Re: Efficiency: Foreach loop and multiple regexs

neilwatson said:

foreach $x (@regex){
   @bad_matches = grep ! m/$x/, @array;
}
[download]

After the grep how would I remove the entries in @bad_matches from @array before the next $x? Would this be faster?

I would suggest that you use the following mechanism (incorporating some ideas from other contributors):

foreach my $regex (@regexes) {
  @array = grep { ! m/$regex/ } @array;
}
[download]

If you really need to keep @bad_matches (those that *did* match the regexes) then you probably want to run the loop manually, as somebody else suggested:

my (@array); # is populated however you originally did
my (@keepers);
my (@bad);
ITEM: # label added
while (@array) {
  my $item = shift @array;
  foreach my $regex (@regexes) {
    if ($item =~ $regex) {
       push @bad, $item;
       next ITEM; # return to label
    }
  }
  push @keepers, $item;
}
# if you want it back on @array, just uncomment:
# @array = @keepers;
[download]

HTH, -- jkahn

Update:(09:37 PDT) added ITEM: label and next ITEM line. Whoops. Code will be substantially sub-optimal without them.

Comment on Re: Efficiency: Foreach loop and multiple regexs Select or Download Code

Replies are listed 'Best First'.
Re: Re: Efficiency: Foreach loop and multiple regexs by waswas-fng (Curate) on Sep 13, 2002 at 16:42 UTC
I like this, it is pretty fast and yet readable: #!/usr/local/bin/perl my @regexs = ('test12','test2','test3','test4'); my $filename = '/var/tmp/testdata'; my @compregex = (); my (%match, %nomatch) = (); # Pre-compile regexes, sorting smaller to larger # (shorter prolly will match more than a longer =) foreach my $regex (sort { length($a) <=> length($b) } @regexs) { print "Building regex for $regex\n"; push @compregex, qr/$regex/; } open FILENAME, $filename; # could rewrite this with map, used while and foreach cause # most people know how they work, some dont understand map =) while (<FILENAME>) { chomp; foreach my $test (@compregex) { ($match{"$_"})++ if (/$test/) ; last if $match{"$_"}; #found a match no need t +o go further } ($nomatch{"$_"})++ unless $match{"$_"}; } print "Matches:\n"; foreach my $X (keys %match) { print "$X\n"; } print "no Matches:\n"; foreach my $X (keys %nomatch) { print "$X\n"; } [download] -Waswas-fng	[reply] [d/l]

Replies are listed 'Best First'.

Re: Re: Efficiency: Foreach loop and multiple regexs
by waswas-fng (Curate) on Sep 13, 2002 at 16:42 UTC

#!/usr/local/bin/perl

my @regexs = ('test12','test2','test3','test4');
my $filename = '/var/tmp/testdata';
my  @compregex  = ();
my (%match, %nomatch) = ();

#  Pre-compile regexes, sorting smaller to larger
# (shorter prolly will match more than a longer =)
foreach my $regex (sort { length($a) <=> length($b) } @regexs) {
        print "Building regex for $regex\n";
        push @compregex, qr/$regex/;
        }

open FILENAME, $filename;

# could rewrite this with map,  used while and foreach cause
# most people know how they work, some dont understand map =)
while (<FILENAME>) {
        chomp;
        foreach my $test (@compregex) {
                        ($match{"$_"})++  if (/$test/) ;
                        last if $match{"$_"}; #found a match no need t
+o go further
        }
        ($nomatch{"$_"})++ unless $match{"$_"};
}

print "Matches:\n";
foreach my $X (keys %match) {
        print "$X\n";
}
print "no Matches:\n";
foreach my $X (keys %nomatch) {
        print "$X\n";
}
[download]

[reply]
[d/l]