preg_replace everything but alpha numeric and accents

Question

Been searching for a while , but as I do not understand the preg-replace parameters that can be entered, I can't solve my problem by looking at similar questions. (therefore I would like to ask that you could explain each sign in your answer?)

I want to filter a string and replace everything that is not alphanumeric or an accent with a space. (After this I use preg_split to make a list of them).

Under here the symbols I known as accents, showed only the lower-case, but want to keep the upper-case variants too.(might have forgotten a few). I really need the left signs. The right are not obligatory:

é è ë ï ê                ç ü ö ä ú ó á ã õ ñ å û â ô

I need this because I have to search my database(prepared statements) for it.

Currently I use:

 $terms=preg_split('/\s+/', preg_replace('/[^a-zA-Z0-9À-ÿ]+/', ' ', $_GET['ter']));

But the following example shows that it does not leave the signs as they are:

url example: mydomain.com/index.php?ter=léopold

$terms=preg_split('/\s+/', preg_replace('/[^a-zA-Z0-9À-ÿ]+/', ' ', $_GET['ter']));
foreach ($erms as $term){echo " term: ".$term;}


Outputs:
term: l� term: opold

What i want it to be:
term: léopold

So I need to create a preg_replace that does not harm the accents ,especially the ones written on the left in the first code-block

Answer 1

Use the mb_ereg_replace, preg_replace with multibyte support