A problem in my awk script. -condition : using variable in awk. - aim : print specific column started count variable in text

Question

I want to print the whole line with specific column started count variable in text.

file.log format

1 101010 101010 4.0001 my home
2 101010 101010 5.0001 my home
3 101010 101010 6.0001 my home
4 101010 101010 7.0001 my home

my script

count=4
awk -v cnt="$count" '$4 ~ /^[cnt]\./' file.log

I want this result.

1 101010 101010 4.0001 my home

But my result is nothing....

Please tell me the wrong point of my script.

FYI `[cnt]` is a bracket expression which means "a single character which is either of the letters c, n, or t". See https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap09.html#tag_09_03_05. — Ed Morton, Dec 12 '20 at 18:57

score 3 · Answer 1 · edited Dec 12 '20 at 19:02

You need NOT to use awk variable in a literal regexp, use int function here to convert 4th field as interger and then compare it directly with variable like:

count="4"
awk -v cnt="$count" 'int($4)==cnt' Input_file

Explanation: First passing shell variable count to cnt as an awk variable. Then using int function on 4th field to get only integer part eg--> from 4.0001 to 4 to make an exact comparison with cnt and once its equal it will print that line.

score 2 · Answer 2 · answered Dec 12 '20 at 18:59

2

You could use a dynamic (aka computed) instead of literal regexp:

awk -v cnt="$count" '$4 ~ ("^" cnt "[.]")' file.log

See How do I use shell variables in an awk script? and https://www.gnu.org/software/gawk/manual/gawk.html#Computed-Regexps.

answered Dec 12 '20 at 18:59

Ed Morton

188,023
17
78
185

A problem in my awk script. -condition : using variable in awk. - aim : print specific column started count variable in text

2 Answers2