1 | /* |
2 | * Copyright © 2012 Google, Inc. |
3 | * |
4 | * This is part of HarfBuzz, a text shaping library. |
5 | * |
6 | * Permission is hereby granted, without written agreement and without |
7 | * license or royalty fees, to use, copy, modify, and distribute this |
8 | * software and its documentation for any purpose, provided that the |
9 | * above copyright notice and the following two paragraphs appear in |
10 | * all copies of this software. |
11 | * |
12 | * IN NO EVENT SHALL THE COPYRIGHT HOLDER BE LIABLE TO ANY PARTY FOR |
13 | * DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES |
14 | * ARISING OUT OF THE USE OF THIS SOFTWARE AND ITS DOCUMENTATION, EVEN |
15 | * IF THE COPYRIGHT HOLDER HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH |
16 | * DAMAGE. |
17 | * |
18 | * THE COPYRIGHT HOLDER SPECIFICALLY DISCLAIMS ANY WARRANTIES, INCLUDING, |
19 | * BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND |
20 | * FITNESS FOR A PARTICULAR PURPOSE. THE SOFTWARE PROVIDED HEREUNDER IS |
21 | * ON AN "AS IS" BASIS, AND THE COPYRIGHT HOLDER HAS NO OBLIGATION TO |
22 | * PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS. |
23 | * |
24 | * Google Author(s): Behdad Esfahbod |
25 | */ |
26 | |
27 | #ifndef HB_OT_SHAPE_COMPLEX_INDIC_PRIVATE_HH |
28 | #define HB_OT_SHAPE_COMPLEX_INDIC_PRIVATE_HH |
29 | |
30 | #include "hb-private.hh" |
31 | |
32 | |
33 | #include "hb-ot-shape-complex-private.hh" |
34 | #include "hb-ot-shape-private.hh" /* XXX Remove */ |
35 | |
36 | |
37 | #define INDIC_TABLE_ELEMENT_TYPE uint16_t |
38 | |
39 | /* Cateories used in the OpenType spec: |
40 | * https://www.microsoft.com/typography/otfntdev/devanot/shaping.aspx |
41 | */ |
42 | /* Note: This enum is duplicated in the -machine.rl source file. |
43 | * Not sure how to avoid duplication. */ |
44 | enum indic_category_t { |
45 | OT_X = 0, |
46 | OT_C = 1, |
47 | OT_V = 2, |
48 | OT_N = 3, |
49 | OT_H = 4, |
50 | OT_ZWNJ = 5, |
51 | OT_ZWJ = 6, |
52 | OT_M = 7, |
53 | OT_SM = 8, |
54 | OT_VD = 9, |
55 | OT_A = 10, |
56 | OT_PLACEHOLDER = 11, |
57 | OT_DOTTEDCIRCLE = 12, |
58 | OT_RS = 13, /* Register Shifter, used in Khmer OT spec. */ |
59 | OT_Coeng = 14, /* Khmer-style Virama. */ |
60 | OT_Repha = 15, /* Atomically-encoded logical or visual repha. */ |
61 | OT_Ra = 16, |
62 | OT_CM = 17, /* Consonant-Medial. */ |
63 | OT_Symbol = 18, /* Avagraha, etc that take marks (SM,A,VD). */ |
64 | OT_CS = 19 |
65 | }; |
66 | |
67 | #define MEDIAL_FLAGS (FLAG (OT_CM)) |
68 | |
69 | /* Note: |
70 | * |
71 | * We treat Vowels and placeholders as if they were consonants. This is safe because Vowels |
72 | * cannot happen in a consonant syllable. The plus side however is, we can call the |
73 | * consonant syllable logic from the vowel syllable function and get it all right! */ |
74 | #define CONSONANT_FLAGS (FLAG (OT_C) | FLAG (OT_CS) | FLAG (OT_Ra) | MEDIAL_FLAGS | FLAG (OT_V) | FLAG (OT_PLACEHOLDER) | FLAG (OT_DOTTEDCIRCLE)) |
75 | #define JOINER_FLAGS (FLAG (OT_ZWJ) | FLAG (OT_ZWNJ)) |
76 | #define HALANT_OR_COENG_FLAGS (FLAG (OT_H) | FLAG (OT_Coeng)) |
77 | |
78 | |
79 | /* Visual positions in a syllable from left to right. */ |
80 | enum indic_position_t { |
81 | POS_START, |
82 | |
83 | POS_RA_TO_BECOME_REPH, |
84 | POS_PRE_M, |
85 | POS_PRE_C, |
86 | |
87 | POS_BASE_C, |
88 | POS_AFTER_MAIN, |
89 | |
90 | POS_ABOVE_C, |
91 | |
92 | POS_BEFORE_SUB, |
93 | POS_BELOW_C, |
94 | POS_AFTER_SUB, |
95 | |
96 | POS_BEFORE_POST, |
97 | POS_POST_C, |
98 | POS_AFTER_POST, |
99 | |
100 | POS_FINAL_C, |
101 | POS_SMVD, |
102 | |
103 | POS_END |
104 | }; |
105 | |
106 | /* Categories used in IndicSyllabicCategory.txt from UCD. */ |
107 | enum indic_syllabic_category_t { |
108 | INDIC_SYLLABIC_CATEGORY_OTHER = OT_X, |
109 | |
110 | INDIC_SYLLABIC_CATEGORY_AVAGRAHA = OT_Symbol, |
111 | INDIC_SYLLABIC_CATEGORY_BINDU = OT_SM, |
112 | INDIC_SYLLABIC_CATEGORY_BRAHMI_JOINING_NUMBER = OT_PLACEHOLDER, /* Don't care. */ |
113 | INDIC_SYLLABIC_CATEGORY_CANTILLATION_MARK = OT_A, |
114 | INDIC_SYLLABIC_CATEGORY_CONSONANT = OT_C, |
115 | INDIC_SYLLABIC_CATEGORY_CONSONANT_DEAD = OT_C, |
116 | INDIC_SYLLABIC_CATEGORY_CONSONANT_FINAL = OT_CM, |
117 | INDIC_SYLLABIC_CATEGORY_CONSONANT_HEAD_LETTER = OT_C, |
118 | INDIC_SYLLABIC_CATEGORY_CONSONANT_KILLER = OT_M, /* U+17CD only. */ |
119 | INDIC_SYLLABIC_CATEGORY_CONSONANT_MEDIAL = OT_CM, |
120 | INDIC_SYLLABIC_CATEGORY_CONSONANT_PLACEHOLDER = OT_PLACEHOLDER, |
121 | INDIC_SYLLABIC_CATEGORY_CONSONANT_PRECEDING_REPHA = OT_Repha, |
122 | INDIC_SYLLABIC_CATEGORY_CONSONANT_PREFIXED = OT_X, /* Don't care. */ |
123 | INDIC_SYLLABIC_CATEGORY_CONSONANT_SUBJOINED = OT_CM, |
124 | INDIC_SYLLABIC_CATEGORY_CONSONANT_SUCCEEDING_REPHA = OT_N, |
125 | INDIC_SYLLABIC_CATEGORY_CONSONANT_WITH_STACKER = OT_CS, |
126 | INDIC_SYLLABIC_CATEGORY_GEMINATION_MARK = OT_SM, /* https://github.com/harfbuzz/harfbuzz/issues/552 */ |
127 | INDIC_SYLLABIC_CATEGORY_INVISIBLE_STACKER = OT_Coeng, |
128 | INDIC_SYLLABIC_CATEGORY_JOINER = OT_ZWJ, |
129 | INDIC_SYLLABIC_CATEGORY_MODIFYING_LETTER = OT_X, |
130 | INDIC_SYLLABIC_CATEGORY_NON_JOINER = OT_ZWNJ, |
131 | INDIC_SYLLABIC_CATEGORY_NUKTA = OT_N, |
132 | INDIC_SYLLABIC_CATEGORY_NUMBER = OT_PLACEHOLDER, |
133 | INDIC_SYLLABIC_CATEGORY_NUMBER_JOINER = OT_PLACEHOLDER, /* Don't care. */ |
134 | INDIC_SYLLABIC_CATEGORY_PURE_KILLER = OT_M, /* Is like a vowel matra. */ |
135 | INDIC_SYLLABIC_CATEGORY_REGISTER_SHIFTER = OT_RS, |
136 | INDIC_SYLLABIC_CATEGORY_SYLLABLE_MODIFIER = OT_SM, |
137 | INDIC_SYLLABIC_CATEGORY_TONE_LETTER = OT_X, |
138 | INDIC_SYLLABIC_CATEGORY_TONE_MARK = OT_N, |
139 | INDIC_SYLLABIC_CATEGORY_VIRAMA = OT_H, |
140 | INDIC_SYLLABIC_CATEGORY_VISARGA = OT_SM, |
141 | INDIC_SYLLABIC_CATEGORY_VOWEL = OT_V, |
142 | INDIC_SYLLABIC_CATEGORY_VOWEL_DEPENDENT = OT_M, |
143 | INDIC_SYLLABIC_CATEGORY_VOWEL_INDEPENDENT = OT_V |
144 | }; |
145 | |
146 | /* Categories used in IndicSMatraCategory.txt from UCD */ |
147 | enum indic_matra_category_t { |
148 | INDIC_MATRA_CATEGORY_NOT_APPLICABLE = POS_END, |
149 | |
150 | INDIC_MATRA_CATEGORY_LEFT = POS_PRE_C, |
151 | INDIC_MATRA_CATEGORY_TOP = POS_ABOVE_C, |
152 | INDIC_MATRA_CATEGORY_BOTTOM = POS_BELOW_C, |
153 | INDIC_MATRA_CATEGORY_RIGHT = POS_POST_C, |
154 | |
155 | /* These should resolve to the position of the last part of the split sequence. */ |
156 | INDIC_MATRA_CATEGORY_BOTTOM_AND_RIGHT = INDIC_MATRA_CATEGORY_RIGHT, |
157 | INDIC_MATRA_CATEGORY_LEFT_AND_RIGHT = INDIC_MATRA_CATEGORY_RIGHT, |
158 | INDIC_MATRA_CATEGORY_TOP_AND_BOTTOM = INDIC_MATRA_CATEGORY_BOTTOM, |
159 | INDIC_MATRA_CATEGORY_TOP_AND_BOTTOM_AND_RIGHT = INDIC_MATRA_CATEGORY_RIGHT, |
160 | INDIC_MATRA_CATEGORY_TOP_AND_LEFT = INDIC_MATRA_CATEGORY_TOP, |
161 | INDIC_MATRA_CATEGORY_TOP_AND_LEFT_AND_RIGHT = INDIC_MATRA_CATEGORY_RIGHT, |
162 | INDIC_MATRA_CATEGORY_TOP_AND_RIGHT = INDIC_MATRA_CATEGORY_RIGHT, |
163 | |
164 | INDIC_MATRA_CATEGORY_OVERSTRUCK = POS_AFTER_MAIN, |
165 | INDIC_MATRA_CATEGORY_VISUAL_ORDER_LEFT = POS_PRE_M |
166 | }; |
167 | |
168 | #define INDIC_COMBINE_CATEGORIES(S,M) \ |
169 | ( \ |
170 | ASSERT_STATIC_EXPR_ZERO (S < 255 && M < 255) + \ |
171 | ( S | \ |
172 | ( \ |
173 | ( \ |
174 | S == INDIC_SYLLABIC_CATEGORY_CONSONANT_MEDIAL || \ |
175 | S == INDIC_SYLLABIC_CATEGORY_GEMINATION_MARK || \ |
176 | S == INDIC_SYLLABIC_CATEGORY_REGISTER_SHIFTER || \ |
177 | S == INDIC_SYLLABIC_CATEGORY_CONSONANT_SUCCEEDING_REPHA || \ |
178 | S == INDIC_SYLLABIC_CATEGORY_VIRAMA || \ |
179 | S == INDIC_SYLLABIC_CATEGORY_VOWEL_DEPENDENT || \ |
180 | false \ |
181 | ? M : INDIC_MATRA_CATEGORY_NOT_APPLICABLE \ |
182 | ) << 8 \ |
183 | ) \ |
184 | ) \ |
185 | ) |
186 | |
187 | HB_INTERNAL INDIC_TABLE_ELEMENT_TYPE |
188 | hb_indic_get_categories (hb_codepoint_t u); |
189 | |
190 | #endif /* HB_OT_SHAPE_COMPLEX_INDIC_PRIVATE_HH */ |
191 | |