Why my php substr() shows obscure characters when cutting a text?

Question

I\'m using the substr() function to limit the characters in strings. but sometimes, the output text contains some obscure characters and Question marks etc...

Answer 1

Just to extend the Gurmbo is answer. Using mb_substr will solve your problem but still if special characters comes at the end when you trip, it still shows the some special characters. So when I did some research, wordpress having method wp_html_excerpt to solve this problem.

wp_html_excerpt method removes those special characters from the end of line.

Here is the source code from wordpress.

/**
 * Safely extracts not more than the first $count characters from html string.
 *
 * UTF-8, tags and entities safe prefix extraction. Entities inside will *NOT*
 * be counted as one character. For example & will be counted as 4, < as
 * 3, etc.
 *
 * @since 2.5.0
 *
 * @param string $str   String to get the excerpt from.
 * @param int    $count Maximum number of characters to take.
 * @param string $more  Optional. What to append if $str needs to be trimmed. Defaults to empty string.
 * @return string The excerpt.
 */
function wp_html_excerpt( $str, $count, $more = null ) {
    if ( null === $more )
        $more = '';
    $str = wp_strip_all_tags( $str, true );
    $excerpt = mb_substr( $str, 0, $count );
    // remove part of an entity at the end
    $excerpt = preg_replace( '/&[^;\s]{0,6}$/', '', $excerpt );
    if ( $str != $excerpt )
        $excerpt = trim( $excerpt ) . $more;
    return $excerpt;
}

Answer 2

Because you are cutting your characters into half.

Use mb_substr for multibyte character encodings like UTF-8. substr just counts bytes while mb_substr counts characters.

Answer 3

The reason is that you use UTF-8, it's multibyte encoding,and substr() works with singlebyte only! htmlentities() doesn't matter.

You SHOULD use mb_substr() http://php.net/manual/en/function.mb-substr.php and other multibyte functions

Answer 4

If you have encoding problems you can also apply the html_entity_decode() function that convert all HTML entities to their applicable characters. For example:

echo substr(html_entity_decode($string_to_cut), 0, 28) . "...";

That also should work.