Nested Foreach Loop skipping every other element??

ba1688 has asked for the wisdom of the Perl Monks concerning the following question:

So I have a nested for each loop

 
use strict;
use warnings;
use File::Copy;

my @files = glob("*.hml");
my @DIR = glob ("*/*.hml");


my @array = ('', grep -d, glob '*');
print "Here's DIR @DIR";
my $argsucess = 0;
foreach my $el (@ARGV){
    print "\n\n\nheres el $el";
    my $delele=0;
    foreach my $element (@DIR){
       print "\n\n\nHERE'S ELE $element";
       print "\n\n\n\n\n\n HERE'S DIR SO FAR @DIR\n\n\n";
        my $re1='((?:[a-z][a-z]+))'; # Non-greedy match on filler
        

        my $re=$re1;
        my $path;
        if ($element =~m/$re/is){
            $path = $1;
        }
       
        if ($el eq $path){
            
            splice(@DIR, $delele, 1);
            $argsucess++;
             next;     
        }
        $delele++;

    }
    
}
[download]

All I'm trying to make this do splice out the files in the directory I specify. It's doing it but for some reason it's doing it to every other element in the directory. For example if I have @DIR= functions/myfunc.hml perltest/abs.hml perltest/abs3.hml perltest/abs4.hml perltest/acos2.hml perltest/zeros17.hml It finds myfunc abs abs4 and zeros17 yet abs3 and acos2 are in the specified directory. It splices out the abs and abs3 yet leaves everything else in that directory. I need it to delete the array elements in @DIR that contain the "perltest" for example. Somethings wrong but I can't see what.

Comment on Nested Foreach Loop skipping every other element?? Download Code

Replies are listed 'Best First'.
Re: Nested Foreach Loop skipping every other element?? by hdb (Monsignor) on Feb 19, 2014 at 16:36 UTC
See http://perldoc.perl.org/perlsyn.html#Foreach-Loops: If any part of LIST is an array, foreach will get very confused if you add or remove elements within the loop body, for example with splice. So don't do that.	[reply]
Re^2: Nested Foreach Loop skipping every other element?? by ba1688 (Novice) on Feb 19, 2014 at 16:38 UTC
So should how else can I delete that element??	[reply]
Re^3: Nested Foreach Loop skipping every other element?? by hdb (Monsignor) on Feb 19, 2014 at 16:47 UTC
Instead of deleting the unwanted ones, you could push the wanted ones onto a new list. Or try to rewrite your conditions into a grep statement.	[reply]
Re^3: Nested Foreach Loop skipping every other element?? by vinoth.ree (Monsignor) on Feb 19, 2014 at 17:00 UTC
Instead doing splice on the same `@DIR` which you used in the `foreach`, you can have a copy of `@DIR` and you can do splice on that array. For Ex. `@DIR` will be used in `foreach`, `@DIR_COPY` can be used in `splice`. *All is well*	[reply] [d/l] [select]
Re: Nested Foreach Loop skipping every other element?? by Kenosis (Priest) on Feb 19, 2014 at 16:58 UTC
Seems `grep`ping, as hdb suggested, may produce your desired results: `use strict; use warnings; my @DIR = qw{functions/myfunc.hml perltest/abs.hml perltest/abs3.hml p +erltest/abs4.hml perltest/acos2.hml perltest/zeros17.hml}; my @filtered = grep !m{^perltest/}, @DIR; print "$_\n" for @filtered;` [download] Output: `functions/myfunc.hml` [download]	[reply] [d/l] [select]
Re: Nested Foreach Loop skipping every other element?? by kennethk (Abbot) on Feb 19, 2014 at 17:44 UTC
After considering hdb and Kenosis's advice, when I need to iterate over an array I am dropping terms from, I usually use an explicit index and iterate backwards over the indices. Note that you already need to use `$delele` (which I assume should be `$delete`) to track the index. As long as you are already tracking it, you might as well include that in your control structure. So maybe your code would read (untested): use strict; use warnings; use File::Copy; my @files = glob(".hml"); my @DIR = glob ("/.hml"); my @array = ('', grep -d, glob ''); print "Here's DIR @DIR"; my $argsucess = 0; foreach my $el (@ARGV){ print "\n\n\nheres el $el"; foreach my $delele (reverse 0 .. $#DIR){ my $element = $DIR[$delele]; print "\n\n\nHERE'S ELE $element"; print "\n\n\n\n\n\n HERE'S DIR SO FAR @DIR\n\n\n"; my $re1='((?:[a-z][a-z]+))'; # Non-greedy match on filler my $re=$re1; my $path; if ($element =~m/$re/is){ $path = $1; } if ($el eq $path){ splice(@DIR, $delele, 1); $argsucess++; next; } } } [download] The benefit to reverse traversal is that you don't interfere with the indices for future iterations by splicing this one out. You might also consider using `qr//` for your regular expression (see Regexp Quote Like Operators). I'm a little unclear on why you aren't just using a fixed regular expression in `if ($element =~m/$re/is){`, nor why you are using the `my $re=$re1;` misdirection. Also, by initializing `my $path;` to undef instead of empty string (`my $path = '';`), you will likely get a number of unnecessary warnings from `if ($el eq $path){`. #11929 First ask yourself `How would I do this without a computer?' Then have the computer do it the same way.	[reply] [d/l] [select]
Re^2: Nested Foreach Loop skipping every other element?? by AnomalousMonk (Archbishop) on Feb 19, 2014 at 22:36 UTC
`my $re1='((?:[a-z][a-z]+))'; # Non-greedy match on filler` ba1688: Also be aware that the quoted regex (also in the OP) is not "non-greedy" either by its definition or by virtue of the way it is used (Update: well, not in any regex-related sense, anyway).	[reply] [d/l]