ODOC: cut

cut — CUT a specified part form the file and Print it on stdout.

Summary :

Print selected parts of lines from each file(s) to standard output. This command is mostly used in shell scripts(think I should write a tut?)

Examples :

Create file1 with 10 columns (separated by TAB) of data for 10 lines and file2 with 10 columns (separated by space) of data for 10 lines.

$ cut -b2 file1 — Print only the byte in 2 position. Tab & Backspace are treated like a 1 byte char.

$ cut -b1,3,5 file1 — Print only the bytes in 1,3 & 5 positions.

$ cut -c2 file1 — Print only character in 2nd position. Normally same as -b option. But you can feel the difference only with multi-byte encodings (Unicode).

$ cut -f5 file1 — Print only the 6th field. TAB is default separator.

$ cut -f4- file1 — Print from 4th field to last field.

$ cut -f-5 file1 — Print from 1st field to 4th field.

$ cut -d’ ‘ -f3-5 file2 — Print only the fields from 3 to 5, which is separated by spaces.

$ cut -s -d’ ‘ -f2 file2 — Don’t print lines that don’t contain the delimiter,space (line without a delimiter is printed verbatim).

$ cut -f2-5,7-9 –output-delimiter=’,’ file1 — In the output, field will be separated by “,” (Not by the default one).

Note:

  1. Use one, and only one of -b, -c or -f.
  2. Range or many ranges can be separated by commas.
  3. Range of Bytes/Chars/Fields can specified like this:

N == Only Nth byte, character or field.

N- == From Nth byte, character or field, to End of line.

N-M == From Nth to Mth (included) byte, character or field.

-M == From 1st to Mth (included) byte, character or field.

Read : man cut

odoc, cut, linux,gnu/linux, shell+scripts