5

I search a pattern, say 'START OF ARAMBOL' and it is matched in a file. Now I would like to comment every line from the line no 1 to the current matched pattern line no. I have to do it for more than 200 files.

I can do this using perl too but is there any good sed method to do so.

Thanks

Ramneek Singh Kakkar
  • 187
  • 1
  • 5
  • 3
    Please be specific about what constitutes a comment in the context of your files (C-style comment? C++ style? leading `#` or `%`?) – steeldriver May 17 '19 at 10:54

2 Answers2

4

As a one-liner to demonstrate the concept :

echo -e 'a\nb\nc\nPATTERN\nd\ne\nf' | sed '0,/PATTERN/ s/^/#/'

You just have to adapt to your context :

  • as for the 'PATTERN'
  • I assumed '#' as the commenting character
  • and regarding how you can apply this to all your files. If they all are 'fileXXX.txt', you can run : sed -i '0,/PATTERN/ s/^/#/' file*txt
Httqm
  • 1,116
  • 7
  • 16
  • 1
    I would stress the assumption (actually reasonable, looking at the question) that the `PATTERN` is found in each file. This solution will comment any line in files in which `PATTERN` is not found. – fra-san May 17 '19 at 11:37
  • The answer assumes GNU sed. – D. Ben Knoble May 17 '19 at 15:30
1

Your question asks about sed but includes the gvim tag, so here is an ed/ex answer:

ed file <<EOF
0,/"$pattern"/s/^/"$comment_char"/
wq
EOF

That it looks remarkably like the sed answer is because sed and ex descend from ed

D. Ben Knoble
  • 484
  • 3
  • 8