如何在模式匹配之前使用正则表达式将字符串拆分为多行

Question

I have a file with the swift data in the below format that needs to be split into multiple lines using regular expression in python. Original file:

ID        Information

1         :20:Test1  :25:test2:28C:test3

Desired Output:

ID  Information

1     :20:Test1  
1     :25:test2  
1     :28C:test3

Using Notepad++ I am able to break the 'Information'column into multiple lines using

Find: ^:[0-9]{2}:|\s:[0-9]{2}:|\s:[0-9]{2}[A-Za-z]{1}:

Replace: \n$0

Need to replicate the same using python. So far i tried the below code but the result does not contain the pattern. It is splitting after the pattern match:

import re

s = ':20:Test1  :25:test2:28C:test3'

l = re.compile('^:[0-9]{2}:|\s:[0-9]{2}:|\s:[0-9]{2}[A-Za-z]{1}:').split(s)

Result: ['', 'Test1 ', 'test2 ', 'test3']

The result should also contain the regex pattern while splitting the string.

Answer 1

How about this pattern:

import re

s = ':20:Test1  :25:test2:28C:test3'

p = re.compile('(:[0-9A-z]{1,3}:)([0-9A-z]+)')

print(p.findall(s))
#[(':20:', 'Test1'), (':25:', 'test2'), (':28C:', 'test3')]

Answer 2

Given that you have multiple types of output, it may be easier to use a little logic with a regex:

s='''\
ID        Information

1         :20:Test1  :25:test2:28C:test3'''
    

import re 

for line in s.splitlines():
    if m:=re.search(r'^(\d+)([ \t]+)(:.*)',line):
        data=re.findall(r'(:[^:]+:[^:]+(?=:|$))', m.group(3))
        for e in data:
            print(m.group(1)+m.group(2)+e.rstrip())
    else:
        print(line)     

Prints:

ID        Information

1         :20:Test1
1         :25:test2
1         :28C:test3

As written, that is Python 3.8+ only. If you want on an earlier Python 3.X:

for line in s.splitlines():
    m=re.search(r'^(\d+)([ \t]+)(:.*)',line)
    if m:
      ...

Answer 3

You may use

import re
text = """ID        Information

1         :20:Test1  :25:test2:28C:test3"""

valid_line_rx = r'^(\d+\s*)(:\d{2}[A-Za-z]?:.*)'
print( re.sub(valid_line_rx, lambda m:
  "\n".join(["{}{}".format(m.group(1),x) for x in re.split(r'(?!^)(?=:\d{2}[A-Za-z]?:)', m.group(2))]),
  text, 
  flags=re.M)
)

See the Python demo , output:

ID        Information

1         :20:Test1  
1         :25:test2
1         :28C:test3

The ^(\d+\s*)(:\d{2}[A-Za-z]?:.*) regex matches

• ^ - start of a line (due to re.M flag)
• (\d+\s*) - Group 1: one or more digits and then 0 or more whitespaces
• (:\d{2}[A-Za-z]?:.*) - Group 2: : , two digits, an optional letter and aa : and then any 0 or more chars other than line break chars as many as possible.

The (??^)(:=?\d{2}[A-Za-z]::) regex matches a location that is not the start of string and is immediately followed with : , 2 digits, an optional letter and a : , and this pattern is used to split the Group 2 value of the above regex match.