utf8_stripspecials optimization

Add preconverted utf-8 string of special characters.

The (once only) conversion of the special character  unicode
array into utf-8 occurs on every DokuWiki page view,
irrespective of action or caching, and takes about one third
of the time involved in delivering a wiki page straight from
cache.

The original unicode array has been left in place in the file
to make any future amendments easier.

darcs-hash:20060923151937-9b6ab-cae0340a95d9596415ef71d7b7e67ef9daca84ef.gz

1 files changed, 23 insertions, 1 deletions

diff --git a/inc/utf8.php b/inc/utf8.php
index dbf09b6fc..0479d8809 100644
--- a/inc/utf8.php
+++ b/inc/utf8.php
@@ -366,10 +366,12 @@ function utf8_romanize($string){
  */
 function utf8_stripspecials($string,$repl='',$additional=''){
   global $UTF8_SPECIAL_CHARS;
+  global $UTF8_SPECIAL_CHARS2;
 
   static $specials = null;
   if(is_null($specials)){
-    $specials = preg_quote(unicode_to_utf8($UTF8_SPECIAL_CHARS), '/');
+#    $specials = preg_quote(unicode_to_utf8($UTF8_SPECIAL_CHARS), '/');
+    $specials = preg_quote($UTF8_SPECIAL_CHARS2, '/');
   }
 
   return preg_replace('/['.$additional.'\x00-\x19'.$specials.']/u',$repl,$string);
@@ -988,6 +990,26 @@ $UTF8_SPECIAL_CHARS = array(
   0xf8fb, 0xf8fc, 0xf8fd, 0xf8fe, 0xfe7c, 0xfe7d,
 );
 
+// utf8 version of above data
+global $UTF8_SPECIAL_CHARS2;
+$UTF8_SPECIAL_CHARS2 = 
+    ' !"#$%&\'()+,/;<=>?@[\]^`{|}~€‚ƒ„…†‡ˆ‰Š‹ŒŽ‘’“”•�'.
+    '�—˜™š›œžŸ ¡¢£¤¥¦§¨©ª«¬­®¯°±²³´µ¶·¸¹º»¼½�'.
+    '�¿×÷ˇ˘˙˚˛˜˝̣̀́̃̉΄΅·βφϑϒϕϖְֱֲֳִֵֶַָֹֻּֽ־ֿ�'.
+    '�ׁׂ׃׳״،؛؟ـًٌٍَُِّْ٪฿‌‍‎‏–—―‗‘’‚“”�'.
+    '��†‡•…‰′″‹›⁄₧₪₫€№℘™Ωℵ←↑→↓↔↕↵'.
+    '⇐⇑⇒⇓⇔∀∂∃∅∆∇∈∉∋∏∑−∕∗∙√∝∞∠∧∨�'.
+    '�∪∫∴∼≅≈≠≡≤≥⊂⊃⊄⊆⊇⊕⊗⊥⋅⌐⌠⌡〈〉⑩─�'.
+    '��┌┐└┘├┤┬┴┼═║╒╓╔╕╖╗╘╙╚╛╜╝╞╟╠'.
+    '╡╢╣╤╥╦╧╨╩╪╫╬▀▄█▌▐░▒▓■▲▼◆◊●�'.
+    '�★☎☛☞♠♣♥♦✁✂✃✄✆✇✈✉✌✍✎✏✐✑✒✓✔✕�'.
+    '��✗✘✙✚✛✜✝✞✟✠✡✢✣✤✥✦✧✩✪✫✬✭✮✯✰✱'.
+    '✲✳✴✵✶✷✸✹✺✻✼✽✾✿❀❁❂❃❄❅❆❇❈❉❊❋�'.
+    '�❏❐❑❒❖❘❙❚❛❜❝❞❡❢❣❤❥❦❧❿➉➓➔➘➙➚�'.
+    '��➜➝➞➟➠➡➢➣➤➥➦➧➨➩➪➫➬➭➮➯➱➲➳➴➵➶'.
+    '➷➸➹➺➻➼➽➾�'.
+    '�ﹼﹽ';
+
 /**
  * Romanization lookup table
  *