Shell Programming and Scripting

Hi all
I was wondering if you may help me in resolving an issue.
In particular I have a file like this:
the

Code:

...

represent different string and what I wrote

Code:

Cur

or

Code:

Ent

are the constant.

Well, what I would like to obtain is a file in which are reported only the ID in which the second column exclusively start with

Code:

Cur

For instance the ID1 is discarded because the secon colum start either with Ent and Cur.

Hoping everything is clear and tht someone could help me.

Best

Giuliano

Confusing!

Code:

ID1 Ent..
ID1 Cur..
ID2 Cur..
ID3 Ent..
ID4 Cur..
ID3 Ent..
ID3 Ent..

In your example there's nothing different between accepted output
ID2 Cur
ID4 Cur
and
ID1 Cur

No because I could have also these cases:

In this case I do not want keep

Code:

ID1

But I think I resolve the issue with grep command.

But if you have alternative...

Lets hope your sample is meaningful and representative. Try

Code:

awk     '$2 ~ /^Ent/    {ENT[$1]}
         $2 ~ /^Cur/    {CUR[$1]}
         END            {for (i in CUR) if (!(i in ENT)) print i, CUR[i]}
        ' file
ID2 Cur..
ID4 Cur..

What if you have repetitions in your input as in ID5 has "Cur" in the 2nd col twice so what is the desired output then...

Code:

ID3 Ent
ID5 Cur
ID6 Ent
ID7 Ent
ID5 Cur

You need to store the value:

Code:

          $2 ~ /^Cur/    {CUR[$1]=$2}

A variant is

Code:

awk '{if ($2~/^Cur/) {CUR[$1]=$2} else {DEL[$1]}} END {for (i in CUR) if (!(i in DEL)) print i, CUR[i]}' file

Absolutely! Thanks. I edited my initial post and missed that one. Rats!