If you have a text file and want to search and replace a part of a matched pattern, how would you do it with perl one-liners, sed or python?
For example:
"653433,78" match with [0-9]{6},[0-9]{2}.
Change , to ..
Add a comment |
2 Answers
You can use numbered sub groups. i.e:
Find:
([0-9]{6}),([0-9]{2}) Replace:
\1.\2 Example in Python:
import re inp = '653433,78' out = re.sub(r'([0-9]{6}),([0-9]{2})', r'\1.\2', inp) print out This would give you:
>>>'653433.78' 
- For using number variables in replacement strings, named reference syntax can be used:
re.sub(pat, r'\g<1>' + f'{my_number}', str)aksh1618– aksh16182019-04-25 18:38:11 +00:00Commented Apr 25, 2019 at 18:38
You can capture the matched part around the pattern you want to replace and then reuse it in the replace statement.
cat yourfile | perl -p -e "s/([0-9]{6}),([0-9]{2})/\\1.\\2/g" A more complex solution with lookahead and lookbacks can also be used (but this will not support wildcards such as * and + for your pattern)
cat yourfile | perl -p -e "s/(?<=[0-9]{6}),(?=[0-9]{2})/./g" 