Nonbreaking spaces appearing oddly in plain text excerpts

I have the same problem…

@dtweney

This might be to do with expected character encoding, in the plugin class/class-s2-core.php file, in the mail() function, find the following line:
$message = html_entity_decode( strip_tags( $message ), ENT_NOQUOTES, 'UTF-8' );

Try replacing it with this:
$message = strip_tags( html_entity_decode( $message, ENT_NOQUOTES ) );

@mattyrob

I think that would just change the encoding. What I really want to do is strip out the nonbreaking spaces.

But I can try it! I will see if that helps.

Actually, the code looks a little different in my version, and it seems to be UTF-8 encoding the text already. Relevant lines:

			$message = html_entity_decode($message, ENT_NOQUOTES, 'UTF-8');
			$message = wordwrap(strip_tags($message), $this->word_wrap, "\n");

And that sorta makes sense, because those B! thingies seems like something that might appear when a one-byte character set tries to interpret two-byte unicode.

So I’m going to try changing that UTF-8 into ISO-8859-1.

Will report back.

@dtweney

I think it might work, try the following on a page on your site:

$string = "  ";
var_dump( $string );
var_dump( html_entity_decode( $string ) );

The first should report the string length as 12 and hopefully the latter should report the string length lower.

OK, so in the class-s2-core.php file, I changed:

$message = html_entity_decode($message, ENT_NOQUOTES, 'UTF-8');
$message = wordwrap(strip_tags($message), $this->word_wrap, "\n");

to

$message = html_entity_decode($message, ENT_NOQUOTES, 'ISO-8859-1');
$message = wordwrap(strip_tags($message), $this->word_wrap, "\n");

And it seems to have worked! No more “B!” appearing in place of nonbreaking spaces in some mail clients!

I will continue to monitor to see if this breaks anything else, but for now it seems to have done the trick.

Dylan.