summaryrefslogtreecommitdiff
path: root/intl/unicharutil/nsUnicodeNormalizer.cpp
blob: 16c3a969c59009ca435466a1de5fad4ff861e520 (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
/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */

/* This file is modified from JPNIC's mDNKit, it is under both MPL and 
 * JPNIC's license.
 */

/* This Source Code Form is subject to the terms of the Mozilla Public
 * License, v. 2.0. If a copy of the MPL was not distributed with this
 * file, You can obtain one at http://mozilla.org/MPL/2.0/. */

/*
 * Copyright (c) 2000,2002 Japan Network Information Center.
 * All rights reserved.
 *  
 * By using this file, you agree to the terms and conditions set forth bellow.
 * 
 * 			LICENSE TERMS AND CONDITIONS 
 * 
 * The following License Terms and Conditions apply, unless a different
 * license is obtained from Japan Network Information Center ("JPNIC"),
 * a Japanese association, Kokusai-Kougyou-Kanda Bldg 6F, 2-3-4 Uchi-Kanda,
 * Chiyoda-ku, Tokyo 101-0047, Japan.
 * 
 * 1. Use, Modification and Redistribution (including distribution of any
 *    modified or derived work) in source and/or binary forms is permitted
 *    under this License Terms and Conditions.
 * 
 * 2. Redistribution of source code must retain the copyright notices as they
 *    appear in each source code file, this License Terms and Conditions.
 * 
 * 3. Redistribution in binary form must reproduce the Copyright Notice,
 *    this License Terms and Conditions, in the documentation and/or other
 *    materials provided with the distribution.  For the purposes of binary
 *    distribution the "Copyright Notice" refers to the following language:
 *    "Copyright (c) 2000-2002 Japan Network Information Center.  All rights reserved."
 * 
 * 4. The name of JPNIC may not be used to endorse or promote products
 *    derived from this Software without specific prior written approval of
 *    JPNIC.
 * 
 * 5. Disclaimer/Limitation of Liability: THIS SOFTWARE IS PROVIDED BY JPNIC
 *    "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
 *    LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
 *    PARTICULAR PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL JPNIC BE LIABLE
 *    FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 *    CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 *    SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR
 *    BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
 *    WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
 *    OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
 *    ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
 */

#include <string.h>

#include "nsMemory.h"
#include "nsUnicodeNormalizer.h"
#include "nsString.h"
#include "mozilla/BinarySearch.h"

NS_IMPL_ISUPPORTS(nsUnicodeNormalizer, nsIUnicodeNormalizer)


nsUnicodeNormalizer::nsUnicodeNormalizer()
{
}

nsUnicodeNormalizer::~nsUnicodeNormalizer()
{
}



#define END_BIT		0x80000000


/*
 * Some constants for Hangul decomposition/composition.
 * These things were taken from unicode book. 
 */
#define SBase		0xac00
#define LBase		0x1100
#define VBase		0x1161
#define TBase		0x11a7
#define LCount		19
#define VCount		21
#define TCount		28
#define SLast		(SBase + LCount * VCount * TCount)

struct composition {
	uint32_t c2;	/* 2nd character */
	uint32_t comp;	/* composed character */
};


#include "normalization_data.h"

/*
 * Macro for multi-level index table.
 */
#define LOOKUPTBL(vprefix, mprefix, v) \
	DMAP(vprefix)[\
		IMAP(vprefix)[\
			IMAP(vprefix)[IDX0(mprefix, v)] + IDX1(mprefix, v)\
		]\
	].tbl[IDX2(mprefix, v)]

#define IDX0(mprefix, v) IDX_0(v, BITS1(mprefix), BITS2(mprefix))
#define IDX1(mprefix, v) IDX_1(v, BITS1(mprefix), BITS2(mprefix))
#define IDX2(mprefix, v) IDX_2(v, BITS1(mprefix), BITS2(mprefix))

#define IDX_0(v, bits1, bits2)	((v) >> ((bits1) + (bits2)))
#define IDX_1(v, bits1, bits2)	(((v) >> (bits2)) & ((1 << (bits1)) - 1))
#define IDX_2(v, bits1, bits2)	((v) & ((1 << (bits2)) - 1))

#define BITS1(mprefix)	mprefix ## _BITS_1
#define BITS2(mprefix)	mprefix ## _BITS_2

#define IMAP(vprefix)	vprefix ## _imap
#define DMAP(vprefix)	vprefix ## _table
#define SEQ(vprefix)	vprefix ## _seq

static int32_t
canonclass(uint32_t c) {
	/* Look up canonicalclass table. */
	return (LOOKUPTBL(canon_class, CANON_CLASS, c));
}

static int32_t
decompose_char(uint32_t c, const uint32_t **seqp)
{
	/* Look up decomposition table. */
	int32_t seqidx = LOOKUPTBL(decompose, DECOMP, c);
	*seqp = SEQ(decompose) + (seqidx & ~DECOMP_COMPAT);
	return (seqidx);
}

static int32_t
compose_char(uint32_t c,
				const struct composition **compp)
{
	/* Look up composition table. */
	int32_t seqidx = LOOKUPTBL(compose, CANON_COMPOSE, c);
	*compp = SEQ(compose) + (seqidx & 0xffff);
	return (seqidx >> 16);
}

static nsresult
mdn__unicode_decompose(int32_t compat, uint32_t *v, size_t vlen,
		       uint32_t c, int32_t *decomp_lenp)
{
	uint32_t *vorg = v;
	int32_t seqidx;
	const uint32_t *seq;

	//assert(v != nullptr && vlen >= 0 && decomp_lenp != nullptr);

	/*
	 * First, check for Hangul.
	 */
	if (SBase <= c && c < SLast) {
		int32_t idx, t_offset, v_offset, l_offset;

		idx = c - SBase;
		t_offset = idx % TCount;
		idx /= TCount;
		v_offset = idx % VCount;
		l_offset = idx / VCount;
		if ((t_offset == 0 && vlen < 2) || (t_offset > 0 && vlen < 3))
			return (NS_ERROR_UNORM_MOREOUTPUT);
		*v++ = LBase + l_offset;
		*v++ = VBase + v_offset;
		if (t_offset > 0)
			*v++ = TBase + t_offset;
		*decomp_lenp = v - vorg;
		return (NS_OK);
	}

	/*
	 * Look up decomposition table.  If no decomposition is defined
	 * or if it is a compatibility decomosition when canonical
	 * decomposition requested, return 'NS_SUCCESS_UNORM_NOTFOUND'.
	 */
	seqidx = decompose_char(c, &seq);
	if (seqidx == 0 || (compat == 0 && (seqidx & DECOMP_COMPAT) != 0))
		return (NS_SUCCESS_UNORM_NOTFOUND);
	
	/*
	 * Copy the decomposed sequence.  The end of the sequence are
	 * marked with END_BIT.
	 */
	do {
		uint32_t c;
		int32_t dlen;
		nsresult r;

		c = *seq & ~END_BIT;

		/* Decompose recursively. */
		r = mdn__unicode_decompose(compat, v, vlen, c, &dlen);
		if (r == NS_OK) {
			v += dlen;
			vlen -= dlen;
		} else if (r == NS_SUCCESS_UNORM_NOTFOUND) {
			if (vlen < 1)
				return (NS_ERROR_UNORM_MOREOUTPUT);
			*v++ = c;
			vlen--;
		} else {
			return (r);
		}

	} while ((*seq++ & END_BIT) == 0);
	
	*decomp_lenp = v - vorg;

	return (NS_OK);
}

static int32_t
mdn__unicode_iscompositecandidate(uint32_t c)
{
	const struct composition *dummy;

	/* Check for Hangul */
	if ((LBase <= c && c < LBase + LCount) || (SBase <= c && c < SLast))
		return (1);

	/*
	 * Look up composition table.  If there are no composition
	 * that begins with the given character, it is not a
	 * composition candidate.
	 */
	if (compose_char(c, &dummy) == 0)
		return (0);
	else
		return (1);
}

namespace {

struct SequenceAdaptor
{
	const composition* const mSequence;
	explicit SequenceAdaptor(const composition* aSequence) : mSequence(aSequence) {}
	uint32_t operator[](size_t aIdx) const {
		return mSequence[aIdx].c2;
	}
};

} // namespace

static nsresult
mdn__unicode_compose(uint32_t c1, uint32_t c2, uint32_t *compp)
{
	int32_t n;
	const struct composition *cseq;

	//assert(compp != nullptr);

	/*
	 * Check for Hangul.
	 */
	if (LBase <= c1 && c1 < LBase + LCount &&
	    VBase <= c2 && c2 < VBase + VCount) {
		/*
		 * Hangul L and V.
		 */
		*compp = SBase +
			((c1 - LBase) * VCount + (c2 - VBase)) * TCount;
		return (NS_OK);
	} else if (SBase <= c1 && c1 < SLast &&
		   TBase <= c2 && c2 < TBase + TCount &&
		   (c1 - SBase) % TCount == 0) {
		/*
		 * Hangul LV and T.
		 */
		*compp = c1 + (c2 - TBase);
		return (NS_OK);
	}

	/*
	 * Look up composition table.  If the result is 0, no composition
	 * is defined.  Otherwise, upper 16bits of the result contains
	 * the number of composition that begins with 'c1', and the lower
	 * 16bits is the offset in 'compose_seq'.
	 */
	if ((n = compose_char(c1, &cseq)) == 0)
		return (NS_SUCCESS_UNORM_NOTFOUND);

	/*
	 * The composite sequences are sorted by the 2nd character 'c2'.
	 * So we can use binary search.
	 */
	size_t idx;
	if (mozilla::BinarySearch(SequenceAdaptor(cseq), 0, n, c2, &idx)) {
		*compp = cseq[idx].comp;
		return (NS_OK);
	}

	return (NS_SUCCESS_UNORM_NOTFOUND);
}


#define WORKBUF_SIZE		128
#define WORKBUF_SIZE_MAX	10000

typedef struct {
	int32_t cur;		/* pointing now processing character */
	int32_t last;		/* pointing just after the last character */
	int32_t size;		/* size of UCS and CLASS array */
	uint32_t *ucs;	/* UCS-4 characters */
	int32_t *cclass;		/* and their canonical classes */
	uint32_t ucs_buf[WORKBUF_SIZE];	/* local buffer */
	int32_t class_buf[WORKBUF_SIZE];		/* ditto */
} workbuf_t;

static nsresult	decompose(workbuf_t *wb, uint32_t c, int32_t compat);
static void		get_class(workbuf_t *wb);
static void		reorder(workbuf_t *wb);
static void		compose(workbuf_t *wb);
static nsresult flush_before_cur(workbuf_t *wb, nsAString& aToStr);
static void		workbuf_init(workbuf_t *wb);
static void		workbuf_free(workbuf_t *wb);
static nsresult	workbuf_extend(workbuf_t *wb);
static nsresult	workbuf_append(workbuf_t *wb, uint32_t c);
static void		workbuf_shift(workbuf_t *wb, int32_t shift);
static void		workbuf_removevoid(workbuf_t *wb);


static nsresult
mdn_normalize(bool do_composition, bool compat,
	  const nsAString& aSrcStr, nsAString& aToStr)
{
	workbuf_t wb;
	nsresult r = NS_OK;
	/*
	 * Initialize working buffer.
	 */
	workbuf_init(&wb);

	nsAString::const_iterator start, end;
	aSrcStr.BeginReading(start); 
	aSrcStr.EndReading(end); 

	while (start != end) {
		uint32_t c;
		char16_t curChar;

		//assert(wb.cur == wb.last);

		/*
		 * Get one character from 'from'.
		 */
		curChar= *start++;

		if (NS_IS_HIGH_SURROGATE(curChar) && start != end && NS_IS_LOW_SURROGATE(*(start)) ) {
			c = SURROGATE_TO_UCS4(curChar, *start);
			++start;
		} else {
			c = curChar;
		}

		/*
		 * Decompose it.
		 */
		if ((r = decompose(&wb, c, compat)) != NS_OK)
			break;

		/*
		 * Get canonical class.
		 */
		get_class(&wb);

		/*
		 * Reorder & compose.
		 */
		for (; wb.cur < wb.last; wb.cur++) {
			if (wb.cur == 0) {
				continue;
			} else if (wb.cclass[wb.cur] > 0) {
				/*
				 * This is not a starter. Try reordering.
				 * Note that characters up to it are
				 * already in canonical order.
				 */
				reorder(&wb);
				continue;
			}

			/*
			 * This is a starter character, and there are
			 * some characters before it.  Those characters
			 * have been reordered properly, and
			 * ready for composition.
			 */
			if (do_composition && wb.cclass[0] == 0)
				compose(&wb);

			/*
			 * If CUR points to a starter character,
			 * then process of characters before CUR are
			 * already finished, because any further
			 * reordering/composition for them are blocked
			 * by the starter CUR points.
			 */
			if (wb.cur > 0 && wb.cclass[wb.cur] == 0) {
				/* Flush everything before CUR. */
				r = flush_before_cur(&wb, aToStr);
				if (r != NS_OK)
					break;
			}
		}
	}

	if (r == NS_OK) {
		if (do_composition && wb.cur > 0 && wb.cclass[0] == 0) {
			/*
			 * There is some characters left in WB.
			 * They are ordered, but not composed yet.
			 * Now CUR points just after the last character in WB,
			 * and since compose() tries to compose characters
			 * between top and CUR inclusive, we must make CUR
			 * one character back during compose().
			 */
			wb.cur--;
			compose(&wb);
			wb.cur++;
		}
		/*
		 * Call this even when WB.CUR == 0, to make TO
		 * NUL-terminated.
		 */
		r = flush_before_cur(&wb, aToStr);
	}

	workbuf_free(&wb);

	return (r);
}

static nsresult
decompose(workbuf_t *wb, uint32_t c, int32_t compat) {
	nsresult r;
	int32_t dec_len;

again:
	r = mdn__unicode_decompose(compat, wb->ucs + wb->last,
				   wb->size - wb->last, c, &dec_len);
	switch (r) {
	case NS_OK:
		wb->last += dec_len;
		return (NS_OK);
	case NS_SUCCESS_UNORM_NOTFOUND:
		return (workbuf_append(wb, c));
	case NS_ERROR_UNORM_MOREOUTPUT:
		if ((r = workbuf_extend(wb)) != NS_OK)
			return (r);
		if (wb->size > WORKBUF_SIZE_MAX) {
			// "mdn__unormalize_form*: " "working buffer too large\n"
			return (NS_ERROR_FAILURE);
		}
		goto again;
	default:
		return (r);
	}
	/* NOTREACHED */
}

static void		
get_class(workbuf_t *wb) {
	int32_t i;

	for (i = wb->cur; i < wb->last; i++)
		wb->cclass[i] = canonclass(wb->ucs[i]);
}

static void
reorder(workbuf_t *wb) {
	uint32_t c;
	int32_t i;
	int32_t cclass;

	//assert(wb != nullptr);

	i = wb->cur;
	c = wb->ucs[i];
	cclass = wb->cclass[i];

	while (i > 0 && wb->cclass[i - 1] > cclass) {
		wb->ucs[i] = wb->ucs[i - 1];
		wb->cclass[i] =wb->cclass[i - 1];
		i--;
		wb->ucs[i] = c;
		wb->cclass[i] = cclass;
	}
}

static void
compose(workbuf_t *wb) {
	int32_t cur;
	uint32_t *ucs;
	int32_t *cclass;
	int32_t last_class;
	int32_t nvoids;
	int32_t i;

	//assert(wb != nullptr && wb->cclass[0] == 0);

	cur = wb->cur;
	ucs = wb->ucs;
	cclass = wb->cclass;

	/*
	 * If there are no decomposition sequence that begins with
	 * the top character, composition is impossible.
	 */
	if (!mdn__unicode_iscompositecandidate(ucs[0]))
		return;

	last_class = 0;
	nvoids = 0;
	for (i = 1; i <= cur; i++) {
		uint32_t c;
		int32_t cl = cclass[i];

		if ((last_class < cl || cl == 0) &&
		    mdn__unicode_compose(ucs[0], ucs[i],
					 &c) == NS_OK) {
			/*
			 * Replace the top character with the composed one.
			 */
			ucs[0] = c;
			cclass[0] = canonclass(c);

			cclass[i] = -1;	/* void this character */
			nvoids++;
		} else {
			last_class = cl;
		}
	}

	/* Purge void characters, if any. */
	if (nvoids > 0)
		workbuf_removevoid(wb);
}

static nsresult
flush_before_cur(workbuf_t *wb, nsAString& aToStr) 
{
	int32_t i;

	for (i = 0; i < wb->cur; i++) {
		if (!IS_IN_BMP(wb->ucs[i])) {
			aToStr.Append((char16_t)H_SURROGATE(wb->ucs[i]));
			aToStr.Append((char16_t)L_SURROGATE(wb->ucs[i]));
		} else {
			aToStr.Append((char16_t)(wb->ucs[i]));
		}
	}

	workbuf_shift(wb, wb->cur);

	return (NS_OK);
}

static void
workbuf_init(workbuf_t *wb) {
	wb->cur = 0;
	wb->last = 0;
	wb->size = WORKBUF_SIZE;
	wb->ucs = wb->ucs_buf;
	wb->cclass = wb->class_buf;
}

static void
workbuf_free(workbuf_t *wb) {
	if (wb->ucs != wb->ucs_buf) {
		free(wb->ucs);
		free(wb->cclass);
	}
}

static nsresult
workbuf_extend(workbuf_t *wb) {
	int32_t newsize = wb->size * 3;

	if (wb->ucs == wb->ucs_buf) {
		wb->ucs = (uint32_t*)moz_xmalloc(sizeof(wb->ucs[0]) * newsize);
		if (!wb->ucs)
			return NS_ERROR_OUT_OF_MEMORY;
		wb->cclass = (int32_t*)moz_xmalloc(sizeof(wb->cclass[0]) * newsize);
		if (!wb->cclass) {
			free(wb->ucs);
			wb->ucs = nullptr;
			return NS_ERROR_OUT_OF_MEMORY;
		}
	} else {
		void* buf = moz_xrealloc(wb->ucs, sizeof(wb->ucs[0]) * newsize);
		if (!buf)
			return NS_ERROR_OUT_OF_MEMORY;
		wb->ucs = (uint32_t*)buf;
		buf = moz_xrealloc(wb->cclass, sizeof(wb->cclass[0]) * newsize);
		if (!buf)
			return NS_ERROR_OUT_OF_MEMORY;
		wb->cclass = (int32_t*)buf;
	}
	return (NS_OK);
}

static nsresult
workbuf_append(workbuf_t *wb, uint32_t c) {
	nsresult r;

	if (wb->last >= wb->size && (r = workbuf_extend(wb)) != NS_OK)
		return (r);
	wb->ucs[wb->last++] = c;
	return (NS_OK);
}

static void
workbuf_shift(workbuf_t *wb, int32_t shift) {
	int32_t nmove;

	//assert(wb != nullptr && wb->cur >= shift);

	nmove = wb->last - shift;
	memmove(&wb->ucs[0], &wb->ucs[shift],
		      nmove * sizeof(wb->ucs[0]));
	memmove(&wb->cclass[0], &wb->cclass[shift],
		      nmove * sizeof(wb->cclass[0]));
	wb->cur -= shift;
	wb->last -= shift;
}

static void
workbuf_removevoid(workbuf_t *wb) {
	int32_t i, j;
	int32_t last = wb->last;

	for (i = j = 0; i < last; i++) {
		if (wb->cclass[i] >= 0) {
			if (j < i) {
				wb->ucs[j] = wb->ucs[i];
				wb->cclass[j] = wb->cclass[i];
			}
			j++;
		}
	}
	wb->cur -= last - j;
	wb->last = j;
}

nsresult  
nsUnicodeNormalizer::NormalizeUnicodeNFD( const nsAString& aSrc, nsAString& aDest)
{
  return mdn_normalize(false, false, aSrc, aDest);
}

nsresult  
nsUnicodeNormalizer::NormalizeUnicodeNFC( const nsAString& aSrc, nsAString& aDest)
{
  return mdn_normalize(true, false, aSrc, aDest);
}

nsresult  
nsUnicodeNormalizer::NormalizeUnicodeNFKD( const nsAString& aSrc, nsAString& aDest)
{
  return mdn_normalize(false, true, aSrc, aDest);
}

nsresult  
nsUnicodeNormalizer::NormalizeUnicodeNFKC( const nsAString& aSrc, nsAString& aDest)
{
  return mdn_normalize(true, true, aSrc, aDest);
}

bool
nsUnicodeNormalizer::Compose(uint32_t a, uint32_t b, uint32_t *ab)
{
  return mdn__unicode_compose(a, b, ab) == NS_OK;
}

bool
nsUnicodeNormalizer::DecomposeNonRecursively(uint32_t c, uint32_t *c1, uint32_t *c2)
{
  // We can't use mdn__unicode_decompose here, because that does a recursive
  // decomposition that may yield more than two characters, but the harfbuzz
  // callback wants just a single-step decomp that is guaranteed to produce
  // no more than two characters. So we do a low-level lookup in the table
  // of decomp sequences.
  const uint32_t *seq;
  uint32_t seqidx = decompose_char(c, &seq);
  if (seqidx == 0 || ((seqidx & DECOMP_COMPAT) != 0)) {
    return false;
  }
  *c1 = *seq & ~END_BIT;
  if (*seq & END_BIT) {
    *c2 = 0;
  } else {
    *c2 = *++seq & ~END_BIT;
  }
  return true;
}