windows下python3 使用cx_Oracle，xlrd插件进行excel数据清洗录入

1. 读取xlsx数据进行清洗
2. cx_Oracle批量入库

create table temp_table
(
importtime varchar2(128),
carrier varchar2(32),

);

select * from temp_table

# -*- coding: utf-8 -*-

import xlrd
import datetime
import cx_Oracle
import time
from itertools import islice
import os
os.environ['NLS_LANG']='SIMPLIFIED CHINESE_CHINA.ZHS16GBK'



LineName = ['1号线','2号线']
StationName = []

########################链接数据库相关######################################

def getConnOracle(username,password,ip,service_name):
    try:
        conn = cx_Oracle.connect(username+'/'+password+'@'+ip+'/'+service_name)  # 连接数据库
        return conn
    except Exception:
        print(Exception)

#######################进行数据批量插入#######################

def insertOracle(conn,data,input_file_name):
    sheetnumber = getSheetNumber(data)
    cursor = conn.cursor()
    try:
        for x in range(0,sheetnumber):
            templist = excel_table_byindex(input_file_name,0,x)
            cursor.prepare('insert into temp_table(importtime ,carrier) values(:1,:2)') 
             # 使用cursor进行各种操作，templist数值需要和表temp_table对应
             cursor.executemany(None,templist)

        conn.commit()
    except cx_Oracle.DatabaseError as msg:
        print(msg)
    finally:
        cursor.close()
        conn.close()

###########################打开excel文件########################
def openXLS(path):
    try:
        data = xlrd.open_workbook(path)
        return data
    except Exception:
        print(Exception)

def getSheetNumber(data):
    sheet_num = len(data.sheets())
    return sheet_num
#######################一些数据清洗工作########################
def getlineName(str):
    for x in LineName:
        if x in str:
            return  x

def getStationName(str):
    for x in StationName:
        if x in str:
            return x
##########将excel中除去表头的一个sheet读出来,返回一个list#############
def excel_table_byindex(path,colnameindex = 0,by_index = 0):
    today = time.strftime('%Y%m%d', time.localtime(time.time()))
    data = openXLS(path)
    table = data.sheets()[by_index]
    nrows = table.nrows
    ncols = table.ncols

    colnames = table.row_values(colnameindex)
    list = []
    for rownum in range(1,nrows):
        row = table.row_values(rownum)
        temp_lineName = getlineName(row[6])
        temp_stationName = getStationName(row[6])
        if row:
            app = [today, str(row[1]), str(row[2]),temp_stationName,temp_lineName]
            # for i in range(len(colnames)):
            #     app[colnames[i]] = row[i]
            list.append(app)
    return list

###################一个可以从文件第二行开始读的办法#############

def getAllStationName(path):
    StationName_file = open(path, 'r', encoding='utf-8')
    #count = len(StationName_file.readlines())

    for line in islice(StationName_file,1,None):
        str_temp = line.strip('\n')
        if str_temp not in LineName and str_temp !='----'and str_temp!='':
            StationName.append(str_temp)

####################################################################
def getStationNamefromexcel(path):

    data = openXLS(path)
    table = data.sheets()[0]
    nrows = table.nrows
    ncols = table.ncols
    colnames = table.row_values(0)
    list = []
    for rownum in range(0,nrows):
        row = table.row_values(rownum)[0]
        if row:
            list.append(row)
    return list

#################################################################
def main():
    username = 'xx'
    password = 'xx'
    ip = '192.168.1.1'
    service_name = 'iop'
    #获取数据库链接
    conn = getConnOracle(username,password,ip,service_name)

    input_file_name = (r"E:\code\python\findS\subwayBase\xx.xlsx")
    #output_file_name = input("Enter the output file name:")
    getAllStationName(r"E:\code\python\findS\subwayBase\站点.txt")


    begin = datetime.datetime.now()

    insertOracle(conn,openXLS(input_file_name),input_file_name)

    # x.fetchone()
    # c.close()  # 关闭cursor
    # conn.close()  # 关闭连接

    end = datetime.datetime.now()
    print((end - begin).seconds)

if __name__ =='__main__':
    main()

Traceback (most recent call last):
  File "E:/code/python/findS/findSubwayBase.py", line 134, in <module>
    main()
  File "E:/code/python/findS/findSubwayBase.py", line 124, in main
    insertOracle(conn,openXLS(input_file_name),input_file_name)
  File "E:/code/python/findS/findSubwayBase.py", line 32, in insertOracle
    cursor.executemany(None,templist)
UnicodeEncodeError: 'ascii' codec can't encode characters in position 1-6: ordinal not in range(128)

Process finished with exit code 1

#在开头加上
import sys
reload(sys)
sys.setdefaultencoding( "utf-8" )

import os
os.environ['NLS_LANG']='SIMPLIFIED CHINESE_CHINA.ZHS16GBK'

#获取一个工作表

table = data.sheets()[0]          #通过索引顺序获取

table = data.sheet_by_index(0) #通过索引顺序获取

table = data.sheet_by_name(u'Sheet1')#通过名称获取

#获取整行和整列的值（数组）
 　　
table.row_values(i)

table.col_values(i)

#获取行数和列数
　　
nrows = table.nrows
ncols = table.ncols

#循环行列表数据
for i in range(nrows ):
      print table.row_values(i)

#单元格
cell_A1 = table.cell(0,0).value

cell_C4 = table.cell(2,3).value

#使用行列索引
cell_A1 = table.row(0)[0].value

cell_A2 = table.col(1)[0].value

#简单的写入
row = 0

col = 0

# 类型 0 empty,1 string, 2 number, 3 date, 4 boolean, 5 error
ctype = 1 value = '单元格的值'

xf = 0 # 扩展的格式化

table.put_cell(row, col, ctype, value, xf)

table.cell(0,0)  #单元格的值'

table.cell(0,0).value #单元格的值'

[OS.ENVIRON详解]: http://blog.csdn.net/junweifan/article/details/7615591
[python编码]:http://www.cnblogs.com/fkissx/p/5417363.html