aboutsummaryrefslogtreecommitdiffstats
path: root/contrib/libs/icu/common/brkeng.cpp
blob: 78492db66201f1cbdd42767ea41c276dc20b0dac (plain) (blame)
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
// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
 ************************************************************************************
 * Copyright (C) 2006-2016, International Business Machines Corporation
 * and others. All Rights Reserved.
 ************************************************************************************
 */

#include "unicode/utypes.h"

#if !UCONFIG_NO_BREAK_ITERATION

#include "unicode/uchar.h"
#include "unicode/uniset.h"
#include "unicode/chariter.h"
#include "unicode/ures.h"
#include "unicode/udata.h"
#include "unicode/putil.h"
#include "unicode/ustring.h"
#include "unicode/uscript.h"
#include "unicode/ucharstrie.h"
#include "unicode/bytestrie.h"

#include "brkeng.h"
#include "cmemory.h"
#include "dictbe.h"
#include "charstr.h"
#include "dictionarydata.h"
#include "mutex.h"
#include "uvector.h"
#include "umutex.h"
#include "uresimp.h"
#include "ubrkimpl.h"

U_NAMESPACE_BEGIN

/*
 ******************************************************************
 */

LanguageBreakEngine::LanguageBreakEngine() {
}

LanguageBreakEngine::~LanguageBreakEngine() {
}

/*
 ******************************************************************
 */

LanguageBreakFactory::LanguageBreakFactory() {
}

LanguageBreakFactory::~LanguageBreakFactory() {
}

/*
 ******************************************************************
 */

UnhandledEngine::UnhandledEngine(UErrorCode &status) : fHandled(nullptr) {
    (void)status;
}

UnhandledEngine::~UnhandledEngine() {
    delete fHandled;
    fHandled = nullptr;
}

UBool
UnhandledEngine::handles(UChar32 c) const {
    return fHandled && fHandled->contains(c);
}

int32_t
UnhandledEngine::findBreaks( UText *text,
                             int32_t /* startPos */,
                             int32_t endPos,
                             UVector32 &/*foundBreaks*/ ) const {
    UChar32 c = utext_current32(text); 
    while((int32_t)utext_getNativeIndex(text) < endPos && fHandled->contains(c)) {
        utext_next32(text);            // TODO:  recast loop to work with post-increment operations.
        c = utext_current32(text);
    }
    return 0;
}

void
UnhandledEngine::handleCharacter(UChar32 c) {
    if (fHandled == nullptr) {
        fHandled = new UnicodeSet();
        if (fHandled == nullptr) {
            return;
        }
    }
    if (!fHandled->contains(c)) {
        UErrorCode status = U_ZERO_ERROR;
        // Apply the entire script of the character.
        int32_t script = u_getIntPropertyValue(c, UCHAR_SCRIPT);
        fHandled->applyIntPropertyValue(UCHAR_SCRIPT, script, status);
    }
}

/*
 ******************************************************************
 */

ICULanguageBreakFactory::ICULanguageBreakFactory(UErrorCode &/*status*/) {
    fEngines = 0;
}

ICULanguageBreakFactory::~ICULanguageBreakFactory() {
    if (fEngines != 0) {
        delete fEngines;
    }
}

U_NAMESPACE_END
U_CDECL_BEGIN
static void U_CALLCONV _deleteEngine(void *obj) {
    delete (const icu::LanguageBreakEngine *) obj;
}
U_CDECL_END
U_NAMESPACE_BEGIN

const LanguageBreakEngine *
ICULanguageBreakFactory::getEngineFor(UChar32 c) {
    const LanguageBreakEngine *lbe = NULL;
    UErrorCode  status = U_ZERO_ERROR;

    static UMutex gBreakEngineMutex;
    Mutex m(&gBreakEngineMutex);

    if (fEngines == NULL) {
        UStack  *engines = new UStack(_deleteEngine, NULL, status);
        if (U_FAILURE(status) || engines == NULL) {
            // Note: no way to return error code to caller.
            delete engines;
            return NULL;
        }
        fEngines = engines;
    } else {
        int32_t i = fEngines->size();
        while (--i >= 0) {
            lbe = (const LanguageBreakEngine *)(fEngines->elementAt(i));
            if (lbe != NULL && lbe->handles(c)) {
                return lbe;
            }
        }
    }
    
    // We didn't find an engine. Create one.
    lbe = loadEngineFor(c);
    if (lbe != NULL) {
        fEngines->push((void *)lbe, status);
    }
    return lbe;
}

const LanguageBreakEngine *
ICULanguageBreakFactory::loadEngineFor(UChar32 c) {
    UErrorCode status = U_ZERO_ERROR;
    UScriptCode code = uscript_getScript(c, &status);
    if (U_SUCCESS(status)) {
        DictionaryMatcher *m = loadDictionaryMatcherFor(code);
        if (m != NULL) {
            const LanguageBreakEngine *engine = NULL;
            switch(code) {
            case USCRIPT_THAI:
                engine = new ThaiBreakEngine(m, status);
                break;
            case USCRIPT_LAO:
                engine = new LaoBreakEngine(m, status);
                break;
            case USCRIPT_MYANMAR:
                engine = new BurmeseBreakEngine(m, status);
                break;
            case USCRIPT_KHMER:
                engine = new KhmerBreakEngine(m, status);
                break;

#if !UCONFIG_NO_NORMALIZATION
                // CJK not available w/o normalization
            case USCRIPT_HANGUL:
                engine = new CjkBreakEngine(m, kKorean, status);
                break;

            // use same BreakEngine and dictionary for both Chinese and Japanese
            case USCRIPT_HIRAGANA:
            case USCRIPT_KATAKANA:
            case USCRIPT_HAN:
                engine = new CjkBreakEngine(m, kChineseJapanese, status);
                break;
#if 0
            // TODO: Have to get some characters with script=common handled
            // by CjkBreakEngine (e.g. U+309B). Simply subjecting
            // them to CjkBreakEngine does not work. The engine has to
            // special-case them.
            case USCRIPT_COMMON:
            {
                UBlockCode block = ublock_getCode(code);
                if (block == UBLOCK_HIRAGANA || block == UBLOCK_KATAKANA)
                   engine = new CjkBreakEngine(dict, kChineseJapanese, status);
                break;
            }
#endif
#endif

            default:
                break;
            }
            if (engine == NULL) {
                delete m;
            }
            else if (U_FAILURE(status)) {
                delete engine;
                engine = NULL;
            }
            return engine;
        }
    }
    return NULL;
}

DictionaryMatcher *
ICULanguageBreakFactory::loadDictionaryMatcherFor(UScriptCode script) { 
    UErrorCode status = U_ZERO_ERROR;
    // open root from brkitr tree.
    UResourceBundle *b = ures_open(U_ICUDATA_BRKITR, "", &status);
    b = ures_getByKeyWithFallback(b, "dictionaries", b, &status);
    int32_t dictnlength = 0;
    const UChar *dictfname =
        ures_getStringByKeyWithFallback(b, uscript_getShortName(script), &dictnlength, &status);
    if (U_FAILURE(status)) {
        ures_close(b);
        return NULL;
    }
    CharString dictnbuf;
    CharString ext;
    const UChar *extStart = u_memrchr(dictfname, 0x002e, dictnlength);  // last dot
    if (extStart != NULL) {
        int32_t len = (int32_t)(extStart - dictfname);
        ext.appendInvariantChars(UnicodeString(FALSE, extStart + 1, dictnlength - len - 1), status);
        dictnlength = len;
    }
    dictnbuf.appendInvariantChars(UnicodeString(FALSE, dictfname, dictnlength), status);
    ures_close(b);

    UDataMemory *file = udata_open(U_ICUDATA_BRKITR, ext.data(), dictnbuf.data(), &status);
    if (U_SUCCESS(status)) {
        // build trie
        const uint8_t *data = (const uint8_t *)udata_getMemory(file);
        const int32_t *indexes = (const int32_t *)data;
        const int32_t offset = indexes[DictionaryData::IX_STRING_TRIE_OFFSET];
        const int32_t trieType = indexes[DictionaryData::IX_TRIE_TYPE] & DictionaryData::TRIE_TYPE_MASK;
        DictionaryMatcher *m = NULL;
        if (trieType == DictionaryData::TRIE_TYPE_BYTES) {
            const int32_t transform = indexes[DictionaryData::IX_TRANSFORM];
            const char *characters = (const char *)(data + offset);
            m = new BytesDictionaryMatcher(characters, transform, file);
        }
        else if (trieType == DictionaryData::TRIE_TYPE_UCHARS) {
            const UChar *characters = (const UChar *)(data + offset);
            m = new UCharsDictionaryMatcher(characters, file);
        }
        if (m == NULL) {
            // no matcher exists to take ownership - either we are an invalid 
            // type or memory allocation failed
            udata_close(file);
        }
        return m;
    } else if (dictfname != NULL) {
        // we don't have a dictionary matcher.
        // returning NULL here will cause us to fail to find a dictionary break engine, as expected
        status = U_ZERO_ERROR;
        return NULL;
    }
    return NULL;
}

U_NAMESPACE_END

#endif /* #if !UCONFIG_NO_BREAK_ITERATION */