diff options
| author | Urban Müller | 2026-07-13 14:44:41 +0200 |
|---|---|---|
| committer | Urban Müller | 2026-07-13 14:44:41 +0200 |
| commit | 6524748e66c87ec899b6e24d02508ddd31265c17 (patch) | |
| tree | 8fec7a36d7ef159f50842be7c30af0eea0fbf4c8 /it.class | |
| parent | b1202bc02df27381c63c36085dd2fb2af028b25f (diff) | |
| download | itools-6524748e66c87ec899b6e24d02508ddd31265c17.tar.gz itools-6524748e66c87ec899b6e24d02508ddd31265c17.tar.bz2 itools-6524748e66c87ec899b6e24d02508ddd31265c17.zip | |
combining chars not well defined after blanks and cause more trouble after word split
Diffstat (limited to 'it.class')
| -rw-r--r-- | it.class | 4 |
1 files changed, 2 insertions, 2 deletions
@@ -609,8 +609,8 @@ static function any2utf8($value, $errprefix = "") list($value, $error) = array(preg_replace('/\xef\xb7[\x90-\xaf]|\xef\xbf[\xbe\xbf]/', " ", $value), "forbidden utf-8 character. input=$value"); $value = preg_replace('/\xc2\xad/', '', $value); # Kill invisible soft hyphens $value = normalizer_normalize($value, Normalizer::FORM_C); - if (preg_match('/^\pM/u', $value)) # Remove combining characters (e.g. U+0338 "Combining Long Solidus Overlay") at beginning of string - list($value, $error) = array(preg_replace('/^\pM+/u', "", $value), "combining character at beginning of string. input=$value"); + if (preg_match('/(^| )\pM/u', $value)) # Remove combining characters (e.g. U+0338 "Combining Long Solidus Overlay") at start of after blank + list($value, $error) = array(preg_replace('/(^| )\pM+/u', '$1', $value), "combining character at beginning of string. input=$value"); if ($error && $errprefix) it::error(array('title' => "$errprefix: " . trim($error))); } |