Shell Programming and Scripting

Hi,

I have a file like this.

please notice that ./usr/orders1/order_new_2627 appears more than once, thus needs to be merged.

I would like to merge the lines where the first column match

so the output should be like this:
Please help

<rant mode>
It's not a UNIX.com rule at present, but personally I really dislike seeing 'URGENT' in the subject without any explanation of why
</rant mode>

That said, I see you've also been active in helping others so I should probably just calm down

Not a very elegant solution but it should do the job (note that this will go all wrong if you have any spaces in the "/usr/orders1/order..." names in your list).

Code:

#!/usr/bin/perl -w
my %lines;
while (<>) {
  if (/^\s*([^\s]+) (.*)$/) {
    $lines{$1}.="$lines{$1}:$2";
  }
}
foreach $key keys(%lines) {
  print "$key $lines{$key}\n";
}

(Untested)

Code:

awk '{
arr[$1]=sprintf("%s:%s",arr[$1],substr($0,length($1)+1))
}
END{
for ( i in arr)
        print i" " arr[i]
}' file