How to remove html special chars?

Either decode them using html_entity_decode or remove them using preg_replace:

$Content = preg_replace("/&#?[a-z0-9]+;/i","",$Content); 

(From here)

EDIT: Alternative according to Jacco's comment

might be nice to replace the '+' with {2,8} or something. This will limit the chance of replacing entire sentences when an unencoded '&' is present.

$Content = preg_replace("/&#?[a-z0-9]{2,8};/i","",$Content); 

Use html_entity_decode to convert HTML entities.

You'll need to set charset to make it work correctly.

You may want take a look at htmlentities() and html_entity_decode() here

$orig = "I'll \"walk\" the <b>dog</b> now";

$a = htmlentities($orig);

$b = html_entity_decode($a);

echo $a; // I'll &quot;walk&quot; the &lt;b&gt;dog&lt;/b&gt; now

echo $b; // I'll "walk" the <b>dog</b> now

If you want to convert the HTML special characters and not just remove them as well as strip things down and prepare for plain text this was the solution that worked for me...

function htmlToPlainText($str){
    $str = str_replace(' ', ' ', $str);
    $str = html_entity_decode($str, ENT_QUOTES | ENT_COMPAT , 'UTF-8');
    $str = html_entity_decode($str, ENT_HTML5, 'UTF-8');
    $str = html_entity_decode($str);
    $str = htmlspecialchars_decode($str);
    $str = strip_tags($str);

    return $str;
}

$string = '<p>this is (&nbsp;) a test</p>
<div>Yes this is! &amp; does it get "processed"? </div>'

htmlToPlainText($string);
// "this is ( ) a test. Yes this is! & does it get processed?"`

html_entity_decode w/ ENT_QUOTES | ENT_XML1 converts things like ' htmlspecialchars_decode converts things like & html_entity_decode converts things like '< and strip_tags removes any HTML tags left over.

EDIT - Added str_replace(' ', ' ', $str); and several other html_entity_decode() as continued testing has shown a need for them.

A plain vanilla strings way to do it without engaging the preg regex engine:

function remEntities($str) {
  if(substr_count($str, '&') && substr_count($str, ';')) {
    // Find amper
    $amp_pos = strpos($str, '&');
    //Find the ;
    $semi_pos = strpos($str, ';');
    // Only if the ; is after the &
    if($semi_pos > $amp_pos) {
      //is a HTML entity, try to remove
      $tmp = substr($str, 0, $amp_pos);
      $tmp = $tmp. substr($str, $semi_pos + 1, strlen($str));
      $str = $tmp;
      //Has another entity in it?
      if(substr_count($str, '&') && substr_count($str, ';'))
        $str = remEntities($tmp);
    }
  }
  return $str;
}

It looks like what you really want is:

function xmlEntities($string) {
    $translationTable = get_html_translation_table(HTML_ENTITIES, ENT_QUOTES);

    foreach ($translationTable as $char => $entity) {
        $from[] = $entity;
        $to[] = '&#'.ord($char).';';
    }
    return str_replace($from, $to, $string);
}

It replaces the named-entities with their number-equivalent.

If you are working in WordPress and are like me and simply need to check for an empty field (and there are a copious amount of random html entities in what seems like a blank string) then take a look at:

sanitize_title_with_dashes( string $title, string $raw_title = '', string $context = 'display' )

Link to wordpress function page

For people not working on WordPress, I found this function REALLY useful to create my own sanitizer, take a look at the full code and it's really in depth!

What If By "Remove HTML Special Chars" You Meant "Replace Appropriately"?

After all, just look at your example...

  & ©

If you're stripping this for an RSS feed, shouldn't you want the equivalents?

" ", &, ©

Or maybe you don't exactly want the equivalents. Maybe you'd want to have   just be ignored (to prevent too much space), but then have © actually get replaced. Let's work out a solution that solves anyone's version of this problem...

How to SELECTIVELY-REPLACE HTML Special Chars

The logic is simple: preg_match_all('/(&#[0-9]+;)/' grabs all of the matches, and then we simply build a list of matchables and replaceables, such as str_replace([searchlist], [replacelist], $term). Before we do this, we also need to convert named entities to their numeric counterparts, i.e., " " is unacceptable, but "&#00A0;" is fine. (Thanks to it-alien's solution to this part of the problem.)

Working Demo

In this demo, I replace { with "HTML Entity #123". Of course, you can fine-tune this to any kind of find-replace you want for your case.

Why did I make this? I use it with generating Rich Text Format from UTF8-character-encoded HTML.

See full working demo:

Full Online Working Demo

    function FixUTF8($args) {
        $output = $args['input'];
        
        $output = convertNamedHTMLEntitiesToNumeric(['input'=>$output]);
        
        preg_match_all('/(&#[0-9]+;)/', $output, $matches, PREG_OFFSET_CAPTURE);
        $full_matches = $matches[0];
        
        $found = [];
        $search = [];
        $replace = [];
        
        for($i = 0; $i < count($full_matches); $i++) {
            $match = $full_matches[$i];
            $word = $match[0];
            if(!$found[$word]) {
                $found[$word] = TRUE;
                $search[] = $word;
                $replacement = str_replace(['&#', ';'], ['HTML Entity #', ''], $word);
                $replace[] = $replacement;
            }
        }

        $new_output = str_replace($search, $replace, $output);
        
        return $new_output;
    }
    
    function convertNamedHTMLEntitiesToNumeric($args) {
        $input = $args['input'];
        return preg_replace_callback("/(&[a-zA-Z][a-zA-Z0-9]*;)/",function($m){
            $c = html_entity_decode($m[0],ENT_HTML5,"UTF-8");
            # return htmlentities($c,ENT_XML1,"UTF-8"); -- see update below
            
            $convmap = array(0x80, 0xffff, 0, 0xffff);
            return mb_encode_numericentity($c, $convmap, 'UTF-8');
        }, $input);
    }

print(FixUTF8(['input'=>"Oggi &egrave; un bel&nbsp;giorno"]));

Input:

"Oggi è un bel giorno"

Output:

Oggi HTML Entity #232 un belHTML Entity #160giorno