removing array elements

[AMiSM]

Hi, everybody!

I need to remove the elements of list number 2 from list number 1. Can somebody please tell why this doesn't work:

$outputFile = "C:\\Documents and Settings\\a\\Desktop\\Output.doc";

unlink ( $outputFile );
while ( -e $outputFile ) {}


open ( LIST1, "<", "D:\\ATLinks.doc" );
@list1 = <LIST1>;
close ( LIST1 );
$list1 = "@list1";

open ( LIST2, "<", "D:\\ATLog1.doc" );
@list2 = <LIST2>;
close ( LIST2 );
chomp ( @list2 );

for ( @list2 ) {
$list1 =~ s/$_//gms;
}

open ( FIL, ">>", $outputFile );
print FIL $list1;
close ( FIL );

 

[PaulTEG]

Look into the array splice function

Paul
------------------------------------
Spend an hour a week on CPAN, helps cure all known programming ailments ;-)

 

[AMiSM]

I've seen that mentioned elsewhere, but I'm not clear on how it applies. I've seen how it can be used to overwrite one array over the other, from a specific point onward, but how do you use it to remove one array from another, especially when the overlap is fragmented?

Is this a drop-in solution, or do you incorporate it into an algorithm?

Please excuse my larval ignorance, here.

 

[cfmartin2000]

Try something like this:

my %list_1;
my %list_2; #contains the elements to remove from list 1

open (FILE1, "file1") or die $!;
while(<FILE1>){
chomp($_);
$list_1{$_}=1;
}
close(FILE1);

open (FILE2, "file2") or die $!;
while(<FILE2>){
chomp($_);
$list_2{$_}=1;
}
close(FILE1);


delete @list_1{keys %list_2};


print join("\n", keys %list_1);

 

[AMiSM]

Whoa!! That's works so cool!! I'll be looking at this for awhile. It doesn't quite make sense to me, especially 'delete'.

The '%' denotes a hash, right? I've read that you use hashes for this kind of thing, but hashes seem a bit archane to me, just yet. Onward and upward, I guess.

Thanks, man!

 

[KevinADC]

'delete' does just that, it deletes a key/value pair from a hash. It does work on arrays but not the same as it does on hashes:

http://perldoc.perl.org/functions/delete.html

 

[AMiSM]

cfmartin, why is the hash 'list_1' accessed as an array in your code?

==> delete @list_1{keys %list_2};

Is it one of those context things? Does delete operate only on the keys this way, or something?

 

[cfmartin2000]

It's a hash slice which I believe is a list context. This is similar to $hash{'an_item'} being in a scalar context. The clue here is the curly brackets which let you know you are working with a hash.

You could write the same thing as follows keeping everything in a scalar context:

foreach my $invalid (keys %list_2){
delete ($list_1{$invalid});
}

 

[AMiSM]

So, essentially, to get an array from a hash,...

@some_array = @some_hash{ [some_statement_that_returns_an_array_context] };

correct?

 

[cfmartin2000]

Yes exactly. Your @some_array contains the values from your %some_hash slice.

Like this:

my %hash;
@hash{'A'..'Z'} = (1..26); #Same as %hash = (A=>1, B=>2, C=>3 ....)

my @list;
@list = @hash{ qw/A B D F/ };

#@list = (1, 2, 4, 6);

 

[AMiSM]

I think I understand why hashes are used for the two lists problem. Between the 'delete' and 'grep' functions, I'm sensing a way of thinking and doing things.
This is some very cool stuff! I'm liking Perl more and more!