这个正则表达式/ ^（[^-] *）（-*。）有什么问题？/？

Question

Consider this:

my @str = ("Farbkeil","L 0AA61 Rec","L 0AA61 Rec - 150 dpi",,"L 0AA61 Rec - 400 dpi");
for my $s (@str) {
#   my ($m) = ($s =~ /^([^-]*)(?=-)/);
#   my ($m) = ($s =~ /^([^-]*) (?=-)/);
    my ($m) = ($s =~ /^([^-]*)( - *.)?/);
    print "$s => -$m-\n";
}

It produces this output

Farbkeil => -Farbkeil-
L 0AA61 Rec => -L 0AA61 Rec-
L 0AA61 Rec - 150 dpi => -L 0AA61 Rec -
L 0AA61 Rec - 400 dpi => -L 0AA61 Rec -

What I want is to get rid of the space before the optional "-", so that it looks like

Farbkeil => -Farbkeil-
L 0AA61 Rec => -L 0AA61 Rec-
L 0AA61 Rec - 150 dpi => -L 0AA61 Rec-
L 0AA61 Rec - 400 dpi => -L 0AA61 Rec-

I've tried the above regexes, the last one is the closest but not quite. Guessing didn't work either, surprisingly... Of course I could just trim the result, but there must be a more elegant way?

Any ideas?

Answer 1

You can use this regex for match:

/^([^-]+)(?=\s-|$)/

RegEx Demo

(?=\\s-|$) is a lookahead that makes sure first part ([^-]+) is always followed by a space + hyphen OR end of input .

Answer 2

And another one. Just because it requires only one small change to your initial Regex. (note the \\b outside the round brackets). \\b is for word boundary.

my @str = ("Farbkeil","L 0AA61 Rec","L 0AA61 Rec - 150 dpi",,"L 0AA61 Rec - 400 dpi");
for my $s (@str) {
    my ($m) = ($s =~ /^([^-]*)\b(-)?/);
    print "$s => -$m-\n";
}

As for why your Regex failed:

([^-]*) greedily matches everything that is not a - . Thus it always matches the trailing space if it's there. The \\b forces a stop and can match amongst others a space or end of line.

Answer 3

I'd just remove / - .*/

for my $s (@str) {
    (my $m = $s) =~ s/ - .*//;
    print "$s => [$m]\n";
}

In more recent Perls, you can use /r :

my $m = $s =~ s/ - .*//r;