Scripting columns

Shell Programming and Scripting
1

I've got the following data in a file - basically hundred of passes.

     Pass Id                   : PASSID1
     Pass Name              : 7 day pass 10 Gb
     Pass Group              : uk
     Short Name             : pmbbrh2
     type                       : restrict
     Validity                    : 168 hours
     Pass Id                     : BBFFF009
     Pass Name                : World pass 24hr
     Pass Group               : World
     Short Name               : CsRHhh3
     type                        : block
     Validity                    : 24 hours

What id like is something like the following.

Pass Id    Pass Name       Pass Group   Short Name     type    Validity
PASSID1  7 day pass 1 Gb   uk         pmbbrh2        restrict   168 hours
BBFFF009 World pass 24hr   EU        CsRHhh3        block     24 hours

I know this is something simple, but it actually hurts my mind trying to solve it. I cant use grep or awk or sed for this. any other ideas ?

The data in the first column is the same throughout the file but the data in the second column is sometimes the same.

Thanks in advanced

Jeffers

2

just wondering the reasons behind such restrictions?

3

Certain passes can only use certain protocols. if the use protocols that use p2p or skype they are restricted

thanks

4

I think vgersh99 meant why the restrictions prohibiting the use of grep, sed, and awk.

5

maybe its my understanding of these commands.

I've tried using awk and grep with the cut command and using ":" as the delimitar but after that i get lost.

some of the values have more than 4 or 5 words and so becomes a complete mess.

6

Perhaps this will do:

$ awk -F'^ +| *: *' 'function pr(a) {printf("%-10s %-20s %-12s %-12s %-10s %-10s\n", a[1],a[2],a[3],a[4],a[5],a[0])}
{i=NR%6; k=$2; v=$3} NR==6 {pr(k)} !i{pr(v)}' data

Pass Id    Pass Name            Pass Group   Short Name   type       Validity  
PASSID1    7 day pass 10 Gb     uk           pmbbrh2      restrict   168 hours 
BBFFF009   World pass 24hr      World        CsRHhh3      block      24 hours  
PASSID1    7 day pass 10 Gb     uk           pmbbrh2      restrict   168 hours 
BBFFF009   World pass 24hr      World        CsRHhh3      block      24 hours  
PASSID1    7 day pass 10 Gb     uk           pmbbrh2      restrict   168 hours 
BBFFF009   World pass 24hr      World        CsRHhh3      block      24 hours

Cheers,
Alister

7

Assuming the order of columns is persistent between 'Pass Id'-s and there're no missing columns.

nawk -f jeff.awk myFile

jeff.awk:

BEGIN {
  FS=" *: *"
  tab=sprintf("\t")

  width=20
}

function trim(str)
{
    sub("^([ ]*|" tab "*)", "", str);
    sub("([ ]*|" tab "*)" "$", "", str);
    return str;
}

{ $1=trim($1) }
!($1 in header){ header[$1]; headerL=(headerL)?headerL OFS sprintf("%-*s", width,$1): sprintf("%-*s", width,$1) }
$1 ~ /^ *Pass Id/ {pass=trim($2)}
{ v[pass]=(v[pass])? v[pass] OFS sprintf("%-*s", width,trim($2)) : sprintf("%-*s", width,trim($2)) }
END {
  print headerL
  for(i in v)
     print v
}
8

Holy S*!t

Dam that looks complicated, i cant even begin to understand that, just when you think you're getting somewhere !!!!!

Thanks i'll give them both a try and get back to you.

thanks again

Jeffers