = []
# generate_history_csv()
# import wendu_trend_chart
# wendu_trend_chart.generate(STOCK_NAMES[1:], date_index, y_axises)
def get_result(date):
result = {}
# 遍历指数集合，计算每一个指数的温度及50日均线信号（指定日期50均值与收盘价比较）
for index, stock in enumerate(constants.STOCK_CODES):
# 读取数据目录下的指数csv文件，获取当前指数数据
df = read_data(stock)
# 取得指定日期行的数据
if date is None:
date = date_utils.csvDate2lxrDate(df[constants.DATE][0])
# 取得指定日期之前最新一天的数据
data_result = df.loc[
df[constants.DATE].map(lambda x: date_utils.csvDate2lxrDate(x)) <=
date_utils.lxrDate2csvDate(date)]
if len(data_result.values) == 0:
print("指定日期不存在", stock, date)
return date, None
# 买点1：判断50日均线突破，当天收盘价是否突破50日均线
# 买点2：判断100日均线突破，当天收盘价是否突破100日均线
# 卖点1，卖点2，同上述相反
# 连续突破情况，首日突破表示买入，后续情况多仓、持仓
# 上一交易日50日均线信号
data_index = data_result.index.values[0]
data = data_result.loc[data_index]
data_before = data_result.loc[data_index + 1]
fifty_signal = dp.check_signal(data_before, data,
constants.FIFTY_MEDIAN, 1)
hundred_signal = dp.check_signal(data_before, data,
constants.HUNDRED_MEDIAN, 2)
result_stock = {
'name': constants.STOCK_NAMES[index],
constants.DATE: data[constants.DATE],
constants.PB_PERCENTILE: data[constants.PB_PERCENTILE],
constants.PE_PERCENTILE: data[constants.PE_PERCENTILE],
constants.FIFTY_SIGNAL: fifty_signal,
constants.HUNDRED_SIGNAL: hundred_signal
}
print('指数数据:', date, result_stock)
result[stock] = result_stock
return date, result
def data_download_stock(stock, url):
# 请求数据
request_data = {
"token": constants.TOKEN,
"stockCodes": [stock],
"metricsList": ["pb", "pe_ttm", "sp"],
"startDate": "2000-01-01"
}
result = requests.post(url, json=request_data)
pbs = []
pes = []
cps = []
date = []
if result.status_code == 200 and result.json()['message'] == 'success':
for data in result.json()['data']:
if 'sp' in data.keys():
cp = data['sp']
cps.append(cp)
else:
continue
split_date = data['date'].split('T')
date.append(dateutils.lxrDate2csvDate(split_date[0]))
pb = ''
if 'pb' in data.keys():
pb = data['pb']
pbs.append(pb)
pe = ''
if 'pe_ttm' in data.keys():
pe = data['pe_ttm']
pes.append(pe)
df = pd.DataFrame()
df.insert(0, constants.DATE, date)
df.insert(1, constants.CP, cps)
df.insert(2, constants.PB, pbs)
df.insert(3, constants.PE, pes)
return df
def generate(stock, input_date, show_signal):
stock_name, df = download(stock)
if stock_name is None or df is None:
print("没有获取到数据，请确认代码输入是否正确")
return
print('生成股票温度图：', stock_name)
date = input_date
# 取得指定日期行的数据
if date is None:
date = date_utils.csvDate2lxrDate(df[constants.DATE][0])
data = df.loc[df[constants.DATE] == date_utils.lxrDate2csvDate(date)]
if len(data.values) == 0:
print("指定日期不存在", stock, date)
return
# 买点1：判断50日均线突破，当天收盘价是否突破50日均线
# 买点2：判断100日均线突破，当天收盘价是否突破100日均线
# 卖点1，卖点2，同上述相反
# 连续突破情况，首日突破表示买入，后续情况多仓、持仓
# 上一交易日50日均线信号
data_index = data.index.values[0]
data = df.loc[data_index]
data_before = df.loc[data_index + 1]
fifty_signal = dp.check_signal(data_before, data, constants.FIFTY_MEDIAN,
1)
hundred_signal = dp.check_signal(data_before, data,
constants.HUNDRED_MEDIAN, 2)
result_stock = {
constants.PB_PERCENTILE: data[constants.PB_PERCENTILE],
constants.PE_PERCENTILE: data[constants.PE_PERCENTILE],
constants.FIFTY_SIGNAL: fifty_signal,
constants.HUNDRED_SIGNAL: hundred_signal
}
generate_chart(stock, stock_name, date, result_stock, show_signal)
def get_new_data(stock, start_date_fix):
new_data = pd.DataFrame()
result_data = download_data(stock, start_date_fix)
dates = []
pbs = []
pes = []
cps = []
for data in result_data:
if start_date_fix == data[constants.DATE]:
# 开始日期的数据在csv中已存在，不需要再放入新的data frame
continue
pes.append(data[constants.PE])
pbs.append(data[constants.PB])
cps.append(data[constants.CP])
dates.append(date_utils.lxrDate2csvDate(data[constants.DATE]))
# 将新的数据插入到data frame，然后返回
new_data.insert(0, constants.DATE, dates)
new_data.insert(1, constants.CP, cps)
new_data.insert(2, constants.PB, pbs)
new_data.insert(3, constants.PE, pes)
return new_data