summaryrefslogtreecommitdiff
path: root/i18npool/source/breakiterator/breakiterator_unicode.cxx
blob: 4e5df75d27015115ed3ce2c0dfec5b8a67a5e9ba (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
/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
/*
 * This file is part of the LibreOffice project.
 *
 * 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/.
 *
 * This file incorporates work covered by the following license notice:
 *
 *   Licensed to the Apache Software Foundation (ASF) under one or more
 *   contributor license agreements. See the NOTICE file distributed
 *   with this work for additional information regarding copyright
 *   ownership. The ASF licenses this file to you under the Apache
 *   License, Version 2.0 (the "License"); you may not use this file
 *   except in compliance with the License. You may obtain a copy of
 *   the License at http://www.apache.org/licenses/LICENSE-2.0 .
 */

#include <breakiterator_unicode.hxx>
#include <cppuhelper/supportsservice.hxx>
#include <localedata.hxx>
#include <i18nlangtag/languagetag.hxx>
#include <i18nlangtag/languagetagicu.hxx>
#include <unicode/uchar.h>
#include <unicode/locid.h>
#include <unicode/rbbi.h>
#include <unicode/udata.h>
#include <rtl/strbuf.hxx>
#include <rtl/ustring.hxx>

#include <com/sun/star/i18n/BreakType.hpp>
#include <com/sun/star/i18n/CharacterIteratorMode.hpp>
#include <com/sun/star/i18n/WordType.hpp>

U_CDECL_BEGIN
extern const char OpenOffice_dat[];
U_CDECL_END

using namespace ::com::sun::star;
using namespace ::com::sun::star::i18n;
using namespace ::com::sun::star::lang;

namespace i18npool {

// Cache map of breakiterators, stores state information so has to be
// thread_local.
thread_local static BreakIterator_Unicode::BIMap theBIMap;

BreakIterator_Unicode::BreakIterator_Unicode()
    : cBreakIterator( u"com.sun.star.i18n.BreakIterator_Unicode"_ustr )    // implementation name
    , lineRule( "line" )
    , icuBI( nullptr )
{
}

BreakIterator_Unicode::~BreakIterator_Unicode()
{
}

namespace {

/*
    Wrapper class to provide public access to the icu::RuleBasedBreakIterator's
    setbreakType method.
*/
class OOoRuleBasedBreakIterator : public icu::RuleBasedBreakIterator
{
    public:
    OOoRuleBasedBreakIterator(UDataMemory* image,
                              UErrorCode &status)
        : icu::RuleBasedBreakIterator(image, status)
        { };

};

}

// loading ICU breakiterator on demand.
void BreakIterator_Unicode::loadICUBreakIterator(const css::lang::Locale& rLocale,
        sal_Int16 rBreakType, sal_Int16 nWordType, const char *rule, const OUString& rText)
{
    bool bNewBreak = false;
    UErrorCode status = U_ZERO_ERROR;
    sal_Int16 breakType = 0;
    switch (rBreakType) {
        case LOAD_CHARACTER_BREAKITERATOR: icuBI=&character; breakType = 3; break;
        case LOAD_WORD_BREAKITERATOR:
            assert (nWordType >= 0 && nWordType<= WordType::WORD_COUNT);
            icuBI=&words[nWordType];
            switch (nWordType) {
                case WordType::ANY_WORD: break; // odd but previous behavior
                case WordType::ANYWORD_IGNOREWHITESPACES:
                    breakType = 0; rule = "edit_word"; break;
                case WordType::DICTIONARY_WORD:
                    breakType = 1; rule = "dict_word"; break;
                default:
                case WordType::WORD_COUNT:
                    breakType = 2; rule = "count_word"; break;
            }
            break;
        case LOAD_SENTENCE_BREAKITERATOR: icuBI=&sentence; breakType = 5; break;
        case LOAD_LINE_BREAKITERATOR: icuBI=&line; breakType = 4; break;
    }

    // Using the cache map prevents accessing the file system for each
    // udata_open() where ICU tries first files then data objects. And that for
    // two fallbacks worst case... for each new allocated EditEngine, layout
    // cell, ... *ouch*  Also non-rule locale based iterators can be mapped.
    // This also speeds up loading iterators for alternating or generally more
    // than one language/locale in that iterators are not constructed and
    // destroyed en masse.
    // Four possible keys, locale rule based with break type, locale rule based
    // only, rule based only, locale based with break type. A fifth global key
    // for the initial lookup.
    // Multiple global keys may map to identical value data.
    // All enums used here should be in the range 0..9 so assert that and avoid
    // expensive numeric conversion in append() for faster construction of the
    // always used global key.
    assert( 0 <= breakType && breakType <= 9 && 0 <= rBreakType && rBreakType <= 9 && 0 <= nWordType && nWordType <= 9);
    const OString aLangtagStr( LanguageTag::convertToBcp47( rLocale).toUtf8());
    OStringBuffer aKeyBuf(64);
    aKeyBuf.append( aLangtagStr + ";" );
    if (rule)
        aKeyBuf.append(rule);
    aKeyBuf.append(";" + OStringChar(static_cast<char>('0'+breakType)) + ";"
        + OStringChar(static_cast<char>('0'+rBreakType)) + ";"
        + OStringChar( static_cast<char>('0'+nWordType)));
    // langtag;rule;breakType;rBreakType;nWordType
    const OString aBIMapGlobalKey( aKeyBuf.makeStringAndClear());

    if (icuBI->maBIMapKey != aBIMapGlobalKey || !icuBI->mpValue || !icuBI->mpValue->mpBreakIterator)
    {

        auto aMapIt( theBIMap.find( aBIMapGlobalKey));
        bool bInMap = (aMapIt != theBIMap.end());
        if (bInMap)
            icuBI->mpValue = aMapIt->second;
        else
            icuBI->mpValue.reset();

        if (!bInMap && rule)
            do
            {
                const uno::Sequence< OUString > breakRules = LocaleDataImpl::get()->getBreakIteratorRules(rLocale);

                status = U_ZERO_ERROR;
                udata_setAppData("OpenOffice", OpenOffice_dat, &status);
                if ( !U_SUCCESS(status) )
                    throw uno::RuntimeException("udata_setAppData returned error " + OUString::createFromAscii(u_errorName(status)));

                std::shared_ptr<OOoRuleBasedBreakIterator> rbi;

                if (breakRules.getLength() > breakType && !breakRules[breakType].isEmpty())
                {
                    // langtag;rule;breakType
                    const OString aBIMapRuleTypeKey( aLangtagStr + ";" + rule + ";" + OString::number(breakType));
                    aMapIt = theBIMap.find( aBIMapRuleTypeKey);
                    bInMap = (aMapIt != theBIMap.end());
                    if (bInMap)
                    {
                        icuBI->mpValue = aMapIt->second;
                        icuBI->maBIMapKey = aBIMapGlobalKey;
                        theBIMap.insert( std::make_pair( aBIMapGlobalKey, icuBI->mpValue));
                        break;  // do
                    }

                    rbi = std::make_shared<OOoRuleBasedBreakIterator>(udata_open("OpenOffice", "brk",
                        OUStringToOString(breakRules[breakType], RTL_TEXTENCODING_ASCII_US).getStr(), &status), status);

                    if (U_SUCCESS(status))
                    {
                        icuBI->mpValue = std::make_shared<BI_ValueData>();
                        icuBI->mpValue->mpBreakIterator = rbi;
                        theBIMap.insert( std::make_pair( aBIMapRuleTypeKey, icuBI->mpValue));
                    }
                    else
                    {
                        rbi.reset();
                    }
                }
                else
                {
                    // language;rule (not langtag, unless we'd actually load such)
                    OString aLanguage( LanguageTag( rLocale).getLanguage().toUtf8());
                    const OString aBIMapRuleKey( aLanguage + ";" + rule);
                    aMapIt = theBIMap.find( aBIMapRuleKey);
                    bInMap = (aMapIt != theBIMap.end());
                    if (bInMap)
                    {
                        icuBI->mpValue = aMapIt->second;
                        icuBI->maBIMapKey = aBIMapGlobalKey;
                        theBIMap.insert( std::make_pair( aBIMapGlobalKey, icuBI->mpValue));
                        break;  // do
                    }

                    status = U_ZERO_ERROR;
                    OString aUDName = OString::Concat(rule) + "_" + aLanguage;
                    UDataMemory* pUData = udata_open("OpenOffice", "brk", aUDName.getStr(), &status);
                    if( U_SUCCESS(status) )
                        rbi = std::make_shared<OOoRuleBasedBreakIterator>( pUData, status);
                    if ( U_SUCCESS(status) )
                    {
                        icuBI->mpValue = std::make_shared<BI_ValueData>();
                        icuBI->mpValue->mpBreakIterator = rbi;
                        theBIMap.insert( std::make_pair( aBIMapRuleKey, icuBI->mpValue));
                    }
                    else
                    {
                        rbi.reset();

                        // ;rule (only)
                        const OString aBIMapRuleOnlyKey( OString::Concat(";") + rule);
                        aMapIt = theBIMap.find( aBIMapRuleOnlyKey);
                        bInMap = (aMapIt != theBIMap.end());
                        if (bInMap)
                        {
                            icuBI->mpValue = aMapIt->second;
                            icuBI->maBIMapKey = aBIMapGlobalKey;
                            theBIMap.insert( std::make_pair( aBIMapGlobalKey, icuBI->mpValue));
                            break;  // do
                        }

                        status = U_ZERO_ERROR;
                        pUData = udata_open("OpenOffice", "brk", rule, &status);
                        if( U_SUCCESS(status) )
                            rbi = std::make_shared<OOoRuleBasedBreakIterator>( pUData, status);
                        if ( U_SUCCESS(status) )
                        {
                            icuBI->mpValue = std::make_shared<BI_ValueData>();
                            icuBI->mpValue->mpBreakIterator = rbi;
                            theBIMap.insert( std::make_pair( aBIMapRuleOnlyKey, icuBI->mpValue));
                        }
                        else
                        {
                            rbi.reset();
                        }
                    }
                }
            } while (false);

        if (!icuBI->mpValue || !icuBI->mpValue->mpBreakIterator)
            do
            {
                // langtag;;;rBreakType (empty rule; empty breakType)
                const OString aBIMapLocaleTypeKey( aLangtagStr + ";;;" + OString::number(rBreakType));
                aMapIt = theBIMap.find( aBIMapLocaleTypeKey);
                bInMap = (aMapIt != theBIMap.end());
                if (bInMap)
                {
                    icuBI->mpValue = aMapIt->second;
                    icuBI->maBIMapKey = aBIMapGlobalKey;
                    theBIMap.insert( std::make_pair( aBIMapGlobalKey, icuBI->mpValue));
                    break;  // do
                }

                icu::Locale icuLocale( LanguageTagIcu::getIcuLocale( LanguageTag( rLocale)));
                std::shared_ptr< icu::BreakIterator > pBI;

                status = U_ZERO_ERROR;
                switch (rBreakType) {
                    case LOAD_CHARACTER_BREAKITERATOR:
                        pBI.reset( icu::BreakIterator::createCharacterInstance(icuLocale, status) );
                        break;
                    case LOAD_WORD_BREAKITERATOR:
                        pBI.reset( icu::BreakIterator::createWordInstance(icuLocale, status) );
                        break;
                    case LOAD_SENTENCE_BREAKITERATOR:
                        pBI.reset( icu::BreakIterator::createSentenceInstance(icuLocale, status) );
                        break;
                    case LOAD_LINE_BREAKITERATOR:
                        pBI.reset( icu::BreakIterator::createLineInstance(icuLocale, status) );
                        break;
                }
                if ( !U_SUCCESS(status) || !pBI ) {
                    throw uno::RuntimeException("Failed to create ICU BreakIterator: error " + OUString::createFromAscii(u_errorName(status)));
                }
                icuBI->mpValue = std::make_shared<BI_ValueData>();
                icuBI->mpValue->mpBreakIterator = std::move(pBI);
                theBIMap.insert( std::make_pair( aBIMapLocaleTypeKey, icuBI->mpValue));
            } while (false);
        if (!icuBI->mpValue || !icuBI->mpValue->mpBreakIterator) {
            throw uno::RuntimeException(u"ICU BreakIterator is not properly initialized"_ustr);
        }
        icuBI->maBIMapKey = aBIMapGlobalKey;
        if (!bInMap)
            theBIMap.insert( std::make_pair( aBIMapGlobalKey, icuBI->mpValue));
        bNewBreak=true;
    }

    if (!(bNewBreak || icuBI->mpValue->maICUText.pData != rText.pData))
        return;

    const UChar *pText = reinterpret_cast<const UChar *>(rText.getStr());

    status = U_ZERO_ERROR;
    icuBI->mpValue->mpUt = utext_openUChars(icuBI->mpValue->mpUt, pText, rText.getLength(), &status);

    if (!U_SUCCESS(status))
        throw uno::RuntimeException("utext_openUChars returned error " + OUString::createFromAscii(u_errorName(status)));

    icuBI->mpValue->mpBreakIterator->setText(icuBI->mpValue->mpUt, status);

    if (!U_SUCCESS(status))
        throw uno::RuntimeException("Failed to set text for ICU BreakIterator: error " + OUString::createFromAscii(u_errorName(status)));

    icuBI->mpValue->maICUText = rText;
}

sal_Int32 SAL_CALL BreakIterator_Unicode::nextCharacters( const OUString& Text,
        sal_Int32 nStartPos, const lang::Locale &rLocale,
        sal_Int16 nCharacterIteratorMode, sal_Int32 nCount, sal_Int32& nDone )
{
    if (nCharacterIteratorMode == CharacterIteratorMode::SKIPCELL ) { // for CELL mode
        loadICUBreakIterator(rLocale, LOAD_CHARACTER_BREAKITERATOR, 0, "char", Text);
        icu::BreakIterator* pBI = character.mpValue->mpBreakIterator.get();
        for (nDone = 0; nDone < nCount; nDone++) {
            nStartPos = pBI->following(nStartPos);
            if (nStartPos == icu::BreakIterator::DONE)
                return Text.getLength();
        }
    } else { // for CHARACTER mode
        for (nDone = 0; nDone < nCount && nStartPos < Text.getLength(); nDone++)
            Text.iterateCodePoints(&nStartPos);
    }
    return nStartPos;
}

sal_Int32 SAL_CALL BreakIterator_Unicode::previousCharacters( const OUString& Text,
        sal_Int32 nStartPos, const lang::Locale& rLocale,
        sal_Int16 nCharacterIteratorMode, sal_Int32 nCount, sal_Int32& nDone )
{
    if (nCharacterIteratorMode == CharacterIteratorMode::SKIPCELL ) { // for CELL mode
        loadICUBreakIterator(rLocale, LOAD_CHARACTER_BREAKITERATOR, 0, "char", Text);
        icu::BreakIterator* pBI = character.mpValue->mpBreakIterator.get();
        for (nDone = 0; nDone < nCount; nDone++) {
            nStartPos = pBI->preceding(nStartPos);
            if (nStartPos == icu::BreakIterator::DONE)
                return 0;
        }
    } else { // for BS to delete one char and CHARACTER mode.
        for (nDone = 0; nDone < nCount && nStartPos > 0; nDone++)
            Text.iterateCodePoints(&nStartPos, -1);
    }
    return nStartPos;
}


Boundary SAL_CALL BreakIterator_Unicode::nextWord( const OUString& Text, sal_Int32 nStartPos,
    const lang::Locale& rLocale, sal_Int16 rWordType )
{
    loadICUBreakIterator(rLocale, LOAD_WORD_BREAKITERATOR, rWordType, nullptr, Text);

    Boundary rv;
    rv.startPos = icuBI->mpValue->mpBreakIterator->following(nStartPos);
    if( rv.startPos >= Text.getLength() || rv.startPos == icu::BreakIterator::DONE )
        rv.endPos = rv.startPos;
    else {
        if ((rWordType == WordType::ANYWORD_IGNOREWHITESPACES
             && u_isUWhiteSpace(Text.iterateCodePoints(&rv.startPos, 0)))
            || (rWordType == WordType::DICTIONARY_WORD
                && u_isWhitespace(Text.iterateCodePoints(&rv.startPos, 0))))
            rv.startPos = icuBI->mpValue->mpBreakIterator->following(rv.startPos);

        rv.endPos = icuBI->mpValue->mpBreakIterator->following(rv.startPos);
        if(rv.endPos == icu::BreakIterator::DONE)
            rv.endPos = rv.startPos;
    }
    return rv;
}


Boundary SAL_CALL BreakIterator_Unicode::previousWord(const OUString& Text, sal_Int32 nStartPos,
        const lang::Locale& rLocale, sal_Int16 rWordType)
{
    loadICUBreakIterator(rLocale, LOAD_WORD_BREAKITERATOR, rWordType, nullptr, Text);

    Boundary rv;
    rv.startPos = icuBI->mpValue->mpBreakIterator->preceding(nStartPos);
    if( rv.startPos < 0)
        rv.endPos = rv.startPos;
    else {

        if ((rWordType == WordType::ANYWORD_IGNOREWHITESPACES
             && u_isUWhiteSpace(Text.iterateCodePoints(&rv.startPos, 0)))
            || (rWordType == WordType::DICTIONARY_WORD
                && u_isWhitespace(Text.iterateCodePoints(&rv.startPos, 0))))
            rv.startPos = icuBI->mpValue->mpBreakIterator->preceding(rv.startPos);

        rv.endPos = icuBI->mpValue->mpBreakIterator->following(rv.startPos);
        if(rv.endPos == icu::BreakIterator::DONE)
            rv.endPos = rv.startPos;
    }
    return rv;
}


Boundary SAL_CALL BreakIterator_Unicode::getWordBoundary( const OUString& Text, sal_Int32 nPos, const lang::Locale& rLocale,
        sal_Int16 rWordType, sal_Bool bDirection )
{
    loadICUBreakIterator(rLocale, LOAD_WORD_BREAKITERATOR, rWordType, nullptr, Text);
    sal_Int32 len = Text.getLength();

    Boundary rv;
    if(icuBI->mpValue->mpBreakIterator->isBoundary(nPos)) {
        rv.startPos = rv.endPos = nPos;
        if((bDirection || nPos == 0) && nPos < len) //forward
            rv.endPos = icuBI->mpValue->mpBreakIterator->following(nPos);
        else
            rv.startPos = icuBI->mpValue->mpBreakIterator->preceding(nPos);
    } else {
        if(nPos <= 0) {
            rv.startPos = 0;
            rv.endPos = len ? icuBI->mpValue->mpBreakIterator->following(sal_Int32(0)) : 0;
        } else if(nPos >= len) {
            rv.startPos = icuBI->mpValue->mpBreakIterator->preceding(len);
            rv.endPos = len;
        } else {
            rv.startPos = icuBI->mpValue->mpBreakIterator->preceding(nPos);
            rv.endPos = icuBI->mpValue->mpBreakIterator->following(nPos);
        }
    }
    if (rv.startPos == icu::BreakIterator::DONE)
        rv.startPos = rv.endPos;
    else if (rv.endPos == icu::BreakIterator::DONE)
        rv.endPos = rv.startPos;

    return rv;
}


sal_Int32 SAL_CALL BreakIterator_Unicode::beginOfSentence( const OUString& Text, sal_Int32 nStartPos,
        const lang::Locale &rLocale )
{
    loadICUBreakIterator(rLocale, LOAD_SENTENCE_BREAKITERATOR, 0, "sent", Text);

    sal_Int32 len = Text.getLength();
    if (len > 0 && nStartPos == len)
        Text.iterateCodePoints(&nStartPos, -1); // issue #i27703# treat end position as part of last sentence
    if (!sentence.mpValue->mpBreakIterator->isBoundary(nStartPos))
        nStartPos = sentence.mpValue->mpBreakIterator->preceding(nStartPos);

    // skip preceding space.
    sal_uInt32 ch = Text.iterateCodePoints(&nStartPos);
    while (nStartPos < len && u_isWhitespace(ch)) ch = Text.iterateCodePoints(&nStartPos);
    Text.iterateCodePoints(&nStartPos, -1);

    return nStartPos;
}

sal_Int32 SAL_CALL BreakIterator_Unicode::endOfSentence( const OUString& Text, sal_Int32 nStartPos,
        const lang::Locale &rLocale )
{
    loadICUBreakIterator(rLocale, LOAD_SENTENCE_BREAKITERATOR, 0, "sent", Text);

    sal_Int32 len = Text.getLength();
    if (len > 0 && nStartPos == len)
        Text.iterateCodePoints(&nStartPos, -1); // issue #i27703# treat end position as part of last sentence
    nStartPos = sentence.mpValue->mpBreakIterator->following(nStartPos);

    sal_Int32 nPos=nStartPos;
    while (nPos > 0 && u_isWhitespace(Text.iterateCodePoints(&nPos, -1))) nStartPos=nPos;

    return nStartPos;
}

LineBreakResults SAL_CALL BreakIterator_Unicode::getLineBreak(
        const OUString& Text, sal_Int32 nStartPos,
        const lang::Locale& rLocale, sal_Int32 nMinBreakPos,
        const LineBreakHyphenationOptions& hOptions,
        const LineBreakUserOptions& /*rOptions*/ )
{
    LineBreakResults lbr;

    if (nStartPos >= Text.getLength()) {
        lbr.breakIndex = Text.getLength();
        lbr.breakType = BreakType::WORDBOUNDARY;
        return lbr;
    }

    loadICUBreakIterator(rLocale, LOAD_LINE_BREAKITERATOR, 0, lineRule, Text);

    icu::BreakIterator* pLineBI = line.mpValue->mpBreakIterator.get();
    bool GlueSpace=true;
    while (GlueSpace) {
        // don't break with Slash U+002F SOLIDUS at end of line; see "else" below!
        if (pLineBI->preceding(nStartPos + 1) == nStartPos
                && (nStartPos == 0 || Text[nStartPos - 1] != '/'))
        { //Line boundary break
            lbr.breakIndex = nStartPos;
            lbr.breakType = BreakType::WORDBOUNDARY;
        } else if (hOptions.rHyphenator.is()) { //Hyphenation break
            sal_Int32 boundary_with_punctuation = (pLineBI->next() != icu::BreakIterator::DONE) ? pLineBI->current() : 0;
            pLineBI->preceding(nStartPos + 1); // reset to check correct hyphenation of "word-word"

            sal_Int32 nStartPosWordEnd = nStartPos;
            while (pLineBI->current() < nStartPosWordEnd && u_ispunct(static_cast<sal_uInt32>(Text[nStartPosWordEnd]))) // starting punctuation
                nStartPosWordEnd --;

            Boundary wBoundary = getWordBoundary( Text, nStartPosWordEnd, rLocale,
                WordType::DICTIONARY_WORD, false);

            nStartPosWordEnd = wBoundary.endPos;
            while (nStartPosWordEnd < Text.getLength() && (u_ispunct(static_cast<sal_uInt32>(Text[nStartPosWordEnd])))) // ending punctuation
                nStartPosWordEnd ++;
            nStartPosWordEnd = nStartPosWordEnd - wBoundary.endPos;
            if (hOptions.hyphenIndex - wBoundary.startPos < nStartPosWordEnd) nStartPosWordEnd = hOptions.hyphenIndex - wBoundary.startPos;
#define SPACE 0x0020
            while (boundary_with_punctuation > wBoundary.endPos && Text[--boundary_with_punctuation] == SPACE);
            uno::Reference< linguistic2::XHyphenatedWord > aHyphenatedWord = hOptions.rHyphenator->hyphenate(Text.copy(wBoundary.startPos,
                        wBoundary.endPos - wBoundary.startPos), rLocale,
                    static_cast<sal_Int16>(hOptions.hyphenIndex - wBoundary.startPos - ((hOptions.hyphenIndex == wBoundary.endPos)? nStartPosWordEnd : 0)), hOptions.aHyphenationOptions);
            if (aHyphenatedWord.is()) {
                lbr.rHyphenatedWord = aHyphenatedWord;
                if(wBoundary.startPos + aHyphenatedWord->getHyphenationPos() + 1 < nMinBreakPos )
                    lbr.breakIndex = -1;
                else
                    lbr.breakIndex = wBoundary.startPos; //aHyphenatedWord->getHyphenationPos();
                lbr.breakType = BreakType::HYPHENATION;

                // check not optimal hyphenation of "word-word" (word with hyphens)
                if (lbr.breakIndex > -1 && wBoundary.startPos + aHyphenatedWord->getHyphenationPos() < pLineBI->current()) {
                    lbr.breakIndex = pLineBI->current();
                    lbr.breakType = BreakType::WORDBOUNDARY;
                }

            } else {
                lbr.breakIndex = pLineBI->preceding(nStartPos);
                lbr.breakType = BreakType::WORDBOUNDARY;
            }
        } else { //word boundary break
            lbr.breakIndex = pLineBI->preceding(nStartPos);
            lbr.breakType = BreakType::WORDBOUNDARY;

            // Special case for Slash U+002F SOLIDUS in URI and path names.
            // TR14 defines that as SY: Symbols Allowing Break After (A).
            // This is unwanted in paths, see also i#17155
            if (lbr.breakIndex > 0 && Text[lbr.breakIndex-1] == '/')
            {
                // Look backward and take any whitespace before as a break
                // opportunity. This also glues something like "w/o".
                // Avoid an overly long path and break it as was indicated.
                // Overly long here is arbitrarily defined.
                const sal_Int32 nOverlyLong = 66;
                sal_Int32 nPos = lbr.breakIndex - 1;
                while (nPos > 0 && lbr.breakIndex - nPos < nOverlyLong)
                {
                    if (u_isWhitespace(Text.iterateCodePoints( &nPos, -1)))
                    {
                        lbr.breakIndex = nPos + 1;
                        break;
                    }
                }
            }
        }

#define WJ 0x2060   // Word Joiner
        GlueSpace=false;
        if (lbr.breakType == BreakType::WORDBOUNDARY) {
            nStartPos = lbr.breakIndex;
            if (nStartPos >= 0 && Text[nStartPos--] == WJ)
                GlueSpace=true;
            while (nStartPos >= 0 &&
                    (u_isWhitespace(Text.iterateCodePoints(&nStartPos, 0)) || Text[nStartPos] == WJ)) {
                if (Text[nStartPos--] == WJ)
                    GlueSpace=true;
            }
            if (GlueSpace && nStartPos < 0)  {
                lbr.breakIndex = 0;
                break;
            }
        }
    }

    return lbr;
}

OUString SAL_CALL
BreakIterator_Unicode::getImplementationName()
{
    return cBreakIterator;
}

sal_Bool SAL_CALL
BreakIterator_Unicode::supportsService(const OUString& rServiceName)
{
    return cppu::supportsService(this, rServiceName);
}

uno::Sequence< OUString > SAL_CALL
BreakIterator_Unicode::getSupportedServiceNames()
{
    return{ cBreakIterator };
}

}

extern "C" SAL_DLLPUBLIC_EXPORT css::uno::XInterface *
com_sun_star_i18n_BreakIterator_Unicode_get_implementation(
    css::uno::XComponentContext *,
    css::uno::Sequence<css::uno::Any> const &)
{
    return cppu::acquire(new i18npool::BreakIterator_Unicode());
}

/* vim:set shiftwidth=4 softtabstop=4 expandtab: */