check numeric or alphabetic

[louis_coetzee]

hi, new to php, can anyone help me, i need a function to check tel no input, to check if it is numeric or text if text, ask to please put in valid no. thanks

[Maq]

is_numeric($variable)  //numeric

if (trim($variable) == ""))  //empty

 

[genericnumber1]

Alphanumeric:

<?php
function is_alphanumeric($string) {
return !preg_match('#[^A-Za-z0-9]#', $string);
}

 

Alphabetical:

<?php
function is_alphabetical($string) {
return !preg_match('#[^A-Za-z]#', $string);
}

[.josh]

Alphanumeric:

<?php
function is_alphanumeric($string) {
return !preg_match('#[^A-Za-z0-9]#', $string);
}

 

Alphabetical:

<?php
function is_alphabetical($string) {
return !preg_match('#[^A-Za-z]#', $string);
}

 

whoa there buddy, step away from the regex.  There's no reason to be doing double negatives like that. ^ inside a character class will only match something that is not listed, so you're having to turn around and use ! to do the opposite of that negative.

 

Also, you're only checking for 1 character.  The first character.  So if a string is '1abc' it will return positive for numeric. You need to add a quantifier and start and end of string assertion.

 

if preg_match('~^[A-Za-z]+$~', $string)
   // is alpha
if preg_match('~^[A-Za-z0-9]+$~', $string)
   // is alphanumeric

 

you could also use ctype_digit ctype_alnum ctype_alpha

[genericnumber1]

Actually, give it a try, it will search every character and function as expected until it finds a match (which it shouldn't, because matches are bad in this case), you should know that! But you're right about the double negatives, it can be confusing. As far as your solution goes, I've seen an example where a parser will stop at a new line character that matches between the start-line and end-line character. Eg. "test\n@" will return true for your alphanumeric pattern. I doubt that's true in this case, but I didn't feel up to trying it... hell, I could be mistaken, but I went with what I knew would work.

[.josh]

Actually, give it a try, it will search every character and function as expected until it finds a match (which it shouldn't, because matches are bad in this case), you should know that! But you're right about the double negatives, it can be confusing. As far as your solution goes, I've seen an example where a parser will stop at a new line character that matches between the start-line and end-line character. Eg. "test\n@" will return true for your alphanumeric pattern. I doubt that's true in this case, but I didn't feel up to trying it... hell, I could be mistaken, but I went with what I knew would work.

 

well damn...upon further inspection, I was wrong even about double negative.  Apparently 2 wrongs do make a right.  It does search until it finds a match. So if you match

 

return !preg_match('#[^A-Za-z]#', $string);

 

against

asdf

2asdf

sdf2

2323

you will get

 

false

true

true

true

 

from the negated class, but the ! will invert it.  Okay my bad, it does work.  But still, you don't need to make it so complicated, lol.  As far as mine returning a true on a new line:

 

$string = "test\n@";

if (preg_match('~^[A-Za-z]+$~', $string))
   echo "$string true";
else
   echo "$string false";
echo "<br/>";
if (preg_match('~^[A-Za-z0-9]+$~', $string))
   echo "$string true";
else
   echo "$string false";

 

output:

 

test @ false

test @ false

[genericnumber1]

So we were both wrong, call it a draw! (I try to add Monty Python references to every day conversation at every opportunity.)

 

Yours is certainly more readable.