No illegal characters?

[Perad]

Hello,

 

Could someone show me what line of code I would need to add to a script to check a string to see if it contains anything other than 0-9, a-z, A-Z

 

Thanks

[jackpf]

'/[^a-zA-Z0-9]/'

I think would do it.

 

Or

if(preg_match('\W', $string))

would match any non-word characters.

[Ken2k7]

Or

if(preg_match('\W', $string))

would match any non-word characters.

Wouldn't that just match one non-word character? Same with your other Regex. It only match one character.

[jackpf]

Yeah, but if the string contains one letter that isn't alphanumeric then it's not alphanumeric duhh

 

And it should match all non-alphanumeric chars anyway.

 

Oh, and I made a mistake on the second one, it should be

'/\W/'

(with delimiters). My bad.

[Ken2k7]

Don't you mean:

/\W+/

?

 

Or possibly:

/\W*/

[jackpf]

Well it doesn't matter. It could contain 1 full stop or ten thousand hiphens, it would still contain

anything other than 0-9, a-z, A-Z

 

But yeah, either would work

[Ken2k7]

Oh right. My sincere apology. Early in the morning. Hehe...

[jackpf]

Lol. Half three here

[nrg_alpha]

The problem with the involvement of \w is that this also includes the underscore, as well as characters like é, à etc (depending on the locale from my understanding).

So the use \W, won't eliminate those characters (which falls outside of a-zA-Z0-9). jackpf, you're initial character would do it. One could also use:

 

#[^a-z0-9]#i

 

or one could make use of !ctype_alnum as well...

[Ken2k7]

Wouldn't this be good?

preg_match('/\W/', $str, $match);
if (empty($match)) // all valid

 

 

[nrg_alpha]

My bad on !ctype_alnum.. (as this includes characters like é...[perhaps again, locale dependent?]).

 

 

Wouldn't this be good?

preg_match('/\W/', $str, $match);
if (empty($match)) // all valid

 

You tell me

$str = '123aBc_dé';
echo $str . "<br />\n";
echo (preg_match('/\W/', $str))? 'Does not only contain a-zA-Z0-9' : 'Does only contain a-zA-Z0-9';

 

 

 

[nrg_alpha]

And of course, both ctype_alnum and \w also include exponents (1E²7³). So from where I see it, the only difference I can tell between ctype_alnum and \w is that cytpe_alnum lacks support for underscore.. so in either case, I would personally stick to #[^a-z0-9]#i

 

Damn you, ctype_alnum! *shakes fist*

 

EDIT - I guess one would have to set the proper locale for the desired effect for cype_alnum.

[jackpf]

preg_match('/\W/', $str, $match);

if (empty($match)) // all valid

I don't think that would work since $match would be an array. Idk if empty() checks arrays. If it does, forgive me.

 

If it doesn't, I think

preg_match('/\W/', $str, $match);
if (count($match) == 0) // all valid

would be better. Or not. Idk

[Ken2k7]

empty checks a mixed var. So pretty much anything.

[jackpf]

Oh right. Well ignore me then.

[premiso]

If you want to do a check, why not use the return of the preg_match

 

Return Values

 

preg_match() returns the number of times pattern matches. That will be either 0 times (no match) or 1 time because preg_match() will stop searching after the first match. preg_match_all() on the contrary will continue until it reaches the end of subject . preg_match() returns FALSE if an error occurred.

 

So...

 

preg_match('/\W/', $str, $match);
if (count($match) == 0) // all valid

 

Becomes:

if (preg_match('/\W/', $str, $match) > 0) // all valid

 

There is a reason they gave it a return value

[Ken2k7]

Hehe... yeah I did that really early in the morning. You caught me.