From 3413fe6bfd423d45d083ff4ed62c0f72c2cc272f Mon Sep 17 00:00:00 2001 From: Bharat Mediratta Date: Thu, 28 May 2009 06:11:53 +0800 Subject: Rename 'kohana' to 'system' to conform to the Kohana filesystem layout. I'm comfortable with us not clearly drawing the distinction about the fact that it's Kohana. Signed-off-by: Gallery Role Account --- system/core/utf8/from_unicode.php | 68 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 68 insertions(+) create mode 100644 system/core/utf8/from_unicode.php (limited to 'system/core/utf8/from_unicode.php') diff --git a/system/core/utf8/from_unicode.php b/system/core/utf8/from_unicode.php new file mode 100644 index 00000000..66c6742d --- /dev/null +++ b/system/core/utf8/from_unicode.php @@ -0,0 +1,68 @@ += 0) AND ($arr[$k] <= 0x007f)) + { + echo chr($arr[$k]); + } + // 2 byte sequence + elseif ($arr[$k] <= 0x07ff) + { + echo chr(0xc0 | ($arr[$k] >> 6)); + echo chr(0x80 | ($arr[$k] & 0x003f)); + } + // Byte order mark (skip) + elseif ($arr[$k] == 0xFEFF) + { + // nop -- zap the BOM + } + // Test for illegal surrogates + elseif ($arr[$k] >= 0xD800 AND $arr[$k] <= 0xDFFF) + { + // Found a surrogate + trigger_error('utf8::from_unicode: Illegal surrogate at index: '.$k.', value: '.$arr[$k], E_USER_WARNING); + return FALSE; + } + // 3 byte sequence + elseif ($arr[$k] <= 0xffff) + { + echo chr(0xe0 | ($arr[$k] >> 12)); + echo chr(0x80 | (($arr[$k] >> 6) & 0x003f)); + echo chr(0x80 | ($arr[$k] & 0x003f)); + } + // 4 byte sequence + elseif ($arr[$k] <= 0x10ffff) + { + echo chr(0xf0 | ($arr[$k] >> 18)); + echo chr(0x80 | (($arr[$k] >> 12) & 0x3f)); + echo chr(0x80 | (($arr[$k] >> 6) & 0x3f)); + echo chr(0x80 | ($arr[$k] & 0x3f)); + } + // Out of range + else + { + trigger_error('utf8::from_unicode: Codepoint out of Unicode range at index: '.$k.', value: '.$arr[$k], E_USER_WARNING); + return FALSE; + } + } + + $result = ob_get_contents(); + ob_end_clean(); + return $result; +} -- cgit v1.2.3