Split columns based on patterns in fields

Question

Example: I'd like to split $2 in input file, putting a tab after pattern surname

Input:

name    surname1    
name    surname30000
name    surname456

Desired output:

name    surname    1
name    surname    30000
name    surname    456

Kira · Accepted Answer · 2015-11-13 13:05:02Z

up vote 2 down vote accepted

An awk alternative:

awk 'match($2,/[0-9]+$/) {printf("%s\t%s\t%s\n", $1, substr($2,0,RSTART), substr($2,RSTART,RLENGTH))}' filename

This will match only the numeric string at the end of the second column.

answered yesterday

Kira
64612

add a comment |

Lambert · Answer 2 · 2015-11-13 11:19:08Z

up vote 2 down vote

You can lookup the first digit and replace it by a tab followed by the digit using:

sed 's/[0-9]/\t&/' file

answered yesterday

Lambert
3,97813

add a comment |

cas · Answer 3 · 2015-11-13 11:19:16Z

up vote 2 down vote

sed -e 's/\([0-9]\)/\t\1/' input.txt

adds a tab before the first digit on every line.

Output:

name    surname 1    
name    surname 30000
name    surname 456

answered yesterday

cas
13.2k1342

add a comment |

Thomas Dickey · Answer 4 · 2015-11-13 11:28:20Z

The data might be in a specific column, e.g., 16:

sed -e 's/^\(.\{15,15\}\)/\1 /'

Alternatively, the data of interest in the third field may not be numeric:

sed -e 's/^\([[:alpha:]]\+[[:space:]]\+\)\([[:alpha:]]\+\)/\1\2 /'

Either of these will put a space before the digits (or other non-alphabetic) characters in the data.

For reference:

4 Answers 4