[RESOLVED] url exclude characters

nickford

Well I haven't been using php for very long now, but I've been trying to build a url input page. I need php to check that no illegal characters are being used. I dont know how I can make explode use several different characters.
this is the source that I have attempted this with:

$host    = $_POST['whois'];
$suff    = $_POST['suffix'];
$whois   = $host . $suff;
$explode = explode ("|", $whois);

It's quite obvious that i need it to explode it at many more characters e.g.
-<>#%\,{}^[]
I have tried many different things like putting "or" between them all

$char = "|" or " -" or "<" or ">" or "#" or "%" or "\"" or "," or "{" or "}" or "^" or "[" or "]";
$explode = explode ($char, $whois);

, but this didn't work. :mad: Nether did any of the other things that I tried.
O yes, i also tried

$parse = parse_url($whois);
$disp = $parse['host'];

to get no luck from it.
If anyone could help I would be very gatefull. 😃
Nick.

drew010

[man]preg_match[/man] will check for illegal characters for you...

if (preg_match("/[\-<>#%\\,\{\}\^\[\]]/", $whois)) {
  echo "Illegal characters used.";
} else {
  echo "Input is okay.";
}

konsu

or better yet, write a regular expression that matches a valid url and use it to parse the url and check for errors at the same time.

nickford

What do you mean by that? Cold you elaberate please?
Many thanks
Nick.

konsu

regular expressions allow you to check if a string of characters conforms to a particular format, and if it does, to, optionally, extract substrings of this string. take a look at http://www.google.com/search?hl=en&q=regular+expressions

konsu

but, wait, there is parse_url() that you apparently used. what was wrong with it?

drew010

parse_url doesnt validate the url it just tries its best to break a string into url parts.

php -r 'print_r(parse_url("abcdefghi"));'
Array
(
    [path] => abcdefghi
)

konsu

right, i just read it too in the manual ; -). thanks. so to validate a url, i would just write a regular expression and use it to split the url at the same time.

nickford

Yer. When i tried parse_url I was trying to get it to display host. this didn't work because I was a whois query e.g. bladestream.com(my site)

LoganK

Well, I had prepared this whole elaborate reply, but since I opened tha tab in FF half-an-hour ago, and only got to it now (I open up many tabs in FF on PHPBuilder, then shuffle through them, seeing what to reply to), I didn't realize others had posted in the time it took me to get to this tab.

What a shame 🙁

How embarrasing! :o

And nick, about your problem - I'd go with preg_match(). Drew even gave you everything you need!

P.S. I also had a great joke in my posts with smilies, but the board won't let you use more than 3 smilies at a time.

nickford

Cheers Drew, worked like a charm.

drew010 wrote:

if (preg_match("/[\-<>#%\\,\{\}\^\[\]]/", $whois)) {
  echo "Illegal characters used.";
} else {
  echo "Input is okay.";
}

LoganK

Knew it'd work! 🙂

Mark thread resolved please! 😃