From: Jarkko Hietaniemi Date: Wed, 3 Mar 2004 09:37:21 +0000 (+0200) Subject: Speed up the unicode case mappings (for /i, lc, uc, etc). X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=commitdiff_plain;h=b08cf34e02b6d68e83a46f7566341e6914ff1a2e;p=p5sagit%2Fp5-mst-13.2.git Speed up the unicode case mappings (for /i, lc, uc, etc). Subject: [PATCH] [perl #24826] Message-Id: <9B5CBF96-6CE5-11D8-83B0-00039362CB92@iki.fi> p4raw-id: //depot/perl@22427 --- diff --git a/lib/unicore/To/Fold.pl b/lib/unicore/To/Fold.pl index ee60833..33c5d46 100644 --- a/lib/unicore/To/Fold.pl +++ b/lib/unicore/To/Fold.pl @@ -3,111 +3,112 @@ # Any changes made here will be lost! +# The ke UTF-8 _bytes_, the value UTF-8 (speed hack) %utf8::ToSpecFold = ( -'00DF' => "\x{0073}\x{0073}", -'0130' => "\x{0069}\x{0307}", -'0149' => "\x{02BC}\x{006E}", -'01F0' => "\x{006A}\x{030C}", -'0390' => "\x{03B9}\x{0308}\x{0301}", -'03B0' => "\x{03C5}\x{0308}\x{0301}", -'0587' => "\x{0565}\x{0582}", -'1E96' => "\x{0068}\x{0331}", -'1E97' => "\x{0074}\x{0308}", -'1E98' => "\x{0077}\x{030A}", -'1E99' => "\x{0079}\x{030A}", -'1E9A' => "\x{0061}\x{02BE}", -'1F50' => "\x{03C5}\x{0313}", -'1F52' => "\x{03C5}\x{0313}\x{0300}", -'1F54' => "\x{03C5}\x{0313}\x{0301}", -'1F56' => "\x{03C5}\x{0313}\x{0342}", -'1F80' => "\x{1F00}\x{03B9}", -'1F81' => "\x{1F01}\x{03B9}", -'1F82' => "\x{1F02}\x{03B9}", -'1F83' => "\x{1F03}\x{03B9}", -'1F84' => "\x{1F04}\x{03B9}", -'1F85' => "\x{1F05}\x{03B9}", -'1F86' => "\x{1F06}\x{03B9}", -'1F87' => "\x{1F07}\x{03B9}", -'1F88' => "\x{1F00}\x{03B9}", -'1F89' => "\x{1F01}\x{03B9}", -'1F8A' => "\x{1F02}\x{03B9}", -'1F8B' => "\x{1F03}\x{03B9}", -'1F8C' => "\x{1F04}\x{03B9}", -'1F8D' => "\x{1F05}\x{03B9}", -'1F8E' => "\x{1F06}\x{03B9}", -'1F8F' => "\x{1F07}\x{03B9}", -'1F90' => "\x{1F20}\x{03B9}", -'1F91' => "\x{1F21}\x{03B9}", -'1F92' => "\x{1F22}\x{03B9}", -'1F93' => "\x{1F23}\x{03B9}", -'1F94' => "\x{1F24}\x{03B9}", -'1F95' => "\x{1F25}\x{03B9}", -'1F96' => "\x{1F26}\x{03B9}", -'1F97' => "\x{1F27}\x{03B9}", -'1F98' => "\x{1F20}\x{03B9}", -'1F99' => "\x{1F21}\x{03B9}", -'1F9A' => "\x{1F22}\x{03B9}", -'1F9B' => "\x{1F23}\x{03B9}", -'1F9C' => "\x{1F24}\x{03B9}", -'1F9D' => "\x{1F25}\x{03B9}", -'1F9E' => "\x{1F26}\x{03B9}", -'1F9F' => "\x{1F27}\x{03B9}", -'1FA0' => "\x{1F60}\x{03B9}", -'1FA1' => "\x{1F61}\x{03B9}", -'1FA2' => "\x{1F62}\x{03B9}", -'1FA3' => "\x{1F63}\x{03B9}", -'1FA4' => "\x{1F64}\x{03B9}", -'1FA5' => "\x{1F65}\x{03B9}", -'1FA6' => "\x{1F66}\x{03B9}", -'1FA7' => "\x{1F67}\x{03B9}", -'1FA8' => "\x{1F60}\x{03B9}", -'1FA9' => "\x{1F61}\x{03B9}", -'1FAA' => "\x{1F62}\x{03B9}", -'1FAB' => "\x{1F63}\x{03B9}", -'1FAC' => "\x{1F64}\x{03B9}", -'1FAD' => "\x{1F65}\x{03B9}", -'1FAE' => "\x{1F66}\x{03B9}", -'1FAF' => "\x{1F67}\x{03B9}", -'1FB2' => "\x{1F70}\x{03B9}", -'1FB3' => "\x{03B1}\x{03B9}", -'1FB4' => "\x{03AC}\x{03B9}", -'1FB6' => "\x{03B1}\x{0342}", -'1FB7' => "\x{03B1}\x{0342}\x{03B9}", -'1FBC' => "\x{03B1}\x{03B9}", -'1FC2' => "\x{1F74}\x{03B9}", -'1FC3' => "\x{03B7}\x{03B9}", -'1FC4' => "\x{03AE}\x{03B9}", -'1FC6' => "\x{03B7}\x{0342}", -'1FC7' => "\x{03B7}\x{0342}\x{03B9}", -'1FCC' => "\x{03B7}\x{03B9}", -'1FD2' => "\x{03B9}\x{0308}\x{0300}", -'1FD3' => "\x{03B9}\x{0308}\x{0301}", -'1FD6' => "\x{03B9}\x{0342}", -'1FD7' => "\x{03B9}\x{0308}\x{0342}", -'1FE2' => "\x{03C5}\x{0308}\x{0300}", -'1FE3' => "\x{03C5}\x{0308}\x{0301}", -'1FE4' => "\x{03C1}\x{0313}", -'1FE6' => "\x{03C5}\x{0342}", -'1FE7' => "\x{03C5}\x{0308}\x{0342}", -'1FF2' => "\x{1F7C}\x{03B9}", -'1FF3' => "\x{03C9}\x{03B9}", -'1FF4' => "\x{03CE}\x{03B9}", -'1FF6' => "\x{03C9}\x{0342}", -'1FF7' => "\x{03C9}\x{0342}\x{03B9}", -'1FFC' => "\x{03C9}\x{03B9}", -'FB00' => "\x{0066}\x{0066}", -'FB01' => "\x{0066}\x{0069}", -'FB02' => "\x{0066}\x{006C}", -'FB03' => "\x{0066}\x{0066}\x{0069}", -'FB04' => "\x{0066}\x{0066}\x{006C}", -'FB05' => "\x{0073}\x{0074}", -'FB06' => "\x{0073}\x{0074}", -'FB13' => "\x{0574}\x{0576}", -'FB14' => "\x{0574}\x{0565}", -'FB15' => "\x{0574}\x{056B}", -'FB16' => "\x{057E}\x{0576}", -'FB17' => "\x{0574}\x{056D}", +"\xC3\x9F" => "\x{0073}\x{0073}", +"\xC4\xB0" => "\x{0069}\x{0307}", +"\xC5\x89" => "\x{02BC}\x{006E}", +"\xC7\xB0" => "\x{006A}\x{030C}", +"\xCE\x90" => "\x{03B9}\x{0308}\x{0301}", +"\xCE\xB0" => "\x{03C5}\x{0308}\x{0301}", +"\xD6\x87" => "\x{0565}\x{0582}", +"\xE1\xBA\x96" => "\x{0068}\x{0331}", +"\xE1\xBA\x97" => "\x{0074}\x{0308}", +"\xE1\xBA\x98" => "\x{0077}\x{030A}", +"\xE1\xBA\x99" => "\x{0079}\x{030A}", +"\xE1\xBA\x9A" => "\x{0061}\x{02BE}", +"\xE1\xBD\x90" => "\x{03C5}\x{0313}", +"\xE1\xBD\x92" => "\x{03C5}\x{0313}\x{0300}", +"\xE1\xBD\x94" => "\x{03C5}\x{0313}\x{0301}", +"\xE1\xBD\x96" => "\x{03C5}\x{0313}\x{0342}", +"\xE1\xBE\x80" => "\x{1F00}\x{03B9}", +"\xE1\xBE\x81" => "\x{1F01}\x{03B9}", +"\xE1\xBE\x82" => "\x{1F02}\x{03B9}", +"\xE1\xBE\x83" => "\x{1F03}\x{03B9}", +"\xE1\xBE\x84" => "\x{1F04}\x{03B9}", +"\xE1\xBE\x85" => "\x{1F05}\x{03B9}", +"\xE1\xBE\x86" => "\x{1F06}\x{03B9}", +"\xE1\xBE\x87" => "\x{1F07}\x{03B9}", +"\xE1\xBE\x88" => "\x{1F00}\x{03B9}", +"\xE1\xBE\x89" => "\x{1F01}\x{03B9}", +"\xE1\xBE\x8A" => "\x{1F02}\x{03B9}", +"\xE1\xBE\x8B" => "\x{1F03}\x{03B9}", +"\xE1\xBE\x8C" => "\x{1F04}\x{03B9}", +"\xE1\xBE\x8D" => "\x{1F05}\x{03B9}", +"\xE1\xBE\x8E" => "\x{1F06}\x{03B9}", +"\xE1\xBE\x8F" => "\x{1F07}\x{03B9}", +"\xE1\xBE\x90" => "\x{1F20}\x{03B9}", +"\xE1\xBE\x91" => "\x{1F21}\x{03B9}", +"\xE1\xBE\x92" => "\x{1F22}\x{03B9}", +"\xE1\xBE\x93" => "\x{1F23}\x{03B9}", +"\xE1\xBE\x94" => "\x{1F24}\x{03B9}", +"\xE1\xBE\x95" => "\x{1F25}\x{03B9}", +"\xE1\xBE\x96" => "\x{1F26}\x{03B9}", +"\xE1\xBE\x97" => "\x{1F27}\x{03B9}", +"\xE1\xBE\x98" => "\x{1F20}\x{03B9}", +"\xE1\xBE\x99" => "\x{1F21}\x{03B9}", +"\xE1\xBE\x9A" => "\x{1F22}\x{03B9}", +"\xE1\xBE\x9B" => "\x{1F23}\x{03B9}", +"\xE1\xBE\x9C" => "\x{1F24}\x{03B9}", +"\xE1\xBE\x9D" => "\x{1F25}\x{03B9}", +"\xE1\xBE\x9E" => "\x{1F26}\x{03B9}", +"\xE1\xBE\x9F" => "\x{1F27}\x{03B9}", +"\xE1\xBE\xA0" => "\x{1F60}\x{03B9}", +"\xE1\xBE\xA1" => "\x{1F61}\x{03B9}", +"\xE1\xBE\xA2" => "\x{1F62}\x{03B9}", +"\xE1\xBE\xA3" => "\x{1F63}\x{03B9}", +"\xE1\xBE\xA4" => "\x{1F64}\x{03B9}", +"\xE1\xBE\xA5" => "\x{1F65}\x{03B9}", +"\xE1\xBE\xA6" => "\x{1F66}\x{03B9}", +"\xE1\xBE\xA7" => "\x{1F67}\x{03B9}", +"\xE1\xBE\xA8" => "\x{1F60}\x{03B9}", +"\xE1\xBE\xA9" => "\x{1F61}\x{03B9}", +"\xE1\xBE\xAA" => "\x{1F62}\x{03B9}", +"\xE1\xBE\xAB" => "\x{1F63}\x{03B9}", +"\xE1\xBE\xAC" => "\x{1F64}\x{03B9}", +"\xE1\xBE\xAD" => "\x{1F65}\x{03B9}", +"\xE1\xBE\xAE" => "\x{1F66}\x{03B9}", +"\xE1\xBE\xAF" => "\x{1F67}\x{03B9}", +"\xE1\xBE\xB2" => "\x{1F70}\x{03B9}", +"\xE1\xBE\xB3" => "\x{03B1}\x{03B9}", +"\xE1\xBE\xB4" => "\x{03AC}\x{03B9}", +"\xE1\xBE\xB6" => "\x{03B1}\x{0342}", +"\xE1\xBE\xB7" => "\x{03B1}\x{0342}\x{03B9}", +"\xE1\xBE\xBC" => "\x{03B1}\x{03B9}", +"\xE1\xBF\x82" => "\x{1F74}\x{03B9}", +"\xE1\xBF\x83" => "\x{03B7}\x{03B9}", +"\xE1\xBF\x84" => "\x{03AE}\x{03B9}", +"\xE1\xBF\x86" => "\x{03B7}\x{0342}", +"\xE1\xBF\x87" => "\x{03B7}\x{0342}\x{03B9}", +"\xE1\xBF\x8C" => "\x{03B7}\x{03B9}", +"\xE1\xBF\x92" => "\x{03B9}\x{0308}\x{0300}", +"\xE1\xBF\x93" => "\x{03B9}\x{0308}\x{0301}", +"\xE1\xBF\x96" => "\x{03B9}\x{0342}", +"\xE1\xBF\x97" => "\x{03B9}\x{0308}\x{0342}", +"\xE1\xBF\xA2" => "\x{03C5}\x{0308}\x{0300}", +"\xE1\xBF\xA3" => "\x{03C5}\x{0308}\x{0301}", +"\xE1\xBF\xA4" => "\x{03C1}\x{0313}", +"\xE1\xBF\xA6" => "\x{03C5}\x{0342}", +"\xE1\xBF\xA7" => "\x{03C5}\x{0308}\x{0342}", +"\xE1\xBF\xB2" => "\x{1F7C}\x{03B9}", +"\xE1\xBF\xB3" => "\x{03C9}\x{03B9}", +"\xE1\xBF\xB4" => "\x{03CE}\x{03B9}", +"\xE1\xBF\xB6" => "\x{03C9}\x{0342}", +"\xE1\xBF\xB7" => "\x{03C9}\x{0342}\x{03B9}", +"\xE1\xBF\xBC" => "\x{03C9}\x{03B9}", +"\xEF\xAC\x80" => "\x{0066}\x{0066}", +"\xEF\xAC\x81" => "\x{0066}\x{0069}", +"\xEF\xAC\x82" => "\x{0066}\x{006C}", +"\xEF\xAC\x83" => "\x{0066}\x{0066}\x{0069}", +"\xEF\xAC\x84" => "\x{0066}\x{0066}\x{006C}", +"\xEF\xAC\x85" => "\x{0073}\x{0074}", +"\xEF\xAC\x86" => "\x{0073}\x{0074}", +"\xEF\xAC\x93" => "\x{0574}\x{0576}", +"\xEF\xAC\x94" => "\x{0574}\x{0565}", +"\xEF\xAC\x95" => "\x{0574}\x{056B}", +"\xEF\xAC\x96" => "\x{057E}\x{0576}", +"\xEF\xAC\x97" => "\x{0574}\x{056D}", ); return <<'END'; diff --git a/lib/unicore/To/Lower.pl b/lib/unicore/To/Lower.pl index 8bf064a..7e5adfe 100644 --- a/lib/unicore/To/Lower.pl +++ b/lib/unicore/To/Lower.pl @@ -3,36 +3,37 @@ # Any changes made here will be lost! +# The key UTF-8 _bytes_, the value UTF-8 (speed hack) %utf8::ToSpecLower = ( -'0130' => "\x{0069}\x{0307}", -'1F88' => "\x{1F80}", -'1F89' => "\x{1F81}", -'1F8A' => "\x{1F82}", -'1F8B' => "\x{1F83}", -'1F8C' => "\x{1F84}", -'1F8D' => "\x{1F85}", -'1F8E' => "\x{1F86}", -'1F8F' => "\x{1F87}", -'1F98' => "\x{1F90}", -'1F99' => "\x{1F91}", -'1F9A' => "\x{1F92}", -'1F9B' => "\x{1F93}", -'1F9C' => "\x{1F94}", -'1F9D' => "\x{1F95}", -'1F9E' => "\x{1F96}", -'1F9F' => "\x{1F97}", -'1FA8' => "\x{1FA0}", -'1FA9' => "\x{1FA1}", -'1FAA' => "\x{1FA2}", -'1FAB' => "\x{1FA3}", -'1FAC' => "\x{1FA4}", -'1FAD' => "\x{1FA5}", -'1FAE' => "\x{1FA6}", -'1FAF' => "\x{1FA7}", -'1FBC' => "\x{1FB3}", -'1FCC' => "\x{1FC3}", -'1FFC' => "\x{1FF3}", +"\xC4\xB0" => "\x{0069}\x{0307}", +"\xE1\xBE\x88" => "\x{1F80}", +"\xE1\xBE\x89" => "\x{1F81}", +"\xE1\xBE\x8A" => "\x{1F82}", +"\xE1\xBE\x8B" => "\x{1F83}", +"\xE1\xBE\x8C" => "\x{1F84}", +"\xE1\xBE\x8D" => "\x{1F85}", +"\xE1\xBE\x8E" => "\x{1F86}", +"\xE1\xBE\x8F" => "\x{1F87}", +"\xE1\xBE\x98" => "\x{1F90}", +"\xE1\xBE\x99" => "\x{1F91}", +"\xE1\xBE\x9A" => "\x{1F92}", +"\xE1\xBE\x9B" => "\x{1F93}", +"\xE1\xBE\x9C" => "\x{1F94}", +"\xE1\xBE\x9D" => "\x{1F95}", +"\xE1\xBE\x9E" => "\x{1F96}", +"\xE1\xBE\x9F" => "\x{1F97}", +"\xE1\xBE\xA8" => "\x{1FA0}", +"\xE1\xBE\xA9" => "\x{1FA1}", +"\xE1\xBE\xAA" => "\x{1FA2}", +"\xE1\xBE\xAB" => "\x{1FA3}", +"\xE1\xBE\xAC" => "\x{1FA4}", +"\xE1\xBE\xAD" => "\x{1FA5}", +"\xE1\xBE\xAE" => "\x{1FA6}", +"\xE1\xBE\xAF" => "\x{1FA7}", +"\xE1\xBE\xBC" => "\x{1FB3}", +"\xE1\xBF\x8C" => "\x{1FC3}", +"\xE1\xBF\xBC" => "\x{1FF3}", ); return <<'END'; diff --git a/lib/unicore/To/Title.pl b/lib/unicore/To/Title.pl index cfeccab..2223f7b 100644 --- a/lib/unicore/To/Title.pl +++ b/lib/unicore/To/Title.pl @@ -3,83 +3,84 @@ # Any changes made here will be lost! +# The key UTF-8 _bytes_, the value UTF-8 (speed hack) %utf8::ToSpecTitle = ( -'00DF' => "\x{0053}\x{0073}", -'0149' => "\x{02BC}\x{004E}", -'01F0' => "\x{004A}\x{030C}", -'0390' => "\x{0399}\x{0308}\x{0301}", -'03B0' => "\x{03A5}\x{0308}\x{0301}", -'0587' => "\x{0535}\x{0582}", -'1E96' => "\x{0048}\x{0331}", -'1E97' => "\x{0054}\x{0308}", -'1E98' => "\x{0057}\x{030A}", -'1E99' => "\x{0059}\x{030A}", -'1E9A' => "\x{0041}\x{02BE}", -'1F50' => "\x{03A5}\x{0313}", -'1F52' => "\x{03A5}\x{0313}\x{0300}", -'1F54' => "\x{03A5}\x{0313}\x{0301}", -'1F56' => "\x{03A5}\x{0313}\x{0342}", -'1F80' => "\x{1F88}", -'1F81' => "\x{1F89}", -'1F82' => "\x{1F8A}", -'1F83' => "\x{1F8B}", -'1F84' => "\x{1F8C}", -'1F85' => "\x{1F8D}", -'1F86' => "\x{1F8E}", -'1F87' => "\x{1F8F}", -'1F90' => "\x{1F98}", -'1F91' => "\x{1F99}", -'1F92' => "\x{1F9A}", -'1F93' => "\x{1F9B}", -'1F94' => "\x{1F9C}", -'1F95' => "\x{1F9D}", -'1F96' => "\x{1F9E}", -'1F97' => "\x{1F9F}", -'1FA0' => "\x{1FA8}", -'1FA1' => "\x{1FA9}", -'1FA2' => "\x{1FAA}", -'1FA3' => "\x{1FAB}", -'1FA4' => "\x{1FAC}", -'1FA5' => "\x{1FAD}", -'1FA6' => "\x{1FAE}", -'1FA7' => "\x{1FAF}", -'1FB2' => "\x{1FBA}\x{0345}", -'1FB3' => "\x{1FBC}", -'1FB4' => "\x{0386}\x{0345}", -'1FB6' => "\x{0391}\x{0342}", -'1FB7' => "\x{0391}\x{0342}\x{0345}", -'1FC2' => "\x{1FCA}\x{0345}", -'1FC3' => "\x{1FCC}", -'1FC4' => "\x{0389}\x{0345}", -'1FC6' => "\x{0397}\x{0342}", -'1FC7' => "\x{0397}\x{0342}\x{0345}", -'1FD2' => "\x{0399}\x{0308}\x{0300}", -'1FD3' => "\x{0399}\x{0308}\x{0301}", -'1FD6' => "\x{0399}\x{0342}", -'1FD7' => "\x{0399}\x{0308}\x{0342}", -'1FE2' => "\x{03A5}\x{0308}\x{0300}", -'1FE3' => "\x{03A5}\x{0308}\x{0301}", -'1FE4' => "\x{03A1}\x{0313}", -'1FE6' => "\x{03A5}\x{0342}", -'1FE7' => "\x{03A5}\x{0308}\x{0342}", -'1FF2' => "\x{1FFA}\x{0345}", -'1FF3' => "\x{1FFC}", -'1FF4' => "\x{038F}\x{0345}", -'1FF6' => "\x{03A9}\x{0342}", -'1FF7' => "\x{03A9}\x{0342}\x{0345}", -'FB00' => "\x{0046}\x{0066}", -'FB01' => "\x{0046}\x{0069}", -'FB02' => "\x{0046}\x{006C}", -'FB03' => "\x{0046}\x{0066}\x{0069}", -'FB04' => "\x{0046}\x{0066}\x{006C}", -'FB05' => "\x{0053}\x{0074}", -'FB06' => "\x{0053}\x{0074}", -'FB13' => "\x{0544}\x{0576}", -'FB14' => "\x{0544}\x{0565}", -'FB15' => "\x{0544}\x{056B}", -'FB16' => "\x{054E}\x{0576}", -'FB17' => "\x{0544}\x{056D}", +"\xC3\x9F" => "\x{0053}\x{0073}", +"\xC5\x89" => "\x{02BC}\x{004E}", +"\xC7\xB0" => "\x{004A}\x{030C}", +"\xCE\x90" => "\x{0399}\x{0308}\x{0301}", +"\xCE\xB0" => "\x{03A5}\x{0308}\x{0301}", +"\xD6\x87" => "\x{0535}\x{0582}", +"\xE1\xBA\x96" => "\x{0048}\x{0331}", +"\xE1\xBA\x97" => "\x{0054}\x{0308}", +"\xE1\xBA\x98" => "\x{0057}\x{030A}", +"\xE1\xBA\x99" => "\x{0059}\x{030A}", +"\xE1\xBA\x9A" => "\x{0041}\x{02BE}", +"\xE1\xBD\x90" => "\x{03A5}\x{0313}", +"\xE1\xBD\x92" => "\x{03A5}\x{0313}\x{0300}", +"\xE1\xBD\x94" => "\x{03A5}\x{0313}\x{0301}", +"\xE1\xBD\x96" => "\x{03A5}\x{0313}\x{0342}", +"\xE1\xBE\x80" => "\x{1F88}", +"\xE1\xBE\x81" => "\x{1F89}", +"\xE1\xBE\x82" => "\x{1F8A}", +"\xE1\xBE\x83" => "\x{1F8B}", +"\xE1\xBE\x84" => "\x{1F8C}", +"\xE1\xBE\x85" => "\x{1F8D}", +"\xE1\xBE\x86" => "\x{1F8E}", +"\xE1\xBE\x87" => "\x{1F8F}", +"\xE1\xBE\x90" => "\x{1F98}", +"\xE1\xBE\x91" => "\x{1F99}", +"\xE1\xBE\x92" => "\x{1F9A}", +"\xE1\xBE\x93" => "\x{1F9B}", +"\xE1\xBE\x94" => "\x{1F9C}", +"\xE1\xBE\x95" => "\x{1F9D}", +"\xE1\xBE\x96" => "\x{1F9E}", +"\xE1\xBE\x97" => "\x{1F9F}", +"\xE1\xBE\xA0" => "\x{1FA8}", +"\xE1\xBE\xA1" => "\x{1FA9}", +"\xE1\xBE\xA2" => "\x{1FAA}", +"\xE1\xBE\xA3" => "\x{1FAB}", +"\xE1\xBE\xA4" => "\x{1FAC}", +"\xE1\xBE\xA5" => "\x{1FAD}", +"\xE1\xBE\xA6" => "\x{1FAE}", +"\xE1\xBE\xA7" => "\x{1FAF}", +"\xE1\xBE\xB2" => "\x{1FBA}\x{0345}", +"\xE1\xBE\xB3" => "\x{1FBC}", +"\xE1\xBE\xB4" => "\x{0386}\x{0345}", +"\xE1\xBE\xB6" => "\x{0391}\x{0342}", +"\xE1\xBE\xB7" => "\x{0391}\x{0342}\x{0345}", +"\xE1\xBF\x82" => "\x{1FCA}\x{0345}", +"\xE1\xBF\x83" => "\x{1FCC}", +"\xE1\xBF\x84" => "\x{0389}\x{0345}", +"\xE1\xBF\x86" => "\x{0397}\x{0342}", +"\xE1\xBF\x87" => "\x{0397}\x{0342}\x{0345}", +"\xE1\xBF\x92" => "\x{0399}\x{0308}\x{0300}", +"\xE1\xBF\x93" => "\x{0399}\x{0308}\x{0301}", +"\xE1\xBF\x96" => "\x{0399}\x{0342}", +"\xE1\xBF\x97" => "\x{0399}\x{0308}\x{0342}", +"\xE1\xBF\xA2" => "\x{03A5}\x{0308}\x{0300}", +"\xE1\xBF\xA3" => "\x{03A5}\x{0308}\x{0301}", +"\xE1\xBF\xA4" => "\x{03A1}\x{0313}", +"\xE1\xBF\xA6" => "\x{03A5}\x{0342}", +"\xE1\xBF\xA7" => "\x{03A5}\x{0308}\x{0342}", +"\xE1\xBF\xB2" => "\x{1FFA}\x{0345}", +"\xE1\xBF\xB3" => "\x{1FFC}", +"\xE1\xBF\xB4" => "\x{038F}\x{0345}", +"\xE1\xBF\xB6" => "\x{03A9}\x{0342}", +"\xE1\xBF\xB7" => "\x{03A9}\x{0342}\x{0345}", +"\xEF\xAC\x80" => "\x{0046}\x{0066}", +"\xEF\xAC\x81" => "\x{0046}\x{0069}", +"\xEF\xAC\x82" => "\x{0046}\x{006C}", +"\xEF\xAC\x83" => "\x{0046}\x{0066}\x{0069}", +"\xEF\xAC\x84" => "\x{0046}\x{0066}\x{006C}", +"\xEF\xAC\x85" => "\x{0053}\x{0074}", +"\xEF\xAC\x86" => "\x{0053}\x{0074}", +"\xEF\xAC\x93" => "\x{0544}\x{0576}", +"\xEF\xAC\x94" => "\x{0544}\x{0565}", +"\xEF\xAC\x95" => "\x{0544}\x{056B}", +"\xEF\xAC\x96" => "\x{054E}\x{0576}", +"\xEF\xAC\x97" => "\x{0544}\x{056D}", ); return <<'END'; diff --git a/lib/unicore/To/Upper.pl b/lib/unicore/To/Upper.pl index 1a42789..6c8fd30 100644 --- a/lib/unicore/To/Upper.pl +++ b/lib/unicore/To/Upper.pl @@ -3,110 +3,111 @@ # Any changes made here will be lost! +# The key UTF-8 _bytes_, the value UTF-8 (speed hack) %utf8::ToSpecUpper = ( -'00DF' => "\x{0053}\x{0053}", -'0149' => "\x{02BC}\x{004E}", -'01F0' => "\x{004A}\x{030C}", -'0390' => "\x{0399}\x{0308}\x{0301}", -'03B0' => "\x{03A5}\x{0308}\x{0301}", -'0587' => "\x{0535}\x{0552}", -'1E96' => "\x{0048}\x{0331}", -'1E97' => "\x{0054}\x{0308}", -'1E98' => "\x{0057}\x{030A}", -'1E99' => "\x{0059}\x{030A}", -'1E9A' => "\x{0041}\x{02BE}", -'1F50' => "\x{03A5}\x{0313}", -'1F52' => "\x{03A5}\x{0313}\x{0300}", -'1F54' => "\x{03A5}\x{0313}\x{0301}", -'1F56' => "\x{03A5}\x{0313}\x{0342}", -'1F80' => "\x{1F08}\x{0399}", -'1F81' => "\x{1F09}\x{0399}", -'1F82' => "\x{1F0A}\x{0399}", -'1F83' => "\x{1F0B}\x{0399}", -'1F84' => "\x{1F0C}\x{0399}", -'1F85' => "\x{1F0D}\x{0399}", -'1F86' => "\x{1F0E}\x{0399}", -'1F87' => "\x{1F0F}\x{0399}", -'1F88' => "\x{1F08}\x{0399}", -'1F89' => "\x{1F09}\x{0399}", -'1F8A' => "\x{1F0A}\x{0399}", -'1F8B' => "\x{1F0B}\x{0399}", -'1F8C' => "\x{1F0C}\x{0399}", -'1F8D' => "\x{1F0D}\x{0399}", -'1F8E' => "\x{1F0E}\x{0399}", -'1F8F' => "\x{1F0F}\x{0399}", -'1F90' => "\x{1F28}\x{0399}", -'1F91' => "\x{1F29}\x{0399}", -'1F92' => "\x{1F2A}\x{0399}", -'1F93' => "\x{1F2B}\x{0399}", -'1F94' => "\x{1F2C}\x{0399}", -'1F95' => "\x{1F2D}\x{0399}", -'1F96' => "\x{1F2E}\x{0399}", -'1F97' => "\x{1F2F}\x{0399}", -'1F98' => "\x{1F28}\x{0399}", -'1F99' => "\x{1F29}\x{0399}", -'1F9A' => "\x{1F2A}\x{0399}", -'1F9B' => "\x{1F2B}\x{0399}", -'1F9C' => "\x{1F2C}\x{0399}", -'1F9D' => "\x{1F2D}\x{0399}", -'1F9E' => "\x{1F2E}\x{0399}", -'1F9F' => "\x{1F2F}\x{0399}", -'1FA0' => "\x{1F68}\x{0399}", -'1FA1' => "\x{1F69}\x{0399}", -'1FA2' => "\x{1F6A}\x{0399}", -'1FA3' => "\x{1F6B}\x{0399}", -'1FA4' => "\x{1F6C}\x{0399}", -'1FA5' => "\x{1F6D}\x{0399}", -'1FA6' => "\x{1F6E}\x{0399}", -'1FA7' => "\x{1F6F}\x{0399}", -'1FA8' => "\x{1F68}\x{0399}", -'1FA9' => "\x{1F69}\x{0399}", -'1FAA' => "\x{1F6A}\x{0399}", -'1FAB' => "\x{1F6B}\x{0399}", -'1FAC' => "\x{1F6C}\x{0399}", -'1FAD' => "\x{1F6D}\x{0399}", -'1FAE' => "\x{1F6E}\x{0399}", -'1FAF' => "\x{1F6F}\x{0399}", -'1FB2' => "\x{1FBA}\x{0399}", -'1FB3' => "\x{0391}\x{0399}", -'1FB4' => "\x{0386}\x{0399}", -'1FB6' => "\x{0391}\x{0342}", -'1FB7' => "\x{0391}\x{0342}\x{0399}", -'1FBC' => "\x{0391}\x{0399}", -'1FC2' => "\x{1FCA}\x{0399}", -'1FC3' => "\x{0397}\x{0399}", -'1FC4' => "\x{0389}\x{0399}", -'1FC6' => "\x{0397}\x{0342}", -'1FC7' => "\x{0397}\x{0342}\x{0399}", -'1FCC' => "\x{0397}\x{0399}", -'1FD2' => "\x{0399}\x{0308}\x{0300}", -'1FD3' => "\x{0399}\x{0308}\x{0301}", -'1FD6' => "\x{0399}\x{0342}", -'1FD7' => "\x{0399}\x{0308}\x{0342}", -'1FE2' => "\x{03A5}\x{0308}\x{0300}", -'1FE3' => "\x{03A5}\x{0308}\x{0301}", -'1FE4' => "\x{03A1}\x{0313}", -'1FE6' => "\x{03A5}\x{0342}", -'1FE7' => "\x{03A5}\x{0308}\x{0342}", -'1FF2' => "\x{1FFA}\x{0399}", -'1FF3' => "\x{03A9}\x{0399}", -'1FF4' => "\x{038F}\x{0399}", -'1FF6' => "\x{03A9}\x{0342}", -'1FF7' => "\x{03A9}\x{0342}\x{0399}", -'1FFC' => "\x{03A9}\x{0399}", -'FB00' => "\x{0046}\x{0046}", -'FB01' => "\x{0046}\x{0049}", -'FB02' => "\x{0046}\x{004C}", -'FB03' => "\x{0046}\x{0046}\x{0049}", -'FB04' => "\x{0046}\x{0046}\x{004C}", -'FB05' => "\x{0053}\x{0054}", -'FB06' => "\x{0053}\x{0054}", -'FB13' => "\x{0544}\x{0546}", -'FB14' => "\x{0544}\x{0535}", -'FB15' => "\x{0544}\x{053B}", -'FB16' => "\x{054E}\x{0546}", -'FB17' => "\x{0544}\x{053D}", +"\xC3\x9F" => "\x{0053}\x{0053}", +"\xC5\x89" => "\x{02BC}\x{004E}", +"\xC7\xB0" => "\x{004A}\x{030C}", +"\xCE\x90" => "\x{0399}\x{0308}\x{0301}", +"\xCE\xB0" => "\x{03A5}\x{0308}\x{0301}", +"\xD6\x87" => "\x{0535}\x{0552}", +"\xE1\xBA\x96" => "\x{0048}\x{0331}", +"\xE1\xBA\x97" => "\x{0054}\x{0308}", +"\xE1\xBA\x98" => "\x{0057}\x{030A}", +"\xE1\xBA\x99" => "\x{0059}\x{030A}", +"\xE1\xBA\x9A" => "\x{0041}\x{02BE}", +"\xE1\xBD\x90" => "\x{03A5}\x{0313}", +"\xE1\xBD\x92" => "\x{03A5}\x{0313}\x{0300}", +"\xE1\xBD\x94" => "\x{03A5}\x{0313}\x{0301}", +"\xE1\xBD\x96" => "\x{03A5}\x{0313}\x{0342}", +"\xE1\xBE\x80" => "\x{1F08}\x{0399}", +"\xE1\xBE\x81" => "\x{1F09}\x{0399}", +"\xE1\xBE\x82" => "\x{1F0A}\x{0399}", +"\xE1\xBE\x83" => "\x{1F0B}\x{0399}", +"\xE1\xBE\x84" => "\x{1F0C}\x{0399}", +"\xE1\xBE\x85" => "\x{1F0D}\x{0399}", +"\xE1\xBE\x86" => "\x{1F0E}\x{0399}", +"\xE1\xBE\x87" => "\x{1F0F}\x{0399}", +"\xE1\xBE\x88" => "\x{1F08}\x{0399}", +"\xE1\xBE\x89" => "\x{1F09}\x{0399}", +"\xE1\xBE\x8A" => "\x{1F0A}\x{0399}", +"\xE1\xBE\x8B" => "\x{1F0B}\x{0399}", +"\xE1\xBE\x8C" => "\x{1F0C}\x{0399}", +"\xE1\xBE\x8D" => "\x{1F0D}\x{0399}", +"\xE1\xBE\x8E" => "\x{1F0E}\x{0399}", +"\xE1\xBE\x8F" => "\x{1F0F}\x{0399}", +"\xE1\xBE\x90" => "\x{1F28}\x{0399}", +"\xE1\xBE\x91" => "\x{1F29}\x{0399}", +"\xE1\xBE\x92" => "\x{1F2A}\x{0399}", +"\xE1\xBE\x93" => "\x{1F2B}\x{0399}", +"\xE1\xBE\x94" => "\x{1F2C}\x{0399}", +"\xE1\xBE\x95" => "\x{1F2D}\x{0399}", +"\xE1\xBE\x96" => "\x{1F2E}\x{0399}", +"\xE1\xBE\x97" => "\x{1F2F}\x{0399}", +"\xE1\xBE\x98" => "\x{1F28}\x{0399}", +"\xE1\xBE\x99" => "\x{1F29}\x{0399}", +"\xE1\xBE\x9A" => "\x{1F2A}\x{0399}", +"\xE1\xBE\x9B" => "\x{1F2B}\x{0399}", +"\xE1\xBE\x9C" => "\x{1F2C}\x{0399}", +"\xE1\xBE\x9D" => "\x{1F2D}\x{0399}", +"\xE1\xBE\x9E" => "\x{1F2E}\x{0399}", +"\xE1\xBE\x9F" => "\x{1F2F}\x{0399}", +"\xE1\xBE\xA0" => "\x{1F68}\x{0399}", +"\xE1\xBE\xA1" => "\x{1F69}\x{0399}", +"\xE1\xBE\xA2" => "\x{1F6A}\x{0399}", +"\xE1\xBE\xA3" => "\x{1F6B}\x{0399}", +"\xE1\xBE\xA4" => "\x{1F6C}\x{0399}", +"\xE1\xBE\xA5" => "\x{1F6D}\x{0399}", +"\xE1\xBE\xA6" => "\x{1F6E}\x{0399}", +"\xE1\xBE\xA7" => "\x{1F6F}\x{0399}", +"\xE1\xBE\xA8" => "\x{1F68}\x{0399}", +"\xE1\xBE\xA9" => "\x{1F69}\x{0399}", +"\xE1\xBE\xAA" => "\x{1F6A}\x{0399}", +"\xE1\xBE\xAB" => "\x{1F6B}\x{0399}", +"\xE1\xBE\xAC" => "\x{1F6C}\x{0399}", +"\xE1\xBE\xAD" => "\x{1F6D}\x{0399}", +"\xE1\xBE\xAE" => "\x{1F6E}\x{0399}", +"\xE1\xBE\xAF" => "\x{1F6F}\x{0399}", +"\xE1\xBE\xB2" => "\x{1FBA}\x{0399}", +"\xE1\xBE\xB3" => "\x{0391}\x{0399}", +"\xE1\xBE\xB4" => "\x{0386}\x{0399}", +"\xE1\xBE\xB6" => "\x{0391}\x{0342}", +"\xE1\xBE\xB7" => "\x{0391}\x{0342}\x{0399}", +"\xE1\xBE\xBC" => "\x{0391}\x{0399}", +"\xE1\xBF\x82" => "\x{1FCA}\x{0399}", +"\xE1\xBF\x83" => "\x{0397}\x{0399}", +"\xE1\xBF\x84" => "\x{0389}\x{0399}", +"\xE1\xBF\x86" => "\x{0397}\x{0342}", +"\xE1\xBF\x87" => "\x{0397}\x{0342}\x{0399}", +"\xE1\xBF\x8C" => "\x{0397}\x{0399}", +"\xE1\xBF\x92" => "\x{0399}\x{0308}\x{0300}", +"\xE1\xBF\x93" => "\x{0399}\x{0308}\x{0301}", +"\xE1\xBF\x96" => "\x{0399}\x{0342}", +"\xE1\xBF\x97" => "\x{0399}\x{0308}\x{0342}", +"\xE1\xBF\xA2" => "\x{03A5}\x{0308}\x{0300}", +"\xE1\xBF\xA3" => "\x{03A5}\x{0308}\x{0301}", +"\xE1\xBF\xA4" => "\x{03A1}\x{0313}", +"\xE1\xBF\xA6" => "\x{03A5}\x{0342}", +"\xE1\xBF\xA7" => "\x{03A5}\x{0308}\x{0342}", +"\xE1\xBF\xB2" => "\x{1FFA}\x{0399}", +"\xE1\xBF\xB3" => "\x{03A9}\x{0399}", +"\xE1\xBF\xB4" => "\x{038F}\x{0399}", +"\xE1\xBF\xB6" => "\x{03A9}\x{0342}", +"\xE1\xBF\xB7" => "\x{03A9}\x{0342}\x{0399}", +"\xE1\xBF\xBC" => "\x{03A9}\x{0399}", +"\xEF\xAC\x80" => "\x{0046}\x{0046}", +"\xEF\xAC\x81" => "\x{0046}\x{0049}", +"\xEF\xAC\x82" => "\x{0046}\x{004C}", +"\xEF\xAC\x83" => "\x{0046}\x{0046}\x{0049}", +"\xEF\xAC\x84" => "\x{0046}\x{0046}\x{004C}", +"\xEF\xAC\x85" => "\x{0053}\x{0054}", +"\xEF\xAC\x86" => "\x{0053}\x{0054}", +"\xEF\xAC\x93" => "\x{0544}\x{0546}", +"\xEF\xAC\x94" => "\x{0544}\x{0535}", +"\xEF\xAC\x95" => "\x{0544}\x{053B}", +"\xEF\xAC\x96" => "\x{054E}\x{0546}", +"\xEF\xAC\x97" => "\x{0544}\x{053D}", ); return <<'END'; diff --git a/lib/unicore/mktables b/lib/unicore/mktables index 5fdac52..09b8175 100644 --- a/lib/unicore/mktables +++ b/lib/unicore/mktables @@ -1658,16 +1658,18 @@ sub SpecialCasing_txt() { my $NormalCase = do "To/$case.pl" || die "$0: $@\n"; - my @OUT = ( - $HEADER, "\n", - "%utf8::ToSpec$case =\n(\n", - ); + my @OUT = + ( + $HEADER, "\n", + "# The key UTF-8 _bytes_, the value UTF-8 (speed hack)\n", + "%utf8::ToSpec$case =\n(\n", + ); for my $prop (sort { $a->[0] <=> $b->[0] } @{$CaseInfo{$case}}) { my ($ix, $code, $to) = @$prop; my $tostr = join "", map { sprintf "\\x{%s}", $_ } split ' ', $to; - push @OUT, sprintf qq['%04X' => "$tostr",\n], $ix; + push @OUT, sprintf qq["%s" => "$tostr",\n], join("", map { sprintf "\\x%02X", $_ } unpack("U0C*", pack("U", $ix))); # Remove any single-character mappings for # the same character since we are going for # the special casing rules. @@ -1719,14 +1721,16 @@ sub CaseFolding_txt() # my $CommonFold = do "To/Fold.pl" || die "$0: To/Fold.pl: $!\n"; - my @OUT = ( - $HEADER, "\n", - "%utf8::ToSpecFold =\n(\n", - ); + my @OUT = + ( + $HEADER, "\n", + "# The ke UTF-8 _bytes_, the value UTF-8 (speed hack)\n", + "%utf8::ToSpecFold =\n(\n", + ); for my $code (sort { $a <=> $b } keys %Fold) { my $foldstr = join "", map { sprintf "\\x{%s}", $_ } split ' ', $Fold{$code}; - push @OUT, sprintf qq['%04X' => "$foldstr",\n], $code; + push @OUT, sprintf qq["%s" => "$foldstr",\n], join("", map { sprintf "\\x%02X", $_ } unpack("U0C*", pack("U", $code))); } push @OUT, ( ");\n\n", diff --git a/t/uni/case.pl b/t/uni/case.pl index b6df5a8..0402be4 100644 --- a/t/uni/case.pl +++ b/t/uni/case.pl @@ -20,15 +20,15 @@ sub casetest { my %seen; for my $i (sort keys %simple) { - $seen{hex $i}++; + $seen{$i}++; } print "# ", scalar keys %simple, " simple mappings\n"; my $both; for my $i (sort keys %$spec) { - if (++$seen{hex $i} == 2) { - warn "$base: $i seen twice\n"; + if (++$seen{$i} == 2) { + warn sprintf "$base: $i seen twice\n"; $both++; } } @@ -52,7 +52,7 @@ sub casetest { my $test = 1; - for my $i (sort { hex $a <=> hex $b } keys %simple) { + for my $i (sort keys %simple) { my $w = $simple{$i}; my $c = pack "U0U", hex $i; my $d = $func->($c); @@ -62,9 +62,11 @@ sub casetest { $test++; } - for my $i (sort { hex $a <=> hex $b } keys %$spec) { + for my $i (sort keys %$spec) { my $w = unidump($spec->{$i}); - my $c = pack "U0U", hex $i; + my $u = unpack "U0U", $i; + my $h = sprintf "%04X", $u; + my $c = chr($u); $c .= chr(0x100); chop $c; my $d = $func->($c); my $e = unidump($d); if (ord "A" == 193) { # EBCDIC @@ -116,7 +118,7 @@ sub casetest { # just undo our remapping. } print $w eq $e ? - "ok $test # $i -> $w\n" : "not ok $test # $i -> $e ($w)\n"; + "ok $test # $i -> $w\n" : "not ok $test # $h -> $e ($w)\n"; $test++; } diff --git a/utf8.c b/utf8.c index 6a665eb..6b2527c 100644 --- a/utf8.c +++ b/utf8.c @@ -1401,21 +1401,19 @@ Perl_to_utf8_case(pTHX_ U8 *p, U8* ustrp, STRLEN *lenp, SV **swashp, char *norma if (!*swashp) /* load on-demand */ *swashp = swash_init("utf8", normal, &PL_sv_undef, 4, 0); - if (special) { + /* The 0xDF is the only special casing Unicode code point below 0x100. */ + if (special && (uv1 == 0xDF || uv1 > 0xFF)) { /* It might be "special" (sometimes, but not always, * a multicharacter mapping) */ HV *hv; - SV *keysv; - HE *he; - SV *val; - - if ((hv = get_hv(special, FALSE)) && - (keysv = sv_2mortal(Perl_newSVpvf(aTHX_ "%04"UVXf, uv1))) && - (he = hv_fetch_ent(hv, keysv, FALSE, 0)) && - (val = HeVAL(he))) { - char *s; + SV **svp; + + if ((hv = get_hv(special, FALSE)) && + (svp = hv_fetch(hv, (const char*)tmpbuf, UNISKIP(uv1), FALSE)) && + (*svp)) { + char *s; - s = SvPV(val, len); + s = SvPV(*svp, len); if (len == 1) len = uvuni_to_utf8(ustrp, NATIVE_TO_UNI(*(U8*)s)) - ustrp; else { @@ -1426,7 +1424,7 @@ Perl_to_utf8_case(pTHX_ U8 *p, U8* ustrp, STRLEN *lenp, SV **swashp, char *norma U8 *t = (U8*)s, *tend = t + len, *d; d = tmpbuf; - if (SvUTF8(val)) { + if (SvUTF8(*svp)) { STRLEN tlen = 0; while (t < tend) {