summaryrefslogtreecommitdiff
path: root/modules/khmer/khmer-fc.c
blob: 75cebf87c0551ad424ad85228aed32769dc1f4bf (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
/* Pango
 * khmer-fc.c: Shaper for Khmer script
 *
 * Copyright (C) 2004 Open Forum of Cambodia (www.forum.org.kh / www.khmeros.info)
 * Authors: Jens Herden <jens@khmeros.info> and Javier Sola <javier@khmeros.info>
 *
 * Based on code from other shapers
 * Copyright (C) 1999-2004 Red Hat Software
 * Author: Owen Taylor <otaylor@redhat.com>

 * Partially based on Indic shaper
 * Copyright (C) 2001, 2002 IBM Corporation
 * Author: Eric Mader <mader@jtcsv.com>
 *
 * This library is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Library General Public
 * License as published by the Free Software Foundation; either
 * version 2 of the License, or (at your option) any later version.
 *
 * This library is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.	 See the GNU
 * Library General Public License for more details.
 *
 * You should have received a copy of the GNU Library General Public
 * License along with this library; if not, write to the
 * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
 * Boston, MA 02111-1307, USA.
 *
 * The license on the original Indic shaper code is as follows:
 *
 *  * Permission is hereby granted, free of charge, to any person
 * obtaining a copy of this software and associated documentation
 * files (the "Software"), to deal in the Software without
 * restriction, including without limitation the rights to use, copy,
 * modify, merge, publish, distribute, and/or sell copies of the
 * Software, and to permit persons to whom the Software is furnished
 * to do so, provided that the above copyright notice(s) and this
 * permission notice appear in all copies of the Software and that
 * both the above copyright notice(s) and this permission notice
 * appear in supporting documentation.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
 * NONINFRINGEMENT OF THIRD PARTY RIGHTS. IN NO EVENT SHALL THE
 * COPYRIGHT HOLDER OR HOLDERS INCLUDED IN THIS NOTICE BE LIABLE FOR
 * ANY CLAIM, OR ANY SPECIAL INDIRECT OR CONSEQUENTIAL DAMAGES, OR ANY
 * DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS,
 * WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS
 * ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE
 * OF THIS SOFTWARE.
 *
 * Except as contained in this notice, the name of a copyright holder
 * shall not be used in advertising or otherwise to promote the sale,
 * use or other dealings in this Software without prior written
 * authorization of the copyright holder.
 */
#include <config.h>
#include <string.h>

#include "pango-engine.h"
#include "pango-ot.h"
#include "pango-utils.h"
#include "pangofc-font.h"


#define SCRIPT_ENGINE_NAME "KhmerScriptEngineFc"
#define RENDER_TYPE PANGO_RENDER_TYPE_FC


typedef PangoEngineShape      KhmerEngineFc;
typedef PangoEngineShapeClass KhmerEngineFcClass ;


static PangoEngineScriptInfo khmer_scripts[] =
{
  { PANGO_SCRIPT_KHMER, "*" }
};

static PangoEngineInfo script_engines[] =
{
  {
    SCRIPT_ENGINE_NAME,
    PANGO_ENGINE_TYPE_SHAPE,
    RENDER_TYPE,
    khmer_scripts, G_N_ELEMENTS (khmer_scripts)
  }
};


/* Vocabulary
 *     Base ->         A consonant or an independent vowel in its full (not subscript) form. It is the
 *                     center of the syllable, it can be surrounded by coeng (subscript) consonants, vowels,
 *                     split vowels, signs... but there is only one base in a syllable, it has to be coded as
 *                     the first character of the syllable.
 *     split vowel --> vowel that has two parts placed separately (e.g. Before and after the consonant).
 *                     Khmer language has five of them. Khmer split vowels either have one part before the
 *                     base and one after the base or they have a part before the base and a part above the base.
 *                     The first part of all Khmer split vowels is the same character, identical to
 *                     the glyph of Khmer dependent vowel SRA EI
 *     coeng -->  modifier used in Khmer to construct coeng (subscript) consonants
 *                Differently than indian languages, the coeng modifies the consonant that follows it,
 *                not the one preceding it  Each consonant has two forms, the base form and the subscript form
 *                the base form is the normal one (using the consonants code-point), the subscript form is
 *                displayed when the combination coeng + consonant is encountered.
 *     Consonant of type 1 -> A consonant which has subscript for that only occupies space under a base consonant
 *     Consonant of type 2.-> Its subscript form occupies space under and before the base (only one, RO)
 *     Consonant of Type 3 -> Its subscript form occupies space under and after the base (KHO, CHHO, THHO, BA, YO, SA)
 *     Consonant shifter -> Khmer has to series of consonants. The same dependent vowel has different sounds
 *                          if it is attached to a consonant of the first series or a consonant of the second series
 *                          Most consonants have an equivalent in the other series, but some of theme exist only in
 *                          one series (for example SA). If we want to use the consonant SA with a vowel sound that
 *                          can only be done with a vowel sound that corresponds to a vowel accompanying a consonant
 *                          of the other series, then we need to use a consonant shifter: TRIISAP or MUSIKATOAN
 *                          x17C9 y x17CA. TRIISAP changes a first series consonant to second series sound and
 *                          MUSIKATOAN a second series consonant to have a first series vowel sound.
 *                          Consonant shifter are both normally supercript marks, but, when they are followed by a
 *                          superscript, they change shape and take the form of subscript dependent vowel SRA U.
 *                          If they are in the same syllable as a coeng consonant, Unicode 3.0 says that they
 *                          should be typed before the coeng. Unicode 4.0 breaks the standard and says that it should
 *                          be placed after the coeng consonant.
 *     Dependent vowel ->   In khmer dependent vowels can be placed above, below, before or after the base
 *                          Each vowel has its own position. Only one vowel per syllable is allowed.
 *     Signs            ->  Khmer has above signs and post signs. Only one above sign and/or one post sign are
 *                          Allowed in a syllable.
 *
 *
 *  order is important here! This order must be the same that is found in each horizontal
 *  line in the statetable for Khmer (see khmerStateTable) .
 */
enum KhmerCharClassValues
{
  CC_RESERVED             =  0,
  CC_CONSONANT            =  1, /* Consonant of type 1 or independent vowel */
  CC_CONSONANT2           =  2, /* Consonant of type 2 */
  CC_CONSONANT3           =  3, /* Consonant of type 3 */
  CC_ZERO_WIDTH_NJ_MARK   =  4, /* Zero Width non joiner character (0x200C) */
  CC_CONSONANT_SHIFTER    =  5,
  CC_ROBAT                =  6, /* Khmer special diacritic accent -treated differently in state table */
  CC_COENG                =  7, /* Subscript consonant combining character */
  CC_DEPENDENT_VOWEL      =  8,
  CC_SIGN_ABOVE           =  9,
  CC_SIGN_AFTER           = 10,
  CC_ZERO_WIDTH_J_MARK    = 11, /* Zero width joiner character */
  CC_COUNT                = 12  /* This is the number of character classes */
};


enum KhmerCharClassFlags
{
  CF_CLASS_MASK    = 0x0000FFFF,

  CF_CONSONANT     = 0x01000000,  /* flag to speed up comparing */
  CF_SPLIT_VOWEL   = 0x02000000,  /* flag for a split vowel -> the first part is added in front of the syllable */
  CF_DOTTED_CIRCLE = 0x04000000,  /* add a dotted circle if a character with this flag is the first in a syllable */
  CF_COENG         = 0x08000000,  /* flag to speed up comparing */
  CF_SHIFTER       = 0x10000000,  /* flag to speed up comparing */
  CF_ABOVE_VOWEL   = 0x20000000,  /* flag to speed up comparing */

  /* position flags */
  CF_POS_BEFORE    = 0x00080000,
  CF_POS_BELOW     = 0x00040000,
  CF_POS_ABOVE     = 0x00020000,
  CF_POS_AFTER     = 0x00010000,
  CF_POS_MASK      = 0x000f0000
};


/* Characters that get refrered to by name */
enum KhmerChar
{
  C_SIGN_ZWNJ     = 0x200C,
  C_SIGN_ZWJ      = 0x200D,
  C_DOTTED_CIRCLE = 0x25CC,
  C_RO            = 0x179A,
  C_VOWEL_AA      = 0x17B6,
  C_SIGN_NIKAHIT  = 0x17C6,
  C_VOWEL_E       = 0x17C1,
  C_COENG         = 0x17D2
};


enum
{
  /* simple classes, they are used in the state table (in this file) to control the length of a syllable
   * they are also used to know where a character should be placed (location in reference to the base character)
   * and also to know if a character, when independently displayed, should be displayed with a dotted-circle to
   * indicate error in syllable construction
   */
  _xx = CC_RESERVED,
  _sa = CC_SIGN_ABOVE | CF_DOTTED_CIRCLE | CF_POS_ABOVE,
  _sp = CC_SIGN_AFTER | CF_DOTTED_CIRCLE| CF_POS_AFTER,
  _c1 = CC_CONSONANT | CF_CONSONANT,
  _c2 = CC_CONSONANT2 | CF_CONSONANT,
  _c3 = CC_CONSONANT3 | CF_CONSONANT,
  _rb = CC_ROBAT | CF_POS_ABOVE | CF_DOTTED_CIRCLE,
  _cs = CC_CONSONANT_SHIFTER | CF_DOTTED_CIRCLE | CF_SHIFTER,
  _dl = CC_DEPENDENT_VOWEL | CF_POS_BEFORE | CF_DOTTED_CIRCLE,
  _db = CC_DEPENDENT_VOWEL | CF_POS_BELOW | CF_DOTTED_CIRCLE,
  _da = CC_DEPENDENT_VOWEL | CF_POS_ABOVE | CF_DOTTED_CIRCLE | CF_ABOVE_VOWEL,
  _dr = CC_DEPENDENT_VOWEL | CF_POS_AFTER | CF_DOTTED_CIRCLE,
  _co = CC_COENG | CF_COENG | CF_DOTTED_CIRCLE,

  /* split vowel */
  _va = _da | CF_SPLIT_VOWEL,
  _vr = _dr | CF_SPLIT_VOWEL
};


/* Character class: a character class value
 * ORed with character class flags.
 */
typedef glong KhmerCharClass;


/* Character class tables
 * _xx character does not combine into syllable, such as numbers, puntuation marks, non-Khmer signs...
 * _sa Sign placed above the base
 * _sp Sign placed after the base
 * _c1 Consonant of type 1 or independent vowel (independent vowels behave as type 1 consonants)
 * _c2 Consonant of type 2 (only RO)
 * _c3 Consonant of type 3
 * _rb Khmer sign robat u17CC. combining mark for subscript consonants
 * _cd Consonant-shifter
 * _dl Dependent vowel placed before the base (left of the base)
 * _db Dependent vowel placed below the base
 * _da Dependent vowel placed above the base
 * _dr Dependent vowel placed behind the base (right of the base)
 * _co Khmer combining mark COENG u17D2, combines with the consonant or independent vowel following
 *     it to create a subscript consonant or independent vowel
 * _va Khmer split vowel in wich the first part is before the base and the second one above the base
 * _vr Khmer split vowel in wich the first part is before the base and the second one behind (right of) the base
 */
static const KhmerCharClass khmerCharClasses[] =
{
  _c1, _c1, _c1, _c3, _c1, _c1, _c1, _c1, _c3, _c1, _c1, _c1, _c1, _c3, _c1, _c1, /* 1780 - 178F */
  _c1, _c1, _c1, _c1, _c3, _c1, _c1, _c1, _c1, _c3, _c2, _c1, _c1, _c1, _c3, _c3, /* 1790 - 179F */
  _c1, _c3, _c1, _c1, _c1, _c1, _c1, _c1, _c1, _c1, _c1, _c1, _c1, _c1, _c1, _c1, /* 17A0 - 17AF */
  _c1, _c1, _c1, _c1, _dr, _dr, _dr, _da, _da, _da, _da, _db, _db, _db, _va, _vr, /* 17B0 - 17BF */
  _vr, _dl, _dl, _dl, _vr, _vr, _sa, _sp, _sp, _cs, _cs, _sa, _rb, _sa, _sa, _sa, /* 17C0 - 17CF */
  _sa, _sa, _co, _sa, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _sa, _xx, _xx, /* 17D0 - 17DF */
};

/* this define must reflect the range of khmerCharClasses */
#define firstChar 0x1780
#define lastChar 0x17df



/* The stateTable is used to calculate the end (the length) of a well
 * formed Khmer Syllable.
 *
 * Each horizontal line is ordered exactly the same way as the values in KhmerClassTable
 * CharClassValues. This coincidence of values allows the follow up of the table.
 *
 * Each line corresponds to a state, which does not necessarily need to be a type
 * of component... for example, state 2 is a base, with is always a first character
 * in the syllable, but the state could be produced a consonant of any type when
 * it is the first character that is analysed (in ground state).
 *
 * Differentiating 3 types of consonants is necessary in order to
 * forbid the use of certain combinations, such as having a second
 * coeng after a coeng RO,
 * The inexistent possibility of having a type 3 after another type 3 is permitted,
 * eliminating it would very much complicate the table, and it does not create typing
 * problems, as the case above.
 *
 * The table is quite complex, in order to limit the number of coeng consonants
 * to 2 (by means of the table).
 *
 * There a peculiarity, as far as Unicode is concerned:
 * - The consonant-shifter is considered in two possible different
 *   locations, the one considered in Unicode 3.0 and the one considered in
 *   Unicode 4.0. (there is a backwards compatibility problem in this standard).
 *
 *
 * xx    independent character, such as a number, punctuation sign or non-khmer char
 *
 * c1    Khmer consonant of type 1 or an independent vowel
 *       that is, a letter in which the subscript for is only under the
 *       base, not taking any space to the right or to the left
 *
 * c2    Khmer consonant of type 2, the coeng form takes space under
 *       and to the left of the base (only RO is of this type)
 *
 * c3    Khmer consonant of type 3. Its subscript form takes space under
 *       and to the right of the base.
 *
 * cs    Khmer consonant shifter
 *
 * rb    Khmer robat
 *
 * co    coeng character (u17D2)
 *
 * dv    dependent vowel (including split vowels, they are treated in the same way).
 *       even if dv is not defined above, the component that is really tested for is
 *       KhmerClassTable::CC_DEPENDENT_VOWEL, which is common to all dependent vowels
 *
 * zwj   Zero Width joiner
 *
 * zwnj  Zero width non joiner
 *
 * sa    above sign
 *
 * sp    post sign
 *
 * there are lines with equal content but for an easier understanding
 * (and maybe change in the future) we did not join them
 */
static const gint8 khmerStateTable[][CC_COUNT] =
{
/* xx  c1  c2  c3 zwnj cs  rb  co  dv  sa  sp zwj  */
  { 1,  2,  2,  2,  1,  1,  1,  6,  1,  1,  1,  2}, /*  0 - ground state */
  {-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1}, /*  1 - exit state (or sign to the right of the syllable) */
  {-1, -1, -1, -1,  3,  4,  5,  6, 16, 17,  1, -1}, /*  2 - Base consonant */
  {-1, -1, -1, -1, -1,  4, -1, -1, 16, -1, -1, -1}, /*  3 - First ZWNJ before a register shifter
                                                            It can only be followed by a shifter or a vowel */
  {-1, -1, -1, -1, 15, -1, -1,  6, 16, 17,  1, 14}, /*  4 - First register shifter */
  {-1, -1, -1, -1, -1, -1, -1, -1, 20, -1,  1, -1}, /*  5 - Robat */
  {-1,  7,  8,  9, -1, -1, -1, -1, -1, -1, -1, -1}, /*  6 - First Coeng */
  {-1, -1, -1, -1, 12, 13, -1, 10, 16, 17,  1, 14}, /*  7 - First consonant of type 1 after coeng */
  {-1, -1, -1, -1, 12, 13, -1, -1, 16, 17,  1, 14}, /*  8 - First consonant of type 2 after coeng */
  {-1, -1, -1, -1, 12, 13, -1, 10, 16, 17,  1, 14}, /*  9 - First consonant or type 3 after ceong */
  {-1, 11, 11, 11, -1, -1, -1, -1, -1, -1, -1, -1}, /* 10 - Second Coeng (no register shifter before) */
  {-1, -1, -1, -1, 15, -1, -1, -1, 16, 17,  1, 14}, /* 11 - Second coeng consonant (or ind. vowel) no register shifter before */
  {-1, -1, -1, -1, -1, 13, -1, -1, 16, -1, -1, -1}, /* 12 - Second ZWNJ before a register shifter */
  {-1, -1, -1, -1, 15, -1, -1, -1, 16, 17,  1, 14}, /* 13 - Second register shifter */
  {-1, -1, -1, -1, -1, -1, -1, -1, 16, -1, -1, -1}, /* 14 - ZWJ before vowel */
  {-1, -1, -1, -1, -1, -1, -1, -1, 16, -1, -1, -1}, /* 15 - ZWNJ before vowel */
  {-1, -1, -1, -1, -1, -1, -1, -1, -1, 17,  1, 18}, /* 16 - dependent vowel */
  {-1, -1, -1, -1, -1, -1, -1, -1, -1, -1,  1, 18}, /* 17 - sign above */
  {-1, -1, -1, -1, -1, -1, -1, 19, -1, -1, -1, -1}, /* 18 - ZWJ after vowel */
  {-1,  1, -1,  1, -1, -1, -1, -1, -1, -1, -1, -1}, /* 19 - Third coeng */
  {-1, -1, -1, -1, -1, -1, -1, -1, -1, -1,  1, -1}, /* 20 - dependent vowel after a Robat */
};


enum property_flags
{
  abvf = 0x0001,
  pref = 0x0002,
  pstf = 0x0004,
  blwf = 0x0008,

  pres = 0x0010,
  blws = 0x0020,
  abvs = 0x0040,
  psts = 0x0080,
  clig = 0x0100,

  dist = 0x0200,
  blwm = 0x0400,
  abvm = 0x0800,
  mkmk = 0x1000
};


enum properties
{
  blwf_p    = /*(blwf | blws | clig | dist | blwm | mkmk)*/ (abvf | pref | pstf | pres | abvs | psts | abvm),
  pstf_p    = /*(blwf | blws | pref | pres | pstf | psts | clig | dist | blwm)*/ (abvf | abvs | abvm | mkmk),
  abvf_p    = /*(abvf | abvs | clig | dist | abvm | mkmk)*/ (pref | pstf | blwf | pres | blws | psts | blwm),
  pref_p    = /*(pref | pres | clig | dist)*/ (abvf | pstf | blwf | blws | abvs | psts | blwm | abvm | mkmk),
  default_p = /*(pres | blws | clig | dist | abvm | blwm | mkmk)*/ (pref | blwf |abvf | pstf | abvs | psts)
};


/* Below we define how a character in the input string is either in the khmerCharClasses table
 * (in which case we get its type back), a ZWJ or ZWNJ (two characters that may appear
 * within the syllable, but are not in the table) we also get their type back, or an unknown object
 * in which case we get _xx (CC_RESERVED) back
 */
static KhmerCharClass
get_char_class (gunichar ch)
{
  if (ch == C_SIGN_ZWJ)
    return CC_ZERO_WIDTH_J_MARK;

  if (ch == C_SIGN_ZWNJ)
    return CC_ZERO_WIDTH_NJ_MARK;

  if (ch < firstChar || ch > lastChar)
    return CC_RESERVED;

  return khmerCharClasses[ch - firstChar];
}


/* Given an input string of characters and a location in which to start looking
 * calculate, using the state table, which one is the last character of the syllable
 * that starts in the starting position.
 */
static glong
find_syllable (const gunichar *chars,
	       glong           start,
	       glong           char_count)
{
  glong cursor = start;
  gint8 state = 0;
  KhmerCharClass charClass;

  while (cursor < char_count)
    {
      charClass = get_char_class (chars[cursor]) & CF_CLASS_MASK;
      state = khmerStateTable[state][charClass];

      if (state < 0)
	break;

      cursor += 1;
    }

  return cursor;
}


static void
maybe_add_GSUB_feature (PangoOTRuleset *ruleset,
                        PangoOTInfo    *info,
                        guint           script_index,
                        PangoOTTag      tag,
                        gulong          property_bit)
{
  guint feature_index;

  /* 0xffff == default language system */
  if (pango_ot_info_find_feature (info, PANGO_OT_TABLE_GSUB,
                                  tag, script_index, 0xffff, &feature_index))
    pango_ot_ruleset_add_feature (ruleset, PANGO_OT_TABLE_GSUB, feature_index,
                                  property_bit);
}


static void
maybe_add_GPOS_feature (PangoOTRuleset *ruleset,
                        PangoOTInfo    *info,
                        guint           script_index,
                        PangoOTTag      tag,
                        gulong          property_bit)
{
  guint feature_index;

  /* 0xffff == default language system */
  if (pango_ot_info_find_feature (info, PANGO_OT_TABLE_GPOS,
                                  tag, script_index, 0xffff, &feature_index))
    pango_ot_ruleset_add_feature (ruleset, PANGO_OT_TABLE_GPOS, feature_index,
                                  property_bit);
}


static PangoOTRuleset *
get_ruleset (FT_Face face)
{
  PangoOTRuleset *ruleset;
  static GQuark ruleset_quark = 0;

  PangoOTInfo *info = pango_ot_info_get (face);

  if (!ruleset_quark)
    ruleset_quark = g_quark_from_string ("pango-khmer-ruleset");

  if (!info)
    return NULL;

  ruleset = g_object_get_qdata (G_OBJECT (info), ruleset_quark);

  if (!ruleset)
    {
      PangoOTTag khmer_tag = FT_MAKE_TAG ('k', 'h', 'm', 'r');
      guint script_index;

      ruleset = pango_ot_ruleset_new (info);

      if (pango_ot_info_find_script (info, PANGO_OT_TABLE_GSUB,
                                      khmer_tag, &script_index))
        {
          maybe_add_GSUB_feature (ruleset, info, script_index, FT_MAKE_TAG ('p','r','e','f'), pref);
          maybe_add_GSUB_feature (ruleset, info, script_index, FT_MAKE_TAG ('b','l','w','f'), blwf);
          maybe_add_GSUB_feature (ruleset, info, script_index, FT_MAKE_TAG ('a','b','v','f'), abvf);
          maybe_add_GSUB_feature (ruleset, info, script_index, FT_MAKE_TAG ('p','s','t','f'), pstf);

          maybe_add_GSUB_feature (ruleset, info, script_index, FT_MAKE_TAG ('p','r','e','s'), pres);
          maybe_add_GSUB_feature (ruleset, info, script_index, FT_MAKE_TAG ('b','l','w','s'), blws);
          maybe_add_GSUB_feature (ruleset, info, script_index, FT_MAKE_TAG ('a','b','v','s'), abvs);
          maybe_add_GSUB_feature (ruleset, info, script_index, FT_MAKE_TAG ('p','s','t','s'), psts);
          maybe_add_GSUB_feature (ruleset, info, script_index, FT_MAKE_TAG ('c','l','i','g'), clig);
        }

      if (pango_ot_info_find_script (info, PANGO_OT_TABLE_GPOS,
                                      khmer_tag, &script_index))
        {
          maybe_add_GPOS_feature (ruleset, info, script_index, FT_MAKE_TAG ('d','i','s','t'), dist);
          maybe_add_GPOS_feature (ruleset, info, script_index, FT_MAKE_TAG ('b','l','w','m'), blwm);
          maybe_add_GPOS_feature (ruleset, info, script_index, FT_MAKE_TAG ('a','b','v','m'), abvm);
          maybe_add_GPOS_feature (ruleset, info, script_index, FT_MAKE_TAG ('m','k','m','k'), mkmk);
        }

      g_object_set_qdata_full (G_OBJECT (info), ruleset_quark, ruleset,
                                (GDestroyNotify)g_object_unref);
    }

  return ruleset;
}


static PangoGlyph
get_index (PangoFcFont *fc_font, gunichar wc)
{
  PangoGlyph index = pango_fc_font_get_glyph (fc_font, wc);
  if (!index)
    index = pango_fc_font_get_unknown_glyph (fc_font, wc);
  return index;
}


static void
khmer_engine_shape (PangoEngineShape *engine,
                    PangoFont        *font,
                    const char       *text,
                    int               length,
                    PangoAnalysis    *analysis,
                    PangoGlyphString *glyphs)
{
  PangoFcFont *fc_font;
  FT_Face face;
  PangoOTRuleset *ruleset;
  PangoOTBuffer *buffer;
  glong n_chars;
  gunichar *wcs;
  const char *p;
  int i;
  glong syllable;
  KhmerCharClass charClass;
  glong cursor = 0;

  g_return_if_fail (font != NULL);
  g_return_if_fail (text != NULL);
  g_return_if_fail (length >= 0);
  g_return_if_fail (analysis != NULL);

  fc_font = PANGO_FC_FONT (font);
  face = pango_fc_font_lock_face (fc_font);
  g_return_if_fail (face != NULL);

  buffer = pango_ot_buffer_new (fc_font);

  wcs = g_utf8_to_ucs4_fast (text, length, &n_chars);

  p = text;
  /* This loop only exits when we reach the end of a run, which may contain
   * several syllables.
   */
  while (cursor < n_chars)
    {
      /* write a pre vowel or the pre part of a split vowel first
       * and look out for coeng + ro. RO is the only vowel of type 2, and
       * therefore the only one that requires saving space before the base.
       */
      glong coengRo = -1;  /* There is no Coeng Ro, if found this value will change */

      syllable = find_syllable (wcs, cursor, n_chars);

      for (i = cursor; i < syllable; i += 1)
        {
          charClass = get_char_class (wcs[i]);

          /* if a split vowel, write the pre part. In Khmer the pre part
           * is the same for all split vowels, same glyph as pre vowel C_VOWEL_E
           */
          if (charClass & CF_SPLIT_VOWEL)
            {
              pango_ot_buffer_add_glyph (buffer, get_index (fc_font, C_VOWEL_E), pref_p, p - text);
              break; /* there can be only one vowel */
            }

          /* if a vowel with pos before write it out */
          if (charClass & CF_POS_BEFORE)
            {
              pango_ot_buffer_add_glyph (buffer, get_index (fc_font, wcs[i]), pref_p, p - text);
              break; /* there can be only one vowel */
            }

          /* look for coeng + ro and remember position
           * works because coeng + ro is always in front of a vowel (if there is a vowel)
           * and because CC_CONSONANT2 is enough to identify it, as it is the only consonant
           * with this flag
           */
          if ((charClass & CF_COENG) && (i + 1 < syllable) &&
             ((get_char_class (wcs[i + 1]) & CF_CLASS_MASK) == CC_CONSONANT2))
            {
              coengRo = i;
            }
        }

      /* write coeng + ro if found  */
      if (coengRo > -1)
        {
          pango_ot_buffer_add_glyph (buffer, get_index (fc_font, C_COENG), pref_p, p - text);
          pango_ot_buffer_add_glyph (buffer, get_index (fc_font, C_RO), pref_p, p - text);
        }

      /* shall we add a dotted circle?
      * If in the position in which the base should be (first char in the string) there is
      * a character that has the Dotted circle flag (a character that cannot be a base)
      * then write a dotted circle
      */
      if (get_char_class (wcs[cursor]) & CF_DOTTED_CIRCLE)
        {
          pango_ot_buffer_add_glyph (buffer, get_index (fc_font, C_DOTTED_CIRCLE), default_p, p - text);
        }

      /* copy what is left to the output, skipping before vowels and
      * coeng Ro if they are present
      */
      for (i = cursor; i < syllable; i += 1)
        {
          charClass = get_char_class (wcs[i]);

          /* skip a before vowel, it was already processed */
          if (charClass & CF_POS_BEFORE)
            {
              p = g_utf8_next_char (p);
              continue;
            }

          /* skip coeng + ro, it was already processed */
          if (i == coengRo)
            {
              p = g_utf8_next_char (p);
              i += 1;
              p = g_utf8_next_char (p);
              continue;
            }

          switch (charClass & CF_POS_MASK)
            {
              case CF_POS_ABOVE :
		pango_ot_buffer_add_glyph (buffer, get_index (fc_font, wcs[i]), abvf_p, p - text);
		break;

              case CF_POS_AFTER :
		pango_ot_buffer_add_glyph (buffer, get_index (fc_font, wcs[i]), pstf_p, p - text);
		break;
		
              case CF_POS_BELOW :
		pango_ot_buffer_add_glyph (buffer, get_index (fc_font, wcs[i]), blwf_p, p - text);
		break;

              default:
                  /* assign the correct flags to a coeng consonant
                  * Consonants of type 3 are taged as Post forms and those type 1 as below forms
                  */
		if ((charClass & CF_COENG) && i + 1 < syllable)
		  {
		    if ((get_char_class (wcs[i + 1]) & CF_CLASS_MASK) == CC_CONSONANT3)
		      {
			pango_ot_buffer_add_glyph (buffer, get_index (fc_font, wcs[i]), pstf_p, p - text);
			p = g_utf8_next_char (p);
			i += 1;
			pango_ot_buffer_add_glyph (buffer, get_index (fc_font, wcs[i]), pstf_p, p - text);
			break;
		      }
		    else
		      {
			pango_ot_buffer_add_glyph (buffer, get_index (fc_font, wcs[i]), blwf_p, p - text);
			p = g_utf8_next_char (p);
			i += 1;
			pango_ot_buffer_add_glyph (buffer, get_index (fc_font, wcs[i]), blwf_p, p - text);
			break;
		      }
		  }

                  /* if a shifter is followed by an above vowel change the shifter to below form,
                  * an above vowel can have two possible positions i + 1 or i + 3
                  * (position i+1 corresponds to unicode 3, position i+3 to Unicode 4)
                  * and there is an extra rule for C_VOWEL_AA + C_SIGN_NIKAHIT also for two
                  * different positions, right after the shifter or after a vowel (Unicode 4)
                  */
                  if ((charClass & CF_SHIFTER) && (i + 1 < syllable))
                    {
                      if (get_char_class (wcs[i + 1]) & CF_ABOVE_VOWEL)
                        {
                          pango_ot_buffer_add_glyph (buffer, get_index (fc_font, wcs[i]), blwf_p, p - text);
                          break;
                        }
                      if (i + 2 < syllable &&
                          (wcs[i + 1] == C_VOWEL_AA) &&
                          (wcs[i + 2] == C_SIGN_NIKAHIT) )
                        {
                          pango_ot_buffer_add_glyph (buffer, get_index (fc_font, wcs[i]), blwf_p, p - text);
                          break;
                        }
                      if (i + 3 < syllable && (get_char_class (wcs[i + 3]) & CF_ABOVE_VOWEL) )
                        {
                          pango_ot_buffer_add_glyph (buffer, get_index (fc_font, wcs[i]), blwf_p, p - text);
                          break;
                        }
                      if (i + 4 < syllable &&
                          (wcs[i + 3] == C_VOWEL_AA) &&
                          (wcs[i + 4] == C_SIGN_NIKAHIT) )
                        {
                          pango_ot_buffer_add_glyph (buffer, get_index (fc_font, wcs[i]), blwf_p, p - text);
                          break;
                        }

                    }

                  /* default - any other characters  */
                  pango_ot_buffer_add_glyph (buffer, get_index (fc_font, wcs[i]), default_p, p - text);
                  break;
            } /* switch */
          p = g_utf8_next_char (p);
        } /* for */

      cursor = syllable; /* move the pointer to the start of next syllable */
    } /* while */

  /* do gsub processing */
  ruleset = get_ruleset (face);
  if (ruleset != NULL)
    {
      pango_ot_ruleset_substitute (ruleset, buffer);
      pango_ot_ruleset_position (ruleset, buffer);
    }

  pango_ot_buffer_output (buffer, glyphs);

  g_free (wcs);
  pango_ot_buffer_destroy (buffer);

  pango_fc_font_unlock_face (fc_font);
}


static void
khmer_engine_fc_class_init (PangoEngineShapeClass *class)
{
  class->script_shape = khmer_engine_shape;
}

PANGO_ENGINE_SHAPE_DEFINE_TYPE (KhmerEngineFc, khmer_engine_fc,
                                khmer_engine_fc_class_init, NULL);


void
PANGO_MODULE_ENTRY(init) (GTypeModule *module)
{
  khmer_engine_fc_register_type (module);
}


void
PANGO_MODULE_ENTRY(exit) (void)
{
}


void
PANGO_MODULE_ENTRY(list) (PangoEngineInfo **engines,
                          int              *n_engines)
{
  *engines = script_engines;
  *n_engines = G_N_ELEMENTS (script_engines);
}


PangoEngine *
PANGO_MODULE_ENTRY(create) (const char *id)
{
  if (!strcmp (id, SCRIPT_ENGINE_NAME))
    return g_object_new (khmer_engine_fc_type, NULL);
  else
    return NULL;
}