Deleting elements of one list using another

Sj03rd has asked for the wisdom of the Perl Monks concerning the following question:

Greetings to you, oh great Monks

I would like to compare two arrays and keep only those elements that PARTIALLY intersect. For example, let the two array's be:

@arr1 = ("0000007 | John | ABC.txt | 42","0000014 | Jane | XYZ.txt | 34","0000017 | Jessica | GHI.txt | 21", etc);
@arr2 = (0000007, 0000014);
This should result in a text file arr3.dat that contains something like
0000007 | John | ABC.txt | 42
0000014 | Jane | XYZ.txt | 34

I know I should use hashes and know how to split element-by-element but can't figure this one out.

EDIT: CODE ADDED, SORRY FOR BEING UNSPECIFIC. THE FINAL ROW SUGGESTS THAT THE HASHING GOES WRONG...
EDIT AGAIN: SOLUTION BY nemesdani WORKS FINE (BUT SEE MY REPLY).
Thanks!


use feature ':5.10';

open (KEYFILE, "arr2.txt");
    @arr2 = <KEYFILE>;
close(KEYFILE);

# Declare and fill up the hash
my %elements;
foreach (@arr2) {
    $elements{$_} = 1;
};

#I'm repeating this for several quarterly files:
for($year=2006; $year<2012; $year=$year+1){
    for($i=1; $i<5; $i=$i+1){
    
        # Load each quarterly file
        $filea = $year . "QTR" . $i . "arr1.txt";
        open(MYINFILE, $filea);
            @arr1 = <MYINFILE>;
        close(MYINFILE);
        $sizegn = @arr1;

        # For each entry in the quarterly file
        for($j=0; $j<$sizegn; $j++){
            
            # Pick only lines that contain the string "txt"
            if($arr1[$j] =~ m/txt/){
            
                # splits the elements of @arr1, and format the first e
+lement into 7-digit number:
                @arraydata = split(/\|/, $arr1[$j]); 
                my $element = sprintf("%07d", $arraydata[0]); 
                
                # now check if $element is in hash :
                if (exists $elements{$element}){
                    open(MYOUTFILE, ">>Arr3.dat");
                    print MYOUTFILE $arr1[$j];
                    close(MYOUTFILE);
                }
                print "D'oh - '$element' not found\n" unless (exists $
+elements{$element});
            }
        }
    }
}
[download]

Comment on Deleting elements of one list using another Download Code

Replies are listed 'Best First'.
Re: Deleting elements of one list using another by nemesdani (Friar) on Aug 02, 2012 at 08:40 UTC
Untested and unnecessary verbose, but I think you'll get the idea: `use strict; use warnings; my @arr1 = ("John, ABC, 42","Jane, XYZ, 34","Jessica, GHI, 21"); my @arr2 = ("ABC", "XYZ"); my @arr3; my %seen; foreach my $arr1(@arr1) { my @splitted = split (/,/, $arr1); foreach my $arr2(@arr2) { if (grep(/$arr2/, @splitted)) { unless ($seen{$arr1} ){ push (@arr3, $arr1); } } } } for (@arr3) {print "$_\n";}` [download] I'm too lazy to be proud of being impatient.	[reply] [d/l]
Re^2: Deleting elements of one list using another by Sj03rd (Initiate) on Aug 02, 2012 at 12:16 UTC
Great, this works fine for me. Flawless code except for `@splitted` in the `grep` command which should be `@splitted[0]`. Thanks!	[reply] [d/l] [select]
Re^3: Deleting elements of one list using another by nemesdani (Friar) on Aug 02, 2012 at 20:25 UTC
You mean $splitted[0]? :) But no. splitted[0] contains "John" in the first loop, than "Jane", finally "Jessica". You'd like to grep the whole "John ABC 42" for ABC, don't you? Otherwise I haven't understand the problem well enough. I'm too lazy to be proud of being impatient.	[reply] [d/l]
Re: Deleting elements of one list using another by Anonymous Monk on Aug 02, 2012 at 08:11 UTC
I know I should use hashes and know how to split element-by-element but can't figure this one out. Figure what out, exactly (be specific)? Anyway, its FAQ $ perldoc -q duplicate perlfaq4 How can I remove duplicate elements from a list or array? $ perldoc -q substring perlfaq4 How can I count the number of occurrences of a substring within a string? Also see perlintro#Regular expressions the Simple matching paragraph	[reply]
Re: Deleting elements of one list using another by jwkrahn (Abbot) on Aug 02, 2012 at 09:45 UTC
`$ perl -le' use Data::Dumper; my @arr1 = ( "John, ABC, 42", "Jane, XYZ, 34", "Jessica, GHI, 21" ); my @arr2 = ( "ABC", "XYZ" ); my $pattern = join "\|", @arr2; my @arr3 = grep /$pattern/, @arr1; print Dumper \@arr3; ' $VAR1 = [ 'John, ABC, 42', 'Jane, XYZ, 34' ];` [download]	[reply] [d/l]
Re^2: Deleting elements of one list using another by suaveant (Parson) on Aug 02, 2012 at 13:39 UTC
Right idea but you aren't using anything to anchor the pattern, so you would also match 'JohnABC, CDE, 14' `@arr1 = ("0000007 \| John \| ABC.txt \| 42","0000014 \| Jane \| XYZ.txt \| 3 +4","0000017 \| Jessica \| GHI.txt \| 21", etc); @arr2 = ('0000007', '0000014'); my $re = join('\|',@arr2); my @arr3 = grep /^(?:$re)\s/, @arr1; print join("\n",@arr3),"\n";` [download] - Ant - Some of my best work - (1 2 3)	[reply] [d/l]
Re: Deleting elements of one list using another by aaron_baugher (Curate) on Aug 02, 2012 at 09:18 UTC
In pseudo-code: `make a hash of @arr2 instantiate @arr3 for each element of @arr1 split it to get the second field if that field is a key in the hash push the element onto @arr3` [download] Aaron B. Available for small or large Perl jobs; see my home node.	[reply] [d/l]