filename contains:

123 4
1234 5
12345 689

i want result:
1234
12345
12345679

cat filename | awk '{print $1, $2, $3}' will get my result, but too many $$s.

sed -e 's/[ ]//g' filename will take care one space, [ ] will take 2 spaces.

is it a short way to do this ?

Regards

i am new to unix world. if you don't mind, would you please explain the syntax.

many thanks to you, and this site.

The gsub function substitutes all spaces with nothing and awk evaluates the 1 as true and prints the entire line.

Regards

tr replaces characters with other characters, or, with the -d option, with nothing.

sed -e 's/ //g' filename

I like the awk response. ta.

I think this is the easiest answer and also good implementation of RE.