股票数据自动入库hive到oracle

来源：互联网发布：java 文件存在判断编辑：程序博客网时间：2024/05/01 08:43

------import_stock_d.py-----------------------------------

#!/usr/bin/python

import tushare as ts
import os
import re
stocklistpath = '/home/cloudera/data/list/stocklist.txt'
savepath='/home/cloudera/data/data/'
openstock = open(stocklistpath,'r+')
for i in openstock:
r = r"S[ZH]\d\d\d\d\d\d"
stocklist = re.findall(r,i)

for i in stocklist:
stocknum = i[2:8]
df = ts.get_hist_data(stocknum)
df.to_csv(savepath + i +'.txt')
print i

for path,d,filelist in os.walk('/home/cloudera/data/data/'):
for filename in filelist:
filepath = os.path.join(path,filename)
print filepath
file = open(filepath,'r+')
file.seek(0,0)
filename1 = filename[0:8]+','
print filename1
for line in file.readlines():

print file.writelines(filename1 + line)

file.close()

--------------------------------StockRun.sh------------------------------------------------------

python /home/cloudera/python/import_stock_d.py

hadoop fs -put /home/cloudera/data/data /stock

hive -e "LOAD DATA INPATH '/stock/data/*' OVERWRITE INTO TABLE import_stock_d";

hive -e "insert overwrite table import_stock_d select * from import_stock_d where turnover is not null"

sqoop export --table import_stock_d -connect jdbc:oracle:thin:@192.168.1.10:1521:orcl --username stock --password stock --export-dir '/user/hive/warehouse/import_stock_d/*' --input-fields-terminated-by ',' --input-lines-terminated-by '\n' --columns 'code,T_DATE,OPEN,HIGH,CLOSE,LOW,VOLUME,PRICE_CHANGE,P_CHANGE,MA5,MA10,MA20,V_MA5,V_MA10,V_MA20,TURNOVER'

0 0