Shell Programming and Scripting

This awk script should group emails based upon the data following @:

Code:

awk '
{
for (x=1; x<=NF; x++) {
        split($x,em,"@")
        all[em[2]] = all[em[2]] ? all[em[2]] FS $x : $x
        }
}
END{
for (i in all) {
    c++
    print "Group: "c FS all[i]
    }
}' inputfile

@Don Cragun : As i mentioned in my requirement, i want asdf@blh.com in one group; and the remaining id's in other group.

@RudiC : Thanks for your input.

Regards,

Let's be perfectly clear. From your first post:
So, no matter what may appear in any file somewhere, what you want is:

Code:

group1='emailid1@blh.com emaild2@blh.com emailid3@blh.com'
group2='asdf@blah.com'

or with your changes since then:

Code:

group2='asdf@blh.com'

We all mistakenly assumed that you were trying to read a list of email addresses from a file and split them into groups based on some criteria derived from the addresses themselves. And since you had three addresses ending with "@blh.com" and one address ending with "@blah.com", we all assumed that that was the criteria to be used to determine which group should contain an address.

It could have been that you wanted addresses starting with "a" in one list and addresses starting with "e" in the other list. It could have been that you wanted addresses with a digit before the "@" in one list and addresses with an alphabetic before the "@" in the other list. It could have been that you wanted the 3rd address in one list and the other addresses in the other list. It could have been some other pattern none of us saw.

But, since you just want two lists containing exactly those names, why did you give us the red herring talking about a file containing your list of names??? The file is irrelevant.

@Don Cragun : My sincere apologies for the confusion about the requirement !
The reason i gave the file name containing email address is, that is the way the shell script is set up [by the clients] to read email addresses.