Question

我正在写一个csv文件。 row [3]和row [4]都是我原始csv文件中的日期列。行[3]和行[4]显示浮动。在csv文件中，它们都是整数，但是当我运行Python代码时，它会显示一个突出显示小数的错误。

import csv
import xlrd
import time
import datetime

workbook = xlrd.open_workbook('T:/SUNDAY REPORT 12.30.16.xlsm')
for sheet in workbook.sheets():
with open('{}.csv'.format(sheet.name), 'wb') as f:
    writer = csv.writer(f)
    writer.writerows(sheet.row_values(row) for row in range(sheet.nrows))

print "CSV converted"

time.sleep(5)

def xldate_to_datetime(xldate):
    tempDate = datetime.datetime(1900, 1, 1)
    deltaDays = datetime.timedelta(days=float(xldate)-2)
    TheTime = (tempDate + deltaDays )
    return TheTime.strftime("%m/%d/%Y")

with open("InfoCenterTracker.csv","rb") as source:
    rdr= csv.reader( source )
    with open("result.csv","wb") as result:
        wtr= csv.writer( result )
        next(rdr, None) # skip the headers 
        for r in rdr:
             r[3] = xlrd.xldate.xldate_as_datetime(float(r[3]), workbook.datemode)
             r[4] = xlrd.xldate.xldate_as_datetime(float(r[4]), workbook.datemode)                 if r[1] == "": 
                 pass
             else:
                 wtr.writerow(r[1:19])

        print "Columns converted to datetime"

错误：

r[3] = xlrd.xldate.xldate_as_datetime(float(r[3]), workbook.datemode)
ValueError: could not convert string to float:

CSV中的值：

 Week(Row[3]):    Date(row[4]):
  42735.0         42730.0
  42735.0         42730.0
  42735.0         42731.0
  42735.0         42731.0

Answer 1

这应该与Imran的代码一起发挥作用

r[3] = datetime.datetime.strptime(xldate_to_datetime(r[3]), "%m/%d/%Y")

r[4] = datetime.datetime.strptime(xldate_to_datetime(r[3]), "%m/%d/%Y")

Answer 2

使用以下函数将excel数字格式转换为datetime对象

def xldate_to_datetime(xldate):
   tempDate = datetime.datetime(1900, 1, 1)
   deltaDays = datetime.timedelta(days=int(xldate)-2)
   TheTime = (tempDate + deltaDays )
   return TheTime.strftime("%m/%d/%Y")

然后将date_format更改为

date_format = datetime.datetime.strptime(xldate_to_datetime(r[3:5][0]), "%m/%d/%Y")

将列转换为datetime

2 个答案: