CODEWITHSUNDEEP
linuxsed
Igor Boguslavetz
Igor Boguslavetz

Reputation: 63

Please help to remove specific lines from text

I need to remove :

guide_change_names \
  -design { lalala } \
  { 
     }

from text file. Rules: there should be no word between {} lalala could be any word.

Text before:

guide_change_names \
  -design { lalala } \
  { 
        ha hah haha
     }
guide_change_names \
  -design { lalala } \
  { 

     }
guide_change_names \
  -design { lalala } \
  { 
            lal lal lal
     }
guide_change_names \
  -design { lalala } \
  { 
     }

Text After:

  guide_change_names \
      -design { lalala } \
      { 
            ha hah haha
         }

    guide_change_names \
      -design { lalala } \
      { 
                lal lal lal
         }

Upvotes: 3

Views: 74

Answers (2)

jkshah
jkshah

Reputation: 11713

Try using GNU awk

awk 'BEGIN { RS="guide_change_names[^{]*{[^}]*}[^{]*{[ \n]*}[ \n]*" } 1' file

Here idea is to set Record Separator (RS) as the pattern to be removed.

Output:

guide_change_names \
  -design { lalala } \
  {
        ha hah haha
     }

guide_change_names \
  -design { lalala } \
  {
            lal lal lal
     }

Upvotes: 3

iruvar
iruvar

Reputation: 23394

You could use . Set the record separator RS to guide_change_names...}..} and overwrite it to "" if only spaces are contained between the second pair of curly braces 

awk --re-interval -v RS='guide_change_names([^}]+[}]){2}\\n' \
    'RT ~ /[{][[:space:]]+[}]\n$/{RT=""};{printf "%s%s", $0, RT}' file.txt

Upvotes: 3

Related Questions