0

I have a file named file.txt and it contains several lines containing string "NaN". How can I delete lines containing "Nan" and one line before and after it. I know sed －i '/pattern/d' file.txt can delete the matched line, but how can I delete neaby lines of the matched line. Best regards

Winston Pan
  • 47
  • 6
  • Please do share samples of input and expected output in your question to make your question more clear, thank you. – RavinderSingh13 Aug 26 '21 at 12:04
  • `ed` is probably a more suitable command for tasks like this, though I am not sure how to handle the case where `NaN` is the first or last line in the file. `printf 'g/NaN/-,+d\nwq\n' | ed file.txt`. (When there is no previous or next line for `-` or `+` to match, the script fails.) – chepner Aug 26 '21 at 12:27

1 Answers1

0

This is a most inelegant solution, but works. Perhaps, it will anger a UNIX guru to come and provide a real answer.

grep 'Nan' -n -C 1 target_file.txt | awk -F '[-:]' '{print $2}' | sed '2d' | paste -d, - - | sed 's/$/d/' > del_lines.sed && sed -f del_lines.sed target_file.txt > output_file.txt

fyi, The word "perticular" is misspelled, its "par-" like in golf!

dannylee8
  • 505
  • 7
  • 17