[RESOLVED] Pattern question

scialom

Hello all,

I have a quick question - I want to check a string to see that it is only has letters (a-z A-Z) and numbers (0-9) and not any other chars

So $str = "formula1"; is 1
but $str = "formula+"; is 0

I know that I should use patterns but I don't really know how to do so.
Any idea?
Thanks!

laserlight

In this case use [man]ctype_alnum/man.

scialom

thanks!

nrg_alpha

I'm going to fathom a guess in that PCRE doesn't have a special sequence character for just that exact purpose.. (at least, not that I have come across as of yet).

Closest thing would be [a-z\d] with the i modifier after the closing delimiter. It seems like it would be a handy thing to be able to use in one simple step (much like \s is shorthand for a character class that matches tabs, spaces, newline and carriage return).

Then again, perhaps ctype_alnum() is faster?

laserlight

I'm going to fathom a guess in that PCRE doesn't have a special sequence character for just that exact purpose.. (at least, not that I have come across as of yet).

Yes, I cannot find such a character type either, even among the Unicode character properties.

Closest thing would be [a-z\d] with the i modifier after the closing delimiter. It seems like it would be a handy thing to be able to use in one simple step (much like \s is shorthand for a character class that matches tabs, spaces, newline and carriage return).

If the character class included the underscore, \w would be the one to use.

Then again, perhaps ctype_alnum() is faster?

Yes, I believe that is the case.

nrg_alpha

laserlight;10883736 wrote:
If the character class included the underscore, \w would be the one to use.

The issue with \w is not only an underscore.. it technically also matches characters such as é ä Ä Ü Ö ü to mention a few, (though I'm not sure if this is a locale issue or not)...

On my screen, \w != pure [a-zA-Z0-9_] (but I suppose for general purpose stuff, people use this for general default listed purposes). <-- I just re-read this... not sure if I made any sense or not.

laserlight

The issue with \w is not only an underscore.. it technically also matches characters such as é ä Ä Ü Ö ü to mention a few, (though I'm not sure if this is a locale issue or not)...

Yes, depending on the locale, accented letters may also be matched by \w.