How to work with substrings on Linux

$ echo "Focus on Peace on Earth" | cut -d' ' -f3,5
Peace Earth
$ echo "one two three 4 5 6" | cut -d' ' -f1-3,6
one two three 6

To use an alternate delimiter (in this case, a colon), use a command like this:

$ cut -d':' -f1-3,5,6 /etc/passwd | tail -n 5
justme:x:1004:JustMe:/home/justme
lola:x:1006::/home/lola
dumdum:x:1007::/home/dumdum

With awk, you can use more than one delimiter. In the following example, two delimiters are specified, so the awk command accepts either a colon or a blank to separate fields. The first two lines display the file, and the last two lines show the command and result.

$ cat file
Monday:1 Tuesday:2 Wednesday:3 Thursday:4 Friday:5
$ awk -F'[: ]' '{OFS=" ";print $1,$3,$4}' file
Monday Tuesday 2

Selecting substrings

To select an arbitrary sequence or characters from a string, you can use an awk command like the one below in which the $0 represents the entire phrase, 10 represents the first character position to be grabbed and 5 is the length of the string to be displayed.

$ echo "Focus on Peace" | awk '{print substr($0,10,5)}'
Peace

To do the same kind of thing with the cut command, you would use a command like this in which the 13^th through 22^nd characters are extracted from the phrase and displayed.

$ echo "Linux is an impressive OS" | cut -c 13-22
impressive

In this next command, the cut command displays the 7^th-12^th characters from the lines in a file. The head command simply limits the display to the first 4 lines of output.

$ cut -c 7-12 sayings | head -4
with 3
and ov
nd be
and be

Using grep

You can use the grep command to select multiple words from a file. In this example, only the selected words are displayed, not the entire lines. This is because the -o (display only the matched items) option is being used.