aboutsummaryrefslogtreecommitdiffstats
path: root/contrib/libs/icu/i18n/rbt_rule.h
blob: b927f5d6c05c110d281719bdf73fbc942dabd7c9 (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
308
309
310
// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
* Copyright (C) {1999-2001}, International Business Machines Corporation and others. All Rights Reserved.
**********************************************************************
*   Date        Name        Description
*   11/17/99    aliu        Creation.
**********************************************************************
*/
#ifndef RBT_RULE_H
#define RBT_RULE_H

#include "unicode/utypes.h"

#if !UCONFIG_NO_TRANSLITERATION

#include "unicode/uobject.h"
#include "unicode/unistr.h"
#include "unicode/utrans.h"
#include "unicode/unimatch.h"

U_NAMESPACE_BEGIN

class Replaceable;
class TransliterationRuleData;
class StringMatcher;
class UnicodeFunctor;

/**
 * A transliteration rule used by
 * <code>RuleBasedTransliterator</code>.
 * <code>TransliterationRule</code> is an immutable object.
 *
 * <p>A rule consists of an input pattern and an output string.  When
 * the input pattern is matched, the output string is emitted.  The
 * input pattern consists of zero or more characters which are matched
 * exactly (the key) and optional context.  Context must match if it
 * is specified.  Context may be specified before the key, after the
 * key, or both.  The key, preceding context, and following context
 * may contain variables.  Variables represent a set of Unicode
 * characters, such as the letters <i>a</i> through <i>z</i>.
 * Variables are detected by looking up each character in a supplied
 * variable list to see if it has been so defined.
 *
 * <p>A rule may contain segments in its input string and segment
 * references in its output string.  A segment is a substring of the
 * input pattern, indicated by an offset and limit.  The segment may
 * be in the preceding or following context.  It may not span a
 * context boundary.  A segment reference is a special character in
 * the output string that causes a segment of the input string (not
 * the input pattern) to be copied to the output string.  The range of
 * special characters that represent segment references is defined by
 * RuleBasedTransliterator.Data.
 *
 * @author Alan Liu
 */
class TransliterationRule : public UMemory {

private:

    // TODO Eliminate the pattern and keyLength data members.  They
    // are used only by masks() and getIndexValue() which are called
    // only during build time, not during run-time.  Perhaps these
    // methods and pattern/keyLength can be isolated into a separate
    // object.

    /**
     * The match that must occur before the key, or null if there is no
     * preceding context.
     */
    StringMatcher *anteContext;

    /**
     * The matcher object for the key.  If null, then the key is empty.
     */
    StringMatcher *key;

    /**
     * The match that must occur after the key, or null if there is no
     * following context.
     */
    StringMatcher *postContext;

    /**
     * The object that performs the replacement if the key,
     * anteContext, and postContext are matched.  Never null.
     */
    UnicodeFunctor* output;

    /**
     * The string that must be matched, consisting of the anteContext, key,
     * and postContext, concatenated together, in that order.  Some components
     * may be empty (zero length).
     * @see anteContextLength
     * @see keyLength
     */
    UnicodeString pattern;

    /**
     * An array of matcher objects corresponding to the input pattern
     * segments.  If there are no segments this is null.  N.B. This is
     * a UnicodeMatcher for generality, but in practice it is always a
     * StringMatcher.  In the future we may generalize this, but for
     * now we sometimes cast down to StringMatcher.
     *
     * The array is owned, but the pointers within it are not.
     */
    UnicodeFunctor** segments;

    /**
     * The number of elements in segments[] or zero if segments is NULL.
     */
    int32_t segmentsCount;

    /**
     * The length of the string that must match before the key.  If
     * zero, then there is no matching requirement before the key.
     * Substring [0,anteContextLength) of pattern is the anteContext.
     */
    int32_t anteContextLength;

    /**
     * The length of the key.  Substring [anteContextLength,
     * anteContextLength + keyLength) is the key.

     */
    int32_t keyLength;

    /**
     * Miscellaneous attributes.
     */
    int8_t flags;

    /**
     * Flag attributes.
     */
    enum {
        ANCHOR_START = 1,
        ANCHOR_END   = 2
    };

    /**
     * An alias pointer to the data for this rule.  The data provides
     * lookup services for matchers and segments.
     */
    const TransliterationRuleData* data;

public:

    /**
     * Construct a new rule with the given input, output text, and other
     * attributes.  A cursor position may be specified for the output text.
     * @param input          input string, including key and optional ante and
     *                       post context.
     * @param anteContextPos offset into input to end of ante context, or -1 if
     *                       none.  Must be <= input.length() if not -1.
     * @param postContextPos offset into input to start of post context, or -1
     *                       if none.  Must be <= input.length() if not -1, and must be >=
     *                       anteContextPos.
     * @param outputStr      output string.
     * @param cursorPosition offset into output at which cursor is located, or -1 if
     *                       none.  If less than zero, then the cursor is placed after the
     *                       <code>output</code>; that is, -1 is equivalent to
     *                       <code>output.length()</code>.  If greater than
     *                       <code>output.length()</code> then an exception is thrown.
     * @param cursorOffset   an offset to be added to cursorPos to position the
     *                       cursor either in the ante context, if < 0, or in the post context, if >
     *                       0.  For example, the rule "abc{def} > | @@@ xyz;" changes "def" to
     *                       "xyz" and moves the cursor to before "a".  It would have a cursorOffset
     *                       of -3.
     * @param segs           array of UnicodeMatcher corresponding to input pattern
     *                       segments, or null if there are none.  The array itself is adopted,
     *                       but the pointers within it are not.
     * @param segsCount      number of elements in segs[].
     * @param anchorStart    true if the the rule is anchored on the left to
     *                       the context start.
     * @param anchorEnd      true if the rule is anchored on the right to the
     *                       context limit.
     * @param data           the rule data.
     * @param status         Output parameter filled in with success or failure status.
     */
    TransliterationRule(const UnicodeString& input,
                        int32_t anteContextPos, int32_t postContextPos,
                        const UnicodeString& outputStr,
                        int32_t cursorPosition, int32_t cursorOffset,
                        UnicodeFunctor** segs,
                        int32_t segsCount,
                        UBool anchorStart, UBool anchorEnd,
                        const TransliterationRuleData* data,
                        UErrorCode& status);

    /**
     * Copy constructor.
     * @param other    the object to be copied.
     */
    TransliterationRule(TransliterationRule& other);

    /**
     * Destructor.
     */
    virtual ~TransliterationRule();

    /**
     * Change the data object that this rule belongs to.  Used
     * internally by the TransliterationRuleData copy constructor.
     * @param data    the new data value to be set.
     */
    void setData(const TransliterationRuleData* data);

    /**
     * Return the preceding context length.  This method is needed to
     * support the <code>Transliterator</code> method
     * <code>getMaximumContextLength()</code>.  Internally, this is
     * implemented as the anteContextLength, optionally plus one if
     * there is a start anchor.  The one character anchor gap is
     * needed to make repeated incremental transliteration with
     * anchors work.
     * @return    the preceding context length.
     */
    virtual int32_t getContextLength(void) const;

    /**
     * Internal method.  Returns 8-bit index value for this rule.
     * This is the low byte of the first character of the key,
     * unless the first character of the key is a set.  If it's a
     * set, or otherwise can match multiple keys, the index value is -1.
     * @return    8-bit index value for this rule.
     */
    int16_t getIndexValue() const;

    /**
     * Internal method.  Returns true if this rule matches the given
     * index value.  The index value is an 8-bit integer, 0..255,
     * representing the low byte of the first character of the key.
     * It matches this rule if it matches the first character of the
     * key, or if the first character of the key is a set, and the set
     * contains any character with a low byte equal to the index
     * value.  If the rule contains only ante context, as in foo)>bar,
     * then it will match any key.
     * @param v    the given index value.
     * @return     true if this rule matches the given index value.
     */
    UBool matchesIndexValue(uint8_t v) const;

    /**
     * Return true if this rule masks another rule.  If r1 masks r2 then
     * r1 matches any input string that r2 matches.  If r1 masks r2 and r2 masks
     * r1 then r1 == r2.  Examples: "a>x" masks "ab>y".  "a>x" masks "a[b]>y".
     * "[c]a>x" masks "[dc]a>y".
     * @param r2  the given rule to be compared with.
     * @return    true if this rule masks 'r2'
     */
    virtual UBool masks(const TransliterationRule& r2) const;

    /**
     * Attempt a match and replacement at the given position.  Return
     * the degree of match between this rule and the given text.  The
     * degree of match may be mismatch, a partial match, or a full
     * match.  A mismatch means at least one character of the text
     * does not match the context or key.  A partial match means some
     * context and key characters match, but the text is not long
     * enough to match all of them.  A full match means all context
     * and key characters match.
     * 
     * If a full match is obtained, perform a replacement, update pos,
     * and return U_MATCH.  Otherwise both text and pos are unchanged.
     * 
     * @param text the text
     * @param pos the position indices
     * @param incremental if true, test for partial matches that may
     * be completed by additional text inserted at pos.limit.
     * @return one of <code>U_MISMATCH</code>,
     * <code>U_PARTIAL_MATCH</code>, or <code>U_MATCH</code>.  If
     * incremental is false then U_PARTIAL_MATCH will not be returned.
     */
    UMatchDegree matchAndReplace(Replaceable& text,
                                 UTransPosition& pos,
                                 UBool incremental) const;

    /**
     * Create a rule string that represents this rule object.  Append
     * it to the given string.
     */
    virtual UnicodeString& toRule(UnicodeString& pat,
                                  UBool escapeUnprintable) const;

    /**
     * Union the set of all characters that may be modified by this rule
     * into the given set.
     */
    void addSourceSetTo(UnicodeSet& toUnionTo) const;

    /**
     * Union the set of all characters that may be emitted by this rule
     * into the given set.
     */
    void addTargetSetTo(UnicodeSet& toUnionTo) const;

 private:

    friend class StringMatcher;

    TransliterationRule &operator=(const TransliterationRule &other); // forbid copying of this class
};

U_NAMESPACE_END

#endif /* #if !UCONFIG_NO_TRANSLITERATION */

#endif