def process_dir2file(w, mydir, myfile):
df = read_data_dataframe(myfile)
all_set = set(df.index)
for dirpath, dirnames, filenames in os.walk(mydir):
for filename in filenames:
# 这个日期需要记得修改
if filename < "2017-01-01.csv":
continue
filepath = os.path.join(dirpath, filename)
df1 = read_constituent(filepath)
# print(filepath)
if df1 is None:
continue
if df1.empty:
continue
curr_set = set(df1['wind_code'])
diff_set = curr_set - all_set
if len(diff_set) == 0:
continue
print(filepath)
df2 = download_ipo_last_trade_trading(w, list(diff_set))
df = pd.concat([df, df2])
all_set = set(df.index)
# 出于安全考虑，还是每次都保存
write_data_dataframe(myfile, df)
df['wind_code'] = df.index
df.sort_values(by=['ipo_date', 'wind_code'], inplace=True)
del df['wind_code']
write_data_dataframe(myfile, df)
def download_000300(w, trading_days):
# 4月8号是指数的发布日期
trading_days = trading_days['2005-04-08':date_str]
# 下载多天数据,以另一数据做为标准来下载
# 比如交易数据是10月8号，那就得取10月7号，然后再平移到8号，如果7号没有数据那就得9月30号
path = os.path.join(__CONFIG_H5_STK_FACTOR_DIR__, 'test.csv')
date_index = read_data_dataframe(path)
# dates = list(date_index.index)
download_index_weight(w, trading_days, "000300.SH",
__CONFIG_H5_STK_WEIGHT_DIR__)
from kquant_data.api import all_instruments
from kquant_data.wind_resume.wsd import download_daily_at
from kquant_data.config import __CONFIG_H5_STK_FACTOR_DIR__, __CONFIG_H5_STK_DIR__
if __name__ == '__main__':
w.start()
path = os.path.join(__CONFIG_H5_STK_DIR__, '1day', 'Symbol.csv')
Symbols = all_instruments(path)
wind_codes = list(Symbols['wind_code'])
# 下载多天数据,以另一数据做为标准来下载
# 比如交易数据是10月8号，那就得取10月7号，然后再平移到8号，如果7号没有数据那就得9月30号
path = os.path.join(__CONFIG_H5_STK_FACTOR_DIR__, 'roe.csv')
date_index = read_data_dataframe(path)
field = 'pb_lf'
df = None
for i in range(len(date_index)):
print(date_index.index[i])
date_str = date_index.index[i].strftime('%Y-%m-%d')
df_new = download_daily_at(w, wind_codes, field, date_str,
"Days=Alldays")
if df is None:
df = df_new
else:
df = pd.concat([df, df_new])
path = os.path.join(__CONFIG_H5_STK_FACTOR_DIR__, '%s.csv' % field)
if __name__ == '__main__':
w.start()
root_path = os.path.join(__CONFIG_H5_OPT_DIR__, 'optioncontractbasicinfo',
'510050.SH.csv')
# 这个只要执行一次就可以了，以后每次更新时都跳过即可
if False:
df_new = download_optioncontractbasicinfo(w, status='delisted')
df_new = df_new.set_index('wind_code')
write_data_dataframe(root_path, df_new)
# 下载新数据，并合并上旧数据
if True:
df_old = read_data_dataframe(root_path)
df_old['listed_date'] = pd.to_datetime(df_old['listed_date'])
df_old['expire_date'] = pd.to_datetime(df_old['expire_date'])
df_new = download_optioncontractbasicinfo(w, status='trading')
df_new = df_new.set_index('wind_code')
df_new['listed_date'] = pd.to_datetime(df_new['listed_date'])
df_new['expire_date'] = pd.to_datetime(df_new['expire_date'])
df = pd.concat([df_old, df_new])
# df = df_old
# 将老数据删除，因为合约可能因为除权而变数据
df = df[~df.index.duplicated(keep='last')]
write_data_dataframe(root_path, df)
df = read_optioncontractbasicinfo(root_path)
"""
import os
import pandas as pd
from kquant_data.config import __CONFIG_H5_FUT_FACTOR_DIR__
from kquant_data.xio.csv import read_data_dataframe, write_data_dataframe
from kquant_data.future.symbol import wind_code_2_InstrumentID
if __name__ == '__main__':
input_path = os.path.join(__CONFIG_H5_FUT_FACTOR_DIR__, 'trade_hiscode')
output_path = os.path.join(__CONFIG_H5_FUT_FACTOR_DIR__,
'trade_hiscode.csv')
df_csv = pd.DataFrame(columns=['trade_hiscode'])
for dirpath, dirnames, filenames in os.walk(input_path):
for filename in filenames:
shotname, extension = os.path.splitext(filename)
dirpath_filename = os.path.join(dirpath, filename)
print(dirpath_filename)
_df = read_data_dataframe(dirpath_filename)
_df.dropna(inplace=True)
df_csv.loc[shotname] = None
df_csv.loc[shotname]['trade_hiscode'] = _df.iat[-1, 0]
df = wind_code_2_InstrumentID(df_csv, 'trade_hiscode')
df.index.name = 'product'
write_data_dataframe(output_path, df)
print(output_path)
# print(df)
debug = 1