Re: Re: validate URL

Hi, > Davey <davey@php.net> wrote : > From the URI RFC, comes this amazing section: > ... > ^(([^:/?#]+):)?(//([^/?#]*))?([^?#]*)(\?([^#]*))?(#(.*))? > 12 3 4 5 6 7 8 9 > ... > I have yet to actually get this regex to work with ereg... the following version is working with ereg : function ereg_parse_url($s) { if (ereg('^(([^:/?#]+):)?([/]{2}([^/?#]*))?([^?#]*)(\\?([^#]*))?(#(.*))?',$s,$regs)) { return $regs; } else { return ''; } } > I think unless you need to *parse* the URI a lot of the parens could be > removed... and moving it to PCRE instead of POSIX would be beneficial. > This needs a regex guru for sure :D and the following one is working with preg ;D function preg_parse_url($s) { if (preg_match('!^(([^:/?#]+):)?(//([^/?#]*))?([^?#]*)(\?([^#]*))?(#(.*))?!',$s,$matches)) { return $matches; } else { return ''; } } you can notice that the only difference with the generic URI syntax as found in the RFC are just the ! at the start and the end of the regex... no need to be a regex guru ;-) Regards, -------------------- Frederic Poeydomenge