Shell Programming and Scripting

@ Panyam and @ Rakesh thanks a lot.... It did serve my purpose.... Thanks.. But i am stil working on it.. and will get back, as my original input file is different, so trying to impliment on it... I think it wil be better to have a delimiter to differentiate...

@ ghostdog74 ya da its a type error sorry for confusion



---------- Post updated 06-26-09 at 10:10 AM ---------- Previous update was 06-25-09 at 05:59 PM ----------

Originally quoted by Rakesh:

----------------------------------------
/bin/sort inputfile | awk ' BEGIN { FS=OFS=" ";getline;prev=$1" "$2" "$3;s=$4 }

{
curr=$1" "$2" "$3
# print curr
if ( prev == curr )
s=s" "$4
if ( prev != curr )
{
print prev " " s
prev=curr;s=$4
}
}
END {
print prev " " s
}'
------------------------------------------

This piece of code worked fine but thr is one more thing i would like to highlight.. You have sirted entire inputfile hence the occurances are not printed in order they are see the example:

inputfile:
a b c 1
a b c 2
a b c 3
x y z 6
x y z 44
a b c 89
x y z 9
b s c 100
a b c 19

Output of above code:
a b c 1 19 2 3 89
b s c 100
x y z 44 6 9

So here al these values appended ones also sorted..
Instead shud get as below:

a b c 1 2 3 89 19
b s c 100
x y z 6 44 9

I dont mind the order of values if 1,2,3.. N then can sort only til N-1.. Bec last value is one with different values

Please help in this regard.

In such case I will use paynam's code (i really liked that and learnt) with little modification...

Code:

sort -n -k4,4 inputfile | awk '{a[$1" "$2" "$3]=a[$1" "$2" "$3]" "$4}END{for (i in a) print i a[i]}'

---------- Post updated at 10:57 AM ---------- Previous update was at 10:56 AM ----------

i suggest you to produce your code while posting queries... you could have tried ...
man sort.