Re: Pattern matching \b

[Roberto Etcheverry]

[EMAIL PROTECTED] wrote:
$field = "Search, This is part of the code.";
## We need to split by spaces. The issue is that the comma comes along.
local(@words) = split('\s+', $field);
   foreach $word (@words) {
       if ($word =~ /Search/i) {
           $word =~ s/[,\]\)\}]\b//;
           $word =~ s/\b[,\]\)\}]//;
                   push(@new_array, $word);
                   last;
               }
       }        

This code removes the comma. But I am puzzeled. I thought that the first line
   $word =~ s/[,\]\)\}]\b//; 
would delete the comma.
Yet it was the second line that did it.
       $word =~ s/\b[,\]\)\}]//;

The comma is at the end of the word, before the boundary. Shouldn't the first 
line have gottne rid of it.

 

\b matches the word-nonword or nonword-word boundary. This point is 
between the 'h' and the ',' in 'Search,'.
The first substitution would match ',abc', for example.

Won't be easier to split on [,\]\}\s]+? Like:
$field = "Search, This is part of the code.";
my @words =  split(/[,\]\}\s]+/,$field);
print join('|',@words); # prints: Search|This|is|part|of|the|code.
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
<http://learn.perl.org/> <http://learn.perl.org/first-response>