* check if iconv extension is loaded and, if so, use it in UtfNormal::cleanUp()
wfProfileOut( 'djvutxt' );
if( $retval == 0) {
# Get rid of invalid UTF-8, strip control characters
- if( is_callable( 'iconv' ) ) {
- wfSuppressWarnings();
- $txt = iconv( "UTF-8","UTF-8//IGNORE", $txt );
- wfRestoreWarnings();
- } else {
- $txt = UtfNormal::cleanUp( $txt );
- }
+ $txt = UtfNormal::cleanUp( $txt );
$txt = preg_replace( "/[\013\035\037]/", "", $txt );
$reg = <<<EOR
/\(page\s[\d-]*\s[\d-]*\s[\d-]*\s[\d-]*\s*"
* @return string a clean, shiny, normalized UTF-8 string
*/
static function cleanUp( $string ) {
+ if( extension_loaded( 'iconv' ) ) {
+ wfSuppressWarnings();
+ $ret = iconv( "UTF-8","UTF-8//IGNORE", $string );
+ wfRestoreWarnings();
+ return $ret;
+ }
+
if( NORMALIZE_ICU ) {
# We exclude a few chars that ICU would not.
$string = preg_replace(