How to grep from the third column?

Shell Programming and Scripting
1 
PROL\tests_li004_developers:VAS:3543346:q34243,d3hs35,34bdf3,24sfgg,a3s234

Im trying to grep all text after VAS:3543346, so im trying to just have this

q34243,d3hs35,34bdf3,24sfgg,a3s234

Im confused on how I would do this

2

If you have that in a variable:

var='PROL\tests_li004_developers:VAS:3543346:q34243,d3hs35,34bdf3,24sfgg,a3s234'
echo "${var#*VAS:3543346:}"
q34243,d3hs35,34bdf3,24sfgg,a3s234
3

grep can't do that; try

awk -F: '/VAS:3543346/ {print $4}' file
q34243,d3hs35,34bdf3,24sfgg,a3s234
4

It always helps us provide better answers if you tell us what OS and what shell you're using.

I don't think it can be done with grep if there are an unknown number of colons in the text you want to print (but that isn't the case in the given sample). It can't be done with grep using only standard options, and as far as I know it can't be done with a single grep ...

But if you have a grep (such as the BSD version) with a -o option that prints only the matched text, the two grep pipeline:

grep 'VAS:3543346' file | grep -o '[^:]*$'

does what you want with you sample input. And to specifically check field 2 for VAS and field 3 for 3543346 , you could tighten up the 1st grep to:

grep '^[^:]*:VAS:3543346:' file | grep -o '[^:]*$'

But, unless you're working with big files, it is probably faster to invoke awk once than to invoke grep twice. And, even if the input file is large enough to overcome the cost of invoking a second process, you'd have to benchmark whether awk or a two grep pipeline would be faster. (Without benchmarking, I'd guess that awk would beat the pipeline.)

5

If i use this awk method, what would I do when the number next to VAS is diffrent, like if I want it to do that to a bunch of lines

6

You enter a different number.

Seriously: how would you supply that "bunch of" numbers?

7

for colon delimited input file try:

awk '
FILENAME ~ /list/ {vas[$1]=$1; next; }
vas[$3] {$1=$2=$3=""; sub(":::", ""); print $0;}
' list_of_VAS FS=: OFS=: infile

where list_of_VAS is VAS numbers list file

8

Shell globbing: use * to match any characters rather than a specific string:

var=`vastool list group tests_li004_developers`
echo "${var#*:VAS:*:}"
echo "${var#*:*:*:}"

Regular expression: use .* to match any characters, and [^:]* to match any but a : character:

vastool list groups | sed "s/^[^:]*:VAS:[^:]*://"
vastool list groups | sed -n "/^tests_li004_developers:/ s/^[^:]*:VAS:[^:]*://p"
vastool list groups | sed -n "s/^tests_li004_developers:VAS:[^:]*://p"

awk can use regular expressions, too, but divides into fields $1 $2 ... that is often easier to deal with:

vastool list groups | awk -F":" '($1=="tests_li004_developers") {print $4}'

NB some awk versions have restrictions regarding line length.