1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/gfx/graphite2/src/inc/UtfCodec.h Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,207 @@
1.4 +/* GRAPHITE2 LICENSING
1.5 +
1.6 + Copyright 2011, SIL International
1.7 + All rights reserved.
1.8 +
1.9 + This library is free software; you can redistribute it and/or modify
1.10 + it under the terms of the GNU Lesser General Public License as published
1.11 + by the Free Software Foundation; either version 2.1 of License, or
1.12 + (at your option) any later version.
1.13 +
1.14 + This program is distributed in the hope that it will be useful,
1.15 + but WITHOUT ANY WARRANTY; without even the implied warranty of
1.16 + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
1.17 + Lesser General Public License for more details.
1.18 +
1.19 + You should also have received a copy of the GNU Lesser General Public
1.20 + License along with this library in the file named "LICENSE".
1.21 + If not, write to the Free Software Foundation, 51 Franklin Street,
1.22 + Suite 500, Boston, MA 02110-1335, USA or visit their web page on the
1.23 + internet at http://www.fsf.org/licenses/lgpl.html.
1.24 +
1.25 +Alternatively, the contents of this file may be used under the terms of the
1.26 +Mozilla Public License (http://mozilla.org/MPL) or the GNU General Public
1.27 +License, as published by the Free Software Foundation, either version 2
1.28 +of the License or (at your option) any later version.
1.29 +*/
1.30 +#pragma once
1.31 +
1.32 +#include <cstdlib>
1.33 +#include "inc/Main.h"
1.34 +
1.35 +namespace graphite2 {
1.36 +
1.37 +typedef uint32 uchar_t;
1.38 +
1.39 +template <int N>
1.40 +struct _utf_codec
1.41 +{
1.42 + typedef uchar_t codeunit_t;
1.43 +
1.44 + static void put(codeunit_t * cp, const uchar_t , int8 & len) throw();
1.45 + static uchar_t get(const codeunit_t * cp, int8 & len) throw();
1.46 +};
1.47 +
1.48 +
1.49 +template <>
1.50 +struct _utf_codec<32>
1.51 +{
1.52 +private:
1.53 + static const uchar_t limit = 0x110000;
1.54 +public:
1.55 + typedef uint32 codeunit_t;
1.56 +
1.57 + inline
1.58 + static void put(codeunit_t * cp, const uchar_t usv, int8 & l) throw()
1.59 + {
1.60 + *cp = usv; l = 1;
1.61 + }
1.62 +
1.63 + inline
1.64 + static uchar_t get(const codeunit_t * cp, int8 & l) throw()
1.65 + {
1.66 + if (cp[0] < limit) { l = 1; return cp[0]; }
1.67 + else { l = -1; return 0xFFFD; }
1.68 + }
1.69 +};
1.70 +
1.71 +
1.72 +template <>
1.73 +struct _utf_codec<16>
1.74 +{
1.75 +private:
1.76 + static const int32 lead_offset = 0xD800 - (0x10000 >> 10);
1.77 + static const int32 surrogate_offset = 0x10000 - (0xD800 << 10) - 0xDC00;
1.78 +public:
1.79 + typedef uint16 codeunit_t;
1.80 +
1.81 + inline
1.82 + static void put(codeunit_t * cp, const uchar_t usv, int8 & l) throw()
1.83 + {
1.84 + if (usv < 0x10000) { l = 1; cp[0] = codeunit_t(usv); }
1.85 + else
1.86 + {
1.87 + cp[0] = codeunit_t(lead_offset + (usv >> 10));
1.88 + cp[1] = codeunit_t(0xDC00 + (usv & 0x3FF));
1.89 + l = 2;
1.90 + }
1.91 + }
1.92 +
1.93 + inline
1.94 + static uchar_t get(const codeunit_t * cp, int8 & l) throw()
1.95 + {
1.96 + const uint32 uh = cp[0];
1.97 + l = 1;
1.98 +
1.99 + if (0xD800 > uh || uh > 0xDFFF) { return uh; }
1.100 + const uint32 ul = cp[1];
1.101 + if (uh > 0xDBFF || 0xDC00 > ul || ul > 0xDFFF) { l = -1; return 0xFFFD; }
1.102 + ++l;
1.103 + return (uh<<10) + ul + surrogate_offset;
1.104 + }
1.105 +};
1.106 +
1.107 +
1.108 +template <>
1.109 +struct _utf_codec<8>
1.110 +{
1.111 +private:
1.112 + static const int8 sz_lut[16];
1.113 + static const byte mask_lut[5];
1.114 +
1.115 +
1.116 +public:
1.117 + typedef uint8 codeunit_t;
1.118 +
1.119 + inline
1.120 + static void put(codeunit_t * cp, const uchar_t usv, int8 & l) throw()
1.121 + {
1.122 + if (usv < 0x80) {l = 1; cp[0] = usv; return; }
1.123 + if (usv < 0x0800) {l = 2; cp[0] = 0xC0 + (usv >> 6); cp[1] = 0x80 + (usv & 0x3F); return; }
1.124 + if (usv < 0x10000) {l = 3; cp[0] = 0xE0 + (usv >> 12); cp[1] = 0x80 + ((usv >> 6) & 0x3F); cp[2] = 0x80 + (usv & 0x3F); return; }
1.125 + else {l = 4; cp[0] = 0xF0 + (usv >> 18); cp[1] = 0x80 + ((usv >> 12) & 0x3F); cp[2] = 0x80 + ((usv >> 6) & 0x3F); cp[3] = 0x80 + (usv & 0x3F); return; }
1.126 + }
1.127 +
1.128 + inline
1.129 + static uchar_t get(const codeunit_t * cp, int8 & l) throw()
1.130 + {
1.131 + const int8 seq_sz = sz_lut[*cp >> 4];
1.132 + uchar_t u = *cp & mask_lut[seq_sz];
1.133 + l = 1;
1.134 + bool toolong = false;
1.135 +
1.136 + switch(seq_sz) {
1.137 + case 4: u <<= 6; u |= *++cp & 0x3F; if (*cp >> 6 != 2) break; ++l; toolong = (u < 0x10); // no break
1.138 + case 3: u <<= 6; u |= *++cp & 0x3F; if (*cp >> 6 != 2) break; ++l; toolong |= (u < 0x20); // no break
1.139 + case 2: u <<= 6; u |= *++cp & 0x3F; if (*cp >> 6 != 2) break; ++l; toolong |= (u < 0x80); // no break
1.140 + case 1: break;
1.141 + case 0: l = -1; return 0xFFFD;
1.142 + }
1.143 +
1.144 + if (l != seq_sz || toolong)
1.145 + {
1.146 + l = -l;
1.147 + return 0xFFFD;
1.148 + }
1.149 + return u;
1.150 + }
1.151 +};
1.152 +
1.153 +
1.154 +template <typename C>
1.155 +class _utf_iterator
1.156 +{
1.157 + typedef _utf_codec<sizeof(C)*8> codec;
1.158 +
1.159 + C * cp;
1.160 + mutable int8 sl;
1.161 +
1.162 +public:
1.163 + typedef C codeunit_type;
1.164 + typedef uchar_t value_type;
1.165 + typedef uchar_t * pointer;
1.166 +
1.167 + class reference
1.168 + {
1.169 + const _utf_iterator & _i;
1.170 +
1.171 + reference(const _utf_iterator & i): _i(i) {}
1.172 + public:
1.173 + operator value_type () const throw () { return codec::get(_i.cp, _i.sl); }
1.174 + reference & operator = (const value_type usv) throw() { codec::put(_i.cp, usv, _i.sl); return *this; }
1.175 +
1.176 + friend class _utf_iterator;
1.177 + };
1.178 +
1.179 +
1.180 + _utf_iterator(const void * us=0) : cp(reinterpret_cast<C *>(const_cast<void *>(us))), sl(1) { }
1.181 +
1.182 + _utf_iterator & operator ++ () { cp += abs(sl); return *this; }
1.183 + _utf_iterator operator ++ (int) { _utf_iterator tmp(*this); operator++(); return tmp; }
1.184 +
1.185 + bool operator == (const _utf_iterator & rhs) const throw() { return cp >= rhs.cp; }
1.186 + bool operator != (const _utf_iterator & rhs) const throw() { return !operator==(rhs); }
1.187 +
1.188 + reference operator * () const throw() { return *this; }
1.189 + pointer operator ->() const throw() { return &operator *(); }
1.190 +
1.191 + operator codeunit_type * () const throw() { return cp; }
1.192 +
1.193 + bool error() const throw() { return sl < 1; }
1.194 +};
1.195 +
1.196 +template <typename C>
1.197 +struct utf
1.198 +{
1.199 + typedef typename _utf_codec<sizeof(C)*8>::codeunit_t codeunit_t;
1.200 +
1.201 + typedef _utf_iterator<C> iterator;
1.202 + typedef _utf_iterator<const C> const_iterator;
1.203 +};
1.204 +
1.205 +
1.206 +typedef utf<uint32> utf32;
1.207 +typedef utf<uint16> utf16;
1.208 +typedef utf<uint8> utf8;
1.209 +
1.210 +} // namespace graphite2
