Question

我需要解析数据框中的数据，以消除括号内的所有内容，然后将其移动到新列中。理想情况下，如果可以在新列中消除括号，那也很好，但是我认为这两种结果都会创建预期的解决方案：

current column                                  new column
/reports/industry(5315)/2018                    (5315)
/reports/limit/sector(139)/2017                 (139)
/reports/sector/region(147,189 and 132)/2018    (147,189 and 132)

谢谢，您能给出的任何方向都很好！

Answer 1

IIUC提取物

df.current.str.extract('.*\((.*)\).*',expand=True)
Out[785]: 
               0
0           5315
1            139
2147,189 and 132

Answer 2

您可以使用正则表达式来做到这一点：

old_col = ['/reports/industry(5315)/2018', '/reports/limit/sector(139)/2017', '/reports/sector/region(147,189 and 132)/2018']
df = pd.DataFrame(old_col, columns=['current_column'])
df['new_column'] = df['current_column'].str.extract(r'\((.*)\)')

输出如下：

current_column                                       new_column
0   /reports/industry(5315)/2018                        5315
1   /reports/limit/sector(139)/2017                      139
2   /reports/sector/region(147,189 and 132)/2018    147,189 and 132

Answer 3

>>> import re
>>> re.sub('.*(\(.*\)).*', '\\1', '/reports/industry(5315)/2018')
'(5315)'

完整示例

import pandas as pd
import re


old_col = ['/reports/industry(5315)/2018', '/reports/limit/sector(139)/2017', '/reports/sector/region(147,189 and 132)/2018']
df = pd.DataFrame(old_col, columns=['current_column'])


def grab_dat(x):
    dat = re.sub('.*(\(.*\)).*', '\\1', x)
    return(dat)


df['new_col'] =  df['current_column'].apply(grab_dat)

Answer 4

使用正则表达式和熊猫str函数。

df['new_column'] = df['col'].str.extract(r'(?P<new_column>(?<=\().*(?=\)))', expand=False)

正则表达式说，寻找一个匹配任何东西的模式，使其以“（”开头并以“）”结尾，并将其放在名为“ new_column”的命名匹配组中

Python Pandas解析记录

4 个答案: