[Resolved] How to create pattern matching for letters, digits, commas, periods, and spaces ONLY?

sherry

I am trying to use preg_match (or ereg -- whatever works) for form field validation.

In words, I want only letters, digits, commas, periods, spaces, DASHES, and NOT anything else.

With ereg, I tried:
ereg("([^a-zA-z0-9.\, -])", $field1, $arr1)

This didn't detect '[' nor ']'.

With preg_match, I tried:
preg_match("/[^a-zA-z0-9.\, -]/", "$field1", $arr1)

This didn't detect '[', ']', nor '^'. I haven't seen how to use '^' in the same way it is defined in ereg as 'not', so I still need to find this out, also.

How to create the pattern matching statement from 'In words...' above?

I thought this would have been straight-forward.

Thank you.

drew010

try

if (!preg_match("/^[\w\d-\.,\s]+$/", $string)) {
  echo "bad";
} else {
  echo "good";
}

^ can mean the start of a string, but if you use it inside your character class, like you do above, it actually tells regex to match any character except the one(s) after it. for example, [^xyz] matches anything except xyz, but ^[xyz] wants xyz to be the first letters in the string. the $ is used to denote end of line (or string) in this case.

sherry

if (!preg_match("/^{[\w\d-.,\s]+$/",} $field1, $arr1)) {...}

I put the following in the field to test:

]]hjkl^]hjkl{

The result is that $arr1 is empty (count($arr1) = 0).

It seems like this preg_match() function would return 'true' since the field does contain a letter OR digit or .... and would store the letters, in this field's case, in the $arr1. I interpret it (since I'm new to this function), as 'if preg_match doesn't match letters, digits, dashes, periods, commas, nor spaces, then do something {...}.

In words, what is this statement saying?

Thank you.

BuzzLY

[FONT=courier new]$found = preg_match('§[^{a-z0-9.,- ]§i', $thestring)[/FONT]

This line says "count any character that is not a-z, 0-9, period, comma, dash or space." The i modifier makes it case-insensitive, so you don't have to specify a-z and A-Z.

If there are any matches, $found will be > 0. If no matches, it will === 0, and if there was an error, it will === FALSE.}

sherry

Thank you for your reply, as usual.

THIS WORKS PERFECTLY!

Q: What is the char '§' in your post and what does it mean? At first I thought it was '$'.

Thanks.

BuzzLY

It's just the delimiter I like to use. Most people use /, but I like to use §, since the odds of that character occuring in the pattern is VERY slim. If you use / as your delimiter, and you need to search for a / in your text, then you need to escape it: \/. Using § eliminates the need to escape /.

sherry

Thanks to both you and Drew010. This post is now resolved!