Shell Programming and Scripting

Hi,

radoulov's solution is exactly what i wanted. the order of the values is not important at all. Thanks for all the help given to me. I'm totally new in C-Shell, AWK, perl, etc. I'll try all the solution given and let you guys know the outcome. really appreciate your help

Regards,
SL

---------- Post updated at 07:20 PM ---------- Previous update was at 07:07 PM ----------

Hi,

the script work just nice. But i encounter a problem.

Code:

file1: z= /a:/b:/c
file2: z= /b:/c:/d

after using awk -f merge.awk file1 file2
the result i get is something like this:

Code:

z=/a::/b
/c
/d

is it the "/" is causing the problem here? because i tested the script with the following and it works perfectly.

Code:

file1 : z=a:b:c:d
file2 : z=b:c:d:e

Code:

result : z=a:b:c:d:e

Thanks in advance.
SL

Hm,
I've just tested it and it seams to work as expected:

Code:

$ head file[12]
==> file1 <==
z= /a:/b:/c

==> file2 <==
z= /b:/c:/d
$ cat merge.awk 
BEGIN { FS = " *= *" }
END {
  printf "%s = ", f
  for (k in s) printf "%s", k \
    (j++ < n ? ":" : RS)
  }
{
  n = split($2, t, ":"); f = $1
  for (i = 0; ++i <= n;) s[t[i]]
  n > m && m = n
  }
$ awk -f merge.awk file1 file2
z = /a:/b:/c:/d