# -*- coding: utf-8 -*-
"""
Created on Wed Nov 9 11:01:04 2016
@author: admin
"""
from multiprocessing import Queue
import time
import signal
from sdk import *
import sys
import os
import pandas as pd
import numpy as np
from pandas import Series, DataFrame
import glob
import re
import time
# 循环处理各个月份
for i in range(1, 2):
#进行月份的转化
s = str(i).zfill(2)
#选取每一年的时间
time_of_this_year = '2016'
time_of_this_date = 20160203
s1 = 'DZH-SZ' + time_of_this_year + s + '-TXT'
all_date_list = glob.glob(r'E:\股票\股票l1\股票5档分笔%s年TXT\%s\*' % (time_of_this_year,s1)) # 获取一个月里面所有的天数列表
sdk = SDK('https://api.kuaiyudian.com/rpc') # 获取前一日数据
#定义数据表
preClosePriceDic = {}
#对每一天的数据循环进行处理
for each_day_list in all_date_list:
try:
current_date = re.search(r'\d{8}$', each_day_list).group() # 截取每一个地址的日期
current_date_font_change = time.strftime('%Y-%m-%d', time.strptime(current_date, '%Y%m%d')) # 转换日期格式
except:
print('we can not get the right date' + current_date)
continue
#对特定日期进行处理
if int(current_date) < time_of_this_date:
print(current_date + 'is proceeding')
data = sdk.get_bars_by_date(current_date_font_change) # 获取收盘价
for d in data: # 循环data数据，截取昨日收盘价
preClosePriceDic[d['symbol']] = d['preClosePrice']
n2 = r'E:\股票\股票l1\股票5档分笔%s年TXT\%s\%s\*.txt' % (time_of_this_year,s1, current_date) # 将时间导入到新的地址里面
f = glob.glob(n2) # 获取一天里面所有数据的列表
#对一天内所有的数据进行处理
for k in f:
try:
# 由于数据列标题有时候后会出现改变，所以把日期和时间改为位置1和位置2
k2 = re.search(r'\d{6}[_]', k) # 截取数据的名字
stockcode = (re.search(r'\d{6}', k2.group())).group() # 股票名称
# print(stockcode + 'is proceding')
# print('this----->' + k + 'can not been cleaned')
stocksymbol = 'sz' + stockcode
except:
print('we can not get this file' + k)
continue
try:
if stocksymbol in preClosePriceDic.keys():
# 读取每一个txt格式数据到df
df = pd.read_csv(k, sep=',', encoding='gb2312', parse_dates={'timestamp': [0, 1]})
# print('we can read'+ stocksymbol)
df['preClosePrice'] = float(preClosePriceDic[stocksymbol])
df['timestamp'] = df['timestamp'].astype(np.int64) // 10 ** 9
if stocksymbol == 'sh603633':
df['hardenPrice'] = round(df['preClosePrice'] * 1.44, 3) # 新股则乘以1.44
else:
df['hardenPrice'] = round(df['preClosePrice'] * 1.1, 3) # 旧股则乘以1.1
#由于特殊值的出现，需要对数据进行容错化处理
if '额' in df.columns:
df.rename(columns={'额': 'amount'}, inplace=True)
elif '额、属性(持仓增减)' in df.columns:
df.rename(columns={'额、属性(持仓增减)': 'amount'}, inplace=True)
df['dehardenPrice'] = round(df['preClosePrice'] * 0.9, 3)
df['highPrice'] = df['成交价'].expanding().max()
df['lowPrice'] = df['成交价'].expanding().min()
df['openPrice'] = df.get_value(0, '成交价') # 开盘价 如何获取当前日的第一个价格
df['symbol'] = stocksymbol # 手动。。。。。如何取出名字
df['stockName'] = stockcode # 手动。。。。如何将其进行一一对应
df['orderNum'] = 5 # 固定为5
df['status'] = 'PZ' # 分比数据固定为PZ
df.rename(columns={
'成交价': 'nowPrice',
'总量': 'volume',
'B1价': 'bidv1',
'B1量': 'bidp1',
'B2价': 'bidv2',
'B2量': 'bidp2',
'B3价': 'bidv3',
'B3量': 'bidp3',
'B4价': 'bidv4',
'B4量': 'bidp4',
'B5价': 'bidv5',
'B5量': 'bidp5',
'S1价': 'askv1',
'S1量': 'askp1',
'S2价': 'askv2',
'S2量': 'askp2',
'S3价': 'askv3',
'S3量': 'askp3',
'S4价': 'askv4',
'S4量': 'askp4',
'S5价': 'askv5',
'S5量': 'askp5', }, inplace=True)
# 重新排序
df2 = df.reindex(columns=['symbol',
'stockName',
'timestamp',
'preClosePrice',
'hardenPrice',
'dehardenPrice',
'openPrice',
'highPrice',
'lowPrice',
'nowPrice',
'volume',
'amount',
# 'askVolume',
# 'bidVolume',
# 'totalTrades', #当日成交笔数
'orderNum',
'status',
'bidv1',
'bidp1',
'bidv2',
'bidp2',
'bidv3',
'bidp3',
'bidv4',
'bidp4',
'bidv5',
'bidp5',
'askv1',
'askp1',
'askv2',
'askp2',
'askv3',
'askp3',
'askv4',
'askp4',
'askv5',
'askp5',
])
#生产新的文件
newpath = r'E:\stock_data_rx_daily-sparate\rx_data_sz_%s/%s/' % (time_of_this_year,current_date)
if not os.path.exists(newpath):
os.makedirs(newpath)
df2.to_csv(newpath + '/%s.csv' % (stockcode), index=False, header=False)
except:
print('we can not change the right function of this' + k)
continue
#rx_all_data_list = glob.glob(r'E:\stock_data_rx_daily-sparate\rx_data_sh_%s\*'%(time_of_this_year))
print('we are emerging date with ' +current_date)
outfile = open(r'E:\stock_data_rx\SZ\%s\%s.csv'%(time_of_this_year,current_date),'a')
g = glob.glob(r'E:\stock_data_rx_daily-sparate\rx_data_sz_%s\%s\*.csv' % (time_of_this_year,current_date))
for n in g:
inputfile = open(n, 'r')
# print 'we can open this file' + n
for line in inputfile:
outfile.write(line)
outfile.close()
网友评论