|
| 1 | +#ifndef CTYPE_UCA_1400_H |
| 2 | +#define CTYPE_UCA_1400_H |
| 3 | +/* Copyright (c) 2021, MariaDB |
| 4 | +
|
| 5 | + This library is free software; you can redistribute it and/or |
| 6 | + modify it under the terms of the GNU Library General Public |
| 7 | + License as published by the Free Software Foundation; version 2 |
| 8 | + of the License. |
| 9 | +
|
| 10 | + This library is distributed in the hope that it will be useful, |
| 11 | + but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 12 | + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| 13 | + Library General Public License for more details. |
| 14 | +
|
| 15 | + You should have received a copy of the GNU Library General Public |
| 16 | + License along with this library; if not, write to the Free |
| 17 | + Software Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, |
| 18 | + MA 02110-1335 USA */ |
| 19 | + |
| 20 | + |
| 21 | +/* |
| 22 | + 17000..187FF; Tangut [6144] |
| 23 | + 18800..18AFF; Tangut Components [768] |
| 24 | + 18D00..18D7F; Tangut Supplement [128] |
| 25 | +*/ |
| 26 | +static inline my_bool |
| 27 | +my_uca_1400_is_assigned_tangut(my_wc_t code) |
| 28 | +{ |
| 29 | + return (code >= 0x17000 && code <= 0x187FF) || |
| 30 | + (code >= 0x18800 && code <= 0x18AFF) || |
| 31 | + (code >= 0x18D00 && code <= 0x18D7F); |
| 32 | +} |
| 33 | + |
| 34 | +static inline MY_UCA_IMPLICIT_WEIGHT |
| 35 | +my_uca_1400_implicit_weight_primary_tangut(my_wc_t code) |
| 36 | +{ |
| 37 | + MY_UCA_IMPLICIT_WEIGHT res; |
| 38 | + res.weight[0]= 0xFB00; |
| 39 | + res.weight[1]= (uint16) (code - 0x17000) | 0x8000; |
| 40 | + return res; |
| 41 | +} |
| 42 | + |
| 43 | + |
| 44 | +/* |
| 45 | + 1B170..1B2FF; Nushu [400] |
| 46 | +*/ |
| 47 | +static inline my_bool |
| 48 | +my_uca_1400_is_assigned_nushu(my_wc_t code) |
| 49 | +{ |
| 50 | + return code >= 0x1B170 && code <= 0x1B2FF; |
| 51 | +} |
| 52 | + |
| 53 | +static inline MY_UCA_IMPLICIT_WEIGHT |
| 54 | +my_uca_1400_implicit_weight_primary_nushu(my_wc_t code) |
| 55 | +{ |
| 56 | + MY_UCA_IMPLICIT_WEIGHT res; |
| 57 | + res.weight[0]= 0xFB01; |
| 58 | + res.weight[1]= (uint16) (code - 0x1B170) | 0x8000; |
| 59 | + return res; |
| 60 | +} |
| 61 | + |
| 62 | + |
| 63 | +/* |
| 64 | + 18B00..18CFF; Khitan Small Script [512] |
| 65 | +*/ |
| 66 | +static inline my_bool |
| 67 | +my_uca_1400_is_assigned_khitan_small_script(my_wc_t code) |
| 68 | +{ |
| 69 | + return code >= 0x18B00 && code <= 0x18CFF; |
| 70 | +} |
| 71 | + |
| 72 | +static inline MY_UCA_IMPLICIT_WEIGHT |
| 73 | +my_uca_1400_implicit_weight_primary_khitan(my_wc_t code) |
| 74 | +{ |
| 75 | + MY_UCA_IMPLICIT_WEIGHT res; |
| 76 | + res.weight[0]= 0xFB02; |
| 77 | + res.weight[1]= (uint16) (code - 0x18B00) | 0x8000; |
| 78 | + return res; |
| 79 | +} |
| 80 | + |
| 81 | + |
| 82 | +/* |
| 83 | + Unified_Ideograph=True AND |
| 84 | + ((Block=CJK_Unified_Ideograph) OR (Block=CJK_Compatibility_Ideographs)) |
| 85 | +
|
| 86 | + https://www.unicode.org/Public/14.0.0/ucd/Blocks.txt |
| 87 | +
|
| 88 | + 4E00..9FFF; CJK Unified Ideographs |
| 89 | + F900..FAFF; CJK Compatibility Ideographs |
| 90 | +
|
| 91 | + https://www.unicode.org/Public/14.0.0/ucd/PropList.txt |
| 92 | +
|
| 93 | + 4E00..9FFF ; Unified_Ideograph # Lo [20992] CJK UNIFIED IDEOGRAPH-4E00..CJK UNIFIED IDEOGRAPH-9FFF |
| 94 | + FA0E..FA0F ; Unified_Ideograph # Lo [2] CJK COMPATIBILITY IDEOGRAPH-FA0E..CJK COMPATIBILITY IDEOGRAPH-FA0F |
| 95 | + FA11 ; Unified_Ideograph # Lo CJK COMPATIBILITY IDEOGRAPH-FA11 |
| 96 | + FA13..FA14 ; Unified_Ideograph # Lo [2] CJK COMPATIBILITY IDEOGRAPH-FA13..CJK COMPATIBILITY IDEOGRAPH-FA14 |
| 97 | + FA1F ; Unified_Ideograph # Lo CJK COMPATIBILITY IDEOGRAPH-FA1F |
| 98 | + FA21 ; Unified_Ideograph # Lo CJK COMPATIBILITY IDEOGRAPH-FA21 |
| 99 | + FA23..FA24 ; Unified_Ideograph # Lo [2] CJK COMPATIBILITY IDEOGRAPH-FA23..CJK COMPATIBILITY IDEOGRAPH-FA24 |
| 100 | + FA27..FA29 ; Unified_Ideograph # Lo [3] CJK COMPATIBILITY IDEOGRAPH-FA27..CJK COMPATIBILITY IDEOGRAPH-FA29 |
| 101 | +*/ |
| 102 | +static inline my_bool |
| 103 | +my_uca_1400_is_core_han_unified_ideograph(my_wc_t code) |
| 104 | +{ |
| 105 | + return (code >= 0x4E00 && code <= 0x9FFF) || |
| 106 | + (code >= 0xFA0E && code <= 0xFA0F) || |
| 107 | + (code == 0xFA11) || |
| 108 | + (code >= 0xFA13 && code <= 0xFA14) || |
| 109 | + (code == 0xFA1F) || |
| 110 | + (code == 0xFA21) || |
| 111 | + (code >= 0xFA23 && code <= 0xFA24) || |
| 112 | + (code >= 0xFA27 && code <= 0xFA29); |
| 113 | +} |
| 114 | + |
| 115 | + |
| 116 | +/* |
| 117 | + (Unified_Ideograph=True AND NOT |
| 118 | + ((Block=CJK_Unified_Ideograph) OR (Block=CJK_Compatibility_Ideographs)) |
| 119 | +
|
| 120 | + https://www.unicode.org/Public/14.0.0/ucd/Blocks.txt |
| 121 | +
|
| 122 | + 3400..4DBF; CJK Unified Ideographs Extension A |
| 123 | + 20000..2A6DF; CJK Unified Ideographs Extension B |
| 124 | + 2A700..2B73F; CJK Unified Ideographs Extension C |
| 125 | + 2B740..2B81F; CJK Unified Ideographs Extension D |
| 126 | + 2B820..2CEAF; CJK Unified Ideographs Extension E |
| 127 | + 2CEB0..2EBEF; CJK Unified Ideographs Extension F |
| 128 | + 30000..3134F; CJK Unified Ideographs Extension G |
| 129 | +
|
| 130 | + https://www.unicode.org/Public/14.0.0/ucd/PropList.txt |
| 131 | +
|
| 132 | + 3400..4DBF ; Unified_Ideograph # Lo [6592] CJK UNIFIED IDEOGRAPH-3400..CJK UNIFIED IDEOGRAPH-4DBF |
| 133 | + 20000..2A6DF ; Unified_Ideograph # Lo [42720] CJK UNIFIED IDEOGRAPH-20000..CJK UNIFIED IDEOGRAPH-2A6DF |
| 134 | + 2A700..2B738 ; Unified_Ideograph # Lo [4153] CJK UNIFIED IDEOGRAPH-2A700..CJK UNIFIED IDEOGRAPH-2B738 |
| 135 | + 2B740..2B81D ; Unified_Ideograph # Lo [222] CJK UNIFIED IDEOGRAPH-2B740..CJK UNIFIED IDEOGRAPH-2B81D |
| 136 | + 2B820..2CEA1 ; Unified_Ideograph # Lo [5762] CJK UNIFIED IDEOGRAPH-2B820..CJK UNIFIED IDEOGRAPH-2CEA1 |
| 137 | + 2CEB0..2EBE0 ; Unified_Ideograph # Lo [7473] CJK UNIFIED IDEOGRAPH-2CEB0..CJK UNIFIED IDEOGRAPH-2EBE0 |
| 138 | + 30000..3134A ; Unified_Ideograph # Lo [4939] CJK UNIFIED IDEOGRAPH-30000..CJK UNIFIED IDEOGRAPH-3134A |
| 139 | +*/ |
| 140 | +static inline my_bool |
| 141 | +my_uca_1400_is_other_han_unified_ideograph(my_wc_t code) |
| 142 | +{ |
| 143 | + return (code >= 0x3400 && code <= 0x4DBF) || |
| 144 | + (code >= 0x20000 && code <= 0x2A6DF) || |
| 145 | + (code >= 0x2A700 && code <= 0x2B738) || |
| 146 | + (code >= 0x2B740 && code <= 0x2B81D) || |
| 147 | + (code >= 0x2B820 && code <= 0x2CEA1) || |
| 148 | + (code >= 0x2CEB0 && code <= 0x2EBE0) || |
| 149 | + (code >= 0x30000 && code <= 0x3134A); |
| 150 | +} |
| 151 | + |
| 152 | + |
| 153 | +/* |
| 154 | + See section "Computing Implicit Weights" in |
| 155 | + https://unicode.org/reports/tr10/#Values_For_Base_Table |
| 156 | +*/ |
| 157 | +static inline MY_UCA_IMPLICIT_WEIGHT |
| 158 | +my_uca_1400_implicit_weight_primary(my_wc_t code) |
| 159 | +{ |
| 160 | + if (my_uca_1400_is_core_han_unified_ideograph(code)) |
| 161 | + return my_uca_implicit_weight_primary_default(0xFB40, code); |
| 162 | + |
| 163 | + if (my_uca_1400_is_other_han_unified_ideograph(code)) |
| 164 | + return my_uca_implicit_weight_primary_default(0xFB80, code); |
| 165 | + |
| 166 | + if (my_uca_1400_is_assigned_tangut(code)) |
| 167 | + return my_uca_1400_implicit_weight_primary_tangut(code); |
| 168 | + |
| 169 | + if (my_uca_1400_is_assigned_nushu(code)) |
| 170 | + return my_uca_1400_implicit_weight_primary_nushu(code); |
| 171 | + |
| 172 | + if (my_uca_1400_is_assigned_khitan_small_script(code)) |
| 173 | + return my_uca_1400_implicit_weight_primary_khitan(code); |
| 174 | + |
| 175 | + /* Unassigned - Any other code point */ |
| 176 | + return my_uca_implicit_weight_primary_default(0xFBC0, code); |
| 177 | +} |
| 178 | + |
| 179 | +#endif /* CTYPE_UCA_1400_H */ |
0 commit comments