hb-ot-shaper-khmer-machine.rl (3505B)
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_SHAPER_KHMER_MACHINE_HH 28 #define HB_OT_SHAPER_KHMER_MACHINE_HH 29 30 #include "hb.hh" 31 32 #include "hb-ot-layout.hh" 33 #include "hb-ot-shaper-indic.hh" 34 35 /* buffer var allocations */ 36 #define khmer_category() ot_shaper_var_u8_category() /* khmer_category_t */ 37 38 using khmer_category_t = unsigned; 39 40 #define K_Cat(Cat) khmer_syllable_machine_ex_##Cat 41 42 enum khmer_syllable_type_t { 43 khmer_consonant_syllable, 44 khmer_broken_cluster, 45 khmer_non_khmer_cluster, 46 }; 47 48 %%{ 49 machine khmer_syllable_machine; 50 alphtype unsigned char; 51 write exports; 52 write data; 53 }%% 54 55 %%{ 56 57 58 # We use category H for spec category Coeng 59 60 export C = 1; 61 export V = 2; 62 export H = 4; 63 export ZWNJ = 5; 64 export ZWJ = 6; 65 export PLACEHOLDER = 10; 66 export DOTTEDCIRCLE = 11; 67 export Ra = 15; 68 69 export VAbv = 20; 70 export VBlw = 21; 71 export VPre = 22; 72 export VPst = 23; 73 74 export Robatic = 25; 75 export Xgroup = 26; 76 export Ygroup = 27; 77 78 79 c = (C | Ra | V); 80 cn = c.((ZWJ|ZWNJ)?.Robatic)?; 81 joiner = (ZWJ | ZWNJ); 82 xgroup = (joiner*.Xgroup)*; 83 ygroup = Ygroup*; 84 85 # This grammar was experimentally extracted from what Uniscribe allows. 86 87 matra_group = VPre? xgroup VBlw? xgroup (joiner?.VAbv)? xgroup VPst?; 88 syllable_tail = xgroup matra_group xgroup (H.c)? ygroup; 89 90 91 broken_cluster = Robatic? (H.cn)* (H | syllable_tail); 92 consonant_syllable = (cn|PLACEHOLDER|DOTTEDCIRCLE) broken_cluster; 93 other = any; 94 95 main := |* 96 consonant_syllable => { found_syllable (khmer_consonant_syllable); }; 97 broken_cluster => { found_syllable (khmer_broken_cluster); buffer->scratch_flags |= HB_BUFFER_SCRATCH_FLAG_HAS_BROKEN_SYLLABLE; }; 98 other => { found_syllable (khmer_non_khmer_cluster); }; 99 *|; 100 101 102 }%% 103 104 #define found_syllable(syllable_type) \ 105 HB_STMT_START { \ 106 if (0) fprintf (stderr, "syllable %u..%u %s\n", ts, te, #syllable_type); \ 107 for (unsigned int i = ts; i < te; i++) \ 108 info[i].syllable() = (syllable_serial << 4) | syllable_type; \ 109 syllable_serial++; \ 110 if (syllable_serial == 16) syllable_serial = 1; \ 111 } HB_STMT_END 112 113 inline void 114 find_syllables_khmer (hb_buffer_t *buffer) 115 { 116 unsigned int p, pe, eof, ts, te, act HB_UNUSED; 117 int cs; 118 hb_glyph_info_t *info = buffer->info; 119 %%{ 120 write init; 121 getkey info[p].khmer_category(); 122 }%% 123 124 p = 0; 125 pe = eof = buffer->len; 126 127 unsigned int syllable_serial = 1; 128 %%{ 129 write exec; 130 }%% 131 } 132 133 #undef found_syllable 134 135 #endif /* HB_OT_SHAPER_KHMER_MACHINE_HH */