aboutsummaryrefslogtreecommitdiffstats
path: root/contrib/libs/icu/i18n/rbt.cpp
blob: 1ff04fd95b00c6b0a486c8a597b630cf3aa8eeed (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
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
**********************************************************************
*   Copyright (C) 1999-2015, International Business Machines
*   Corporation and others.  All Rights Reserved.
**********************************************************************
*   Date        Name        Description
*   11/17/99    aliu        Creation.
**********************************************************************
*/

#include "unicode/utypes.h"

#if !UCONFIG_NO_TRANSLITERATION

#include "unicode/rep.h"
#include "unicode/uniset.h"
#include "rbt_pars.h"
#include "rbt_data.h"
#include "rbt_rule.h"
#include "rbt.h"
#include "mutex.h"
#include "umutex.h"

U_NAMESPACE_BEGIN

UOBJECT_DEFINE_RTTI_IMPLEMENTATION(RuleBasedTransliterator)

static Replaceable *gLockedText = nullptr;

void RuleBasedTransliterator::_construct(const UnicodeString& rules,
                                         UTransDirection direction,
                                         UParseError& parseError,
                                         UErrorCode& status) {
    fData = nullptr;
    isDataOwned = true;
    if (U_FAILURE(status)) {
        return;
    }

    TransliteratorParser parser(status);
    parser.parse(rules, direction, parseError, status);
    if (U_FAILURE(status)) {
        return;
    }

    if (parser.idBlockVector.size() != 0 ||
        parser.compoundFilter != nullptr ||
        parser.dataVector.size() == 0) {
        status = U_INVALID_RBT_SYNTAX; // ::ID blocks disallowed in RBT
        return;
    }

    fData = (TransliterationRuleData*)parser.dataVector.orphanElementAt(0);
    setMaximumContextLength(fData->ruleSet.getMaximumContextLength());
}

/**
 * Constructs a new transliterator from the given rules.
 * @param id            the id for the transliterator.
 * @param rules         rules, separated by ';'
 * @param direction     either FORWARD or REVERSE.
 * @param adoptedFilter the filter for this transliterator.
 * @param parseError    Struct to receive information on position 
 *                      of error if an error is encountered
 * @param status        Output param set to success/failure code.
 * @exception IllegalArgumentException if rules are malformed
 * or direction is invalid.
 */
RuleBasedTransliterator::RuleBasedTransliterator(
                            const UnicodeString& id,
                            const UnicodeString& rules,
                            UTransDirection direction,
                            UnicodeFilter* adoptedFilter,
                            UParseError& parseError,
                            UErrorCode& status) :
    Transliterator(id, adoptedFilter) {
    _construct(rules, direction,parseError,status);
}

/**
 * Constructs a new transliterator from the given rules.
 * @param id            the id for the transliterator.
 * @param rules         rules, separated by ';'
 * @param direction     either FORWARD or REVERSE.
 * @param adoptedFilter the filter for this transliterator.
 * @param status        Output param set to success/failure code.
 * @exception IllegalArgumentException if rules are malformed
 * or direction is invalid.
 */
/*RuleBasedTransliterator::RuleBasedTransliterator(
                            const UnicodeString& id,
                            const UnicodeString& rules,
                            UTransDirection direction,
                            UnicodeFilter* adoptedFilter,
                            UErrorCode& status) :
    Transliterator(id, adoptedFilter) {
    UParseError parseError;
    _construct(rules, direction,parseError, status);
}*/

/**
 * Convenience constructor with no filter.
 */
/*RuleBasedTransliterator::RuleBasedTransliterator(
                            const UnicodeString& id,
                            const UnicodeString& rules,
                            UTransDirection direction,
                            UErrorCode& status) :
    Transliterator(id, 0) {
    UParseError parseError;
    _construct(rules, direction,parseError, status);
}*/

/**
 * Convenience constructor with no filter and FORWARD direction.
 */
/*RuleBasedTransliterator::RuleBasedTransliterator(
                            const UnicodeString& id,
                            const UnicodeString& rules,
                            UErrorCode& status) :
    Transliterator(id, 0) {
    UParseError parseError;
    _construct(rules, UTRANS_FORWARD, parseError, status);
}*/

/**
 * Convenience constructor with FORWARD direction.
 */
/*RuleBasedTransliterator::RuleBasedTransliterator(
                            const UnicodeString& id,
                            const UnicodeString& rules,
                            UnicodeFilter* adoptedFilter,
                            UErrorCode& status) :
    Transliterator(id, adoptedFilter) {
    UParseError parseError;
    _construct(rules, UTRANS_FORWARD,parseError, status);
}*/

RuleBasedTransliterator::RuleBasedTransliterator(const UnicodeString& id,
                                 const TransliterationRuleData* theData,
                                 UnicodeFilter* adoptedFilter) :
    Transliterator(id, adoptedFilter),
    fData((TransliterationRuleData*)theData), // cast away const
    isDataOwned(false) {
    setMaximumContextLength(fData->ruleSet.getMaximumContextLength());
}

/**
 * Internal constructor.
 */
RuleBasedTransliterator::RuleBasedTransliterator(const UnicodeString& id,
                                                 TransliterationRuleData* theData,
                                                 UBool isDataAdopted) :
    Transliterator(id, nullptr),
    fData(theData),
    isDataOwned(isDataAdopted) {
    setMaximumContextLength(fData->ruleSet.getMaximumContextLength());
}

/**
 * Copy constructor.
 */
RuleBasedTransliterator::RuleBasedTransliterator(
        const RuleBasedTransliterator& other) :
    Transliterator(other), fData(other.fData),
    isDataOwned(other.isDataOwned) {

    // The data object may or may not be owned.  If it is not owned we
    // share it; it is invariant.  If it is owned, it's still
    // invariant, but we need to copy it to prevent double-deletion.
    // If this becomes a performance issue (if people do a lot of RBT
    // copying -- unlikely) we can reference count the data object.

    // Only do a deep copy if this is owned data, that is, data that
    // will be later deleted.  System transliterators contain
    // non-owned data.
    if (isDataOwned) {
        fData = new TransliterationRuleData(*other.fData);
    }
}

/**
 * Destructor.
 */
RuleBasedTransliterator::~RuleBasedTransliterator() {
    // Delete the data object only if we own it.
    if (isDataOwned) {
        delete fData;
    }
}

RuleBasedTransliterator*
RuleBasedTransliterator::clone() const {
    return new RuleBasedTransliterator(*this);
}

/**
 * Implements {@link Transliterator#handleTransliterate}.
 */
void
RuleBasedTransliterator::handleTransliterate(Replaceable& text, UTransPosition& index,
                                             UBool isIncremental) const {
    /* We keep contextStart and contextLimit fixed the entire time,
     * relative to the text -- contextLimit may move numerically if
     * text is inserted or removed.  The start offset moves toward
     * limit, with replacements happening under it.
     *
     * Example: rules 1. ab>x|y
     *                2. yc>z
     *
     * |eabcd   begin - no match, advance start
     * e|abcd   match rule 1 - change text & adjust start
     * ex|ycd   match rule 2 - change text & adjust start
     * exz|d    no match, advance start
     * exzd|    done
     */

    /* A rule like
     *   a>b|a
     * creates an infinite loop. To prevent that, we put an arbitrary
     * limit on the number of iterations that we take, one that is
     * high enough that any reasonable rules are ok, but low enough to
     * prevent a server from hanging.  The limit is 16 times the
     * number of characters n, unless n is so large that 16n exceeds a
     * uint32_t.
     */
    uint32_t loopCount = 0;
    uint32_t loopLimit = index.limit - index.start;
    if (loopLimit >= 0x10000000) {
        loopLimit = 0xFFFFFFFF;
    } else {
        loopLimit <<= 4;
    }

    // Transliterator locking.  Rule-based Transliterators are not thread safe; concurrent
    //   operations must be prevented.  
    // A Complication: compound transliterators can result in recursive entries to this
    //   function, sometimes with different "This" objects, always with the same text. 
    //   Double-locking must be prevented in these cases.
    //   

    UBool    lockedMutexAtThisLevel = false;

    // Test whether this request is operating on the same text string as
    //   some other transliteration that is still in progress and holding the 
    //   transliteration mutex.  If so, do not lock the transliteration
    //    mutex again.
    //
    //  gLockedText variable is protected by the global ICU mutex.
    //  Shared RBT data protected by transliteratorDataMutex.
    //
    // TODO(andy): Need a better scheme for handling this.

    static UMutex transliteratorDataMutex;
    UBool needToLock;
    {
        Mutex m;
        needToLock = (&text != gLockedText);
    }
    if (needToLock) {
        umtx_lock(&transliteratorDataMutex);  // Contention, longish waits possible here.
        Mutex m;
        gLockedText = &text;
        lockedMutexAtThisLevel = true;
    }
    
    // Check to make sure we don't dereference a null pointer.
    if (fData != nullptr) {
	    while (index.start < index.limit &&
	           loopCount <= loopLimit &&
	           fData->ruleSet.transliterate(text, index, isIncremental)) {
	        ++loopCount;
	    }
    }
    if (lockedMutexAtThisLevel) {
        {
            Mutex m;
            gLockedText = nullptr;
        }
        umtx_unlock(&transliteratorDataMutex);
    }
}

UnicodeString& RuleBasedTransliterator::toRules(UnicodeString& rulesSource,
                                                UBool escapeUnprintable) const {
    return fData->ruleSet.toRules(rulesSource, escapeUnprintable);
}

/**
 * Implement Transliterator framework
 */
void RuleBasedTransliterator::handleGetSourceSet(UnicodeSet& result) const {
    fData->ruleSet.getSourceTargetSet(result, false);
}

/**
 * Override Transliterator framework
 */
UnicodeSet& RuleBasedTransliterator::getTargetSet(UnicodeSet& result) const {
    return fData->ruleSet.getSourceTargetSet(result, true);
}

U_NAMESPACE_END

#endif /* #if !UCONFIG_NO_TRANSLITERATION */