extract entries from a table based on values in another table 1

[gussifinknottle]

Hello,

I have two files that have delimited entries as shown below. I would like to use either Perl script to extract all the rows in File 1 corresponding to values in File 2 and output it to another File.

File 1
-------
1 36 24 Object1
2 45 36 Object2
3 96 21 Object3
4 89 32 Object4
5 98 23 Object5
6 99 21 Object6
7 56 21 Object7

File 2
-------
2 45 36
5 98 23
6 99 21


Thanks,
Gussi

 

[Zhris]

Hi,

Something like this should work. I am creating a hash table of input1 (key = what I will check against input2 | value = the full row [what I will output]) etc.

#! /usr/bin/perl
use strict;
use Cwd;

my $in1_path = getcwd.'/input1.txt';
my $in2_path = getcwd.'/input2.txt';
my $out_path = getcwd.'/output.txt';

my(%output);
open(IN1,"<$in1_path") || die "Cannot Open IN1";
while(my $current1 = <IN1>){
	$current1 =~ /^((\d\s\d\d\s\d\d)\s([A-Za-z0-9]*))/;
	$output{$2}=$1;
}
close(IN1);

unlink($out_path);
open(OUT,">>$out_path") || die "Cannot Open OUT";
open(IN2,"<$in2_path") || die "Cannot Open IN2";
while(my $current2 = <IN2>){
	rem_nl($current2);
	print OUT "$output{$current2}\r\n";
}
close(IN2);
close(OUT);

sub rem_nl{
	$_[0] =~ s/\r$//;
	$_[0] =~ s/\n$//;
	$_[0] =~ s/\r\n$//;
	$_[0] =~ s/\n\r$//;
	return $_[0];
}

Chris

 

[Zhris]

A bit cleaner:

my(%output);
open(IN1,"<$in1_path") || die "Cannot Open IN1";
while(my $current1 = <IN1>){
	$current1 =~ /^((\d\s\d\d\s\d\d)\s([A-Za-z0-9]*))/;
	$output{$2} = $1;
}
close(IN1);

unlink($out_path);
open(OUT,">>$out_path") || die "Cannot Open OUT";
open(IN2,"<$in2_path") || die "Cannot Open IN2";
while(my $current2 = <IN2>){
	$current2 =~ /^(\d\s\d\d\s\d\d)/;
	print OUT "$output{$1}\r\n";
}
close(IN2);
close(OUT);

Chris

 

[gussifinknottle]

Hi Chris,

When I ran this with a test case, it ran fine and give me the output I was looking for. However, when I changed the \d to a regular expression, ([0-9]*), to include search for numbers that may go from 1 to abut 100k, I don't see any output.

Thanks,
Gus

 

[KevinADC]

Try \d+ instead of [0-9]*

------------------------------------------
- Kevin, perl coder unexceptional!

 

[Zhris]

Whilst debugging the regex it may be a good idea to print $2 within the first while loop (IN1) and $1 within the second (IN2) because these are the 2 values that are being compared. If they don't match (IN2 key doesn't exist in hash table) then an empty line is outputted.

Chris