Hello there,

I've two files to compare as follows:

File1
1001
1003
1005
1007

File2
1005 1003 1007 1001
1002 1001 1006 1004
1003 1001 1005 1005
1003 1002 1006 1004

I would like to write a new File3 only if all the fields on each row in File2 contains any of the values in File1.

Based on the above my File3 would look like:
1005 1003 1007 1001
1003 1001 1005 1005

Thanks a lot.

awk ' FILENAME=="file1" { arr[$0]++}
        FILENAME=="file2" { ok=1; 
                                 for(i=1; i<=NF && ok==1; i++) 
                                       { if($i in arr) {ok=1} else {ok=0} }
                                 if(ok==1) { print $0}  } ' file1 file2 > file3

Hi Jim,

This works perfect. I would like to request one modification. Instead of comparing all the fields on each row, can we modify the script to compare only the second, third and fourth field on each row of File2 with all the values in File1 and if its a match, then to write out the entire row (resulting in the same output as before).

Regards,
kbirde

Hi Jim,

I got it to work by changing i=2.

Thanks.
kbirde

nawk '
function has(item){
	for(i=1;i<=max;i++)
	{
		if(_[i]==item)
			return 1
	}
	return 0;
}
{
if (NR==FNR)
{
	_[NR]=$0
	max=NR
}	
else{
	flag=1
	for(j=1;j<=NF;j++)
	{
		if(has($j)!=1)
			flag=0		
	}
	if(flag==1)
		print $0	
}
}' file1 file2

my %hash;
while(<DATA>){
	chomp;
	$hash{$_}=1;
}
open $fh,"<","file2";
while(<$fh>){
	my $flag=1;
	my @tmp=split;
	for(my $i=0;$i<=$#tmp;$i++){
		$flag=0 if not exists $hash{$tmp[$i]};
	}
	print if $flag == 1;
}
__DATA__
1001
1003
1005
1007