Date: 05/11/06
From: mike ()
Date: 05/11/06
On 5/11/06, Michael Chaney <> wrote:
> On Thu, May 11, 2006 at 03:04:58PM -0500, Tilghman Lesher wrote:
> > On Thursday 11 May 2006 14:42, mike wrote:
> > > eliminating spaces and allowing for existing dots or dashes:
> > >
> > > $phone =~ s/\s//g;
> > > $phone =~ s/^([0-9]{3})[-.]?([0-9]{3})[-.]?([0-9]{4})$/$1-$2-$3/
> >
> > Slight problem: you need to backslash the hyphen in the second
> > character class, or else it will translate to "all characters between
> > NUL (\0) and period (0x2e)", which, while it includes the hyphen
> > at 0x2d, also includes a lot more characters.
this i did not know.
> While you're at it, please use "\d" instead of "[0-9]".
i had this discussion earlier with my coworkers. are there any
reasons you know of why using \d is better than [0-9]? speed of
compilation is mostly a non-issue, particularly for mod_perl
applications that use pre-compiled regexes.
i prefer [0-9] because it is more readable, easier to alter the range
if neccessary, and reduces the number of backslashes in the regex that
commonly lead to Leaning Toothpick Syndrome.
also, from what i've read, \d and [0-9] are interchangeable,
regardless of locale.
-mike
-- Send all requests to:Put your command in the SUBJECT of the message: "subscribe" or "unsubscribe" Quick unsubscribe: <mailto:
This archive was generated by hypermail 2.1.6 : 05/11/06 CDT