python csv复制列

Posted 2023-02-23

【中文标题】python csv复制列

【英文标题】：python csv copy column

【发布时间】：2013-10-19 22:29:05

【问题描述】：

我有一个包含以下内容的文件

first_name,last_name,uid,email,dep_code,dep_name
john,smith,jsmith,[e-mail],finance,21230
john,king,jking,[e-mail],human resource,31230

我想复制列“email”并创建一个新列“email2”，然后将 gmail.com 从列 email2 替换为 hotmail.com

我是 python 新手，所以需要专家的帮助，我尝试了几个脚本，但如果有更好的方法，请告诉我。原始文件包含 60000 行。

with open('c:\\Python27\\scripts\\colnewfile.csv', 'rb') as fp_in1, open('c:\\Python27\\scripts\\final.csv', 'wb') as fp_out1:
    writer1 = csv.writer(fp_out1, delimiter=",")
    reader1 = csv.reader(fp_in1, delimiter=",")
    domain = "@hotmail.com"
    for row in reader1:
        if row[2:3] == "uid":
            writer1.append("Email2")
        else:
            writer1.writerow(row+[row[2:3]])

这是最终的脚本，唯一的问题是它没有完成整个输出文件，它只显示 61409 行，而在输入文件中有 61438 行。

inFile = 'c:\Python27\scripts\in-093013.csv' outFile = 'c:\Python27\scripts\final.csv'

open(inFile, 'rb') 作为 fp_in1，open(outFile, 'wb') 作为 fp_out1： writer = csv.writer(fp_out1, delimiter=",") reader = csv.reader(fp_in1, delimiter=",") 对于阅读器中的 col： 德尔科尔[6:] writer.writerow(col) 标题 = 下一个（读者） writer.writerow（标题 + ['email2']） 对于阅读器中的行： 如果长度（行）> 3： 电子邮件 = email.split('@', 1)[0] + '@hotmail.com' writer.writerow(row + [email])

【参考方案1】：

如果您在阅读器上调用next()，您一次只能获得一行；用它来复制标题。复制电子邮件栏很容易：

import csv

infilename = r'c:\Python27\scripts\colnewfile.csv'
outfilename = r'c:\Python27\scripts\final.csv'

with open(infilename, 'rb') as fp_in, open(outfilename, 'wb') as fp_out:
    reader = csv.reader(fp_in, delimiter=",")
    headers = next(reader)  # read first row

    writer = csv.writer(fp_out, delimiter=",")
    writer.writerow(headers + ['email2'])

    for row in reader:
        if len(row) > 3:
            # make sure there are at least 4 columns
            email = row[3].split('@', 1)[0] + '@hotmail.com'
        writer.writerow(row + [email])

此代码在第一个@ 符号上拆分电子邮件地址，获取拆分的第一部分并在其后添加@hotmail.com：

>>> '[e-mail]'.split('@', 1)[0]
'example'
>>> '[e-mail]'.split('@', 1)[0] + '@hotmail.com'
'[e-mail]'

以上产生：

first_name,last_name,uid,email,dep_code,dep_name,email2
john,smith,jsmith,[e-mail],finance,21230,[e-mail]
john,king,jking,[e-mail],human resource,31230,[e-mail]

用于您的示例输入。

【讨论】：

这是我收到的错误 email = row[3] IndexError: list index out of range

@user2820987：那么您的输入文件中有空行，或者至少行太短。我会调整的。

很可能 last 行是空的；其余数据写入成功。

inFile = 'c:\\Python27\\scripts\\in-093013.csv' outFile = 'c:\\Python27\\scripts\\final.csv' with open(inFile, ' rb') as fp_in1, open(outFile, 'wb') as fp_out1: writer = csv.writer(fp_out1, delimiter=",") reader = csv.reader(fp_in1, delimiter=",") for col in reader: del col[6:] writer.writerow(col) headers = next(reader) writer.writerow(headers + ['email2']) for row in reader: if len(row) > 3: email = email.split(' @', 1)[0] + '@hotmail.com' writer.writerow(row + [email])

你不能循环一个打开的reader对象两次；如果您想从输出中删除一列，请在一个循环中执行此操作，如果您也添加一列。

【参考方案2】：

这可以使用 pandas 非常干净地完成。就是这样：

In [1]: import pandas as pd

In [3]: df = pd.read_csv('your_csv_file.csv')

In [4]: def rename_email(row):
   ...:     return row.email.replace('gmail.com', 'hotmail.com')
   ...:

In [5]: df['email2'] = df.apply(rename_email, axis=1)

In [6]: """axis = 1 or ‘columns’: apply function to each row"""

In [7]: df
Out[7]:
  first_name last_name     uid             email        dep_code  dep_name              email2
0       john     smith  jsmith  [e-mail]         finance     21230  [e-mail]
1       john      king   jking   [e-mail]  human resource     31230   [e-mail]

In [8]: df.to_csv('new_update_email_file.csv')