# First build a hash we can use to look up lines (keyed on 2nd line)
my %pairs;
open A, "<A";
while(<A>) { $pairs{<A>}=$_ }
close A;

# We use a third file
open C, ">C";
open B, "<B";
while(<B>) {
  # If this line matches a known 2nd line, insert it's 1st
  print C $pairs{$_} if exists $pairs{$_};
  # Always print the original line
  print C $_;
}
close B;
close C;
[download]

Also to add to Gilimanjaro's post, easy way to overwrite the original is by using rename.

rename (C,B);
[download]

Although when overwriting old files, be sure you are doing the right thing as returning to old might prove difficult. This depends on where the original file came from of course.

If u have the patience --can u explain how it works!!!
Thanx
Smbs

Thanx work just great
Smbs

At last I more or less understand it --but you mentioned in first line comment "(keyed on 2nd line)" --how is this done -the file was read in and every 2nd line is the key --is this some sort of default-what if I wanted every odd line (lines 1,3,5 etc) to be the keys??
Thanx
Smbs

It works because of the way the while loop works: The conditional in the while loop always gets evaluated first, before the content of the loop block. The <A> operator, when used in a while conditional, assigns the line gotten from file A to $_, starting with the 1st line.

The body of the loop, again uses the <A> operator, which gets the next line; you can access a line only once using the <> operator. I use it directly as the key to the %pairs hash, so the key is the 2nd line. It assigns the value of $_ (which was set in the while conditional) to this element of %pairs.

The loop evaluates the conditional again, gets the 3rd line, and assigns it to $_, and the body does the same all over again with the fourth line. So the even/odd mechanism is caused by the double use of <A>.

If I wanted to key on the odd lines, I'd've used $pairs{$_}=<A>;

I want to insert into file "B" (...) missing line above matching second line

Could you clarify that?

It's not at all clear (to me?) what want to do here:

1. Is it the first or second of each pair that must be inserted into the second file?
2. Where do you want the new values added to the second file?

   At the end? At some specific position?

3. How do you decide which values need adding?

4. Pls note lines may appear many times.

   Which lines appear many times in which file?

As you can see, your description left me very confused. Often in these cases the best expaination is a simple example?

Explanation by example
File "a"
abc
123
cfg
12345
fghh
1aaaaa


file "b"
ssssss
ffff
33
123
gggggkl
ffggg
1aaaaa

file "c" shoud read
ssssss
ffff
33
abc ----- note this inserted because "123" found in files "A","B"
123
gggggkl
ffggg
fghh ---- note this inserted because 1aaaaa found in files "A","B"
1aaaaa
Thanx
Smbs

use strict;
use warnings;

my ($in, $out, $key, $val, %hash);
my $goodf = 'inp1.txt';
my $badf = 'inp2.txt';
my $outf = 'out.txt';

open($in, $goodf);
while ($val = <$in>, $key = <$in>) {
    chomp($val, $key); $hash{$key} = $val;
}
close($in);

open($in, $badf);
open($out, ">$outf");
while ($key = <$in>) {
    chomp($key);
    print $out "$hash{$key}\n" if exists $hash{$key};
    print $out "$key\n";
}
close($in);
close($out);
[download]