1

i have a file that contains random lines and key words END:

line 1
line 2
...
line 23
END
line 25
....
line 40
END

and i want to split it into multiple files based on the key word END and have it inside each file as: file 1

line 1
line 2
...
line 23
END

file 2

line 25
.....
line 40
END

I tried:

csplit -k file_name '/END/' '{*}' but i do not get the correct output.
Mathew Linton
  • 33
  • 1
  • 2
  • 8

2 Answers2

3

Add an offset of 1 to the regex to include the matching END in the current file. I also added to ^ and $ to anchor the regex.

csplit -k file -f file --elide-empty-files '/^END$/1' '{*}'
  • -f file Sets the output filename prefix
  • --elide-empty-files This is a GNU extension and doesn't output empty files (in this case an empty file file02)

Output:

$ head file0*
==> file00 <==
line 1
line 2
...
line 23
END

==> file01 <==
line 25
....
line 40
END
Freddy
  • 4,548
  • 1
  • 7
  • 17
2

with awk

$ awk '{f= FILENAME"."(c+1); print > f} /^END$/{close(f); c++}' file

$ head file.*

==> file.1 <==
line 1
line 2
...
line 23
END

==> file.2 <==
line 25
....
line 40
END

Edge case behavior: if the input file is empty (no lines), no output will be generated. if the file has blank line(s), exact copy of the file is generated as a partition (same as with no END marker or only one at the end).

karakfa
  • 66,216
  • 7
  • 41
  • 56