Re: Capturing substrings with complex delimiter, up to a maximum

My proposal is to first split and capture on http:// or https:// with an optional comma beforehand and then put back together:

sub getNurls {
    my( $s, $n ) = @_;
    my @urls = split /,?(https?:\/\/)/, $s;
    return map { defined $urls[2*$_-1]?$urls[2*$_-1].$urls[2*$_]:() } 
+(1..$n);
}

my $max = 2;
my $input = q|http://abc.org,http://de,f.org,https://ghi.org|;
print "$_\n" for getNurls( $input, $max );
[download]

(borrowing notation from prior posts)

Comment on Re: Capturing substrings with complex delimiter, up to a maximum Download Code

Replies are listed 'Best First'.
Re^2: Capturing substrings with complex delimiter, up to a maximum by tobyink (Canon) on Oct 30, 2013 at 10:17 UTC
But what if one of the URLs contains "http://"? For example... `http://translator.example.com/?url=http://www.example.fr/` [download] Rather than an optional comma, an improvement would be to look for a comma or `\A`: `/(?:\A\|,)(https?:\/\/)/` [download] `use Moops; class Cow :rw { has name => (default => 'Ermintrude') }; say Cow->new->name`	[reply] [d/l] [select]
Re^3: Capturing substrings with complex delimiter, up to a maximum by hdb (Monsignor) on Oct 30, 2013 at 11:05 UTC
That would be a problem. But can we rule out that a URL contains ",http://"?	[reply]