1 #该脚本可以提取沪深两市上市公司股票信息,并按以下信息分类:(1)当天股价创近10个交易日新高的股票;(2)停牌的股票;(3)复牌不超过一个交易日或者新发行的股票
2 #将分类后的股票及其信息(股价新高、当前状态等)存入mysql数据库
3 from time import *
4 import pandas as pd
5 import tushare as ts
6 from datetime import date
7 import datetime
8 import MySQLdb
9 import sys
10 reload(sys)
11 sys.setdefaultencoding('utf8') #将编码方式默认改为utf-8
12 #-------------------------------定义函数:计算当天往前推Deltadays个交易日所对应的日期--------------------------
13 def com_StartDate(Deltadays):
14 i=0
15 da=date.today()
16 while i!=Deltadays:
17 da=da-datetime.timedelta(days=1)
18 if da.isoweekday()==6 or da.isoweekday()==7: #判断当前日期是否为周末(此处未考虑法定节假日,后续改进)
19 pass
20 else:
21 i+=1
22 return da.strftime("%Y-%m-%d") #返回值数据类型为string
23
24 #--------------定义函数:判断某支股票处于什么状态,并返回股票代码、名字和股价(对于创新高的股票)等信息--------
25 def stock_info(stockID,startdate,index):
26 today=date.today()
27 df=ts.get_hist_data(stockID,start=startdate,end=today.strftime("%Y-%m-%d")) #获取股票信息
28 if type(df)!=type(None) and len(df.index)>1: #抓取时间周期内,有数据天数低于2天的,算作新股
29 df=df.reindex(index=index,method='ffill',fill_value=int(-1))
30 period_high=df['high'].max()
31 today_high=df.ix[today.strftime("%Y-%m-%d")]['high']
32 if today_high>=period_high and today_high>0 :
33 stock_status=int(0) #非新股,未停牌,股价创新高
34 return stock_status,df
35 elif today_high==int(-1):
36 stock_status=int(1) #非新股,但停牌
37 return stock_status,df
38 else:
39 stock_status=int(2) #非新股,未停牌,股价未创新高
40 return stock_status,df
41 else:
42 stock_status=int(3) #新股或复牌不超过1天的股票
43 return stock_status,df
44
45 #-----------------------------------------------获取沪深两市所有 上市公司基本信息---------------------------
46 all_stocks_info=ts.get_stock_basics()
47 Deltadays=10
48 startdate=com_StartDate(Deltadays)
49 newRecStocksID=[] #股票价格创新高的股票代码列表
50 newRecStocksPrice=[] #创新高的股票价格列表
51 newRecStocksName=[] #股票价格创新高的股票名称列表
52 suspendedStocksID=[] #停牌股票代码列表
53 suspendedStocksName=[] #停牌股票名称列表
54 newORresumeStocksID=[] #新发行或者当天复牌的股票代码列表
55 newORresumeStocksName=[] #新发行或者当天复牌的股票名称列表
56 today=date.today()
57 index=ts.get_hist_data('002625',start=startdate,end=today.strftime("%Y-%m-%d"),retry_count=10,pause=1).index
58 for i in all_stocks_info.index:
59 stock_status,stockinfo=stock_info(str(i), startdate,index)
60 if stock_status==int(0): #当日股价创Deltadays天内新高的股票
61 newRecStocksID.append(str(i).zfill(6))
62 newRecStocksName.append(all_stocks_info.ix[str(i),'name'])
63 newRecStocksPrice.append(stockinfo.ix[today.strftime("%Y-%m-%d"),'high'])
64 elif stock_status==int(1): #已停牌的股票
65 suspendedStocksID.append(str(i).zfill(6))
66 suspendedStocksName.append(all_stocks_info.ix[str(i), 'name'])
67 elif stock_status==int(2): #未停牌,且股价未创Deltadays天内新高的股票
68 pass
69 else: #该股票为新发行或者复牌不超过一天的股票
70 newORresumeStocksID.append(str(i).zfill(6))
71 newORresumeStocksName.append(all_stocks_info.ix[str(i), 'name'])
72
73 #-------------------------------------------数据写入DataFrame----------------------------------------------
74 newRecStocks=pd.DataFrame({'stockID':newRecStocksID,'stockname':newRecStocksName,'high_price':newRecStocksPrice,'status':['NewRecord']*len(newRecStocksID)})
75 suspendedStocks=pd.DataFrame({'stockID':suspendedStocksID,'stockname':suspendedStocksName,'status':['Suspended']*len(suspendedStocksID)})
76 newORresumeStocks=pd.DataFrame({'stockID':newORresumeStocksID,'stockname':newORresumeStocksName,'status':['New or Resume']*len(newORresumeStocksID)})
77
78 #-----------------------------------将数据存入MySQL数据库,数据库名称:stocks------------------------------
79 db=MySQLdb.connect('localhost','root','900417','stocks',charset='utf8')
80 pd.io.sql.to_sql(frame=newRecStocks,con=db,name=today.strftime("%Y%m%d")+'newRecordStocks',flavor='mysql',schema=None,if_exists='replace',index=True,index_label=None,chunksize=1000,dtype=None)
81 pd.io.sql.to_sql(frame=suspendedStocks,con=db,name=today.strftime("%Y%m%d")+'suspendedStocks',flavor='mysql',schema=None,if_exists='replace',index=True,index_label=None,chunksize=1000,dtype=None)
82 pd.io.sql.to_sql(frame=newORresumeStocks,con=db,name=today.strftime("%Y%m%d")+'newORresumeStocks',flavor='mysql',schema=None,if_exists='replace',index=True,index_label=None,chunksize=1000,dtype=None)
83 db.commit()
84 db.close()