+<?php return array('A'=>'a','B'=>'b','C'=>'c','D'=>'d','E'=>'e','F'=>'f','G'=>'g','H'=>'h','I'=>'i','J'=>'j','K'=>'k','L'=>'l','M'=>'m','N'=>'n','O'=>'o','P'=>'p','Q'=>'q','R'=>'r','S'=>'s','T'=>'t','U'=>'u','V'=>'v','W'=>'w','X'=>'x','Y'=>'y','Z'=>'z','µ'=>'μ','À'=>'à','Á'=>'á','Â'=>'â','Ã'=>'ã','Ä'=>'ä','Å'=>'å','Æ'=>'æ','Ç'=>'ç','È'=>'è','É'=>'é','Ê'=>'ê','Ë'=>'ë','Ì'=>'ì','Í'=>'í','Î'=>'î','Ï'=>'ï','Ð'=>'ð','Ñ'=>'ñ','Ò'=>'ò','Ó'=>'ó','Ô'=>'ô','Õ'=>'õ','Ö'=>'ö','Ø'=>'ø','Ù'=>'ù','Ú'=>'ú','Û'=>'û','Ü'=>'ü','Ý'=>'ý','Þ'=>'þ','Ā'=>'ā','Ă'=>'ă','Ą'=>'ą','Ć'=>'ć','Ĉ'=>'ĉ','Ċ'=>'ċ','Č'=>'č','Ď'=>'ď','Đ'=>'đ','Ē'=>'ē','Ĕ'=>'ĕ','Ė'=>'ė','Ę'=>'ę','Ě'=>'ě','Ĝ'=>'ĝ','Ğ'=>'ğ','Ġ'=>'ġ','Ģ'=>'ģ','Ĥ'=>'ĥ','Ħ'=>'ħ','Ĩ'=>'ĩ','Ī'=>'ī','Ĭ'=>'ĭ','Į'=>'į','IJ'=>'ij','Ĵ'=>'ĵ','Ķ'=>'ķ','Ĺ'=>'ĺ','Ļ'=>'ļ','Ľ'=>'ľ','Ŀ'=>'ŀ','Ł'=>'ł','Ń'=>'ń','Ņ'=>'ņ','Ň'=>'ň','Ŋ'=>'ŋ','Ō'=>'ō','Ŏ'=>'ŏ','Ő'=>'ő','Œ'=>'œ','Ŕ'=>'ŕ','Ŗ'=>'ŗ','Ř'=>'ř','Ś'=>'ś','Ŝ'=>'ŝ','Ş'=>'ş','Š'=>'š','Ţ'=>'ţ','Ť'=>'ť','Ŧ'=>'ŧ','Ũ'=>'ũ','Ū'=>'ū','Ŭ'=>'ŭ','Ů'=>'ů','Ű'=>'ű','Ų'=>'ų','Ŵ'=>'ŵ','Ŷ'=>'ŷ','Ÿ'=>'ÿ','Ź'=>'ź','Ż'=>'ż','Ž'=>'ž','ſ'=>'s','Ɓ'=>'ɓ','Ƃ'=>'ƃ','Ƅ'=>'ƅ','Ɔ'=>'ɔ','Ƈ'=>'ƈ','Ɖ'=>'ɖ','Ɗ'=>'ɗ','Ƌ'=>'ƌ','Ǝ'=>'ǝ','Ə'=>'ə','Ɛ'=>'ɛ','Ƒ'=>'ƒ','Ɠ'=>'ɠ','Ɣ'=>'ɣ','Ɩ'=>'ɩ','Ɨ'=>'ɨ','Ƙ'=>'ƙ','Ɯ'=>'ɯ','Ɲ'=>'ɲ','Ɵ'=>'ɵ','Ơ'=>'ơ','Ƣ'=>'ƣ','Ƥ'=>'ƥ','Ʀ'=>'ʀ','Ƨ'=>'ƨ','Ʃ'=>'ʃ','Ƭ'=>'ƭ','Ʈ'=>'ʈ','Ư'=>'ư','Ʊ'=>'ʊ','Ʋ'=>'ʋ','Ƴ'=>'ƴ','Ƶ'=>'ƶ','Ʒ'=>'ʒ','Ƹ'=>'ƹ','Ƽ'=>'ƽ','DŽ'=>'dž','Dž'=>'dž','LJ'=>'lj','Lj'=>'lj','NJ'=>'nj','Nj'=>'nj','Ǎ'=>'ǎ','Ǐ'=>'ǐ','Ǒ'=>'ǒ','Ǔ'=>'ǔ','Ǖ'=>'ǖ','Ǘ'=>'ǘ','Ǚ'=>'ǚ','Ǜ'=>'ǜ','Ǟ'=>'ǟ','Ǡ'=>'ǡ','Ǣ'=>'ǣ','Ǥ'=>'ǥ','Ǧ'=>'ǧ','Ǩ'=>'ǩ','Ǫ'=>'ǫ','Ǭ'=>'ǭ','Ǯ'=>'ǯ','DZ'=>'dz','Dz'=>'dz','Ǵ'=>'ǵ','Ƕ'=>'ƕ','Ƿ'=>'ƿ','Ǹ'=>'ǹ','Ǻ'=>'ǻ','Ǽ'=>'ǽ','Ǿ'=>'ǿ','Ȁ'=>'ȁ','Ȃ'=>'ȃ','Ȅ'=>'ȅ','Ȇ'=>'ȇ','Ȉ'=>'ȉ','Ȋ'=>'ȋ','Ȍ'=>'ȍ','Ȏ'=>'ȏ','Ȑ'=>'ȑ','Ȓ'=>'ȓ','Ȕ'=>'ȕ','Ȗ'=>'ȗ','Ș'=>'ș','Ț'=>'ț','Ȝ'=>'ȝ','Ȟ'=>'ȟ','Ƞ'=>'ƞ','Ȣ'=>'ȣ','Ȥ'=>'ȥ','Ȧ'=>'ȧ','Ȩ'=>'ȩ','Ȫ'=>'ȫ','Ȭ'=>'ȭ','Ȯ'=>'ȯ','Ȱ'=>'ȱ','Ȳ'=>'ȳ','Ⱥ'=>'ⱥ','Ȼ'=>'ȼ','Ƚ'=>'ƚ','Ⱦ'=>'ⱦ','Ɂ'=>'ɂ','Ƀ'=>'ƀ','Ʉ'=>'ʉ','Ʌ'=>'ʌ','Ɇ'=>'ɇ','Ɉ'=>'ɉ','Ɋ'=>'ɋ','Ɍ'=>'ɍ','Ɏ'=>'ɏ','ͅ'=>'ι','Ά'=>'ά','Έ'=>'έ','Ή'=>'ή','Ί'=>'ί','Ό'=>'ό','Ύ'=>'ύ','Ώ'=>'ώ','Α'=>'α','Β'=>'β','Γ'=>'γ','Δ'=>'δ','Ε'=>'ε','Ζ'=>'ζ','Η'=>'η','Θ'=>'θ','Ι'=>'ι','Κ'=>'κ','Λ'=>'λ','Μ'=>'μ','Ν'=>'ν','Ξ'=>'ξ','Ο'=>'ο','Π'=>'π','Ρ'=>'ρ','Σ'=>'σ','Τ'=>'τ','Υ'=>'υ','Φ'=>'φ','Χ'=>'χ','Ψ'=>'ψ','Ω'=>'ω','Ϊ'=>'ϊ','Ϋ'=>'ϋ','ς'=>'σ','ϐ'=>'β','ϑ'=>'θ','ϕ'=>'φ','ϖ'=>'π','Ϙ'=>'ϙ','Ϛ'=>'ϛ','Ϝ'=>'ϝ','Ϟ'=>'ϟ','Ϡ'=>'ϡ','Ϣ'=>'ϣ','Ϥ'=>'ϥ','Ϧ'=>'ϧ','Ϩ'=>'ϩ','Ϫ'=>'ϫ','Ϭ'=>'ϭ','Ϯ'=>'ϯ','ϰ'=>'κ','ϱ'=>'ρ','ϴ'=>'θ','ϵ'=>'ε','Ϸ'=>'ϸ','Ϲ'=>'ϲ','Ϻ'=>'ϻ','Ͻ'=>'ͻ','Ͼ'=>'ͼ','Ͽ'=>'ͽ','Ѐ'=>'ѐ','Ё'=>'ё','Ђ'=>'ђ','Ѓ'=>'ѓ','Є'=>'є','Ѕ'=>'ѕ','І'=>'і','Ї'=>'ї','Ј'=>'ј','Љ'=>'љ','Њ'=>'њ','Ћ'=>'ћ','Ќ'=>'ќ','Ѝ'=>'ѝ','Ў'=>'ў','Џ'=>'џ','А'=>'а','Б'=>'б','В'=>'в','Г'=>'г','Д'=>'д','Е'=>'е','Ж'=>'ж','З'=>'з','И'=>'и','Й'=>'й','К'=>'к','Л'=>'л','М'=>'м','Н'=>'н','О'=>'о','П'=>'п','Р'=>'р','С'=>'с','Т'=>'т','У'=>'у','Ф'=>'ф','Х'=>'х','Ц'=>'ц','Ч'=>'ч','Ш'=>'ш','Щ'=>'щ','Ъ'=>'ъ','Ы'=>'ы','Ь'=>'ь','Э'=>'э','Ю'=>'ю','Я'=>'я','Ѡ'=>'ѡ','Ѣ'=>'ѣ','Ѥ'=>'ѥ','Ѧ'=>'ѧ','Ѩ'=>'ѩ','Ѫ'=>'ѫ','Ѭ'=>'ѭ','Ѯ'=>'ѯ','Ѱ'=>'ѱ','Ѳ'=>'ѳ','Ѵ'=>'ѵ','Ѷ'=>'ѷ','Ѹ'=>'ѹ','Ѻ'=>'ѻ','Ѽ'=>'ѽ','Ѿ'=>'ѿ','Ҁ'=>'ҁ','Ҋ'=>'ҋ','Ҍ'=>'ҍ','Ҏ'=>'ҏ','Ґ'=>'ґ','Ғ'=>'ғ','Ҕ'=>'ҕ','Җ'=>'җ','Ҙ'=>'ҙ','Қ'=>'қ','Ҝ'=>'ҝ','Ҟ'=>'ҟ','Ҡ'=>'ҡ','Ң'=>'ң','Ҥ'=>'ҥ','Ҧ'=>'ҧ','Ҩ'=>'ҩ','Ҫ'=>'ҫ','Ҭ'=>'ҭ','Ү'=>'ү','Ұ'=>'ұ','Ҳ'=>'ҳ','Ҵ'=>'ҵ','Ҷ'=>'ҷ','Ҹ'=>'ҹ','Һ'=>'һ','Ҽ'=>'ҽ','Ҿ'=>'ҿ','Ӏ'=>'ӏ','Ӂ'=>'ӂ','Ӄ'=>'ӄ','Ӆ'=>'ӆ','Ӈ'=>'ӈ','Ӊ'=>'ӊ','Ӌ'=>'ӌ','Ӎ'=>'ӎ','Ӑ'=>'ӑ','Ӓ'=>'ӓ','Ӕ'=>'ӕ','Ӗ'=>'ӗ','Ә'=>'ә','Ӛ'=>'ӛ','Ӝ'=>'ӝ','Ӟ'=>'ӟ','Ӡ'=>'ӡ','Ӣ'=>'ӣ','Ӥ'=>'ӥ','Ӧ'=>'ӧ','Ө'=>'ө','Ӫ'=>'ӫ','Ӭ'=>'ӭ','Ӯ'=>'ӯ','Ӱ'=>'ӱ','Ӳ'=>'ӳ','Ӵ'=>'ӵ','Ӷ'=>'ӷ','Ӹ'=>'ӹ','Ӻ'=>'ӻ','Ӽ'=>'ӽ','Ӿ'=>'ӿ','Ԁ'=>'ԁ','Ԃ'=>'ԃ','Ԅ'=>'ԅ','Ԇ'=>'ԇ','Ԉ'=>'ԉ','Ԋ'=>'ԋ','Ԍ'=>'ԍ','Ԏ'=>'ԏ','Ԑ'=>'ԑ','Ԓ'=>'ԓ','Ա'=>'ա','Բ'=>'բ','Գ'=>'գ','Դ'=>'դ','Ե'=>'ե','Զ'=>'զ','Է'=>'է','Ը'=>'ը','Թ'=>'թ','Ժ'=>'ժ','Ի'=>'ի','Լ'=>'լ','Խ'=>'խ','Ծ'=>'ծ','Կ'=>'կ','Հ'=>'հ','Ձ'=>'ձ','Ղ'=>'ղ','Ճ'=>'ճ','Մ'=>'մ','Յ'=>'յ','Ն'=>'ն','Շ'=>'շ','Ո'=>'ո','Չ'=>'չ','Պ'=>'պ','Ջ'=>'ջ','Ռ'=>'ռ','Ս'=>'ս','Վ'=>'վ','Տ'=>'տ','Ր'=>'ր','Ց'=>'ց','Ւ'=>'ւ','Փ'=>'փ','Ք'=>'ք','Օ'=>'օ','Ֆ'=>'ֆ','Ⴀ'=>'ⴀ','Ⴁ'=>'ⴁ','Ⴂ'=>'ⴂ','Ⴃ'=>'ⴃ','Ⴄ'=>'ⴄ','Ⴅ'=>'ⴅ','Ⴆ'=>'ⴆ','Ⴇ'=>'ⴇ','Ⴈ'=>'ⴈ','Ⴉ'=>'ⴉ','Ⴊ'=>'ⴊ','Ⴋ'=>'ⴋ','Ⴌ'=>'ⴌ','Ⴍ'=>'ⴍ','Ⴎ'=>'ⴎ','Ⴏ'=>'ⴏ','Ⴐ'=>'ⴐ','Ⴑ'=>'ⴑ','Ⴒ'=>'ⴒ','Ⴓ'=>'ⴓ','Ⴔ'=>'ⴔ','Ⴕ'=>'ⴕ','Ⴖ'=>'ⴖ','Ⴗ'=>'ⴗ','Ⴘ'=>'ⴘ','Ⴙ'=>'ⴙ','Ⴚ'=>'ⴚ','Ⴛ'=>'ⴛ','Ⴜ'=>'ⴜ','Ⴝ'=>'ⴝ','Ⴞ'=>'ⴞ','Ⴟ'=>'ⴟ','Ⴠ'=>'ⴠ','Ⴡ'=>'ⴡ','Ⴢ'=>'ⴢ','Ⴣ'=>'ⴣ','Ⴤ'=>'ⴤ','Ⴥ'=>'ⴥ','Ḁ'=>'ḁ','Ḃ'=>'ḃ','Ḅ'=>'ḅ','Ḇ'=>'ḇ','Ḉ'=>'ḉ','Ḋ'=>'ḋ','Ḍ'=>'ḍ','Ḏ'=>'ḏ','Ḑ'=>'ḑ','Ḓ'=>'ḓ','Ḕ'=>'ḕ','Ḗ'=>'ḗ','Ḙ'=>'ḙ','Ḛ'=>'ḛ','Ḝ'=>'ḝ','Ḟ'=>'ḟ','Ḡ'=>'ḡ','Ḣ'=>'ḣ','Ḥ'=>'ḥ','Ḧ'=>'ḧ','Ḩ'=>'ḩ','Ḫ'=>'ḫ','Ḭ'=>'ḭ','Ḯ'=>'ḯ','Ḱ'=>'ḱ','Ḳ'=>'ḳ','Ḵ'=>'ḵ','Ḷ'=>'ḷ','Ḹ'=>'ḹ','Ḻ'=>'ḻ','Ḽ'=>'ḽ','Ḿ'=>'ḿ','Ṁ'=>'ṁ','Ṃ'=>'ṃ','Ṅ'=>'ṅ','Ṇ'=>'ṇ','Ṉ'=>'ṉ','Ṋ'=>'ṋ','Ṍ'=>'ṍ','Ṏ'=>'ṏ','Ṑ'=>'ṑ','Ṓ'=>'ṓ','Ṕ'=>'ṕ','Ṗ'=>'ṗ','Ṙ'=>'ṙ','Ṛ'=>'ṛ','Ṝ'=>'ṝ','Ṟ'=>'ṟ','Ṡ'=>'ṡ','Ṣ'=>'ṣ','Ṥ'=>'ṥ','Ṧ'=>'ṧ','Ṩ'=>'ṩ','Ṫ'=>'ṫ','Ṭ'=>'ṭ','Ṯ'=>'ṯ','Ṱ'=>'ṱ','Ṳ'=>'ṳ','Ṵ'=>'ṵ','Ṷ'=>'ṷ','Ṹ'=>'ṹ','Ṻ'=>'ṻ','Ṽ'=>'ṽ','Ṿ'=>'ṿ','Ẁ'=>'ẁ','Ẃ'=>'ẃ','Ẅ'=>'ẅ','Ẇ'=>'ẇ','Ẉ'=>'ẉ','Ẋ'=>'ẋ','Ẍ'=>'ẍ','Ẏ'=>'ẏ','Ẑ'=>'ẑ','Ẓ'=>'ẓ','Ẕ'=>'ẕ','ẛ'=>'ṡ','Ạ'=>'ạ','Ả'=>'ả','Ấ'=>'ấ','Ầ'=>'ầ','Ẩ'=>'ẩ','Ẫ'=>'ẫ','Ậ'=>'ậ','Ắ'=>'ắ','Ằ'=>'ằ','Ẳ'=>'ẳ','Ẵ'=>'ẵ','Ặ'=>'ặ','Ẹ'=>'ẹ','Ẻ'=>'ẻ','Ẽ'=>'ẽ','Ế'=>'ế','Ề'=>'ề','Ể'=>'ể','Ễ'=>'ễ','Ệ'=>'ệ','Ỉ'=>'ỉ','Ị'=>'ị','Ọ'=>'ọ','Ỏ'=>'ỏ','Ố'=>'ố','Ồ'=>'ồ','Ổ'=>'ổ','Ỗ'=>'ỗ','Ộ'=>'ộ','Ớ'=>'ớ','Ờ'=>'ờ','Ở'=>'ở','Ỡ'=>'ỡ','Ợ'=>'ợ','Ụ'=>'ụ','Ủ'=>'ủ','Ứ'=>'ứ','Ừ'=>'ừ','Ử'=>'ử','Ữ'=>'ữ','Ự'=>'ự','Ỳ'=>'ỳ','Ỵ'=>'ỵ','Ỷ'=>'ỷ','Ỹ'=>'ỹ','Ἀ'=>'ἀ','Ἁ'=>'ἁ','Ἂ'=>'ἂ','Ἃ'=>'ἃ','Ἄ'=>'ἄ','Ἅ'=>'ἅ','Ἆ'=>'ἆ','Ἇ'=>'ἇ','Ἐ'=>'ἐ','Ἑ'=>'ἑ','Ἒ'=>'ἒ','Ἓ'=>'ἓ','Ἔ'=>'ἔ','Ἕ'=>'ἕ','Ἠ'=>'ἠ','Ἡ'=>'ἡ','Ἢ'=>'ἢ','Ἣ'=>'ἣ','Ἤ'=>'ἤ','Ἥ'=>'ἥ','Ἦ'=>'ἦ','Ἧ'=>'ἧ','Ἰ'=>'ἰ','Ἱ'=>'ἱ','Ἲ'=>'ἲ','Ἳ'=>'ἳ','Ἴ'=>'ἴ','Ἵ'=>'ἵ','Ἶ'=>'ἶ','Ἷ'=>'ἷ','Ὀ'=>'ὀ','Ὁ'=>'ὁ','Ὂ'=>'ὂ','Ὃ'=>'ὃ','Ὄ'=>'ὄ','Ὅ'=>'ὅ','Ὑ'=>'ὑ','Ὓ'=>'ὓ','Ὕ'=>'ὕ','Ὗ'=>'ὗ','Ὠ'=>'ὠ','Ὡ'=>'ὡ','Ὢ'=>'ὢ','Ὣ'=>'ὣ','Ὤ'=>'ὤ','Ὥ'=>'ὥ','Ὦ'=>'ὦ','Ὧ'=>'ὧ','Ᾰ'=>'ᾰ','Ᾱ'=>'ᾱ','Ὰ'=>'ὰ','Ά'=>'ά','ι'=>'ι','Ὲ'=>'ὲ','Έ'=>'έ','Ὴ'=>'ὴ','Ή'=>'ή','Ῐ'=>'ῐ','Ῑ'=>'ῑ','Ὶ'=>'ὶ','Ί'=>'ί','Ῠ'=>'ῠ','Ῡ'=>'ῡ','Ὺ'=>'ὺ','Ύ'=>'ύ','Ῥ'=>'ῥ','Ὸ'=>'ὸ','Ό'=>'ό','Ὼ'=>'ὼ','Ώ'=>'ώ','Ω'=>'ω','K'=>'k','Å'=>'å','Ⅎ'=>'ⅎ','Ⅰ'=>'ⅰ','Ⅱ'=>'ⅱ','Ⅲ'=>'ⅲ','Ⅳ'=>'ⅳ','Ⅴ'=>'ⅴ','Ⅵ'=>'ⅵ','Ⅶ'=>'ⅶ','Ⅷ'=>'ⅷ','Ⅸ'=>'ⅸ','Ⅹ'=>'ⅹ','Ⅺ'=>'ⅺ','Ⅻ'=>'ⅻ','Ⅼ'=>'ⅼ','Ⅽ'=>'ⅽ','Ⅾ'=>'ⅾ','Ⅿ'=>'ⅿ','Ↄ'=>'ↄ','Ⓐ'=>'ⓐ','Ⓑ'=>'ⓑ','Ⓒ'=>'ⓒ','Ⓓ'=>'ⓓ','Ⓔ'=>'ⓔ','Ⓕ'=>'ⓕ','Ⓖ'=>'ⓖ','Ⓗ'=>'ⓗ','Ⓘ'=>'ⓘ','Ⓙ'=>'ⓙ','Ⓚ'=>'ⓚ','Ⓛ'=>'ⓛ','Ⓜ'=>'ⓜ','Ⓝ'=>'ⓝ','Ⓞ'=>'ⓞ','Ⓟ'=>'ⓟ','Ⓠ'=>'ⓠ','Ⓡ'=>'ⓡ','Ⓢ'=>'ⓢ','Ⓣ'=>'ⓣ','Ⓤ'=>'ⓤ','Ⓥ'=>'ⓥ','Ⓦ'=>'ⓦ','Ⓧ'=>'ⓧ','Ⓨ'=>'ⓨ','Ⓩ'=>'ⓩ','Ⰰ'=>'ⰰ','Ⰱ'=>'ⰱ','Ⰲ'=>'ⰲ','Ⰳ'=>'ⰳ','Ⰴ'=>'ⰴ','Ⰵ'=>'ⰵ','Ⰶ'=>'ⰶ','Ⰷ'=>'ⰷ','Ⰸ'=>'ⰸ','Ⰹ'=>'ⰹ','Ⰺ'=>'ⰺ','Ⰻ'=>'ⰻ','Ⰼ'=>'ⰼ','Ⰽ'=>'ⰽ','Ⰾ'=>'ⰾ','Ⰿ'=>'ⰿ','Ⱀ'=>'ⱀ','Ⱁ'=>'ⱁ','Ⱂ'=>'ⱂ','Ⱃ'=>'ⱃ','Ⱄ'=>'ⱄ','Ⱅ'=>'ⱅ','Ⱆ'=>'ⱆ','Ⱇ'=>'ⱇ','Ⱈ'=>'ⱈ','Ⱉ'=>'ⱉ','Ⱊ'=>'ⱊ','Ⱋ'=>'ⱋ','Ⱌ'=>'ⱌ','Ⱍ'=>'ⱍ','Ⱎ'=>'ⱎ','Ⱏ'=>'ⱏ','Ⱐ'=>'ⱐ','Ⱑ'=>'ⱑ','Ⱒ'=>'ⱒ','Ⱓ'=>'ⱓ','Ⱔ'=>'ⱔ','Ⱕ'=>'ⱕ','Ⱖ'=>'ⱖ','Ⱗ'=>'ⱗ','Ⱘ'=>'ⱘ','Ⱙ'=>'ⱙ','Ⱚ'=>'ⱚ','Ⱛ'=>'ⱛ','Ⱜ'=>'ⱜ','Ⱝ'=>'ⱝ','Ⱞ'=>'ⱞ','Ⱡ'=>'ⱡ','Ɫ'=>'ɫ','Ᵽ'=>'ᵽ','Ɽ'=>'ɽ','Ⱨ'=>'ⱨ','Ⱪ'=>'ⱪ','Ⱬ'=>'ⱬ','Ⱶ'=>'ⱶ','Ⲁ'=>'ⲁ','Ⲃ'=>'ⲃ','Ⲅ'=>'ⲅ','Ⲇ'=>'ⲇ','Ⲉ'=>'ⲉ','Ⲋ'=>'ⲋ','Ⲍ'=>'ⲍ','Ⲏ'=>'ⲏ','Ⲑ'=>'ⲑ','Ⲓ'=>'ⲓ','Ⲕ'=>'ⲕ','Ⲗ'=>'ⲗ','Ⲙ'=>'ⲙ','Ⲛ'=>'ⲛ','Ⲝ'=>'ⲝ','Ⲟ'=>'ⲟ','Ⲡ'=>'ⲡ','Ⲣ'=>'ⲣ','Ⲥ'=>'ⲥ','Ⲧ'=>'ⲧ','Ⲩ'=>'ⲩ','Ⲫ'=>'ⲫ','Ⲭ'=>'ⲭ','Ⲯ'=>'ⲯ','Ⲱ'=>'ⲱ','Ⲳ'=>'ⲳ','Ⲵ'=>'ⲵ','Ⲷ'=>'ⲷ','Ⲹ'=>'ⲹ','Ⲻ'=>'ⲻ','Ⲽ'=>'ⲽ','Ⲿ'=>'ⲿ','Ⳁ'=>'ⳁ','Ⳃ'=>'ⳃ','Ⳅ'=>'ⳅ','Ⳇ'=>'ⳇ','Ⳉ'=>'ⳉ','Ⳋ'=>'ⳋ','Ⳍ'=>'ⳍ','Ⳏ'=>'ⳏ','Ⳑ'=>'ⳑ','Ⳓ'=>'ⳓ','Ⳕ'=>'ⳕ','Ⳗ'=>'ⳗ','Ⳙ'=>'ⳙ','Ⳛ'=>'ⳛ','Ⳝ'=>'ⳝ','Ⳟ'=>'ⳟ','Ⳡ'=>'ⳡ','Ⳣ'=>'ⳣ','Ａ'=>'ａ','Ｂ'=>'ｂ','Ｃ'=>'ｃ','Ｄ'=>'ｄ','Ｅ'=>'ｅ','Ｆ'=>'ｆ','Ｇ'=>'ｇ','Ｈ'=>'ｈ','Ｉ'=>'ｉ','Ｊ'=>'ｊ','Ｋ'=>'ｋ','Ｌ'=>'ｌ','Ｍ'=>'ｍ','Ｎ'=>'ｎ','Ｏ'=>'ｏ','Ｐ'=>'ｐ','Ｑ'=>'ｑ','Ｒ'=>'ｒ','Ｓ'=>'ｓ','Ｔ'=>'ｔ','Ｕ'=>'ｕ','Ｖ'=>'ｖ','Ｗ'=>'ｗ','Ｘ'=>'ｘ','Ｙ'=>'ｙ','Ｚ'=>'ｚ','𐐀'=>'𐐨','𐐁'=>'𐐩','𐐂'=>'𐐪','𐐃'=>'𐐫','𐐄'=>'𐐬','𐐅'=>'𐐭','𐐆'=>'𐐮','𐐇'=>'𐐯','𐐈'=>'𐐰','𐐉'=>'𐐱','𐐊'=>'𐐲','𐐋'=>'𐐳','𐐌'=>'𐐴','𐐍'=>'𐐵','𐐎'=>'𐐶','𐐏'=>'𐐷','𐐐'=>'𐐸','𐐑'=>'𐐹','𐐒'=>'𐐺','𐐓'=>'𐐻','𐐔'=>'𐐼','𐐕'=>'𐐽','𐐖'=>'𐐾','𐐗'=>'𐐿','𐐘'=>'𐑀','𐐙'=>'𐑁','𐐚'=>'𐑂','𐐛'=>'𐑃','𐐜'=>'𐑄','𐐝'=>'𐑅','𐐞'=>'𐑆','𐐟'=>'𐑇','𐐠'=>'𐑈','𐐡'=>'𐑉','𐐢'=>'𐑊','𐐣'=>'𐑋','𐐤'=>'𐑌','𐐥'=>'𐑍','𐐦'=>'𐑎','𐐧'=>'𐑏'); \ No newline at end of file

+<?php return array('ß'=>'ss','İ'=>'i̇','ʼn'=>'ʼn','ǰ'=>'ǰ','ΐ'=>'ΐ','ΰ'=>'ΰ','և'=>'եւ','ẖ'=>'ẖ','ẗ'=>'ẗ','ẘ'=>'ẘ','ẙ'=>'ẙ','ẚ'=>'aʾ','ὐ'=>'ὐ','ὒ'=>'ὒ','ὔ'=>'ὔ','ὖ'=>'ὖ','ᾀ'=>'ἀι','ᾁ'=>'ἁι','ᾂ'=>'ἂι','ᾃ'=>'ἃι','ᾄ'=>'ἄι','ᾅ'=>'ἅι','ᾆ'=>'ἆι','ᾇ'=>'ἇι','ᾈ'=>'ἀι','ᾉ'=>'ἁι','ᾊ'=>'ἂι','ᾋ'=>'ἃι','ᾌ'=>'ἄι','ᾍ'=>'ἅι','ᾎ'=>'ἆι','ᾏ'=>'ἇι','ᾐ'=>'ἠι','ᾑ'=>'ἡι','ᾒ'=>'ἢι','ᾓ'=>'ἣι','ᾔ'=>'ἤι','ᾕ'=>'ἥι','ᾖ'=>'ἦι','ᾗ'=>'ἧι','ᾘ'=>'ἠι','ᾙ'=>'ἡι','ᾚ'=>'ἢι','ᾛ'=>'ἣι','ᾜ'=>'ἤι','ᾝ'=>'ἥι','ᾞ'=>'ἦι','ᾟ'=>'ἧι','ᾠ'=>'ὠι','ᾡ'=>'ὡι','ᾢ'=>'ὢι','ᾣ'=>'ὣι','ᾤ'=>'ὤι','ᾥ'=>'ὥι','ᾦ'=>'ὦι','ᾧ'=>'ὧι','ᾨ'=>'ὠι','ᾩ'=>'ὡι','ᾪ'=>'ὢι','ᾫ'=>'ὣι','ᾬ'=>'ὤι','ᾭ'=>'ὥι','ᾮ'=>'ὦι','ᾯ'=>'ὧι','ᾲ'=>'ὰι','ᾳ'=>'αι','ᾴ'=>'άι','ᾶ'=>'ᾶ','ᾷ'=>'ᾶι','ᾼ'=>'αι','ῂ'=>'ὴι','ῃ'=>'ηι','ῄ'=>'ήι','ῆ'=>'ῆ','ῇ'=>'ῆι','ῌ'=>'ηι','ῒ'=>'ῒ','ΐ'=>'ΐ','ῖ'=>'ῖ','ῗ'=>'ῗ','ῢ'=>'ῢ','ΰ'=>'ΰ','ῤ'=>'ῤ','ῦ'=>'ῦ','ῧ'=>'ῧ','ῲ'=>'ὼι','ῳ'=>'ωι','ῴ'=>'ώι','ῶ'=>'ῶ','ῷ'=>'ῶι','ῼ'=>'ωι','ff'=>'ff','fi'=>'fi','fl'=>'fl','ffi'=>'ffi','ffl'=>'ffl','ſt'=>'st','st'=>'st','ﬓ'=>'մն','ﬔ'=>'մե','ﬕ'=>'մի','ﬖ'=>'վն','ﬗ'=>'մխ'); \ No newline at end of file

+<?php return array('ᾈ'=>'ᾀ','ᾉ'=>'ᾁ','ᾊ'=>'ᾂ','ᾋ'=>'ᾃ','ᾌ'=>'ᾄ','ᾍ'=>'ᾅ','ᾎ'=>'ᾆ','ᾏ'=>'ᾇ','ᾘ'=>'ᾐ','ᾙ'=>'ᾑ','ᾚ'=>'ᾒ','ᾛ'=>'ᾓ','ᾜ'=>'ᾔ','ᾝ'=>'ᾕ','ᾞ'=>'ᾖ','ᾟ'=>'ᾗ','ᾨ'=>'ᾠ','ᾩ'=>'ᾡ','ᾪ'=>'ᾢ','ᾫ'=>'ᾣ','ᾬ'=>'ᾤ','ᾭ'=>'ᾥ','ᾮ'=>'ᾦ','ᾯ'=>'ᾧ','ᾼ'=>'ᾳ','ῌ'=>'ῃ','ῼ'=>'ῳ'); \ No newline at end of file

+// Unset global variables

+// NFC_QC and NFKC_QC values

+// Contains all the ASCII characters appearing in UTF-8, sorted by frequency

+// Contains all the tail bytes that can appear in the composition of a UTF-8 char

+// Constants used by the Hangul [de]composition algorithms

+* @package phpBB3

+ // The string below is the list of all autorized characters, sorted by frequency in latin text

+ // ASCII strings with no special chars return immediately

+ // Note: we do not check for $GLOBALS['utf_canonical_decomp']. It is assumed they are always loaded together

+ // Replace any byte in the range 0x00..0x1F, except for \r, \n and \t

+ // We replace those characters with a 0xFF byte, which is illegal in UTF-8 and will in turn be replaced with a UTF replacement char

+ // ASCII strings return immediately

+ // ASCII strings return immediately

+

+ // ASCII strings return immediately

+ // ASCII strings return immediately

+ // Load some commonly-used tables

+ // Buffer the last ASCII char before the UTF-8 stuff if applicable

+ // UTF char length array

+ // This array is used to determine the length of a UTF character.

+ // Be $c the result of ($str[$pos] & "\xF0") --where $str is the string we're operating on and $pos

+ // the position of the cursor--, if $utf_len_mask[$c] does not exist, the byte is an ASCII char.

+ // Otherwise, if $utf_len_mask[$c] is greater than 0, we have a the leading byte of a multibyte character

+ // whose length is $utf_len_mask[$c] and if it is equal to 0, the byte is a trailing byte.

+ // Leading bytes masks

+ // Trailing bytes masks

+ "\xED" => 1, "\xEF" => 1, "\xC0" => 1, "\xC1" => 1, "\xE0" => 1, "\xF0" => 1,

+ "\xF4" => 1, "\xF5" => 1, "\xF6" => 1, "\xF7" => 1, "\xF8" => 1, "\xF9" => 1,

+ "\xFA" => 1, "\xFB" => 1, "\xFC" => 1, "\xFD" => 1, "\xFE" => 1, "\xFF" => 1

+ 2 => "\xE0\xC0",

+ 3 => "\xF0\xC0\xC0",

+ 4 => "\xF8\xC0\xC0\xC0"

+ 2 => "\xC0\x80",

+ 3 => "\xE0\x80\x80",

+ 4 => "\xF0\x80\x80\x80"

+ // Main loop

+ // STEP 0: Capture the current char and buffer it

+ // Byte at $pos is either a leading byte or a missplaced trailing byte

+ // Capture the char

+ // Let's find out if a thorough check is needed

+ // If the UTF char is in the qc array then it may not be in normal form. We do nothing here, the actual processing is below this "if" block

+ else if (isset($utf_combining_class[$utf_char]))

+ // A combining character that is NOT canonically ordered

+ // A combining character that IS canonically ordered, skip to the next char

+ // At this point, $utf_char holds a UTF char that we know is not a NF[K]C_QC and is not a combining character.

+ // It can be a singleton, a canonical composite, a replacement char or an even an ill-formed bunch of bytes. Let's find out

+ // Check that we have the correct number of trailing bytes

+ // Current char isn't well-formed or legal: either one or several trailing bytes are missing, or the Unicode char

+ // has been encoded in a five- or six- byte sequence

+ else if ($utf_char[0] < "\xFC")

+

+ switch ($c)

+ // Note: 0xED is quite common in Korean

+ // Surrogates (U+D800..U+DFFF) are not allowed in UTF-8 (UTF sequence 0xEDA080..0xEDBFBF)

+ break;

+ // Note: 0xEF is quite common in Japanese

+ // U+FFFE and U+FFFF are explicitly disallowed (UTF sequence 0xEFBFBE..0xEFBFBF)

+ break;

+ // Overlong sequence: Unicode char U+0000..U+007F encoded as a double-byte UTF char

+ break;

+ // Unicode char U+0000..U+07FF encoded in 3 bytes

+ break;

+ // Unicode char U+0000..U+FFFF encoded in 4 bytes

+ break;

+ // Five- and six- byte sequences do not need being checked for here anymore

+ // Out of the Unicode range

+ else if ($utf_char[0] < "\xFC")

+ else if ($utf_char[0] > "\xFD")

+ break;

+ // The char is a valid starter, move the cursor and go on

+ // A trailing byte came out of nowhere, we will advance the cursor and treat the this byte and all following trailing bytes as if

+ // each of them was a Unicode replacement char

+ // STEP 1: Decompose current char

+

+ // We have found a character that is either:

+ // - in the NFC_QC/NFKC_QC list

+ // - a non-starter char that is not canonically ordered

+ //

+ // We are going to capture the shortest UTF sequence that satisfies these two conditions:

+ //

+ // 1 - If the sequence does not start at the begginning of the string, it must begin with a starter,

+ // and that starter must not have the NF[K]C_QC property equal to "MAYBE"

+ //

+ // 2 - If the sequence does not end at the end of the string, it must end with a non-starter and be

+ // immediately followed by a starter that is not on the QC list

+ //

+

+ while ($_pos < $_len);

+ // The char is not decomposable

+ // STEP 2: Capture the starter

+ // Check out the combining class of the first character of the UTF sequence

+ // Not a starter, inspect previous characters

+ // The last 8 characters are kept in a buffer so that we don't have to capture them everytime.

+ // This is enough for all real-life strings but even if it wasn't, we can capture characters in backward mode,

+ // although it is slower than this method.

+ //

+ // In the following loop, $j starts at the previous buffered character ($i - 1, because current character is

+ // at offset $i) and process them in backward mode until we find a starter.

+ //

+ // $k is the index on each UTF character inside of our UTF sequence. At this time, $utf_seq contains one or more

+ // characters numbered 0 to n. $k starts at 0 and for each char we prepend we pre-decrement it and for numbering

+

+ for ($j = $i - 1; $j >= $j_min && $lpos > $tmp_pos; --$j)

+ // The char is a composite, decompose for storage

+

+ while ($_pos < $_len);

+ // Prepend the UTF sequence with our decomposed sequence

+ // The char expanded into several chars

+ $decomp_cnt = sizeof($decomp_seq);

+

+ foreach ($decomp_seq as $decomp_i => $decomp_char)

+ // Decomposed to a single char, easier to prepend

+ // We have found our starter

+ // The starter was not found in the buffer, let's rewind some more

+ // $utf_len_mask contains the masks of both leading bytes and trailing bytes. If $utf_en > 0 then it's a leading byte, otherwise it's a trailing byte.

+ // UTF byte

+ // UTF *leading* byte

+ // Decompose the character

+

+ while ($_pos < $_len);

+ // Prepend the UTF sequence with our decomposed sequence

+ // The char expanded into several chars

+ $decomp_cnt = sizeof($decomp_seq);

+ foreach ($decomp_seq as $decomp_i => $utf_char)

+ // Decomposed to a single char, easier to prepend

+ // ASCII char

+ while ($lpos > $tmp_pos);

+ // STEP 3: Capture following combining modifiers

+ while ($pos < $len)

+ // A trailing byte came out of nowhere

+ // Trailing bytes are replaced with Unicode replacement chars, we will just ignore it for now, break out of the loop

+ // as if it was a starter (replacement chars ARE starters) and let the next loop replace it

+ // Combining character, add it to the sequence and move the cursor

+ // Decompose the character

+

+ while ($_pos < $_len);

+ // Combining class 0 and no QC, break out of the loop

+ // Note: we do not know if that character is valid. If it's not, the next iteration will replace it

+ // ASCII chars are starters

+ // STEP 4: Sort and combine

+

+ // Here we sort...

+ $k_max = $k + sizeof($utf_seq);

+ // There is only one char in the UTF sequence, add it then jump to the next iteration of main loop

+ // Note: the two commented lines below can be enabled under PHP5 for a very small performance gain in most cases

+// if (substr_compare($str, $utf_seq[0], $lpos, $pos - $lpos))

+// {

+// }

+ // ...there we combine

+ // We add an empty char at the end of the UTF char sequence. It will act as a starter and trigger the sort/combine routine

+ // at the end of the string without altering it

+ // No combining characters... check for a composite of the two starters

+ // Good ol' composite character

+ else if (isset($utf_jamo_type[$utf_char]))

+ // Current char is a composable jamo

+ if (isset($utf_jamo_type[$starter]) && $utf_jamo_type[$starter] == UNICODE_JAMO_L && $utf_jamo_type[$utf_char] == UNICODE_JAMO_V)

+ // We have a L jamo followed by a V jamo, we are going to prefetch the next char to see if it's a T jamo

+ // L+V+T jamos, combine to a LVT Hangul syllable ($k is incremented)

+ // L+V jamos, combine to a LV Hangul syllable

+ // Non-composable jamo, just add it to the sequence

+ // No composite, just add the first starter to the sequence then continue with the other one

+ // For each class of combining characters

+ foreach ($utf_sort as $cc => $utf_chars)

+ // Look for a composite

+ // Found a composite, replace the starter

+ // No composite, all following characters in that class are blocked

+ while (isset($utf_sort[$cc][++$j]));

+ // Add the starter to the normalized sequence, followed by non-starters in canonical order

+

+ foreach ($utf_sort as $utf_chars)

+ // Reset the array and go on

+ while ($k <= $k_max);

+ // Only a ASCII char can make the program get here

+ //

+ // First we skip the current byte with ++$pos, then we quickly skip following ASCII chars with strspn().

+ //

+ // The first two "if"'s here can be removed, with the consequences of being faster on latin text (lots of ASCII) and slower on

+ // multi-byte text (where the only ASCII chars are spaces and punctuation)

+ while ($pos < $len);

+ // Now is time to return the string

+ // If the $tmp_pos cursor is not at the beggining of the string then at least one character was not in normal form. Replace $str with the fixed version

+ // The $tmp_pos cursor is at the end of $str, therefore $tmp holds the whole $str

+ // The rightmost chunk of $str has not been appended to $tmp yet

+ // The string was already in normal form

+ // Load some commonly-used tables

+ // UTF char length array

+ // Leading bytes masks

+ // Trailing bytes masks

+ // Some extra checks are triggered on the first byte of a UTF sequence

+ "\xED" => 1, "\xEF" => 1, "\xC0" => 1, "\xC1" => 1, "\xE0" => 1, "\xF0" => 1,

+ "\xF4" => 1, "\xF5" => 1, "\xF6" => 1, "\xF7" => 1, "\xF8" => 1, "\xF9" => 1,

+ "\xFA" => 1, "\xFB" => 1, "\xFC" => 1, "\xFD" => 1, "\xFE" => 1, "\xFF" => 1

+ // These masks are used to check if a UTF sequence is well formed. Here are the only 3 lengths we acknowledge:

+ // - 2-byte: 110? ???? 10?? ????

+ // - 3-byte: 1110 ???? 10?? ???? 10?? ????

+ // - 4-byte: 1111 0??? 10?? ???? 10?? ???? 10?? ????

+ // Note that 5- and 6- byte sequences are automatically discarded

+ 2 => "\xE0\xC0",

+ 3 => "\xF0\xC0\xC0",

+ 4 => "\xF8\xC0\xC0\xC0"

+

+ 2 => "\xC0\x80",

+ 3 => "\xE0\x80\x80",

+ 4 => "\xF0\x80\x80\x80"

+ // Main loop

+ // STEP 0: Capture the current char

+ // Multibyte char

+ // A trailing byte came out of nowhere, we will treat it and all following trailing bytes as if each of them was a Unicode

+ // replacement char and we will advance the cursor

+ // Dump combiners

+ // STEP 1: Decide what to do with current char

+ // Now, in that order:

+ // - check if that character is decomposable

+ // - check if that character is a non-starter

+ // - check if that character requires extra checks to be performed

+ // Decompose the char

+ // The character decomposed to a non-starter, buffer it for sorting

+ // Not canonically ordered, will require sorting

+ // This character decomposition contains a starter, dump the buffer and continue

+ // Dump combiners

+ foreach ($utf_sort as $utf_chars)

+ // This character decomposition contains an ASCII char, which is a starter. Dump the buffer and continue

+

+ // Dump combiners

+ foreach ($utf_sort as $utf_chars)

+ while ($_pos < $_len);

+ else if (isset($utf_combining_class[$utf_char]))

+ // Combining character

+ // Not in canonical order

+ // Non-decomposable starter, check out if it's a Hangul syllable

+ // Nope, regular UTF char, check that we have the correct number of trailing bytes

+ // Current char isn't well-formed or legal: either one or several trailing bytes are missing, or the Unicode char

+ // has been encoded in a five- or six- byte sequence.

+ // Move the cursor back to its original position then advance it to the position it should really be at

+ foreach ($utf_sort as $utf_chars)

+ // Add a replacement char then another replacement char for every trailing byte.

+ //

+ // @todo I'm not entirely sure that's how we're supposed to mark invalidated byte sequences, check this

+ switch ($utf_char[0])

+ // Note: 0xED is quite common in Korean

+ // Surrogates (U+D800..U+DFFF) are not allowed in UTF-8 (UTF sequence 0xEDA080..0xEDBFBF)

+ foreach ($utf_sort as $utf_chars)

+ break;

+ // Note: 0xEF is quite common in Japanese

+ // U+FFFE and U+FFFF are explicitly disallowed (UTF sequence 0xEFBFBE..0xEFBFBF)

+ foreach ($utf_sort as $utf_chars)

+ break;

+ // Overlong sequence: Unicode char U+0000..U+007F encoded as a double-byte UTF char

+ foreach ($utf_sort as $utf_chars)

+ break;

+ // Unicode char U+0000..U+07FF encoded in 3 bytes

+ foreach ($utf_sort as $utf_chars)

+ break;

+ // Unicode char U+0000..U+FFFF encoded in 4 bytes

+ foreach ($utf_sort as $utf_chars)

+ break;

+ // Out of the Unicode range

+ foreach ($utf_sort as $utf_chars)

+ break;

+ // Hangul syllable

+ // LIndex can only range from 0 to 18, therefore it cannot influence the first two bytes of the L Jamo, which allows us to hardcode them (based on LBase).

+ //

+ // The same goes for VIndex, but for TIndex there's a catch: the value of the third byte could exceed 0xBF and we would have to increment the second byte

+ // Just like other decompositions, the resulting Jamos must be dumped to the tmp string

+ // Do we need to dump stuff to the tmp string?

+ // Dump combiners

+ foreach ($utf_sort as $utf_chars)

+ // ASCII char, which happens to be a starter (as any other ASCII char)

+ // Dump combiners

+ foreach ($utf_sort as $utf_chars)

+ while ($pos < $len);

+ // Now is time to return the string

+ // Dump combiners

+ foreach ($utf_sort as $utf_chars)

+ else if ($tmp_pos)

+ // If the $tmp_pos cursor was moved then at least one character was not in normal form. Replace $str with the fixed version

+ // The $tmp_pos cursor is at the end of $str, therefore $tmp holds the whole $str

+ // The rightmost chunk of $str has not been appended to $tmp yet

+ // The string was already in normal form

+?> \ No newline at end of file

+* @copyright (c) 2006 phpBB Group

+* @todo make sure the replacements are called correctly

+* already done: strtolower, strtoupper, ucfirst, str_split, strrpos, strlen (hopefully!), strpos, substr

+* remaining: clean_username, htmlentities (no longer needed for internal data?), htmlspecialchars (using charset)

+* strspn, chr, ord

+*/

+if (!defined('IN_PHPBB'))

+{

+ exit;

+}

+

+/**

+if (!extension_loaded('xml'))

+{

+ /**

+ * Implementation of PHP's native utf8_encode for people without XML support

+ * This function exploits some nice things that ISO-8859-1 and UTF-8 have in common

+ *

+ * @param string $str ISO-8859-1 encoded data

+ * @return string UTF-8 encoded data

+ */

+ function utf8_encode($str)

+ {

+ $out = '';

+ for ($i = 0, $len = strlen($str); $i < $len; $i++)

+ {

+ $letter = $str[$i];

+ $num = ord($letter);

+ if ($num < 0x80)

+ {

+ $out .= $letter;

+ }

+ else if ($num < 0xC0)

+ {

+ $out .= "\xC2" . $letter;

+ }

+ else

+ {

+ $out .= "\xC3" . chr($num - 64);

+ }

+ }

+ return $out;

+ }

+

+ /**

+ * Implementation of PHP's native utf8_decode for people without XML support

+ *

+ * @param string $string UTF-8 encoded data

+ * @return string ISO-8859-1 encoded data

+ */

+ function utf8_decode($str)

+ {

+ $pos = 0;

+ $len = strlen($str);

+ $ret = '';

+

+ while ($pos < $len)

+ {

+ $ord = ord($str[$pos]) & 0xF0;

+ if ($ord === 0xC0 || $ord === 0xD0)

+ {

+ $charval = ((ord($str[$pos]) & 0x1F) << 6) | (ord($str[$pos + 1]) & 0x3F);

+ $pos += 2;

+ $ret .= (($charval < 256) ? chr($charval) : '?');

+ }

+ else if ($ord === 0xE0)

+ {

+ $ret .= '?';

+ $pos += 3;

+ }

+ else if ($ord === 0xF0)

+ {

+ $ret .= '?';

+ $pos += 4;

+ }

+ else

+ {

+ $ret .= $str[$pos];

+ ++$pos;

+ }

+ }

+ return $ret;

+ }

+}

+

+// mbstring is old and has it's functions around for older versions of PHP.

+// if mbstring is not loaded, we go into native mode.

+if (extension_loaded('mbstring'))

+{

+ /**

+ * UTF-8 aware alternative to strrpos

+ * Find position of last occurrence of a char in a string

+ *

+ * Notes:

+ * - offset for mb_strrpos was added in 5.2.0, we emulate if it is lower

+ */

+ if (version_compare(phpversion(), '5.2.0', '>='))

+ {

+ /**

+ * UTF-8 aware alternative to strrpos

+ * @ignore

+ */

+ function utf8_strrpos($str, $needle, $offset = null)

+ {

+ // Emulate behaviour of strrpos rather than raising warning

+ if (empty($str))

+ {

+ return false;

+ }

+

+ return mb_strrpos($str, $search);

+ }

+ }

+ else

+ {

+ /**

+ * UTF-8 aware alternative to strrpos

+ * @ignore

+ */

+ function utf8_strrpos($str, $needle, $offset = null)

+ {

+ // offset for mb_strrpos was added in 5.2.0

+ if ($offset === false)

+ {

+ // Emulate behaviour of strrpos rather than raising warning

+ if (empty($str))

+ {

+ return false;

+ }

+

+ return mb_strrpos($str, $search);

+ }

+ else

+ {

+ if (!is_int($offset))

+ {

+ trigger_error('utf8_strrpos expects parameter 3 to be long', E_USER_WARNING);

+ return false;

+ }

+

+ $str = mb_substr($str, $offset);

+

+ if (false !== ($pos = mb_strrpos($str, $search)))

+ {

+ return $pos + $offset;

+ }

+

+ return false;

+ }

+ }

+ }

+

+ /**

+ * UTF-8 aware alternative to strpos

+ * @ignore

+ */

+ function utf8_strpos($str, $needle, $offset = null)

+ {

+ if ($offset === false)

+ {

+ return mb_strpos($str, $needle);

+ }

+ else

+ {

+ return mb_strpos($str, $needle, $offset);

+ }

+ }

+

+ /**

+ * UTF-8 aware alternative to strtolower

+ * @ignore

+ */

+ function utf8_strtolower($str)

+ {

+ return mb_strtolower($str);

+ }

+

+ /**

+ * UTF-8 aware alternative to strtoupper

+ * @ignore

+ */

+ function utf8_strtoupper($str)

+ {

+ return mb_strtoupper($str);

+ }

+

+ /**

+ * UTF-8 aware alternative to substr

+ * @ignore

+ */

+ function utf8_substr($str, $offset, $length = null)

+ {

+ if ($length === false)

+ {

+ return mb_substr($str, $offset);

+ }

+ else

+ {

+ return mb_substr($str, $offset, $length);

+ }

+ }

+

+ /**

+ * Return the length (in characters) of a UTF-8 string

+ * @ignore

+ */

+ function utf8_strlen($text)

+ {

+ return mb_strlen($text, 'utf-8');

+ }

+}

+else

+{

+ /**

+ * UTF-8 aware alternative to strrpos

+ * Find position of last occurrence of a char in a string

+ *

+ * @author Harry Fuecks

+ * @param string haystack

+ * @param string needle

+ * @param integer (optional) offset (from left)

+ * @return mixed integer position or FALSE on failure

+ */

+ function utf8_strrpos($str, $needle, $offset = null)

+ {

+ if (is_null($offset))

+ {

+ $ar = explode($needle, $str);

+

+ if (sizeof($ar) > 1)

+ {

+ // Pop off the end of the string where the last match was made

+ array_pop($ar);

+ $str = join($needle, $ar);

+

+ return utf8_strlen($str);

+ }

+ return false;

+ }

+ else

+ {

+ if (!is_int($offset))

+ {

+ trigger_error('utf8_strrpos expects parameter 3 to be long', E_USER_WARNING);

+ return false;

+ }

+

+ $str = utf8_substr($str, $offset);

+

+ if (false !== ($pos = utf8_strrpos($str, $needle)))

+ {

+ return $pos + $offset;

+ }

+

+ return false;

+ }

+ }

+

+ /**

+ * UTF-8 aware alternative to strpos

+ * Find position of first occurrence of a string

+ *

+ * @author Harry Fuecks

+ * @param string haystack

+ * @param string needle

+ * @param integer offset in characters (from left)

+ * @return mixed integer position or FALSE on failure

+ */

+ function utf8_strpos($str, $needle, $offset = null)

+ {

+ if (is_null($offset))

+ {

+ $ar = explode($needle, $str);

+ if (sizeof($ar) > 1)

+ {

+ return utf8_strlen($ar[0]);

+ }

+ return false;

+ }

+ else

+ {

+ if (!is_int($offset))

+ {

+ trigger_error('utf8_strpos: Offset must be an integer', E_USER_ERROR);

+ return false;

+ }

+

+ $str = utf8_substr($str, $offset);

+

+ if (false !== ($pos = utf8_strpos($str, $needle)))

+ {

+ return $pos + $offset;

+ }

+

+ return false;

+ }

+ }

+

+ $UTF8_UPPER_TO_LOWER = array(

+ "\x41" => "\x61", "\x42" => "\x62", "\x43" => "\x63", "\x44" => "\x64",

+ "\x45" => "\x65", "\x46" => "\x66", "\x47" => "\x67", "\x48" => "\x68",

+ "\x49" => "\x69", "\x4A" => "\x6A", "\x4B" => "\x6B", "\x4C" => "\x6C",

+ "\x4D" => "\x6D", "\x4E" => "\x6E", "\x4F" => "\x6F", "\x50" => "\x70",

+ "\x51" => "\x71", "\x52" => "\x72", "\x53" => "\x73", "\x54" => "\x74",

+ "\x55" => "\x75", "\x56" => "\x76", "\x57" => "\x77", "\x58" => "\x78",

+ "\x59" => "\x79", "\x5A" => "\x7A", "\xC3\x80" => "\xC3\xA0", "\xC3\x81" => "\xC3\xA1",

+ "\xC3\x82" => "\xC3\xA2", "\xC3\x83" => "\xC3\xA3", "\xC3\x84" => "\xC3\xA4", "\xC3\x85" => "\xC3\xA5",

+ "\xC3\x86" => "\xC3\xA6", "\xC3\x87" => "\xC3\xA7", "\xC3\x88" => "\xC3\xA8", "\xC3\x89" => "\xC3\xA9",

+ "\xC3\x8A" => "\xC3\xAA", "\xC3\x8B" => "\xC3\xAB", "\xC3\x8C" => "\xC3\xAC", "\xC3\x8D" => "\xC3\xAD",

+ "\xC3\x8E" => "\xC3\xAE", "\xC3\x8F" => "\xC3\xAF", "\xC3\x90" => "\xC3\xB0", "\xC3\x91" => "\xC3\xB1",

+ "\xC3\x92" => "\xC3\xB2", "\xC3\x93" => "\xC3\xB3", "\xC3\x94" => "\xC3\xB4", "\xC3\x95" => "\xC3\xB5",

+ "\xC3\x96" => "\xC3\xB6", "\xC3\x98" => "\xC3\xB8", "\xC3\x99" => "\xC3\xB9", "\xC3\x9A" => "\xC3\xBA",

+ "\xC3\x9B" => "\xC3\xBB", "\xC3\x9C" => "\xC3\xBC", "\xC3\x9D" => "\xC3\xBD", "\xC3\x9E" => "\xC3\xBE",

+ "\xC4\x80" => "\xC4\x81", "\xC4\x82" => "\xC4\x83", "\xC4\x84" => "\xC4\x85", "\xC4\x86" => "\xC4\x87",

+ "\xC4\x88" => "\xC4\x89", "\xC4\x8A" => "\xC4\x8B", "\xC4\x8C" => "\xC4\x8D", "\xC4\x8E" => "\xC4\x8F",

+ "\xC4\x90" => "\xC4\x91", "\xC4\x92" => "\xC4\x93", "\xC4\x96" => "\xC4\x97", "\xC4\x98" => "\xC4\x99",

+ "\xC4\x9A" => "\xC4\x9B", "\xC4\x9C" => "\xC4\x9D", "\xC4\x9E" => "\xC4\x9F", "\xC4\xA0" => "\xC4\xA1",

+ "\xC4\xA2" => "\xC4\xA3", "\xC4\xA4" => "\xC4\xA5", "\xC4\xA6" => "\xC4\xA7", "\xC4\xA8" => "\xC4\xA9",

+ "\xC4\xAA" => "\xC4\xAB", "\xC4\xAE" => "\xC4\xAF", "\xC4\xB4" => "\xC4\xB5", "\xC4\xB6" => "\xC4\xB7",

+ "\xC4\xB9" => "\xC4\xBA", "\xC4\xBB" => "\xC4\xBC", "\xC4\xBD" => "\xC4\xBE", "\xC5\x81" => "\xC5\x82",

+ "\xC5\x83" => "\xC5\x84", "\xC5\x85" => "\xC5\x86", "\xC5\x87" => "\xC5\x88", "\xC5\x8A" => "\xC5\x8B",

+ "\xC5\x8C" => "\xC5\x8D", "\xC5\x90" => "\xC5\x91", "\xC5\x94" => "\xC5\x95", "\xC5\x96" => "\xC5\x97",

+ "\xC5\x98" => "\xC5\x99", "\xC5\x9A" => "\xC5\x9B", "\xC5\x9C" => "\xC5\x9D", "\xC5\x9E" => "\xC5\x9F",

+ "\xC5\xA0" => "\xC5\xA1", "\xC5\xA2" => "\xC5\xA3", "\xC5\xA4" => "\xC5\xA5", "\xC5\xA6" => "\xC5\xA7",

+ "\xC5\xA8" => "\xC5\xA9", "\xC5\xAA" => "\xC5\xAB", "\xC5\xAC" => "\xC5\xAD", "\xC5\xAE" => "\xC5\xAF",

+ "\xC5\xB0" => "\xC5\xB1", "\xC5\xB2" => "\xC5\xB3", "\xC5\xB4" => "\xC5\xB5", "\xC5\xB6" => "\xC5\xB7",

+ "\xC5\xB8" => "\xC3\xBF", "\xC5\xB9" => "\xC5\xBA", "\xC5\xBB" => "\xC5\xBC", "\xC5\xBD" => "\xC5\xBE",

+ "\xC6\xA0" => "\xC6\xA1", "\xC6\xAF" => "\xC6\xB0", "\xC8\x98" => "\xC8\x99", "\xC8\x9A" => "\xC8\x9B",

+ "\xCE\x86" => "\xCE\xAC", "\xCE\x88" => "\xCE\xAD", "\xCE\x89" => "\xCE\xAE", "\xCE\x8A" => "\xCE\xAF",

+ "\xCE\x8C" => "\xCF\x8C", "\xCE\x8E" => "\xCF\x8D", "\xCE\x8F" => "\xCF\x8E", "\xCE\x91" => "\xCE\xB1",

+ "\xCE\x92" => "\xCE\xB2", "\xCE\x93" => "\xCE\xB3", "\xCE\x94" => "\xCE\xB4", "\xCE\x95" => "\xCE\xB5",

+ "\xCE\x96" => "\xCE\xB6", "\xCE\x97" => "\xCE\xB7", "\xCE\x98" => "\xCE\xB8", "\xCE\x99" => "\xCE\xB9",

+ "\xCE\x9A" => "\xCE\xBA", "\xCE\x9B" => "\xCE\xBB", "\xCE\x9C" => "\xCE\xBC", "\xCE\x9D" => "\xCE\xBD",

+ "\xCE\x9E" => "\xCE\xBE", "\xCE\x9F" => "\xCE\xBF", "\xCE\xA0" => "\xCF\x80", "\xCE\xA1" => "\xCF\x81",

+ "\xCE\xA3" => "\xCF\x83", "\xCE\xA4" => "\xCF\x84", "\xCE\xA5" => "\xCF\x85", "\xCE\xA6" => "\xCF\x86",

+ "\xCE\xA7" => "\xCF\x87", "\xCE\xA8" => "\xCF\x88", "\xCE\xA9" => "\xCF\x89", "\xCE\xAA" => "\xCF\x8A",

+ "\xCE\xAB" => "\xCF\x8B", "\xD0\x81" => "\xD1\x91", "\xD0\x82" => "\xD1\x92", "\xD0\x83" => "\xD1\x93",

+ "\xD0\x84" => "\xD1\x94", "\xD0\x85" => "\xD1\x95", "\xD0\x86" => "\xD1\x96", "\xD0\x87" => "\xD1\x97",

+ "\xD0\x88" => "\xD1\x98", "\xD0\x89" => "\xD1\x99", "\xD0\x8A" => "\xD1\x9A", "\xD0\x8B" => "\xD1\x9B",

+ "\xD0\x8C" => "\xD1\x9C", "\xD0\x8E" => "\xD1\x9E", "\xD0\x8F" => "\xD1\x9F", "\xD0\x90" => "\xD0\xB0",

+ "\xD0\x91" => "\xD0\xB1", "\xD0\x92" => "\xD0\xB2", "\xD0\x93" => "\xD0\xB3", "\xD0\x94" => "\xD0\xB4",

+ "\xD0\x95" => "\xD0\xB5", "\xD0\x96" => "\xD0\xB6", "\xD0\x97" => "\xD0\xB7", "\xD0\x98" => "\xD0\xB8",

+ "\xD0\x99" => "\xD0\xB9", "\xD0\x9A" => "\xD0\xBA", "\xD0\x9B" => "\xD0\xBB", "\xD0\x9C" => "\xD0\xBC",

+ "\xD0\x9D" => "\xD0\xBD", "\xD0\x9E" => "\xD0\xBE", "\xD0\x9F" => "\xD0\xBF", "\xD0\xA0" => "\xD1\x80",

+ "\xD0\xA1" => "\xD1\x81", "\xD0\xA2" => "\xD1\x82", "\xD0\xA3" => "\xD1\x83", "\xD0\xA4" => "\xD1\x84",

+ "\xD0\xA5" => "\xD1\x85", "\xD0\xA6" => "\xD1\x86", "\xD0\xA7" => "\xD1\x87", "\xD0\xA8" => "\xD1\x88",

+ "\xD0\xA9" => "\xD1\x89", "\xD0\xAA" => "\xD1\x8A", "\xD0\xAB" => "\xD1\x8B", "\xD0\xAC" => "\xD1\x8C",

+ "\xD0\xAD" => "\xD1\x8D", "\xD0\xAE" => "\xD1\x8E", "\xD0\xAF" => "\xD1\x8F", "\xD2\x90" => "\xD2\x91",

+ "\xE1\xB8\x82" => "\xE1\xB8\x83", "\xE1\xB8\x8A" => "\xE1\xB8\x8B", "\xE1\xB8\x9E" => "\xE1\xB8\x9F", "\xE1\xB9\x80" => "\xE1\xB9\x81",

+ "\xE1\xB9\x96" => "\xE1\xB9\x97", "\xE1\xB9\xA0" => "\xE1\xB9\xA1", "\xE1\xB9\xAA" => "\xE1\xB9\xAB", "\xE1\xBA\x80" => "\xE1\xBA\x81",

+ "\xE1\xBA\x82" => "\xE1\xBA\x83", "\xE1\xBA\x84" => "\xE1\xBA\x85", "\xE1\xBB\xB2" => "\xE1\xBB\xB3"

+ );

+

+ $UTF8_LOWER_TO_UPPER = array(

+ "\x61" => "\x41", "\x62" => "\x42", "\x63" => "\x43", "\x64" => "\x44",

+ "\x65" => "\x45", "\x66" => "\x46", "\x67" => "\x47", "\x68" => "\x48",

+ "\x69" => "\x49", "\x6A" => "\x4A", "\x6B" => "\x4B", "\x6C" => "\x4C",

+ "\x6D" => "\x4D", "\x6E" => "\x4E", "\x6F" => "\x4F", "\x70" => "\x50",

+ "\x71" => "\x51", "\x72" => "\x52", "\x73" => "\x53", "\x74" => "\x54",

+ "\x75" => "\x55", "\x76" => "\x56", "\x77" => "\x57", "\x78" => "\x58",

+ "\x79" => "\x59", "\x7A" => "\x5A", "\xC3\xA0" => "\xC3\x80", "\xC3\xA1" => "\xC3\x81",

+ "\xC3\xA2" => "\xC3\x82", "\xC3\xA3" => "\xC3\x83", "\xC3\xA4" => "\xC3\x84", "\xC3\xA5" => "\xC3\x85",

+ "\xC3\xA6" => "\xC3\x86", "\xC3\xA7" => "\xC3\x87", "\xC3\xA8" => "\xC3\x88", "\xC3\xA9" => "\xC3\x89",

+ "\xC3\xAA" => "\xC3\x8A", "\xC3\xAB" => "\xC3\x8B", "\xC3\xAC" => "\xC3\x8C", "\xC3\xAD" => "\xC3\x8D",

+ "\xC3\xAE" => "\xC3\x8E", "\xC3\xAF" => "\xC3\x8F", "\xC3\xB0" => "\xC3\x90", "\xC3\xB1" => "\xC3\x91",

+ "\xC3\xB2" => "\xC3\x92", "\xC3\xB3" => "\xC3\x93", "\xC3\xB4" => "\xC3\x94", "\xC3\xB5" => "\xC3\x95",

+ "\xC3\xB6" => "\xC3\x96", "\xC3\xB8" => "\xC3\x98", "\xC3\xB9" => "\xC3\x99", "\xC3\xBA" => "\xC3\x9A",

+ "\xC3\xBB" => "\xC3\x9B", "\xC3\xBC" => "\xC3\x9C", "\xC3\xBD" => "\xC3\x9D", "\xC3\xBE" => "\xC3\x9E",

+ "\xC3\xBF" => "\xC5\xB8", "\xC4\x81" => "\xC4\x80", "\xC4\x83" => "\xC4\x82", "\xC4\x85" => "\xC4\x84",

+ "\xC4\x87" => "\xC4\x86", "\xC4\x89" => "\xC4\x88", "\xC4\x8B" => "\xC4\x8A", "\xC4\x8D" => "\xC4\x8C",

+ "\xC4\x8F" => "\xC4\x8E", "\xC4\x91" => "\xC4\x90", "\xC4\x93" => "\xC4\x92", "\xC4\x97" => "\xC4\x96",

+ "\xC4\x99" => "\xC4\x98", "\xC4\x9B" => "\xC4\x9A", "\xC4\x9D" => "\xC4\x9C", "\xC4\x9F" => "\xC4\x9E",

+ "\xC4\xA1" => "\xC4\xA0", "\xC4\xA3" => "\xC4\xA2", "\xC4\xA5" => "\xC4\xA4", "\xC4\xA7" => "\xC4\xA6",

+ "\xC4\xA9" => "\xC4\xA8", "\xC4\xAB" => "\xC4\xAA", "\xC4\xAF" => "\xC4\xAE", "\xC4\xB5" => "\xC4\xB4",

+ "\xC4\xB7" => "\xC4\xB6", "\xC4\xBA" => "\xC4\xB9", "\xC4\xBC" => "\xC4\xBB", "\xC4\xBE" => "\xC4\xBD",

+ "\xC5\x82" => "\xC5\x81", "\xC5\x84" => "\xC5\x83", "\xC5\x86" => "\xC5\x85", "\xC5\x88" => "\xC5\x87",

+ "\xC5\x8B" => "\xC5\x8A", "\xC5\x8D" => "\xC5\x8C", "\xC5\x91" => "\xC5\x90", "\xC5\x95" => "\xC5\x94",

+ "\xC5\x97" => "\xC5\x96", "\xC5\x99" => "\xC5\x98", "\xC5\x9B" => "\xC5\x9A", "\xC5\x9D" => "\xC5\x9C",

+ "\xC5\x9F" => "\xC5\x9E", "\xC5\xA1" => "\xC5\xA0", "\xC5\xA3" => "\xC5\xA2", "\xC5\xA5" => "\xC5\xA4",

+ "\xC5\xA7" => "\xC5\xA6", "\xC5\xA9" => "\xC5\xA8", "\xC5\xAB" => "\xC5\xAA", "\xC5\xAD" => "\xC5\xAC",

+ "\xC5\xAF" => "\xC5\xAE", "\xC5\xB1" => "\xC5\xB0", "\xC5\xB3" => "\xC5\xB2", "\xC5\xB5" => "\xC5\xB4",

+ "\xC5\xB7" => "\xC5\xB6", "\xC5\xBA" => "\xC5\xB9", "\xC5\xBC" => "\xC5\xBB", "\xC5\xBE" => "\xC5\xBD",

+ "\xC6\xA1" => "\xC6\xA0", "\xC6\xB0" => "\xC6\xAF", "\xC8\x99" => "\xC8\x98", "\xC8\x9B" => "\xC8\x9A",

+ "\xCE\xAC" => "\xCE\x86", "\xCE\xAD" => "\xCE\x88", "\xCE\xAE" => "\xCE\x89", "\xCE\xAF" => "\xCE\x8A",

+ "\xCE\xB1" => "\xCE\x91", "\xCE\xB2" => "\xCE\x92", "\xCE\xB3" => "\xCE\x93", "\xCE\xB4" => "\xCE\x94",

+ "\xCE\xB5" => "\xCE\x95", "\xCE\xB6" => "\xCE\x96", "\xCE\xB7" => "\xCE\x97", "\xCE\xB8" => "\xCE\x98",

+ "\xCE\xB9" => "\xCE\x99", "\xCE\xBA" => "\xCE\x9A", "\xCE\xBB" => "\xCE\x9B", "\xCE\xBC" => "\xCE\x9C",

+ "\xCE\xBD" => "\xCE\x9D", "\xCE\xBE" => "\xCE\x9E", "\xCE\xBF" => "\xCE\x9F", "\xCF\x80" => "\xCE\xA0",

+ "\xCF\x81" => "\xCE\xA1", "\xCF\x83" => "\xCE\xA3", "\xCF\x84" => "\xCE\xA4", "\xCF\x85" => "\xCE\xA5",

+ "\xCF\x86" => "\xCE\xA6", "\xCF\x87" => "\xCE\xA7", "\xCF\x88" => "\xCE\xA8", "\xCF\x89" => "\xCE\xA9",

+ "\xCF\x8A" => "\xCE\xAA", "\xCF\x8B" => "\xCE\xAB", "\xCF\x8C" => "\xCE\x8C", "\xCF\x8D" => "\xCE\x8E",

+ "\xCF\x8E" => "\xCE\x8F", "\xD0\xB0" => "\xD0\x90", "\xD0\xB1" => "\xD0\x91", "\xD0\xB2" => "\xD0\x92",

+ "\xD0\xB3" => "\xD0\x93", "\xD0\xB4" => "\xD0\x94", "\xD0\xB5" => "\xD0\x95", "\xD0\xB6" => "\xD0\x96",

+ "\xD0\xB7" => "\xD0\x97", "\xD0\xB8" => "\xD0\x98", "\xD0\xB9" => "\xD0\x99", "\xD0\xBA" => "\xD0\x9A",

+ "\xD0\xBB" => "\xD0\x9B", "\xD0\xBC" => "\xD0\x9C", "\xD0\xBD" => "\xD0\x9D", "\xD0\xBE" => "\xD0\x9E",

+ "\xD0\xBF" => "\xD0\x9F", "\xD1\x80" => "\xD0\xA0", "\xD1\x81" => "\xD0\xA1", "\xD1\x82" => "\xD0\xA2",

+ "\xD1\x83" => "\xD0\xA3", "\xD1\x84" => "\xD0\xA4", "\xD1\x85" => "\xD0\xA5", "\xD1\x86" => "\xD0\xA6",

+ "\xD1\x87" => "\xD0\xA7", "\xD1\x88" => "\xD0\xA8", "\xD1\x89" => "\xD0\xA9", "\xD1\x8A" => "\xD0\xAA",

+ "\xD1\x8B" => "\xD0\xAB", "\xD1\x8C" => "\xD0\xAC", "\xD1\x8D" => "\xD0\xAD", "\xD1\x8E" => "\xD0\xAE",

+ "\xD1\x8F" => "\xD0\xAF", "\xD1\x91" => "\xD0\x81", "\xD1\x92" => "\xD0\x82", "\xD1\x93" => "\xD0\x83",

+ "\xD1\x94" => "\xD0\x84", "\xD1\x95" => "\xD0\x85", "\xD1\x96" => "\xD0\x86", "\xD1\x97" => "\xD0\x87",

+ "\xD1\x98" => "\xD0\x88", "\xD1\x99" => "\xD0\x89", "\xD1\x9A" => "\xD0\x8A", "\xD1\x9B" => "\xD0\x8B",

+ "\xD1\x9C" => "\xD0\x8C", "\xD1\x9E" => "\xD0\x8E", "\xD1\x9F" => "\xD0\x8F", "\xD2\x91" => "\xD2\x90",

+ "\xE1\xB8\x83" => "\xE1\xB8\x82", "\xE1\xB8\x8B" => "\xE1\xB8\x8A", "\xE1\xB8\x9F" => "\xE1\xB8\x9E", "\xE1\xB9\x81" => "\xE1\xB9\x80",

+ "\xE1\xB9\x97" => "\xE1\xB9\x96", "\xE1\xB9\xA1" => "\xE1\xB9\xA0", "\xE1\xB9\xAB" => "\xE1\xB9\xAA", "\xE1\xBA\x81" => "\xE1\xBA\x80",

+ "\xE1\xBA\x83" => "\xE1\xBA\x82", "\xE1\xBA\x85" => "\xE1\xBA\x84", "\xE1\xBB\xB3" => "\xE1\xBB\xB2"

+ );

+

+ /**

+ * UTF-8 aware alternative to strtolower

+ * Make a string lowercase

+ * Note: The concept of a characters "case" only exists is some alphabets

+ * such as Latin, Greek, Cyrillic, Armenian and archaic Georgian - it does

+ * not exist in the Chinese alphabet, for example. See Unicode Standard

+ * Annex #21: Case Mappings

+ *

+ * @param string

+ * @return string string in lowercase

+ */

+ function utf8_strtolower($string)

+ {

+ global $UTF8_UPPER_TO_LOWER;

+

+ return strtr($string, $UTF8_UPPER_TO_LOWER);

+ }

+

+ /**

+ * UTF-8 aware alternative to strtoupper

+ * Make a string uppercase

+ * Note: The concept of a characters "case" only exists is some alphabets

+ * such as Latin, Greek, Cyrillic, Armenian and archaic Georgian - it does

+ * not exist in the Chinese alphabet, for example. See Unicode Standard

+ * Annex #21: Case Mappings

+ *

+ * @param string

+ * @return string string in uppercase

+ */

+ function utf8_strtoupper($string)

+ {

+ global $UTF8_LOWER_TO_UPPER;

+

+ return strtr($string, $UTF8_LOWER_TO_UPPER);

+ }

+

+ /**

+ * UTF-8 aware alternative to substr

+ * Return part of a string given character offset (and optionally length)

+ *

+ * Note arguments: comparied to substr - if offset or length are

+ * not integers, this version will not complain but rather massages them

+ * into an integer.

+ *

+ * Note on returned values: substr documentation states false can be

+ * returned in some cases (e.g. offset > string length)

+ * mb_substr never returns false, it will return an empty string instead.

+ * This adopts the mb_substr approach

+ *

+ * Note on implementation: PCRE only supports repetitions of less than

+ * 65536, in order to accept up to MAXINT values for offset and length,

+ * we'll repeat a group of 65535 characters when needed.

+ *

+ * Note on implementation: calculating the number of characters in the

+ * string is a relatively expensive operation, so we only carry it out when

+ * necessary. It isn't necessary for +ve offsets and no specified length

+ *

+ * @author Chris Smith<chris@jalakai.co.uk>

+ * @param string

+ * @param integer number of UTF-8 characters offset (from left)

+ * @param integer (optional) length in UTF-8 characters from offset

+ * @return mixed string or FALSE if failure

+ */

+ function utf8_substr($str, $offset, $length = NULL)

+ {

+ // generates E_NOTICE

+ // for PHP4 objects, but not PHP5 objects

+ $str = (string) $str;

+ $offset = (int) $offset;

+ if (!is_null($length))

+ {

+ $length = (int) $length;

+ }

+

+ // handle trivial cases

+ if ($length === 0 || ($offset < 0 && $length < 0 && $length < $offset))

+ {

+ return '';

+ }

+

+ // normalise negative offsets (we could use a tail

+ // anchored pattern, but they are horribly slow!)

+ if ($offset < 0)

+ {

+ // see notes

+ $strlen = utf8_strlen($str);

+ $offset = $strlen + $offset;

+ if ($offset < 0)

+ {

+ $offset = 0;

+ }

+ }

+

+ $op = '';

+ $lp = '';

+

+ // establish a pattern for offset, a

+ // non-captured group equal in length to offset

+ if ($offset > 0)

+ {

+ $ox = (int) ($offset / 65535);

+ $oy = $offset % 65535;

+

+ if ($ox)

+ {

+ $op = '(?:.{65535}){' . $ox . '}';

+ }

+

+ $op = '^(?:' . $op . '.{' . $oy . '})';

+ }

+ else

+ {

+ // offset == 0; just anchor the pattern

+ $op = '^';

+ }

+

+ // establish a pattern for length

+ if (is_null($length))

+ {

+ // the rest of the string

+ $lp = '(.*)$';

+ }

+ else

+ {

+ if (!isset($strlen))

+ {

+ // see notes

+ $strlen = utf8_strlen($str);

+ }

+

+ // another trivial case

+ if ($offset > $strlen)

+ {

+ return '';

+ }

+

+ if ($length > 0)

+ {

+ // reduce any length that would

+ // go passed the end of the string

+ $length = min($strlen - $offset, $length);

+

+ $lx = (int) ($length / 65535);

+ $ly = $length % 65535;

+

+ // negative length requires a captured group

+ // of length characters

+ if ($lx)

+ {

+ $lp = '(?:.{65535}){' . $lx . '}';

+ }

+ $lp = '(' . $lp . '.{'. $ly . '})';

+ }

+ else if ($length < 0)

+ {

+ if ($length < ($offset - $strlen))

+ {

+ return '';

+ }

+

+ $lx = (int)((-$length) / 65535);

+ $ly = (-$length) % 65535;

+

+ // negative length requires ... capture everything

+ // except a group of -length characters

+ // anchored at the tail-end of the string

+ if ($lx)

+ {

+ $lp = '(?:.{65535}){' . $lx . '}';

+ }

+ $lp = '(.*)(?:' . $lp . '.{' . $ly . '})$';

+ }

+ }

+

+ if (!preg_match('#' . $op . $lp . '#us', $str, $match))

+ {

+ return '';

+ }

+

+ return $match[1];

+ }

+

+ /**

+ * Return the length (in characters) of a UTF-8 string

+ *

+ * @param string $text UTF-8 string

+ * @return integer Length (in chars) of given string

+ */

+ function utf8_strlen($text)

+ {

+ // Since utf8_decode is replacing multibyte characters to ? strlen works fine

+ return strlen(utf8_decode($text));

+ }

+}

+

+* UTF-8 aware alternative to str_split

+* Convert a string to an array

+*

+* @author Harry Fuecks

+* @param string UTF-8 encoded

+* @param int number to characters to split string by

+* @return string characters in string reverses

+function utf8_str_split($str, $split_len = 1)

+ if (!preg_match('/^[0-9]+$/', $split_len) || $split_len < 1)

+ return false;

+ $len = utf8_strlen($str);

+ if ($len <= $split_len)

+ return array($str);

+ }

+

+ preg_match_all('/.{' . $split_len . '}|[^\x00]{1,' . $split_len . '}$/us', $str, $ar);

+ return $ar[0];

+}

+

+/**

+* UTF-8 aware alternative to strcspn

+* Find length of initial segment not matching mask

+*

+* @author Harry Fuecks

+* @param string

+* @return int

+*/

+function utf8_strspn($str, $mask, $start = null, $length = null)

+{

+ $mask = preg_replace('!([\\\\\\-\\]\\[/^])!', '\\\${1}', $mask);

+

+ if ($start !== null || $length !== null)

+ {

+ $str = utf8_substr($str, $start, $length);

+ preg_match('/^[' . $mask . ']+/u', $str, $matches);

+

+ if (isset($matches[0]))

+ {

+ return utf8_strlen($matches[0]);

+ }

+

+ return 0;

+}

+

+/**

+* UTF-8 aware alternative to ucfirst

+* Make a string's first character uppercase

+*

+* @author Harry Fuecks

+* @param string

+* @return string with first character as upper case (if applicable)

+*/

+function utf8_ucfirst($str)

+{

+ switch (utf8_strlen($str))

+ {

+ case 0:

+ return '';

+ break;

+

+ case 1:

+ return utf8_strtoupper($str);

+ break;

+

+ default:

+ preg_match('/^(.{1})(.*)$/us', $str, $matches);

+ return utf8_strtoupper($matches[1]) . $matches[2];

+ break;

+ }

+* @param string $encoding Original encoding (lowered)

+ // start with something simple

+ // First, try iconv()

+ // Try the mb_string extension

+ // Try the recode extension

+ // If nothing works, check if we have a custom transcoder available

+ // Make sure the encoding name is alphanumeric, we don't want it to be abused into loading arbitrary files

+ trigger_error('Unknown encoding: ' . $encoding, E_USER_ERROR);

+

+ die('Finish me!! ' . basename(__FILE__) . ' at line ' . __LINE__);

+ return preg_replace_callback('#[\\xC2-\\xF4][\\x80-\\xBF]{1,3}#', 'utf8_encode_ncr_callback', $text);

+ return '&#' . utf8_ord($m[0]) . ';';

+}

+

+/**

+* Enter description here...

+*

+* @param string $chr UTF-8 char

+* @return integer UNICODE code point

+*/

+function utf8_ord($chr)

+{

+ switch (strlen($chr))

+ return ord($chr);

+ break;

+ return ((ord($chr[0]) & 0x1F) << 6) | (ord($chr[1]) & 0x3F);

+ break;

+ return ((ord($chr[0]) & 0x0F) << 12) | ((ord($chr[1]) & 0x3F) << 6) | (ord($chr[2]) & 0x3F);

+ break;

+ return ((ord($chr[0]) & 0x07) << 18) | ((ord($chr[1]) & 0x3F) << 12) | ((ord($chr[2]) & 0x3F) << 6) | (ord($chr[3]) & 0x3F);

+ break;

+ return $chr;

+ }

+}

+

+/**

+* Converts an NCR to a UTF-8 char

+*

+* @param integer $cp UNICODE code point

+* @return string UTF-8 char

+*/

+function utf8_chr($cp)

+{

+ if ($cp > 0xFFFF)

+ {

+ return chr(0xF0 | ($cp >> 18)) . chr(0x80 | (($cp >> 12) & 0x3F)) . chr(0x80 | (($cp >> 6) & 0x3F)) . chr(0x80 | ($cp & 0x3F));

+ }

+ else if ($cp > 0x7FF)

+ {

+ return chr(0xE0 | ($cp >> 12)) . chr(0x80 | (($cp >> 6) & 0x3F)) . chr(0x80 | ($cp & 0x3F));

+ }

+ else if ($cp > 0x7F)

+ {

+ return chr(0xC0 | ($cp >> 6)) . chr(0x80 | ($cp & 0x3F));

+ }

+ else

+ {

+ return chr($cp);

+ }

+ return utf8_chr($cp);

+}

+

+/**

+* Takes an array of ints representing the Unicode characters and returns

+* a UTF-8 string.

+*

+* @param string $text text to be case folded

+* @param string $option determines how we will fold the cases

+* @return string case folded text

+*/

+function utf8_case_fold($text, $option = 'full')

+{

+ static $uniarray = array();

+ global $phpbb_root_path, $phpEx;

+

+ // common is always set

+ if (!isset($uniarray['c']))

+ $uniarray['c'] = include($phpbb_root_path . 'includes/utf/data/case_fold_c.' . $phpEx);

+

+ // only set full if we need to

+ if ($option === 'full' && !isset($uniarray['f']))

+ $uniarray['f'] = include($phpbb_root_path . 'includes/utf/data/case_fold_f.' . $phpEx);

+

+ // only set simple if we need to

+ if ($option !== 'full' && !isset($uniarray['s']))

+ $uniarray['s'] = include($phpbb_root_path . 'includes/utf/data/case_fold_s.' . $phpEx);

+ }

+

+ $text = strtr($text, $uniarray['c']);

+ if ($option === 'full')

+ {

+ $text = strtr($text, $uniarray['f']);

+ $text = strtr($text, $uniarray['s']);

+ return $text;

+}

+

+/**

+* A wrapper function for the normalizer which takes care of including the class if required and modifies the passed strings

+* to be in NFC (Normalization Form Composition).

+*

+* @param mixed $strings Either an array of references to strings, a reference to an array of strings or a reference to a single string

+*/

+function utf8_normalize_nfc($strings)

+{

+ if (!is_array($strings) || (sizeof($strings) > 0))

+ {

+ if (!class_exists('utf_normalizer'))

+ {

+ global $phpbb_root_path, $phpEx;

+ include($phpbb_root_path . 'includes/utf/utf_normalizer.' . $phpEx);

+ }

+

+ if (is_array($strings))

+ {

+ foreach ($strings as $key => $string)

+ {

+ $strings[$key] = utf_normalizer::nfc($strings[$key]);

+ }

+ }

+ else

+ {

+ $strings = utf_normalizer::nfc($strings);

+ }

+ }

+}

+

+/**

+* This function is used to generate a "clean" version of a string.

+* Clean means that it is a case insensitive form (case folding) and that it is normalized (NFC).

+* Additionally a homographs of one character are transformed into one specific character (preferably ASCII

+* if it is an ASCII character).

+*

+* Please be aware that if you change something within this function or within

+* functions used here you need to rebuild/update the username_clean column in the users table. And all other

+* columns that store a clean string otherwise you will break this functionality.

+*

+* @param $text An unclean string, mabye user input (has to be valid UTF-8!)

+* @return Cleaned up version of the input string

+*/

+function utf8_clean_string($text)

+{

+ $text = utf8_case_fold($text);

+

+ if (!class_exists('utf_normalizer'))

+ {

+ global $phpbb_root_path, $phpEx;

+ include($phpbb_root_path . 'includes/utf/utf_normalizer.' . $phpEx);

+ }

+

+ $text = utf_normalizer::nfc($text);

+

+ static $homographs = array(

+ // cyrllic

+ "\xD0\xB0" => "\x61",

+ "\xD0\xB5" => "\x65",

+ "\xD0\xBE" => "\x6F",

+ "\xD1\x80" => "\x70",

+ "\xD1\x81" => "\x63",

+ "\xD1\x83" => "\x79",

+ "\xD1\x85" => "\x78",

+ "\xD1\x95" => "\x73",

+ "\xD1\x96" => "\x69",

+ "\xD1\x98" => "\x6A",

+ "\xD2\xBB" => "\x68",

+ // greek

+ "\xCE\xB1" => "\x61",

+ "\xCE\xBF" => "\x6F",

+ // other

+ "\xC2\xA1" => "\x69",

+ );

+

+ $text = strtr($text, $homographs);

+

+ return $text;