Regular expression to identify groups

user_13857749 0
Viewed 50
content = "[row1|col1]:{value:{{calculate}}<<report_1>>[Row2|col2];format:float;} [hiddenr0120a|c0012]:{format:float;}"
regex = re.compile(r"(\[.*?\]).*?\{(.*?)\}")
entries = regex.findall(content)
#Output
[('[row1|col1]', 'value:{{calculate'), ('[Row2|col2]', 'format:float;')]
#Expected Output
[("[row1|col]", "{value:{{calculate}}<<report_1>>[Row2|col2];format: float;}"), ("[hiddenr0120a|c0012]", "{format:float;}")]

I have tried regex "([.?]).?{(.*?)}" which actually ends at the first instance of "}", so it fails for the first scenario.

regex-group python regex
3 Answers

You might use

(\[[^\[\]\r\n]*\]):({.*?})(?: (?=\[)|$)

In parts

  • ( Capture group 1
    • \[[^\[\]\r\n]*\] Match from opening till closing square bracket
  • ) Close group 1
  • : Match literally (Or use .*? to match other chars as well)
  • ( Capture group 2
    • {.*?} Match from { till } as least as possible
  • ) Close group 2
  • (?: Non capture group (?=[) Match a space and assert what is on the right is [
    • | Or
    • $ Assert the end of the string
  • ) close non capture group

Regex demo | Python demo

Example code

import re
 
content = "[row1|col1]:{value:{{calculate}}<<report_1>>[Row2|col2];format:float;} [hiddenr0120a|c0012]:{format:float;}"
regex = re.compile(r"(\[[^\[\]\r\n]*\]):({.*?})(?: (?=\[)|$)")
entries = regex.findall(content)
print (entries)

Output

[('[row1|col1]', '{value:{{calculate}}<<report_1>>[Row2|col2];format:float;}'), ('[hiddenr0120a|c0012]', '{format:float;}')]
user_54249880

Use additional capture groups, and extract like this:

import re

text = "[row1|col1]:{value:{{calculate}}<<report_1>>[Row2|col2];format:float;} [hiddenr0120a|c0012]:{format:float;}"
pattern = r'(\[.*?\]).*?\{(.*?\})\s(\[.*\])\:(\{.*\})'
regex = re.compile(pattern)
entries = re.findall(regex, text)

for entry in entries[0]:
    print(entry)

Returning:

[row1|col1]
value:{{calculate}}<<report_1>>[Row2|col2];format:float;}
[hiddenr0120a|c0012]
{format:float;}
user_74455280

If the second group always ended with ;}, then the regular expression may be as simple as
(\[.+?\]):(\{.*?;\})

import re
content = '''
[row1|col1]:{value:{{calculate}}<<report_1>>[Row2|col2];format:float;}
[hiddenr0120a|c0012]:{format:float;}
'''
r = re.compile(r'(\[.+?\]):(\{.*?;\})')
expected = [("[row1|col1]", "{value:{{calculate}}<<report_1>>[Row2|col2];format:float;}",),
            ("[hiddenr0120a|c0012]", "{format:float;}",),]
assert all(t == expected[i][j] for i, x in enumerate(r.findall(content)) for j, t in enumerate(x))

Demo.

user_33504280
Related