|
1 /* |
|
2 ********************************************************************** |
|
3 * Copyright (C) 1999-2013, International Business Machines |
|
4 * Corporation and others. All Rights Reserved. |
|
5 ********************************************************************** |
|
6 * Date Name Description |
|
7 * 11/17/99 aliu Creation. |
|
8 ********************************************************************** |
|
9 */ |
|
10 #ifndef TRANSLIT_H |
|
11 #define TRANSLIT_H |
|
12 |
|
13 #include "unicode/utypes.h" |
|
14 |
|
15 /** |
|
16 * \file |
|
17 * \brief C++ API: Tranforms text from one format to another. |
|
18 */ |
|
19 |
|
20 #if !UCONFIG_NO_TRANSLITERATION |
|
21 |
|
22 #include "unicode/uobject.h" |
|
23 #include "unicode/unistr.h" |
|
24 #include "unicode/parseerr.h" |
|
25 #include "unicode/utrans.h" // UTransPosition, UTransDirection |
|
26 #include "unicode/strenum.h" |
|
27 |
|
28 U_NAMESPACE_BEGIN |
|
29 |
|
30 class UnicodeFilter; |
|
31 class UnicodeSet; |
|
32 class CompoundTransliterator; |
|
33 class TransliteratorParser; |
|
34 class NormalizationTransliterator; |
|
35 class TransliteratorIDParser; |
|
36 |
|
37 /** |
|
38 * |
|
39 * <code>Transliterator</code> is an abstract class that |
|
40 * transliterates text from one format to another. The most common |
|
41 * kind of transliterator is a script, or alphabet, transliterator. |
|
42 * For example, a Russian to Latin transliterator changes Russian text |
|
43 * written in Cyrillic characters to phonetically equivalent Latin |
|
44 * characters. It does not <em>translate</em> Russian to English! |
|
45 * Transliteration, unlike translation, operates on characters, without |
|
46 * reference to the meanings of words and sentences. |
|
47 * |
|
48 * <p>Although script conversion is its most common use, a |
|
49 * transliterator can actually perform a more general class of tasks. |
|
50 * In fact, <code>Transliterator</code> defines a very general API |
|
51 * which specifies only that a segment of the input text is replaced |
|
52 * by new text. The particulars of this conversion are determined |
|
53 * entirely by subclasses of <code>Transliterator</code>. |
|
54 * |
|
55 * <p><b>Transliterators are stateless</b> |
|
56 * |
|
57 * <p><code>Transliterator</code> objects are <em>stateless</em>; they |
|
58 * retain no information between calls to |
|
59 * <code>transliterate()</code>. (However, this does <em>not</em> |
|
60 * mean that threads may share transliterators without synchronizing |
|
61 * them. Transliterators are not immutable, so they must be |
|
62 * synchronized when shared between threads.) This might seem to |
|
63 * limit the complexity of the transliteration operation. In |
|
64 * practice, subclasses perform complex transliterations by delaying |
|
65 * the replacement of text until it is known that no other |
|
66 * replacements are possible. In other words, although the |
|
67 * <code>Transliterator</code> objects are stateless, the source text |
|
68 * itself embodies all the needed information, and delayed operation |
|
69 * allows arbitrary complexity. |
|
70 * |
|
71 * <p><b>Batch transliteration</b> |
|
72 * |
|
73 * <p>The simplest way to perform transliteration is all at once, on a |
|
74 * string of existing text. This is referred to as <em>batch</em> |
|
75 * transliteration. For example, given a string <code>input</code> |
|
76 * and a transliterator <code>t</code>, the call |
|
77 * |
|
78 * \htmlonly<blockquote>\endhtmlonly<code>String result = t.transliterate(input); |
|
79 * </code>\htmlonly</blockquote>\endhtmlonly |
|
80 * |
|
81 * will transliterate it and return the result. Other methods allow |
|
82 * the client to specify a substring to be transliterated and to use |
|
83 * {@link Replaceable } objects instead of strings, in order to |
|
84 * preserve out-of-band information (such as text styles). |
|
85 * |
|
86 * <p><b>Keyboard transliteration</b> |
|
87 * |
|
88 * <p>Somewhat more involved is <em>keyboard</em>, or incremental |
|
89 * transliteration. This is the transliteration of text that is |
|
90 * arriving from some source (typically the user's keyboard) one |
|
91 * character at a time, or in some other piecemeal fashion. |
|
92 * |
|
93 * <p>In keyboard transliteration, a <code>Replaceable</code> buffer |
|
94 * stores the text. As text is inserted, as much as possible is |
|
95 * transliterated on the fly. This means a GUI that displays the |
|
96 * contents of the buffer may show text being modified as each new |
|
97 * character arrives. |
|
98 * |
|
99 * <p>Consider the simple <code>RuleBasedTransliterator</code>: |
|
100 * |
|
101 * \htmlonly<blockquote>\endhtmlonly<code> |
|
102 * th>{theta}<br> |
|
103 * t>{tau} |
|
104 * </code>\htmlonly</blockquote>\endhtmlonly |
|
105 * |
|
106 * When the user types 't', nothing will happen, since the |
|
107 * transliterator is waiting to see if the next character is 'h'. To |
|
108 * remedy this, we introduce the notion of a cursor, marked by a '|' |
|
109 * in the output string: |
|
110 * |
|
111 * \htmlonly<blockquote>\endhtmlonly<code> |
|
112 * t>|{tau}<br> |
|
113 * {tau}h>{theta} |
|
114 * </code>\htmlonly</blockquote>\endhtmlonly |
|
115 * |
|
116 * Now when the user types 't', tau appears, and if the next character |
|
117 * is 'h', the tau changes to a theta. This is accomplished by |
|
118 * maintaining a cursor position (independent of the insertion point, |
|
119 * and invisible in the GUI) across calls to |
|
120 * <code>transliterate()</code>. Typically, the cursor will |
|
121 * be coincident with the insertion point, but in a case like the one |
|
122 * above, it will precede the insertion point. |
|
123 * |
|
124 * <p>Keyboard transliteration methods maintain a set of three indices |
|
125 * that are updated with each call to |
|
126 * <code>transliterate()</code>, including the cursor, start, |
|
127 * and limit. Since these indices are changed by the method, they are |
|
128 * passed in an <code>int[]</code> array. The <code>START</code> index |
|
129 * marks the beginning of the substring that the transliterator will |
|
130 * look at. It is advanced as text becomes committed (but it is not |
|
131 * the committed index; that's the <code>CURSOR</code>). The |
|
132 * <code>CURSOR</code> index, described above, marks the point at |
|
133 * which the transliterator last stopped, either because it reached |
|
134 * the end, or because it required more characters to disambiguate |
|
135 * between possible inputs. The <code>CURSOR</code> can also be |
|
136 * explicitly set by rules in a <code>RuleBasedTransliterator</code>. |
|
137 * Any characters before the <code>CURSOR</code> index are frozen; |
|
138 * future keyboard transliteration calls within this input sequence |
|
139 * will not change them. New text is inserted at the |
|
140 * <code>LIMIT</code> index, which marks the end of the substring that |
|
141 * the transliterator looks at. |
|
142 * |
|
143 * <p>Because keyboard transliteration assumes that more characters |
|
144 * are to arrive, it is conservative in its operation. It only |
|
145 * transliterates when it can do so unambiguously. Otherwise it waits |
|
146 * for more characters to arrive. When the client code knows that no |
|
147 * more characters are forthcoming, perhaps because the user has |
|
148 * performed some input termination operation, then it should call |
|
149 * <code>finishTransliteration()</code> to complete any |
|
150 * pending transliterations. |
|
151 * |
|
152 * <p><b>Inverses</b> |
|
153 * |
|
154 * <p>Pairs of transliterators may be inverses of one another. For |
|
155 * example, if transliterator <b>A</b> transliterates characters by |
|
156 * incrementing their Unicode value (so "abc" -> "def"), and |
|
157 * transliterator <b>B</b> decrements character values, then <b>A</b> |
|
158 * is an inverse of <b>B</b> and vice versa. If we compose <b>A</b> |
|
159 * with <b>B</b> in a compound transliterator, the result is the |
|
160 * indentity transliterator, that is, a transliterator that does not |
|
161 * change its input text. |
|
162 * |
|
163 * The <code>Transliterator</code> method <code>getInverse()</code> |
|
164 * returns a transliterator's inverse, if one exists, or |
|
165 * <code>null</code> otherwise. However, the result of |
|
166 * <code>getInverse()</code> usually will <em>not</em> be a true |
|
167 * mathematical inverse. This is because true inverse transliterators |
|
168 * are difficult to formulate. For example, consider two |
|
169 * transliterators: <b>AB</b>, which transliterates the character 'A' |
|
170 * to 'B', and <b>BA</b>, which transliterates 'B' to 'A'. It might |
|
171 * seem that these are exact inverses, since |
|
172 * |
|
173 * \htmlonly<blockquote>\endhtmlonly"A" x <b>AB</b> -> "B"<br> |
|
174 * "B" x <b>BA</b> -> "A"\htmlonly</blockquote>\endhtmlonly |
|
175 * |
|
176 * where 'x' represents transliteration. However, |
|
177 * |
|
178 * \htmlonly<blockquote>\endhtmlonly"ABCD" x <b>AB</b> -> "BBCD"<br> |
|
179 * "BBCD" x <b>BA</b> -> "AACD"\htmlonly</blockquote>\endhtmlonly |
|
180 * |
|
181 * so <b>AB</b> composed with <b>BA</b> is not the |
|
182 * identity. Nonetheless, <b>BA</b> may be usefully considered to be |
|
183 * <b>AB</b>'s inverse, and it is on this basis that |
|
184 * <b>AB</b><code>.getInverse()</code> could legitimately return |
|
185 * <b>BA</b>. |
|
186 * |
|
187 * <p><b>IDs and display names</b> |
|
188 * |
|
189 * <p>A transliterator is designated by a short identifier string or |
|
190 * <em>ID</em>. IDs follow the format <em>source-destination</em>, |
|
191 * where <em>source</em> describes the entity being replaced, and |
|
192 * <em>destination</em> describes the entity replacing |
|
193 * <em>source</em>. The entities may be the names of scripts, |
|
194 * particular sequences of characters, or whatever else it is that the |
|
195 * transliterator converts to or from. For example, a transliterator |
|
196 * from Russian to Latin might be named "Russian-Latin". A |
|
197 * transliterator from keyboard escape sequences to Latin-1 characters |
|
198 * might be named "KeyboardEscape-Latin1". By convention, system |
|
199 * entity names are in English, with the initial letters of words |
|
200 * capitalized; user entity names may follow any format so long as |
|
201 * they do not contain dashes. |
|
202 * |
|
203 * <p>In addition to programmatic IDs, transliterator objects have |
|
204 * display names for presentation in user interfaces, returned by |
|
205 * {@link #getDisplayName }. |
|
206 * |
|
207 * <p><b>Factory methods and registration</b> |
|
208 * |
|
209 * <p>In general, client code should use the factory method |
|
210 * {@link #createInstance } to obtain an instance of a |
|
211 * transliterator given its ID. Valid IDs may be enumerated using |
|
212 * <code>getAvailableIDs()</code>. Since transliterators are mutable, |
|
213 * multiple calls to {@link #createInstance } with the same ID will |
|
214 * return distinct objects. |
|
215 * |
|
216 * <p>In addition to the system transliterators registered at startup, |
|
217 * user transliterators may be registered by calling |
|
218 * <code>registerInstance()</code> at run time. A registered instance |
|
219 * acts a template; future calls to {@link #createInstance } with the ID |
|
220 * of the registered object return clones of that object. Thus any |
|
221 * object passed to <tt>registerInstance()</tt> must implement |
|
222 * <tt>clone()</tt> propertly. To register a transliterator subclass |
|
223 * without instantiating it (until it is needed), users may call |
|
224 * {@link #registerFactory }. In this case, the objects are |
|
225 * instantiated by invoking the zero-argument public constructor of |
|
226 * the class. |
|
227 * |
|
228 * <p><b>Subclassing</b> |
|
229 * |
|
230 * Subclasses must implement the abstract method |
|
231 * <code>handleTransliterate()</code>. <p>Subclasses should override |
|
232 * the <code>transliterate()</code> method taking a |
|
233 * <code>Replaceable</code> and the <code>transliterate()</code> |
|
234 * method taking a <code>String</code> and <code>StringBuffer</code> |
|
235 * if the performance of these methods can be improved over the |
|
236 * performance obtained by the default implementations in this class. |
|
237 * |
|
238 * @author Alan Liu |
|
239 * @stable ICU 2.0 |
|
240 */ |
|
241 class U_I18N_API Transliterator : public UObject { |
|
242 |
|
243 private: |
|
244 |
|
245 /** |
|
246 * Programmatic name, e.g., "Latin-Arabic". |
|
247 */ |
|
248 UnicodeString ID; |
|
249 |
|
250 /** |
|
251 * This transliterator's filter. Any character for which |
|
252 * <tt>filter.contains()</tt> returns <tt>false</tt> will not be |
|
253 * altered by this transliterator. If <tt>filter</tt> is |
|
254 * <tt>null</tt> then no filtering is applied. |
|
255 */ |
|
256 UnicodeFilter* filter; |
|
257 |
|
258 int32_t maximumContextLength; |
|
259 |
|
260 public: |
|
261 |
|
262 /** |
|
263 * A context integer or pointer for a factory function, passed by |
|
264 * value. |
|
265 * @stable ICU 2.4 |
|
266 */ |
|
267 union Token { |
|
268 /** |
|
269 * This token, interpreted as a 32-bit integer. |
|
270 * @stable ICU 2.4 |
|
271 */ |
|
272 int32_t integer; |
|
273 /** |
|
274 * This token, interpreted as a native pointer. |
|
275 * @stable ICU 2.4 |
|
276 */ |
|
277 void* pointer; |
|
278 }; |
|
279 |
|
280 #ifndef U_HIDE_INTERNAL_API |
|
281 /** |
|
282 * Return a token containing an integer. |
|
283 * @return a token containing an integer. |
|
284 * @internal |
|
285 */ |
|
286 inline static Token integerToken(int32_t); |
|
287 |
|
288 /** |
|
289 * Return a token containing a pointer. |
|
290 * @return a token containing a pointer. |
|
291 * @internal |
|
292 */ |
|
293 inline static Token pointerToken(void*); |
|
294 #endif /* U_HIDE_INTERNAL_API */ |
|
295 |
|
296 /** |
|
297 * A function that creates and returns a Transliterator. When |
|
298 * invoked, it will be passed the ID string that is being |
|
299 * instantiated, together with the context pointer that was passed |
|
300 * in when the factory function was first registered. Many |
|
301 * factory functions will ignore both parameters, however, |
|
302 * functions that are registered to more than one ID may use the |
|
303 * ID or the context parameter to parameterize the transliterator |
|
304 * they create. |
|
305 * @param ID the string identifier for this transliterator |
|
306 * @param context a context pointer that will be stored and |
|
307 * later passed to the factory function when an ID matching |
|
308 * the registration ID is being instantiated with this factory. |
|
309 * @stable ICU 2.4 |
|
310 */ |
|
311 typedef Transliterator* (U_EXPORT2 *Factory)(const UnicodeString& ID, Token context); |
|
312 |
|
313 protected: |
|
314 |
|
315 /** |
|
316 * Default constructor. |
|
317 * @param ID the string identifier for this transliterator |
|
318 * @param adoptedFilter the filter. Any character for which |
|
319 * <tt>filter.contains()</tt> returns <tt>false</tt> will not be |
|
320 * altered by this transliterator. If <tt>filter</tt> is |
|
321 * <tt>null</tt> then no filtering is applied. |
|
322 * @stable ICU 2.4 |
|
323 */ |
|
324 Transliterator(const UnicodeString& ID, UnicodeFilter* adoptedFilter); |
|
325 |
|
326 /** |
|
327 * Copy constructor. |
|
328 * @stable ICU 2.4 |
|
329 */ |
|
330 Transliterator(const Transliterator&); |
|
331 |
|
332 /** |
|
333 * Assignment operator. |
|
334 * @stable ICU 2.4 |
|
335 */ |
|
336 Transliterator& operator=(const Transliterator&); |
|
337 |
|
338 /** |
|
339 * Create a transliterator from a basic ID. This is an ID |
|
340 * containing only the forward direction source, target, and |
|
341 * variant. |
|
342 * @param id a basic ID of the form S-T or S-T/V. |
|
343 * @param canon canonical ID to assign to the object, or |
|
344 * NULL to leave the ID unchanged |
|
345 * @return a newly created Transliterator or null if the ID is |
|
346 * invalid. |
|
347 * @stable ICU 2.4 |
|
348 */ |
|
349 static Transliterator* createBasicInstance(const UnicodeString& id, |
|
350 const UnicodeString* canon); |
|
351 |
|
352 friend class TransliteratorParser; // for parseID() |
|
353 friend class TransliteratorIDParser; // for createBasicInstance() |
|
354 friend class TransliteratorAlias; // for setID() |
|
355 |
|
356 public: |
|
357 |
|
358 /** |
|
359 * Destructor. |
|
360 * @stable ICU 2.0 |
|
361 */ |
|
362 virtual ~Transliterator(); |
|
363 |
|
364 /** |
|
365 * Implements Cloneable. |
|
366 * All subclasses are encouraged to implement this method if it is |
|
367 * possible and reasonable to do so. Subclasses that are to be |
|
368 * registered with the system using <tt>registerInstance()</tt> |
|
369 * are required to implement this method. If a subclass does not |
|
370 * implement clone() properly and is registered with the system |
|
371 * using registerInstance(), then the default clone() implementation |
|
372 * will return null, and calls to createInstance() will fail. |
|
373 * |
|
374 * @return a copy of the object. |
|
375 * @see #registerInstance |
|
376 * @stable ICU 2.0 |
|
377 */ |
|
378 virtual Transliterator* clone() const; |
|
379 |
|
380 /** |
|
381 * Transliterates a segment of a string, with optional filtering. |
|
382 * |
|
383 * @param text the string to be transliterated |
|
384 * @param start the beginning index, inclusive; <code>0 <= start |
|
385 * <= limit</code>. |
|
386 * @param limit the ending index, exclusive; <code>start <= limit |
|
387 * <= text.length()</code>. |
|
388 * @return The new limit index. The text previously occupying <code>[start, |
|
389 * limit)</code> has been transliterated, possibly to a string of a different |
|
390 * length, at <code>[start, </code><em>new-limit</em><code>)</code>, where |
|
391 * <em>new-limit</em> is the return value. If the input offsets are out of bounds, |
|
392 * the returned value is -1 and the input string remains unchanged. |
|
393 * @stable ICU 2.0 |
|
394 */ |
|
395 virtual int32_t transliterate(Replaceable& text, |
|
396 int32_t start, int32_t limit) const; |
|
397 |
|
398 /** |
|
399 * Transliterates an entire string in place. Convenience method. |
|
400 * @param text the string to be transliterated |
|
401 * @stable ICU 2.0 |
|
402 */ |
|
403 virtual void transliterate(Replaceable& text) const; |
|
404 |
|
405 /** |
|
406 * Transliterates the portion of the text buffer that can be |
|
407 * transliterated unambiguosly after new text has been inserted, |
|
408 * typically as a result of a keyboard event. The new text in |
|
409 * <code>insertion</code> will be inserted into <code>text</code> |
|
410 * at <code>index.limit</code>, advancing |
|
411 * <code>index.limit</code> by <code>insertion.length()</code>. |
|
412 * Then the transliterator will try to transliterate characters of |
|
413 * <code>text</code> between <code>index.cursor</code> and |
|
414 * <code>index.limit</code>. Characters before |
|
415 * <code>index.cursor</code> will not be changed. |
|
416 * |
|
417 * <p>Upon return, values in <code>index</code> will be updated. |
|
418 * <code>index.start</code> will be advanced to the first |
|
419 * character that future calls to this method will read. |
|
420 * <code>index.cursor</code> and <code>index.limit</code> will |
|
421 * be adjusted to delimit the range of text that future calls to |
|
422 * this method may change. |
|
423 * |
|
424 * <p>Typical usage of this method begins with an initial call |
|
425 * with <code>index.start</code> and <code>index.limit</code> |
|
426 * set to indicate the portion of <code>text</code> to be |
|
427 * transliterated, and <code>index.cursor == index.start</code>. |
|
428 * Thereafter, <code>index</code> can be used without |
|
429 * modification in future calls, provided that all changes to |
|
430 * <code>text</code> are made via this method. |
|
431 * |
|
432 * <p>This method assumes that future calls may be made that will |
|
433 * insert new text into the buffer. As a result, it only performs |
|
434 * unambiguous transliterations. After the last call to this |
|
435 * method, there may be untransliterated text that is waiting for |
|
436 * more input to resolve an ambiguity. In order to perform these |
|
437 * pending transliterations, clients should call {@link |
|
438 * #finishTransliteration } after the last call to this |
|
439 * method has been made. |
|
440 * |
|
441 * @param text the buffer holding transliterated and untransliterated text |
|
442 * @param index an array of three integers. |
|
443 * |
|
444 * <ul><li><code>index.start</code>: the beginning index, |
|
445 * inclusive; <code>0 <= index.start <= index.limit</code>. |
|
446 * |
|
447 * <li><code>index.limit</code>: the ending index, exclusive; |
|
448 * <code>index.start <= index.limit <= text.length()</code>. |
|
449 * <code>insertion</code> is inserted at |
|
450 * <code>index.limit</code>. |
|
451 * |
|
452 * <li><code>index.cursor</code>: the next character to be |
|
453 * considered for transliteration; <code>index.start <= |
|
454 * index.cursor <= index.limit</code>. Characters before |
|
455 * <code>index.cursor</code> will not be changed by future calls |
|
456 * to this method.</ul> |
|
457 * |
|
458 * @param insertion text to be inserted and possibly |
|
459 * transliterated into the translation buffer at |
|
460 * <code>index.limit</code>. If <code>null</code> then no text |
|
461 * is inserted. |
|
462 * @param status Output param to filled in with a success or an error. |
|
463 * @see #handleTransliterate |
|
464 * @exception IllegalArgumentException if <code>index</code> |
|
465 * is invalid |
|
466 * @see UTransPosition |
|
467 * @stable ICU 2.0 |
|
468 */ |
|
469 virtual void transliterate(Replaceable& text, UTransPosition& index, |
|
470 const UnicodeString& insertion, |
|
471 UErrorCode& status) const; |
|
472 |
|
473 /** |
|
474 * Transliterates the portion of the text buffer that can be |
|
475 * transliterated unambiguosly after a new character has been |
|
476 * inserted, typically as a result of a keyboard event. This is a |
|
477 * convenience method. |
|
478 * @param text the buffer holding transliterated and |
|
479 * untransliterated text |
|
480 * @param index an array of three integers. |
|
481 * @param insertion text to be inserted and possibly |
|
482 * transliterated into the translation buffer at |
|
483 * <code>index.limit</code>. |
|
484 * @param status Output param to filled in with a success or an error. |
|
485 * @see #transliterate(Replaceable&, UTransPosition&, const UnicodeString&, UErrorCode&) const |
|
486 * @stable ICU 2.0 |
|
487 */ |
|
488 virtual void transliterate(Replaceable& text, UTransPosition& index, |
|
489 UChar32 insertion, |
|
490 UErrorCode& status) const; |
|
491 |
|
492 /** |
|
493 * Transliterates the portion of the text buffer that can be |
|
494 * transliterated unambiguosly. This is a convenience method; see |
|
495 * {@link |
|
496 * #transliterate(Replaceable&, UTransPosition&, const UnicodeString&, UErrorCode&) const } |
|
497 * for details. |
|
498 * @param text the buffer holding transliterated and |
|
499 * untransliterated text |
|
500 * @param index an array of three integers. See {@link #transliterate(Replaceable&, UTransPosition&, const UnicodeString*, UErrorCode&) const }. |
|
501 * @param status Output param to filled in with a success or an error. |
|
502 * @see #transliterate(Replaceable, int[], String) |
|
503 * @stable ICU 2.0 |
|
504 */ |
|
505 virtual void transliterate(Replaceable& text, UTransPosition& index, |
|
506 UErrorCode& status) const; |
|
507 |
|
508 /** |
|
509 * Finishes any pending transliterations that were waiting for |
|
510 * more characters. Clients should call this method as the last |
|
511 * call after a sequence of one or more calls to |
|
512 * <code>transliterate()</code>. |
|
513 * @param text the buffer holding transliterated and |
|
514 * untransliterated text. |
|
515 * @param index the array of indices previously passed to {@link |
|
516 * #transliterate } |
|
517 * @stable ICU 2.0 |
|
518 */ |
|
519 virtual void finishTransliteration(Replaceable& text, |
|
520 UTransPosition& index) const; |
|
521 |
|
522 private: |
|
523 |
|
524 /** |
|
525 * This internal method does incremental transliteration. If the |
|
526 * 'insertion' is non-null then we append it to 'text' before |
|
527 * proceeding. This method calls through to the pure virtual |
|
528 * framework method handleTransliterate() to do the actual |
|
529 * work. |
|
530 * @param text the buffer holding transliterated and |
|
531 * untransliterated text |
|
532 * @param index an array of three integers. See {@link |
|
533 * #transliterate(Replaceable, int[], String)}. |
|
534 * @param insertion text to be inserted and possibly |
|
535 * transliterated into the translation buffer at |
|
536 * <code>index.limit</code>. |
|
537 * @param status Output param to filled in with a success or an error. |
|
538 */ |
|
539 void _transliterate(Replaceable& text, |
|
540 UTransPosition& index, |
|
541 const UnicodeString* insertion, |
|
542 UErrorCode &status) const; |
|
543 |
|
544 protected: |
|
545 |
|
546 /** |
|
547 * Abstract method that concrete subclasses define to implement |
|
548 * their transliteration algorithm. This method handles both |
|
549 * incremental and non-incremental transliteration. Let |
|
550 * <code>originalStart</code> refer to the value of |
|
551 * <code>pos.start</code> upon entry. |
|
552 * |
|
553 * <ul> |
|
554 * <li>If <code>incremental</code> is false, then this method |
|
555 * should transliterate all characters between |
|
556 * <code>pos.start</code> and <code>pos.limit</code>. Upon return |
|
557 * <code>pos.start</code> must == <code> pos.limit</code>.</li> |
|
558 * |
|
559 * <li>If <code>incremental</code> is true, then this method |
|
560 * should transliterate all characters between |
|
561 * <code>pos.start</code> and <code>pos.limit</code> that can be |
|
562 * unambiguously transliterated, regardless of future insertions |
|
563 * of text at <code>pos.limit</code>. Upon return, |
|
564 * <code>pos.start</code> should be in the range |
|
565 * [<code>originalStart</code>, <code>pos.limit</code>). |
|
566 * <code>pos.start</code> should be positioned such that |
|
567 * characters [<code>originalStart</code>, <code> |
|
568 * pos.start</code>) will not be changed in the future by this |
|
569 * transliterator and characters [<code>pos.start</code>, |
|
570 * <code>pos.limit</code>) are unchanged.</li> |
|
571 * </ul> |
|
572 * |
|
573 * <p>Implementations of this method should also obey the |
|
574 * following invariants:</p> |
|
575 * |
|
576 * <ul> |
|
577 * <li> <code>pos.limit</code> and <code>pos.contextLimit</code> |
|
578 * should be updated to reflect changes in length of the text |
|
579 * between <code>pos.start</code> and <code>pos.limit</code>. The |
|
580 * difference <code> pos.contextLimit - pos.limit</code> should |
|
581 * not change.</li> |
|
582 * |
|
583 * <li><code>pos.contextStart</code> should not change.</li> |
|
584 * |
|
585 * <li>Upon return, neither <code>pos.start</code> nor |
|
586 * <code>pos.limit</code> should be less than |
|
587 * <code>originalStart</code>.</li> |
|
588 * |
|
589 * <li>Text before <code>originalStart</code> and text after |
|
590 * <code>pos.limit</code> should not change.</li> |
|
591 * |
|
592 * <li>Text before <code>pos.contextStart</code> and text after |
|
593 * <code> pos.contextLimit</code> should be ignored.</li> |
|
594 * </ul> |
|
595 * |
|
596 * <p>Subclasses may safely assume that all characters in |
|
597 * [<code>pos.start</code>, <code>pos.limit</code>) are filtered. |
|
598 * In other words, the filter has already been applied by the time |
|
599 * this method is called. See |
|
600 * <code>filteredTransliterate()</code>. |
|
601 * |
|
602 * <p>This method is <b>not</b> for public consumption. Calling |
|
603 * this method directly will transliterate |
|
604 * [<code>pos.start</code>, <code>pos.limit</code>) without |
|
605 * applying the filter. End user code should call <code> |
|
606 * transliterate()</code> instead of this method. Subclass code |
|
607 * and wrapping transliterators should call |
|
608 * <code>filteredTransliterate()</code> instead of this method.<p> |
|
609 * |
|
610 * @param text the buffer holding transliterated and |
|
611 * untransliterated text |
|
612 * |
|
613 * @param pos the indices indicating the start, limit, context |
|
614 * start, and context limit of the text. |
|
615 * |
|
616 * @param incremental if true, assume more text may be inserted at |
|
617 * <code>pos.limit</code> and act accordingly. Otherwise, |
|
618 * transliterate all text between <code>pos.start</code> and |
|
619 * <code>pos.limit</code> and move <code>pos.start</code> up to |
|
620 * <code>pos.limit</code>. |
|
621 * |
|
622 * @see #transliterate |
|
623 * @stable ICU 2.4 |
|
624 */ |
|
625 virtual void handleTransliterate(Replaceable& text, |
|
626 UTransPosition& pos, |
|
627 UBool incremental) const = 0; |
|
628 |
|
629 public: |
|
630 /** |
|
631 * Transliterate a substring of text, as specified by index, taking filters |
|
632 * into account. This method is for subclasses that need to delegate to |
|
633 * another transliterator, such as CompoundTransliterator. |
|
634 * @param text the text to be transliterated |
|
635 * @param index the position indices |
|
636 * @param incremental if TRUE, then assume more characters may be inserted |
|
637 * at index.limit, and postpone processing to accomodate future incoming |
|
638 * characters |
|
639 * @stable ICU 2.4 |
|
640 */ |
|
641 virtual void filteredTransliterate(Replaceable& text, |
|
642 UTransPosition& index, |
|
643 UBool incremental) const; |
|
644 |
|
645 private: |
|
646 |
|
647 /** |
|
648 * Top-level transliteration method, handling filtering, incremental and |
|
649 * non-incremental transliteration, and rollback. All transliteration |
|
650 * public API methods eventually call this method with a rollback argument |
|
651 * of TRUE. Other entities may call this method but rollback should be |
|
652 * FALSE. |
|
653 * |
|
654 * <p>If this transliterator has a filter, break up the input text into runs |
|
655 * of unfiltered characters. Pass each run to |
|
656 * subclass.handleTransliterate(). |
|
657 * |
|
658 * <p>In incremental mode, if rollback is TRUE, perform a special |
|
659 * incremental procedure in which several passes are made over the input |
|
660 * text, adding one character at a time, and committing successful |
|
661 * transliterations as they occur. Unsuccessful transliterations are rolled |
|
662 * back and retried with additional characters to give correct results. |
|
663 * |
|
664 * @param text the text to be transliterated |
|
665 * @param index the position indices |
|
666 * @param incremental if TRUE, then assume more characters may be inserted |
|
667 * at index.limit, and postpone processing to accomodate future incoming |
|
668 * characters |
|
669 * @param rollback if TRUE and if incremental is TRUE, then perform special |
|
670 * incremental processing, as described above, and undo partial |
|
671 * transliterations where necessary. If incremental is FALSE then this |
|
672 * parameter is ignored. |
|
673 */ |
|
674 virtual void filteredTransliterate(Replaceable& text, |
|
675 UTransPosition& index, |
|
676 UBool incremental, |
|
677 UBool rollback) const; |
|
678 |
|
679 public: |
|
680 |
|
681 /** |
|
682 * Returns the length of the longest context required by this transliterator. |
|
683 * This is <em>preceding</em> context. The default implementation supplied |
|
684 * by <code>Transliterator</code> returns zero; subclasses |
|
685 * that use preceding context should override this method to return the |
|
686 * correct value. For example, if a transliterator translates "ddd" (where |
|
687 * d is any digit) to "555" when preceded by "(ddd)", then the preceding |
|
688 * context length is 5, the length of "(ddd)". |
|
689 * |
|
690 * @return The maximum number of preceding context characters this |
|
691 * transliterator needs to examine |
|
692 * @stable ICU 2.0 |
|
693 */ |
|
694 int32_t getMaximumContextLength(void) const; |
|
695 |
|
696 protected: |
|
697 |
|
698 /** |
|
699 * Method for subclasses to use to set the maximum context length. |
|
700 * @param maxContextLength the new value to be set. |
|
701 * @see #getMaximumContextLength |
|
702 * @stable ICU 2.4 |
|
703 */ |
|
704 void setMaximumContextLength(int32_t maxContextLength); |
|
705 |
|
706 public: |
|
707 |
|
708 /** |
|
709 * Returns a programmatic identifier for this transliterator. |
|
710 * If this identifier is passed to <code>createInstance()</code>, it |
|
711 * will return this object, if it has been registered. |
|
712 * @return a programmatic identifier for this transliterator. |
|
713 * @see #registerInstance |
|
714 * @see #registerFactory |
|
715 * @see #getAvailableIDs |
|
716 * @stable ICU 2.0 |
|
717 */ |
|
718 virtual const UnicodeString& getID(void) const; |
|
719 |
|
720 /** |
|
721 * Returns a name for this transliterator that is appropriate for |
|
722 * display to the user in the default locale. See {@link |
|
723 * #getDisplayName } for details. |
|
724 * @param ID the string identifier for this transliterator |
|
725 * @param result Output param to receive the display name |
|
726 * @return A reference to 'result'. |
|
727 * @stable ICU 2.0 |
|
728 */ |
|
729 static UnicodeString& U_EXPORT2 getDisplayName(const UnicodeString& ID, |
|
730 UnicodeString& result); |
|
731 |
|
732 /** |
|
733 * Returns a name for this transliterator that is appropriate for |
|
734 * display to the user in the given locale. This name is taken |
|
735 * from the locale resource data in the standard manner of the |
|
736 * <code>java.text</code> package. |
|
737 * |
|
738 * <p>If no localized names exist in the system resource bundles, |
|
739 * a name is synthesized using a localized |
|
740 * <code>MessageFormat</code> pattern from the resource data. The |
|
741 * arguments to this pattern are an integer followed by one or two |
|
742 * strings. The integer is the number of strings, either 1 or 2. |
|
743 * The strings are formed by splitting the ID for this |
|
744 * transliterator at the first '-'. If there is no '-', then the |
|
745 * entire ID forms the only string. |
|
746 * @param ID the string identifier for this transliterator |
|
747 * @param inLocale the Locale in which the display name should be |
|
748 * localized. |
|
749 * @param result Output param to receive the display name |
|
750 * @return A reference to 'result'. |
|
751 * @stable ICU 2.0 |
|
752 */ |
|
753 static UnicodeString& U_EXPORT2 getDisplayName(const UnicodeString& ID, |
|
754 const Locale& inLocale, |
|
755 UnicodeString& result); |
|
756 |
|
757 /** |
|
758 * Returns the filter used by this transliterator, or <tt>NULL</tt> |
|
759 * if this transliterator uses no filter. |
|
760 * @return the filter used by this transliterator, or <tt>NULL</tt> |
|
761 * if this transliterator uses no filter. |
|
762 * @stable ICU 2.0 |
|
763 */ |
|
764 const UnicodeFilter* getFilter(void) const; |
|
765 |
|
766 /** |
|
767 * Returns the filter used by this transliterator, or <tt>NULL</tt> if this |
|
768 * transliterator uses no filter. The caller must eventually delete the |
|
769 * result. After this call, this transliterator's filter is set to |
|
770 * <tt>NULL</tt>. |
|
771 * @return the filter used by this transliterator, or <tt>NULL</tt> if this |
|
772 * transliterator uses no filter. |
|
773 * @stable ICU 2.4 |
|
774 */ |
|
775 UnicodeFilter* orphanFilter(void); |
|
776 |
|
777 /** |
|
778 * Changes the filter used by this transliterator. If the filter |
|
779 * is set to <tt>null</tt> then no filtering will occur. |
|
780 * |
|
781 * <p>Callers must take care if a transliterator is in use by |
|
782 * multiple threads. The filter should not be changed by one |
|
783 * thread while another thread may be transliterating. |
|
784 * @param adoptedFilter the new filter to be adopted. |
|
785 * @stable ICU 2.0 |
|
786 */ |
|
787 void adoptFilter(UnicodeFilter* adoptedFilter); |
|
788 |
|
789 /** |
|
790 * Returns this transliterator's inverse. See the class |
|
791 * documentation for details. This implementation simply inverts |
|
792 * the two entities in the ID and attempts to retrieve the |
|
793 * resulting transliterator. That is, if <code>getID()</code> |
|
794 * returns "A-B", then this method will return the result of |
|
795 * <code>createInstance("B-A")</code>, or <code>null</code> if that |
|
796 * call fails. |
|
797 * |
|
798 * <p>Subclasses with knowledge of their inverse may wish to |
|
799 * override this method. |
|
800 * |
|
801 * @param status Output param to filled in with a success or an error. |
|
802 * @return a transliterator that is an inverse, not necessarily |
|
803 * exact, of this transliterator, or <code>null</code> if no such |
|
804 * transliterator is registered. |
|
805 * @see #registerInstance |
|
806 * @stable ICU 2.0 |
|
807 */ |
|
808 Transliterator* createInverse(UErrorCode& status) const; |
|
809 |
|
810 /** |
|
811 * Returns a <code>Transliterator</code> object given its ID. |
|
812 * The ID must be either a system transliterator ID or a ID registered |
|
813 * using <code>registerInstance()</code>. |
|
814 * |
|
815 * @param ID a valid ID, as enumerated by <code>getAvailableIDs()</code> |
|
816 * @param dir either FORWARD or REVERSE. |
|
817 * @param parseError Struct to recieve information on position |
|
818 * of error if an error is encountered |
|
819 * @param status Output param to filled in with a success or an error. |
|
820 * @return A <code>Transliterator</code> object with the given ID |
|
821 * @see #registerInstance |
|
822 * @see #getAvailableIDs |
|
823 * @see #getID |
|
824 * @stable ICU 2.0 |
|
825 */ |
|
826 static Transliterator* U_EXPORT2 createInstance(const UnicodeString& ID, |
|
827 UTransDirection dir, |
|
828 UParseError& parseError, |
|
829 UErrorCode& status); |
|
830 |
|
831 /** |
|
832 * Returns a <code>Transliterator</code> object given its ID. |
|
833 * The ID must be either a system transliterator ID or a ID registered |
|
834 * using <code>registerInstance()</code>. |
|
835 * @param ID a valid ID, as enumerated by <code>getAvailableIDs()</code> |
|
836 * @param dir either FORWARD or REVERSE. |
|
837 * @param status Output param to filled in with a success or an error. |
|
838 * @return A <code>Transliterator</code> object with the given ID |
|
839 * @stable ICU 2.0 |
|
840 */ |
|
841 static Transliterator* U_EXPORT2 createInstance(const UnicodeString& ID, |
|
842 UTransDirection dir, |
|
843 UErrorCode& status); |
|
844 |
|
845 /** |
|
846 * Returns a <code>Transliterator</code> object constructed from |
|
847 * the given rule string. This will be a RuleBasedTransliterator, |
|
848 * if the rule string contains only rules, or a |
|
849 * CompoundTransliterator, if it contains ID blocks, or a |
|
850 * NullTransliterator, if it contains ID blocks which parse as |
|
851 * empty for the given direction. |
|
852 * @param ID the id for the transliterator. |
|
853 * @param rules rules, separated by ';' |
|
854 * @param dir either FORWARD or REVERSE. |
|
855 * @param parseError Struct to recieve information on position |
|
856 * of error if an error is encountered |
|
857 * @param status Output param set to success/failure code. |
|
858 * @stable ICU 2.0 |
|
859 */ |
|
860 static Transliterator* U_EXPORT2 createFromRules(const UnicodeString& ID, |
|
861 const UnicodeString& rules, |
|
862 UTransDirection dir, |
|
863 UParseError& parseError, |
|
864 UErrorCode& status); |
|
865 |
|
866 /** |
|
867 * Create a rule string that can be passed to createFromRules() |
|
868 * to recreate this transliterator. |
|
869 * @param result the string to receive the rules. Previous |
|
870 * contents will be deleted. |
|
871 * @param escapeUnprintable if TRUE then convert unprintable |
|
872 * character to their hex escape representations, \\uxxxx or |
|
873 * \\Uxxxxxxxx. Unprintable characters are those other than |
|
874 * U+000A, U+0020..U+007E. |
|
875 * @stable ICU 2.0 |
|
876 */ |
|
877 virtual UnicodeString& toRules(UnicodeString& result, |
|
878 UBool escapeUnprintable) const; |
|
879 |
|
880 /** |
|
881 * Return the number of elements that make up this transliterator. |
|
882 * For example, if the transliterator "NFD;Jamo-Latin;Latin-Greek" |
|
883 * were created, the return value of this method would be 3. |
|
884 * |
|
885 * <p>If this transliterator is not composed of other |
|
886 * transliterators, then this method returns 1. |
|
887 * @return the number of transliterators that compose this |
|
888 * transliterator, or 1 if this transliterator is not composed of |
|
889 * multiple transliterators |
|
890 * @stable ICU 3.0 |
|
891 */ |
|
892 int32_t countElements() const; |
|
893 |
|
894 /** |
|
895 * Return an element that makes up this transliterator. For |
|
896 * example, if the transliterator "NFD;Jamo-Latin;Latin-Greek" |
|
897 * were created, the return value of this method would be one |
|
898 * of the three transliterator objects that make up that |
|
899 * transliterator: [NFD, Jamo-Latin, Latin-Greek]. |
|
900 * |
|
901 * <p>If this transliterator is not composed of other |
|
902 * transliterators, then this method will return a reference to |
|
903 * this transliterator when given the index 0. |
|
904 * @param index a value from 0..countElements()-1 indicating the |
|
905 * transliterator to return |
|
906 * @param ec input-output error code |
|
907 * @return one of the transliterators that makes up this |
|
908 * transliterator, if this transliterator is made up of multiple |
|
909 * transliterators, otherwise a reference to this object if given |
|
910 * an index of 0 |
|
911 * @stable ICU 3.0 |
|
912 */ |
|
913 const Transliterator& getElement(int32_t index, UErrorCode& ec) const; |
|
914 |
|
915 /** |
|
916 * Returns the set of all characters that may be modified in the |
|
917 * input text by this Transliterator. This incorporates this |
|
918 * object's current filter; if the filter is changed, the return |
|
919 * value of this function will change. The default implementation |
|
920 * returns an empty set. Some subclasses may override {@link |
|
921 * #handleGetSourceSet } to return a more precise result. The |
|
922 * return result is approximate in any case and is intended for |
|
923 * use by tests, tools, or utilities. |
|
924 * @param result receives result set; previous contents lost |
|
925 * @return a reference to result |
|
926 * @see #getTargetSet |
|
927 * @see #handleGetSourceSet |
|
928 * @stable ICU 2.4 |
|
929 */ |
|
930 UnicodeSet& getSourceSet(UnicodeSet& result) const; |
|
931 |
|
932 /** |
|
933 * Framework method that returns the set of all characters that |
|
934 * may be modified in the input text by this Transliterator, |
|
935 * ignoring the effect of this object's filter. The base class |
|
936 * implementation returns the empty set. Subclasses that wish to |
|
937 * implement this should override this method. |
|
938 * @return the set of characters that this transliterator may |
|
939 * modify. The set may be modified, so subclasses should return a |
|
940 * newly-created object. |
|
941 * @param result receives result set; previous contents lost |
|
942 * @see #getSourceSet |
|
943 * @see #getTargetSet |
|
944 * @stable ICU 2.4 |
|
945 */ |
|
946 virtual void handleGetSourceSet(UnicodeSet& result) const; |
|
947 |
|
948 /** |
|
949 * Returns the set of all characters that may be generated as |
|
950 * replacement text by this transliterator. The default |
|
951 * implementation returns the empty set. Some subclasses may |
|
952 * override this method to return a more precise result. The |
|
953 * return result is approximate in any case and is intended for |
|
954 * use by tests, tools, or utilities requiring such |
|
955 * meta-information. |
|
956 * @param result receives result set; previous contents lost |
|
957 * @return a reference to result |
|
958 * @see #getTargetSet |
|
959 * @stable ICU 2.4 |
|
960 */ |
|
961 virtual UnicodeSet& getTargetSet(UnicodeSet& result) const; |
|
962 |
|
963 public: |
|
964 |
|
965 /** |
|
966 * Registers a factory function that creates transliterators of |
|
967 * a given ID. |
|
968 * @param id the ID being registered |
|
969 * @param factory a function pointer that will be copied and |
|
970 * called later when the given ID is passed to createInstance() |
|
971 * @param context a context pointer that will be stored and |
|
972 * later passed to the factory function when an ID matching |
|
973 * the registration ID is being instantiated with this factory. |
|
974 * @stable ICU 2.0 |
|
975 */ |
|
976 static void U_EXPORT2 registerFactory(const UnicodeString& id, |
|
977 Factory factory, |
|
978 Token context); |
|
979 |
|
980 /** |
|
981 * Registers an instance <tt>obj</tt> of a subclass of |
|
982 * <code>Transliterator</code> with the system. When |
|
983 * <tt>createInstance()</tt> is called with an ID string that is |
|
984 * equal to <tt>obj->getID()</tt>, then <tt>obj->clone()</tt> is |
|
985 * returned. |
|
986 * |
|
987 * After this call the Transliterator class owns the adoptedObj |
|
988 * and will delete it. |
|
989 * |
|
990 * @param adoptedObj an instance of subclass of |
|
991 * <code>Transliterator</code> that defines <tt>clone()</tt> |
|
992 * @see #createInstance |
|
993 * @see #registerFactory |
|
994 * @see #unregister |
|
995 * @stable ICU 2.0 |
|
996 */ |
|
997 static void U_EXPORT2 registerInstance(Transliterator* adoptedObj); |
|
998 |
|
999 /** |
|
1000 * Registers an ID string as an alias of another ID string. |
|
1001 * That is, after calling this function, <tt>createInstance(aliasID)</tt> |
|
1002 * will return the same thing as <tt>createInstance(realID)</tt>. |
|
1003 * This is generally used to create shorter, more mnemonic aliases |
|
1004 * for long compound IDs. |
|
1005 * |
|
1006 * @param aliasID The new ID being registered. |
|
1007 * @param realID The ID that the new ID is to be an alias for. |
|
1008 * This can be a compound ID and can include filters and should |
|
1009 * refer to transliterators that have already been registered with |
|
1010 * the framework, although this isn't checked. |
|
1011 * @stable ICU 3.6 |
|
1012 */ |
|
1013 static void U_EXPORT2 registerAlias(const UnicodeString& aliasID, |
|
1014 const UnicodeString& realID); |
|
1015 |
|
1016 protected: |
|
1017 |
|
1018 #ifndef U_HIDE_INTERNAL_API |
|
1019 /** |
|
1020 * @internal |
|
1021 * @param id the ID being registered |
|
1022 * @param factory a function pointer that will be copied and |
|
1023 * called later when the given ID is passed to createInstance() |
|
1024 * @param context a context pointer that will be stored and |
|
1025 * later passed to the factory function when an ID matching |
|
1026 * the registration ID is being instantiated with this factory. |
|
1027 */ |
|
1028 static void _registerFactory(const UnicodeString& id, |
|
1029 Factory factory, |
|
1030 Token context); |
|
1031 |
|
1032 /** |
|
1033 * @internal |
|
1034 */ |
|
1035 static void _registerInstance(Transliterator* adoptedObj); |
|
1036 |
|
1037 /** |
|
1038 * @internal |
|
1039 */ |
|
1040 static void _registerAlias(const UnicodeString& aliasID, const UnicodeString& realID); |
|
1041 |
|
1042 /** |
|
1043 * Register two targets as being inverses of one another. For |
|
1044 * example, calling registerSpecialInverse("NFC", "NFD", true) causes |
|
1045 * Transliterator to form the following inverse relationships: |
|
1046 * |
|
1047 * <pre>NFC => NFD |
|
1048 * Any-NFC => Any-NFD |
|
1049 * NFD => NFC |
|
1050 * Any-NFD => Any-NFC</pre> |
|
1051 * |
|
1052 * (Without the special inverse registration, the inverse of NFC |
|
1053 * would be NFC-Any.) Note that NFD is shorthand for Any-NFD, but |
|
1054 * that the presence or absence of "Any-" is preserved. |
|
1055 * |
|
1056 * <p>The relationship is symmetrical; registering (a, b) is |
|
1057 * equivalent to registering (b, a). |
|
1058 * |
|
1059 * <p>The relevant IDs must still be registered separately as |
|
1060 * factories or classes. |
|
1061 * |
|
1062 * <p>Only the targets are specified. Special inverses always |
|
1063 * have the form Any-Target1 <=> Any-Target2. The target should |
|
1064 * have canonical casing (the casing desired to be produced when |
|
1065 * an inverse is formed) and should contain no whitespace or other |
|
1066 * extraneous characters. |
|
1067 * |
|
1068 * @param target the target against which to register the inverse |
|
1069 * @param inverseTarget the inverse of target, that is |
|
1070 * Any-target.getInverse() => Any-inverseTarget |
|
1071 * @param bidirectional if true, register the reverse relation |
|
1072 * as well, that is, Any-inverseTarget.getInverse() => Any-target |
|
1073 * @internal |
|
1074 */ |
|
1075 static void _registerSpecialInverse(const UnicodeString& target, |
|
1076 const UnicodeString& inverseTarget, |
|
1077 UBool bidirectional); |
|
1078 #endif /* U_HIDE_INTERNAL_API */ |
|
1079 |
|
1080 public: |
|
1081 |
|
1082 /** |
|
1083 * Unregisters a transliterator or class. This may be either |
|
1084 * a system transliterator or a user transliterator or class. |
|
1085 * Any attempt to construct an unregistered transliterator based |
|
1086 * on its ID will fail. |
|
1087 * |
|
1088 * @param ID the ID of the transliterator or class |
|
1089 * @return the <code>Object</code> that was registered with |
|
1090 * <code>ID</code>, or <code>null</code> if none was |
|
1091 * @see #registerInstance |
|
1092 * @see #registerFactory |
|
1093 * @stable ICU 2.0 |
|
1094 */ |
|
1095 static void U_EXPORT2 unregister(const UnicodeString& ID); |
|
1096 |
|
1097 public: |
|
1098 |
|
1099 /** |
|
1100 * Return a StringEnumeration over the IDs available at the time of the |
|
1101 * call, including user-registered IDs. |
|
1102 * @param ec input-output error code |
|
1103 * @return a newly-created StringEnumeration over the transliterators |
|
1104 * available at the time of the call. The caller should delete this object |
|
1105 * when done using it. |
|
1106 * @stable ICU 3.0 |
|
1107 */ |
|
1108 static StringEnumeration* U_EXPORT2 getAvailableIDs(UErrorCode& ec); |
|
1109 |
|
1110 /** |
|
1111 * Return the number of registered source specifiers. |
|
1112 * @return the number of registered source specifiers. |
|
1113 * @stable ICU 2.0 |
|
1114 */ |
|
1115 static int32_t U_EXPORT2 countAvailableSources(void); |
|
1116 |
|
1117 /** |
|
1118 * Return a registered source specifier. |
|
1119 * @param index which specifier to return, from 0 to n-1, where |
|
1120 * n = countAvailableSources() |
|
1121 * @param result fill-in paramter to receive the source specifier. |
|
1122 * If index is out of range, result will be empty. |
|
1123 * @return reference to result |
|
1124 * @stable ICU 2.0 |
|
1125 */ |
|
1126 static UnicodeString& U_EXPORT2 getAvailableSource(int32_t index, |
|
1127 UnicodeString& result); |
|
1128 |
|
1129 /** |
|
1130 * Return the number of registered target specifiers for a given |
|
1131 * source specifier. |
|
1132 * @param source the given source specifier. |
|
1133 * @return the number of registered target specifiers for a given |
|
1134 * source specifier. |
|
1135 * @stable ICU 2.0 |
|
1136 */ |
|
1137 static int32_t U_EXPORT2 countAvailableTargets(const UnicodeString& source); |
|
1138 |
|
1139 /** |
|
1140 * Return a registered target specifier for a given source. |
|
1141 * @param index which specifier to return, from 0 to n-1, where |
|
1142 * n = countAvailableTargets(source) |
|
1143 * @param source the source specifier |
|
1144 * @param result fill-in paramter to receive the target specifier. |
|
1145 * If source is invalid or if index is out of range, result will |
|
1146 * be empty. |
|
1147 * @return reference to result |
|
1148 * @stable ICU 2.0 |
|
1149 */ |
|
1150 static UnicodeString& U_EXPORT2 getAvailableTarget(int32_t index, |
|
1151 const UnicodeString& source, |
|
1152 UnicodeString& result); |
|
1153 |
|
1154 /** |
|
1155 * Return the number of registered variant specifiers for a given |
|
1156 * source-target pair. |
|
1157 * @param source the source specifiers. |
|
1158 * @param target the target specifiers. |
|
1159 * @stable ICU 2.0 |
|
1160 */ |
|
1161 static int32_t U_EXPORT2 countAvailableVariants(const UnicodeString& source, |
|
1162 const UnicodeString& target); |
|
1163 |
|
1164 /** |
|
1165 * Return a registered variant specifier for a given source-target |
|
1166 * pair. |
|
1167 * @param index which specifier to return, from 0 to n-1, where |
|
1168 * n = countAvailableVariants(source, target) |
|
1169 * @param source the source specifier |
|
1170 * @param target the target specifier |
|
1171 * @param result fill-in paramter to receive the variant |
|
1172 * specifier. If source is invalid or if target is invalid or if |
|
1173 * index is out of range, result will be empty. |
|
1174 * @return reference to result |
|
1175 * @stable ICU 2.0 |
|
1176 */ |
|
1177 static UnicodeString& U_EXPORT2 getAvailableVariant(int32_t index, |
|
1178 const UnicodeString& source, |
|
1179 const UnicodeString& target, |
|
1180 UnicodeString& result); |
|
1181 |
|
1182 protected: |
|
1183 |
|
1184 #ifndef U_HIDE_INTERNAL_API |
|
1185 /** |
|
1186 * Non-mutexed internal method |
|
1187 * @internal |
|
1188 */ |
|
1189 static int32_t _countAvailableSources(void); |
|
1190 |
|
1191 /** |
|
1192 * Non-mutexed internal method |
|
1193 * @internal |
|
1194 */ |
|
1195 static UnicodeString& _getAvailableSource(int32_t index, |
|
1196 UnicodeString& result); |
|
1197 |
|
1198 /** |
|
1199 * Non-mutexed internal method |
|
1200 * @internal |
|
1201 */ |
|
1202 static int32_t _countAvailableTargets(const UnicodeString& source); |
|
1203 |
|
1204 /** |
|
1205 * Non-mutexed internal method |
|
1206 * @internal |
|
1207 */ |
|
1208 static UnicodeString& _getAvailableTarget(int32_t index, |
|
1209 const UnicodeString& source, |
|
1210 UnicodeString& result); |
|
1211 |
|
1212 /** |
|
1213 * Non-mutexed internal method |
|
1214 * @internal |
|
1215 */ |
|
1216 static int32_t _countAvailableVariants(const UnicodeString& source, |
|
1217 const UnicodeString& target); |
|
1218 |
|
1219 /** |
|
1220 * Non-mutexed internal method |
|
1221 * @internal |
|
1222 */ |
|
1223 static UnicodeString& _getAvailableVariant(int32_t index, |
|
1224 const UnicodeString& source, |
|
1225 const UnicodeString& target, |
|
1226 UnicodeString& result); |
|
1227 #endif /* U_HIDE_INTERNAL_API */ |
|
1228 |
|
1229 protected: |
|
1230 |
|
1231 /** |
|
1232 * Set the ID of this transliterators. Subclasses shouldn't do |
|
1233 * this, unless the underlying script behavior has changed. |
|
1234 * @param id the new id t to be set. |
|
1235 * @stable ICU 2.4 |
|
1236 */ |
|
1237 void setID(const UnicodeString& id); |
|
1238 |
|
1239 public: |
|
1240 |
|
1241 /** |
|
1242 * Return the class ID for this class. This is useful only for |
|
1243 * comparing to a return value from getDynamicClassID(). |
|
1244 * Note that Transliterator is an abstract base class, and therefor |
|
1245 * no fully constructed object will have a dynamic |
|
1246 * UCLassID that equals the UClassID returned from |
|
1247 * TRansliterator::getStaticClassID(). |
|
1248 * @return The class ID for class Transliterator. |
|
1249 * @stable ICU 2.0 |
|
1250 */ |
|
1251 static UClassID U_EXPORT2 getStaticClassID(void); |
|
1252 |
|
1253 /** |
|
1254 * Returns a unique class ID <b>polymorphically</b>. This method |
|
1255 * is to implement a simple version of RTTI, since not all C++ |
|
1256 * compilers support genuine RTTI. Polymorphic operator==() and |
|
1257 * clone() methods call this method. |
|
1258 * |
|
1259 * <p>Concrete subclasses of Transliterator must use the |
|
1260 * UOBJECT_DEFINE_RTTI_IMPLEMENTATION macro from |
|
1261 * uobject.h to provide the RTTI functions. |
|
1262 * |
|
1263 * @return The class ID for this object. All objects of a given |
|
1264 * class have the same class ID. Objects of other classes have |
|
1265 * different class IDs. |
|
1266 * @stable ICU 2.0 |
|
1267 */ |
|
1268 virtual UClassID getDynamicClassID(void) const = 0; |
|
1269 |
|
1270 private: |
|
1271 static UBool initializeRegistry(UErrorCode &status); |
|
1272 |
|
1273 public: |
|
1274 #ifndef U_HIDE_OBSOLETE_API |
|
1275 /** |
|
1276 * Return the number of IDs currently registered with the system. |
|
1277 * To retrieve the actual IDs, call getAvailableID(i) with |
|
1278 * i from 0 to countAvailableIDs() - 1. |
|
1279 * @return the number of IDs currently registered with the system. |
|
1280 * @obsolete ICU 3.4 use getAvailableIDs() instead |
|
1281 */ |
|
1282 static int32_t U_EXPORT2 countAvailableIDs(void); |
|
1283 |
|
1284 /** |
|
1285 * Return the index-th available ID. index must be between 0 |
|
1286 * and countAvailableIDs() - 1, inclusive. If index is out of |
|
1287 * range, the result of getAvailableID(0) is returned. |
|
1288 * @param index the given ID index. |
|
1289 * @return the index-th available ID. index must be between 0 |
|
1290 * and countAvailableIDs() - 1, inclusive. If index is out of |
|
1291 * range, the result of getAvailableID(0) is returned. |
|
1292 * @obsolete ICU 3.4 use getAvailableIDs() instead; this function |
|
1293 * is not thread safe, since it returns a reference to storage that |
|
1294 * may become invalid if another thread calls unregister |
|
1295 */ |
|
1296 static const UnicodeString& U_EXPORT2 getAvailableID(int32_t index); |
|
1297 #endif /* U_HIDE_OBSOLETE_API */ |
|
1298 }; |
|
1299 |
|
1300 inline int32_t Transliterator::getMaximumContextLength(void) const { |
|
1301 return maximumContextLength; |
|
1302 } |
|
1303 |
|
1304 inline void Transliterator::setID(const UnicodeString& id) { |
|
1305 ID = id; |
|
1306 // NUL-terminate the ID string, which is a non-aliased copy. |
|
1307 ID.append((UChar)0); |
|
1308 ID.truncate(ID.length()-1); |
|
1309 } |
|
1310 |
|
1311 #ifndef U_HIDE_INTERNAL_API |
|
1312 inline Transliterator::Token Transliterator::integerToken(int32_t i) { |
|
1313 Token t; |
|
1314 t.integer = i; |
|
1315 return t; |
|
1316 } |
|
1317 |
|
1318 inline Transliterator::Token Transliterator::pointerToken(void* p) { |
|
1319 Token t; |
|
1320 t.pointer = p; |
|
1321 return t; |
|
1322 } |
|
1323 #endif /* U_HIDE_INTERNAL_API */ |
|
1324 |
|
1325 U_NAMESPACE_END |
|
1326 |
|
1327 #endif /* #if !UCONFIG_NO_TRANSLITERATION */ |
|
1328 |
|
1329 #endif |