Why does this matching fail?

[vijdev]

$input=hjkhk$#@$@;
$test=preg_match_all("/^[A-Za-z]{1,20}/",$input,$match);

 

why does it pass $#@$@, although my definition ensures only alphas?

how do i ensure $#@$@ is not passed, or anything else from the keyboard like :;{"}[]! is not passed?

please help!

[premiso]

If I am not mistaken the {1,20} means 1 OR 20 not one through 20.

 

Chances are that is your issue.

[vijdev]

If I am not mistaken the {1,20} means 1 OR 20 not one through 20.

 

Chances are that is your issue.

 

I want to have minimum one character and maximum 20 characters for that input field.

[premiso]

Better to do that check outside of regex:

 

$stringLen = strlen($string);
if ($stringLen < 1 || $stringLen > 20 || !preg_match("/[A-Za-z]*/",$input) ) {
    echo 'You have provided invalid input. Valid input ......'; 
}

 

Untested, pending any retarded errors I made should work for what you want.

[vijdev]

Better to do that check outside of regex:

 

$stringLen = strlen($string);
if ($stringLen < 1 || $stringLen > 20 || !preg_match("/[A-Za-z]*/",$input) ) {
    echo 'You have provided invalid input. Valid input ......'; 
}

 

Untested, pending any retarded errors I made should work for what you want.

 

thanks, but would this solve my initial problem of not passing $#@$@ ?

[Goldeneye]

What exactly do you want this regular expression to do? You want it to match alphabetic-characters at the beginning of a string? If so, it appears to be working fine for me:

<?php
$input='hjkhk$#@$@';
preg_match_all("/^[A-Za-z]{1,20}/",$input,$match);
print_r($match);
?>

 

produces:

Array
(
    [0] => Array
        (
            [0] => hjkhk
        )

)

[newbtophp]

If I am not mistaken the {1,20} means 1 OR 20 not one through 20.

 

Chances are that is your issue.

 

Hmm I've always taught {1,20} meant the range of characters accepted, so in this case characters between 1-20 are allowed. But thanks for the correction.

[premiso]

 

thanks, but would this solve my initial problem of not passing $#@$@ ?

 

Did you try it?

 

The reason it was returning true was because 1 or 20 had an alpha character, which was a valid response.

[vijdev]

Ok, Let me put again the problem definition for clarity:

 

1.I need to check if input is only alphabets, not one single other type character

2.Needs to be minimum 1 char and max 20 characters long

 

This is what am intending to do with above expression.Any directions?..have to admit it looks so simple! :|

[DavidAM]

Unless I am terribly mistaken, {1,20} means anywhere between 1 and 20 (inclusive) occurances of the preceeding class. In this case, your RegExp will match any string that starts with 1 or more alphabetic characters (upper or lower case) regardless of how long the string is.  It will match this entire paragraph. 

 

I think if you anchor the RegExp to the end of the string as well, then it will do what you are expecting:

 

$input='hjkhk$#@$@';
$test=preg_match_all("/^[A-Za-z]{1,20}$/",$input,$match);

 

The caret ( ^ ) anchors to the start of the string and the dollar-sign ( $ ) anchors to the end. So now it should match a string of 1 to 20 alphabetic characters.

 

I am no RegExp expert, not even a little bit.  But I think this is correct. Give it a go

[ZachMEdwards]

Or using case-insensitive:

$pattern = '/^[A-Z]{1,20}$/i';

[vijdev]

thanks guys, the $ sign did help!...too many things to remember for a newbie like me!

but it was a great learning one!thanks a lot!