1/* 2 * Copyright © 2011,2012 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_MACHINE_HH 28#define HB_OT_SHAPE_COMPLEX_INDIC_MACHINE_HH 29 30#include "hb.hh" 31 32%%{ 33 machine indic_syllable_machine; 34 alphtype unsigned char; 35 write data; 36}%% 37 38%%{ 39 40# Same order as enum indic_category_t. Not sure how to avoid duplication. 41C = 1; 42V = 2; 43N = 3; 44H = 4; 45ZWNJ = 5; 46ZWJ = 6; 47M = 7; 48SM = 8; 49A = 10; 50PLACEHOLDER = 11; 51DOTTEDCIRCLE = 12; 52RS = 13; 53Repha = 15; 54Ra = 16; 55Symbol= 18; 56CS = 19; 57 58c = (C | Ra); # is_consonant 59n = ((ZWNJ?.RS)? (N.N?)?); # is_consonant_modifier 60z = ZWJ|ZWNJ; # is_joiner 61reph = (Ra H | Repha); # possible reph 62 63cn = c.ZWJ?.n?; 64forced_rakar = ZWJ H ZWJ Ra; 65symbol = Symbol.N?; 66matra_group = z{0,3}.M.N?.(H | forced_rakar)?; 67syllable_tail = (z?.SM.SM?.ZWNJ?)? A{0,3}?; 68halant_group = (z?.H.(ZWJ.N?)?); 69final_halant_group = halant_group | H.ZWNJ; 70halant_or_matra_group = (final_halant_group | matra_group{0,4}); 71 72complex_syllable_tail = (halant_group.cn){0,4} halant_or_matra_group syllable_tail; 73 74 75consonant_syllable = (Repha|CS)? cn complex_syllable_tail; 76vowel_syllable = reph? V.n? (ZWJ | complex_syllable_tail); 77standalone_cluster = ((Repha|CS)? PLACEHOLDER | reph? DOTTEDCIRCLE).n? complex_syllable_tail; 78symbol_cluster = symbol syllable_tail; 79broken_cluster = reph? n? complex_syllable_tail; 80other = any; 81 82main := |* 83 consonant_syllable => { found_syllable (consonant_syllable); }; 84 vowel_syllable => { found_syllable (vowel_syllable); }; 85 standalone_cluster => { found_syllable (standalone_cluster); }; 86 symbol_cluster => { found_syllable (symbol_cluster); }; 87 broken_cluster => { found_syllable (broken_cluster); }; 88 other => { found_syllable (non_indic_cluster); }; 89*|; 90 91 92}%% 93 94#define found_syllable(syllable_type) \ 95 HB_STMT_START { \ 96 if (0) fprintf (stderr, "syllable %d..%d %s\n", ts, te, #syllable_type); \ 97 for (unsigned int i = ts; i < te; i++) \ 98 info[i].syllable() = (syllable_serial << 4) | syllable_type; \ 99 syllable_serial++; \ 100 if (unlikely (syllable_serial == 16)) syllable_serial = 1; \ 101 } HB_STMT_END 102 103static void 104find_syllables (hb_buffer_t *buffer) 105{ 106 unsigned int p, pe, eof, ts, te, act; 107 int cs; 108 hb_glyph_info_t *info = buffer->info; 109 %%{ 110 write init; 111 getkey info[p].indic_category(); 112 }%% 113 114 p = 0; 115 pe = eof = buffer->len; 116 117 unsigned int syllable_serial = 1; 118 %%{ 119 write exec; 120 }%% 121} 122 123#endif /* HB_OT_SHAPE_COMPLEX_INDIC_MACHINE_HH */ 124