1 | // © 2016 and later: Unicode, Inc. and others. |
2 | // License & terms of use: http://www.unicode.org/copyright.html |
3 | /* |
4 | ****************************************************************************** |
5 | * |
6 | * Copyright (C) 1999-2013, International Business Machines |
7 | * Corporation and others. All Rights Reserved. |
8 | * |
9 | ****************************************************************************** |
10 | * file name: ubidi.h |
11 | * encoding: UTF-8 |
12 | * tab size: 8 (not used) |
13 | * indentation:4 |
14 | * |
15 | * created on: 1999jul27 |
16 | * created by: Markus W. Scherer, updated by Matitiahu Allouche |
17 | */ |
18 | |
19 | #ifndef UBIDI_H |
20 | #define UBIDI_H |
21 | |
22 | #include "unicode/utypes.h" |
23 | #include "unicode/uchar.h" |
24 | |
25 | #if U_SHOW_CPLUSPLUS_API |
26 | #include "unicode/localpointer.h" |
27 | #endif // U_SHOW_CPLUSPLUS_API |
28 | |
29 | /** |
30 | *\file |
31 | * \brief C API: Bidi algorithm |
32 | * |
33 | * <h2>Bidi algorithm for ICU</h2> |
34 | * |
35 | * This is an implementation of the Unicode Bidirectional Algorithm. |
36 | * The algorithm is defined in the |
37 | * <a href="http://www.unicode.org/unicode/reports/tr9/">Unicode Standard Annex #9</a>.<p> |
38 | * |
39 | * Note: Libraries that perform a bidirectional algorithm and |
40 | * reorder strings accordingly are sometimes called "Storage Layout Engines". |
41 | * ICU's Bidi and shaping (u_shapeArabic()) APIs can be used at the core of such |
42 | * "Storage Layout Engines". |
43 | * |
44 | * <h3>General remarks about the API:</h3> |
45 | * |
46 | * In functions with an error code parameter, |
47 | * the <code>pErrorCode</code> pointer must be valid |
48 | * and the value that it points to must not indicate a failure before |
49 | * the function call. Otherwise, the function returns immediately. |
50 | * After the function call, the value indicates success or failure.<p> |
51 | * |
52 | * The "limit" of a sequence of characters is the position just after their |
53 | * last character, i.e., one more than that position.<p> |
54 | * |
55 | * Some of the API functions provide access to "runs". |
56 | * Such a "run" is defined as a sequence of characters |
57 | * that are at the same embedding level |
58 | * after performing the Bidi algorithm.<p> |
59 | * |
60 | * @author Markus W. Scherer |
61 | * @version 1.0 |
62 | * |
63 | * |
64 | * <h4> Sample code for the ICU Bidi API </h4> |
65 | * |
66 | * <h5>Rendering a paragraph with the ICU Bidi API</h5> |
67 | * |
68 | * This is (hypothetical) sample code that illustrates |
69 | * how the ICU Bidi API could be used to render a paragraph of text. |
70 | * Rendering code depends highly on the graphics system, |
71 | * therefore this sample code must make a lot of assumptions, |
72 | * which may or may not match any existing graphics system's properties. |
73 | * |
74 | * <p>The basic assumptions are:</p> |
75 | * <ul> |
76 | * <li>Rendering is done from left to right on a horizontal line.</li> |
77 | * <li>A run of single-style, unidirectional text can be rendered at once.</li> |
78 | * <li>Such a run of text is passed to the graphics system with |
79 | * characters (code units) in logical order.</li> |
80 | * <li>The line-breaking algorithm is very complicated |
81 | * and Locale-dependent - |
82 | * and therefore its implementation omitted from this sample code.</li> |
83 | * </ul> |
84 | * |
85 | * <pre> |
86 | * \code |
87 | *#include <unicode/ubidi.h> |
88 | * |
89 | *typedef enum { |
90 | * styleNormal=0, styleSelected=1, |
91 | * styleBold=2, styleItalics=4, |
92 | * styleSuper=8, styleSub=16 |
93 | *} Style; |
94 | * |
95 | *typedef struct { int32_t limit; Style style; } StyleRun; |
96 | * |
97 | *int getTextWidth(const UChar *text, int32_t start, int32_t limit, |
98 | * const StyleRun *styleRuns, int styleRunCount); |
99 | * |
100 | * // set *pLimit and *pStyleRunLimit for a line |
101 | * // from text[start] and from styleRuns[styleRunStart] |
102 | * // using ubidi_getLogicalRun(para, ...) |
103 | *void getLineBreak(const UChar *text, int32_t start, int32_t *pLimit, |
104 | * UBiDi *para, |
105 | * const StyleRun *styleRuns, int styleRunStart, int *pStyleRunLimit, |
106 | * int *pLineWidth); |
107 | * |
108 | * // render runs on a line sequentially, always from left to right |
109 | * |
110 | * // prepare rendering a new line |
111 | * void startLine(UBiDiDirection textDirection, int lineWidth); |
112 | * |
113 | * // render a run of text and advance to the right by the run width |
114 | * // the text[start..limit-1] is always in logical order |
115 | * void renderRun(const UChar *text, int32_t start, int32_t limit, |
116 | * UBiDiDirection textDirection, Style style); |
117 | * |
118 | * // We could compute a cross-product |
119 | * // from the style runs with the directional runs |
120 | * // and then reorder it. |
121 | * // Instead, here we iterate over each run type |
122 | * // and render the intersections - |
123 | * // with shortcuts in simple (and common) cases. |
124 | * // renderParagraph() is the main function. |
125 | * |
126 | * // render a directional run with |
127 | * // (possibly) multiple style runs intersecting with it |
128 | * void renderDirectionalRun(const UChar *text, |
129 | * int32_t start, int32_t limit, |
130 | * UBiDiDirection direction, |
131 | * const StyleRun *styleRuns, int styleRunCount) { |
132 | * int i; |
133 | * |
134 | * // iterate over style runs |
135 | * if(direction==UBIDI_LTR) { |
136 | * int styleLimit; |
137 | * |
138 | * for(i=0; i<styleRunCount; ++i) { |
139 | * styleLimit=styleRuns[i].limit; |
140 | * if(start<styleLimit) { |
141 | * if(styleLimit>limit) { styleLimit=limit; } |
142 | * renderRun(text, start, styleLimit, |
143 | * direction, styleRuns[i].style); |
144 | * if(styleLimit==limit) { break; } |
145 | * start=styleLimit; |
146 | * } |
147 | * } |
148 | * } else { |
149 | * int styleStart; |
150 | * |
151 | * for(i=styleRunCount-1; i>=0; --i) { |
152 | * if(i>0) { |
153 | * styleStart=styleRuns[i-1].limit; |
154 | * } else { |
155 | * styleStart=0; |
156 | * } |
157 | * if(limit>=styleStart) { |
158 | * if(styleStart<start) { styleStart=start; } |
159 | * renderRun(text, styleStart, limit, |
160 | * direction, styleRuns[i].style); |
161 | * if(styleStart==start) { break; } |
162 | * limit=styleStart; |
163 | * } |
164 | * } |
165 | * } |
166 | * } |
167 | * |
168 | * // the line object represents text[start..limit-1] |
169 | * void renderLine(UBiDi *line, const UChar *text, |
170 | * int32_t start, int32_t limit, |
171 | * const StyleRun *styleRuns, int styleRunCount, |
172 | * UErrorCode *pErrorCode) { |
173 | * UBiDiDirection direction=ubidi_getDirection(line); |
174 | * if(direction!=UBIDI_MIXED) { |
175 | * // unidirectional |
176 | * if(styleRunCount<=1) { |
177 | * renderRun(text, start, limit, direction, styleRuns[0].style); |
178 | * } else { |
179 | * renderDirectionalRun(text, start, limit, |
180 | * direction, styleRuns, styleRunCount); |
181 | * } |
182 | * } else { |
183 | * // mixed-directional |
184 | * int32_t count, i, length; |
185 | * UBiDiLevel level; |
186 | * |
187 | * count=ubidi_countRuns(line, pErrorCode); |
188 | * if(U_SUCCESS(*pErrorCode)) { |
189 | * if(styleRunCount<=1) { |
190 | * Style style=styleRuns[0].style; |
191 | * |
192 | * // iterate over directional runs |
193 | * for(i=0; i<count; ++i) { |
194 | * direction=ubidi_getVisualRun(line, i, &start, &length); |
195 | * renderRun(text, start, start+length, direction, style); |
196 | * } |
197 | * } else { |
198 | * int32_t j; |
199 | * |
200 | * // iterate over both directional and style runs |
201 | * for(i=0; i<count; ++i) { |
202 | * direction=ubidi_getVisualRun(line, i, &start, &length); |
203 | * renderDirectionalRun(text, start, start+length, |
204 | * direction, styleRuns, styleRunCount); |
205 | * } |
206 | * } |
207 | * } |
208 | * } |
209 | * } |
210 | * |
211 | *void renderParagraph(const UChar *text, int32_t length, |
212 | * UBiDiDirection textDirection, |
213 | * const StyleRun *styleRuns, int styleRunCount, |
214 | * int lineWidth, |
215 | * UErrorCode *pErrorCode) { |
216 | * UBiDi *para; |
217 | * |
218 | * if(pErrorCode==NULL || U_FAILURE(*pErrorCode) || length<=0) { |
219 | * return; |
220 | * } |
221 | * |
222 | * para=ubidi_openSized(length, 0, pErrorCode); |
223 | * if(para==NULL) { return; } |
224 | * |
225 | * ubidi_setPara(para, text, length, |
226 | * textDirection ? UBIDI_DEFAULT_RTL : UBIDI_DEFAULT_LTR, |
227 | * NULL, pErrorCode); |
228 | * if(U_SUCCESS(*pErrorCode)) { |
229 | * UBiDiLevel paraLevel=1&ubidi_getParaLevel(para); |
230 | * StyleRun styleRun={ length, styleNormal }; |
231 | * int width; |
232 | * |
233 | * if(styleRuns==NULL || styleRunCount<=0) { |
234 | * styleRunCount=1; |
235 | * styleRuns=&styleRun; |
236 | * } |
237 | * |
238 | * // assume styleRuns[styleRunCount-1].limit>=length |
239 | * |
240 | * width=getTextWidth(text, 0, length, styleRuns, styleRunCount); |
241 | * if(width<=lineWidth) { |
242 | * // everything fits onto one line |
243 | * |
244 | * // prepare rendering a new line from either left or right |
245 | * startLine(paraLevel, width); |
246 | * |
247 | * renderLine(para, text, 0, length, |
248 | * styleRuns, styleRunCount, pErrorCode); |
249 | * } else { |
250 | * UBiDi *line; |
251 | * |
252 | * // we need to render several lines |
253 | * line=ubidi_openSized(length, 0, pErrorCode); |
254 | * if(line!=NULL) { |
255 | * int32_t start=0, limit; |
256 | * int styleRunStart=0, styleRunLimit; |
257 | * |
258 | * for(;;) { |
259 | * limit=length; |
260 | * styleRunLimit=styleRunCount; |
261 | * getLineBreak(text, start, &limit, para, |
262 | * styleRuns, styleRunStart, &styleRunLimit, |
263 | * &width); |
264 | * ubidi_setLine(para, start, limit, line, pErrorCode); |
265 | * if(U_SUCCESS(*pErrorCode)) { |
266 | * // prepare rendering a new line |
267 | * // from either left or right |
268 | * startLine(paraLevel, width); |
269 | * |
270 | * renderLine(line, text, start, limit, |
271 | * styleRuns+styleRunStart, |
272 | * styleRunLimit-styleRunStart, pErrorCode); |
273 | * } |
274 | * if(limit==length) { break; } |
275 | * start=limit; |
276 | * styleRunStart=styleRunLimit-1; |
277 | * if(start>=styleRuns[styleRunStart].limit) { |
278 | * ++styleRunStart; |
279 | * } |
280 | * } |
281 | * |
282 | * ubidi_close(line); |
283 | * } |
284 | * } |
285 | * } |
286 | * |
287 | * ubidi_close(para); |
288 | *} |
289 | *\endcode |
290 | * </pre> |
291 | */ |
292 | |
293 | /*DOCXX_TAG*/ |
294 | /*@{*/ |
295 | |
296 | /** |
297 | * UBiDiLevel is the type of the level values in this |
298 | * Bidi implementation. |
299 | * It holds an embedding level and indicates the visual direction |
300 | * by its bit 0 (even/odd value).<p> |
301 | * |
302 | * It can also hold non-level values for the |
303 | * <code>paraLevel</code> and <code>embeddingLevels</code> |
304 | * arguments of <code>ubidi_setPara()</code>; there: |
305 | * <ul> |
306 | * <li>bit 7 of an <code>embeddingLevels[]</code> |
307 | * value indicates whether the using application is |
308 | * specifying the level of a character to <i>override</i> whatever the |
309 | * Bidi implementation would resolve it to.</li> |
310 | * <li><code>paraLevel</code> can be set to the |
311 | * pseudo-level values <code>UBIDI_DEFAULT_LTR</code> |
312 | * and <code>UBIDI_DEFAULT_RTL</code>.</li> |
313 | * </ul> |
314 | * |
315 | * @see ubidi_setPara |
316 | * |
317 | * <p>The related constants are not real, valid level values. |
318 | * <code>UBIDI_DEFAULT_XXX</code> can be used to specify |
319 | * a default for the paragraph level for |
320 | * when the <code>ubidi_setPara()</code> function |
321 | * shall determine it but there is no |
322 | * strongly typed character in the input.<p> |
323 | * |
324 | * Note that the value for <code>UBIDI_DEFAULT_LTR</code> is even |
325 | * and the one for <code>UBIDI_DEFAULT_RTL</code> is odd, |
326 | * just like with normal LTR and RTL level values - |
327 | * these special values are designed that way. Also, the implementation |
328 | * assumes that UBIDI_MAX_EXPLICIT_LEVEL is odd. |
329 | * |
330 | * Note: The numeric values of the related constants will not change: |
331 | * They are tied to the use of 7-bit byte values (plus the override bit) |
332 | * and of the UBiDiLevel=uint8_t data type in this API. |
333 | * |
334 | * @see UBIDI_DEFAULT_LTR |
335 | * @see UBIDI_DEFAULT_RTL |
336 | * @see UBIDI_LEVEL_OVERRIDE |
337 | * @see UBIDI_MAX_EXPLICIT_LEVEL |
338 | * @stable ICU 2.0 |
339 | */ |
340 | typedef uint8_t UBiDiLevel; |
341 | |
342 | /** Paragraph level setting.<p> |
343 | * |
344 | * Constant indicating that the base direction depends on the first strong |
345 | * directional character in the text according to the Unicode Bidirectional |
346 | * Algorithm. If no strong directional character is present, |
347 | * then set the paragraph level to 0 (left-to-right).<p> |
348 | * |
349 | * If this value is used in conjunction with reordering modes |
350 | * <code>UBIDI_REORDER_INVERSE_LIKE_DIRECT</code> or |
351 | * <code>UBIDI_REORDER_INVERSE_FOR_NUMBERS_SPECIAL</code>, the text to reorder |
352 | * is assumed to be visual LTR, and the text after reordering is required |
353 | * to be the corresponding logical string with appropriate contextual |
354 | * direction. The direction of the result string will be RTL if either |
355 | * the righmost or leftmost strong character of the source text is RTL |
356 | * or Arabic Letter, the direction will be LTR otherwise.<p> |
357 | * |
358 | * If reordering option <code>UBIDI_OPTION_INSERT_MARKS</code> is set, an RLM may |
359 | * be added at the beginning of the result string to ensure round trip |
360 | * (that the result string, when reordered back to visual, will produce |
361 | * the original source text). |
362 | * @see UBIDI_REORDER_INVERSE_LIKE_DIRECT |
363 | * @see UBIDI_REORDER_INVERSE_FOR_NUMBERS_SPECIAL |
364 | * @stable ICU 2.0 |
365 | */ |
366 | #define UBIDI_DEFAULT_LTR 0xfe |
367 | |
368 | /** Paragraph level setting.<p> |
369 | * |
370 | * Constant indicating that the base direction depends on the first strong |
371 | * directional character in the text according to the Unicode Bidirectional |
372 | * Algorithm. If no strong directional character is present, |
373 | * then set the paragraph level to 1 (right-to-left).<p> |
374 | * |
375 | * If this value is used in conjunction with reordering modes |
376 | * <code>UBIDI_REORDER_INVERSE_LIKE_DIRECT</code> or |
377 | * <code>UBIDI_REORDER_INVERSE_FOR_NUMBERS_SPECIAL</code>, the text to reorder |
378 | * is assumed to be visual LTR, and the text after reordering is required |
379 | * to be the corresponding logical string with appropriate contextual |
380 | * direction. The direction of the result string will be RTL if either |
381 | * the righmost or leftmost strong character of the source text is RTL |
382 | * or Arabic Letter, or if the text contains no strong character; |
383 | * the direction will be LTR otherwise.<p> |
384 | * |
385 | * If reordering option <code>UBIDI_OPTION_INSERT_MARKS</code> is set, an RLM may |
386 | * be added at the beginning of the result string to ensure round trip |
387 | * (that the result string, when reordered back to visual, will produce |
388 | * the original source text). |
389 | * @see UBIDI_REORDER_INVERSE_LIKE_DIRECT |
390 | * @see UBIDI_REORDER_INVERSE_FOR_NUMBERS_SPECIAL |
391 | * @stable ICU 2.0 |
392 | */ |
393 | #define UBIDI_DEFAULT_RTL 0xff |
394 | |
395 | /** |
396 | * Maximum explicit embedding level. |
397 | * Same as the max_depth value in the |
398 | * <a href="http://www.unicode.org/reports/tr9/#BD2">Unicode Bidirectional Algorithm</a>. |
399 | * (The maximum resolved level can be up to <code>UBIDI_MAX_EXPLICIT_LEVEL+1</code>). |
400 | * @stable ICU 2.0 |
401 | */ |
402 | #define UBIDI_MAX_EXPLICIT_LEVEL 125 |
403 | |
404 | /** Bit flag for level input. |
405 | * Overrides directional properties. |
406 | * @stable ICU 2.0 |
407 | */ |
408 | #define UBIDI_LEVEL_OVERRIDE 0x80 |
409 | |
410 | /** |
411 | * Special value which can be returned by the mapping functions when a logical |
412 | * index has no corresponding visual index or vice-versa. This may happen |
413 | * for the logical-to-visual mapping of a Bidi control when option |
414 | * <code>#UBIDI_OPTION_REMOVE_CONTROLS</code> is specified. This can also happen |
415 | * for the visual-to-logical mapping of a Bidi mark (LRM or RLM) inserted |
416 | * by option <code>#UBIDI_OPTION_INSERT_MARKS</code>. |
417 | * @see ubidi_getVisualIndex |
418 | * @see ubidi_getVisualMap |
419 | * @see ubidi_getLogicalIndex |
420 | * @see ubidi_getLogicalMap |
421 | * @stable ICU 3.6 |
422 | */ |
423 | #define UBIDI_MAP_NOWHERE (-1) |
424 | |
425 | /** |
426 | * <code>UBiDiDirection</code> values indicate the text direction. |
427 | * @stable ICU 2.0 |
428 | */ |
429 | enum UBiDiDirection { |
430 | /** Left-to-right text. This is a 0 value. |
431 | * <ul> |
432 | * <li>As return value for <code>ubidi_getDirection()</code>, it means |
433 | * that the source string contains no right-to-left characters, or |
434 | * that the source string is empty and the paragraph level is even. |
435 | * <li> As return value for <code>ubidi_getBaseDirection()</code>, it |
436 | * means that the first strong character of the source string has |
437 | * a left-to-right direction. |
438 | * </ul> |
439 | * @stable ICU 2.0 |
440 | */ |
441 | UBIDI_LTR, |
442 | /** Right-to-left text. This is a 1 value. |
443 | * <ul> |
444 | * <li>As return value for <code>ubidi_getDirection()</code>, it means |
445 | * that the source string contains no left-to-right characters, or |
446 | * that the source string is empty and the paragraph level is odd. |
447 | * <li> As return value for <code>ubidi_getBaseDirection()</code>, it |
448 | * means that the first strong character of the source string has |
449 | * a right-to-left direction. |
450 | * </ul> |
451 | * @stable ICU 2.0 |
452 | */ |
453 | UBIDI_RTL, |
454 | /** Mixed-directional text. |
455 | * <p>As return value for <code>ubidi_getDirection()</code>, it means |
456 | * that the source string contains both left-to-right and |
457 | * right-to-left characters. |
458 | * @stable ICU 2.0 |
459 | */ |
460 | UBIDI_MIXED, |
461 | /** No strongly directional text. |
462 | * <p>As return value for <code>ubidi_getBaseDirection()</code>, it means |
463 | * that the source string is missing or empty, or contains neither left-to-right |
464 | * nor right-to-left characters. |
465 | * @stable ICU 4.6 |
466 | */ |
467 | UBIDI_NEUTRAL |
468 | }; |
469 | |
470 | /** @stable ICU 2.0 */ |
471 | typedef enum UBiDiDirection UBiDiDirection; |
472 | |
473 | /** |
474 | * Forward declaration of the <code>UBiDi</code> structure for the declaration of |
475 | * the API functions. Its fields are implementation-specific.<p> |
476 | * This structure holds information about a paragraph (or multiple paragraphs) |
477 | * of text with Bidi-algorithm-related details, or about one line of |
478 | * such a paragraph.<p> |
479 | * Reordering can be done on a line, or on one or more paragraphs which are |
480 | * then interpreted each as one single line. |
481 | * @stable ICU 2.0 |
482 | */ |
483 | struct UBiDi; |
484 | |
485 | /** @stable ICU 2.0 */ |
486 | typedef struct UBiDi UBiDi; |
487 | |
488 | /** |
489 | * Allocate a <code>UBiDi</code> structure. |
490 | * Such an object is initially empty. It is assigned |
491 | * the Bidi properties of a piece of text containing one or more paragraphs |
492 | * by <code>ubidi_setPara()</code> |
493 | * or the Bidi properties of a line within a paragraph by |
494 | * <code>ubidi_setLine()</code>.<p> |
495 | * This object can be reused for as long as it is not deallocated |
496 | * by calling <code>ubidi_close()</code>.<p> |
497 | * <code>ubidi_setPara()</code> and <code>ubidi_setLine()</code> will allocate |
498 | * additional memory for internal structures as necessary. |
499 | * |
500 | * @return An empty <code>UBiDi</code> object. |
501 | * @stable ICU 2.0 |
502 | */ |
503 | U_CAPI UBiDi * U_EXPORT2 |
504 | ubidi_open(void); |
505 | |
506 | /** |
507 | * Allocate a <code>UBiDi</code> structure with preallocated memory |
508 | * for internal structures. |
509 | * This function provides a <code>UBiDi</code> object like <code>ubidi_open()</code> |
510 | * with no arguments, but it also preallocates memory for internal structures |
511 | * according to the sizings supplied by the caller.<p> |
512 | * Subsequent functions will not allocate any more memory, and are thus |
513 | * guaranteed not to fail because of lack of memory.<p> |
514 | * The preallocation can be limited to some of the internal memory |
515 | * by setting some values to 0 here. That means that if, e.g., |
516 | * <code>maxRunCount</code> cannot be reasonably predetermined and should not |
517 | * be set to <code>maxLength</code> (the only failproof value) to avoid |
518 | * wasting memory, then <code>maxRunCount</code> could be set to 0 here |
519 | * and the internal structures that are associated with it will be allocated |
520 | * on demand, just like with <code>ubidi_open()</code>. |
521 | * |
522 | * @param maxLength is the maximum text or line length that internal memory |
523 | * will be preallocated for. An attempt to associate this object with a |
524 | * longer text will fail, unless this value is 0, which leaves the allocation |
525 | * up to the implementation. |
526 | * |
527 | * @param maxRunCount is the maximum anticipated number of same-level runs |
528 | * that internal memory will be preallocated for. An attempt to access |
529 | * visual runs on an object that was not preallocated for as many runs |
530 | * as the text was actually resolved to will fail, |
531 | * unless this value is 0, which leaves the allocation up to the implementation.<br><br> |
532 | * The number of runs depends on the actual text and maybe anywhere between |
533 | * 1 and <code>maxLength</code>. It is typically small. |
534 | * |
535 | * @param pErrorCode must be a valid pointer to an error code value. |
536 | * |
537 | * @return An empty <code>UBiDi</code> object with preallocated memory. |
538 | * @stable ICU 2.0 |
539 | */ |
540 | U_CAPI UBiDi * U_EXPORT2 |
541 | ubidi_openSized(int32_t maxLength, int32_t maxRunCount, UErrorCode *pErrorCode); |
542 | |
543 | /** |
544 | * <code>ubidi_close()</code> must be called to free the memory |
545 | * associated with a UBiDi object.<p> |
546 | * |
547 | * <strong>Important: </strong> |
548 | * A parent <code>UBiDi</code> object must not be destroyed or reused if |
549 | * it still has children. |
550 | * If a <code>UBiDi</code> object has become the <i>child</i> |
551 | * of another one (its <i>parent</i>) by calling |
552 | * <code>ubidi_setLine()</code>, then the child object must |
553 | * be destroyed (closed) or reused (by calling |
554 | * <code>ubidi_setPara()</code> or <code>ubidi_setLine()</code>) |
555 | * before the parent object. |
556 | * |
557 | * @param pBiDi is a <code>UBiDi</code> object. |
558 | * |
559 | * @see ubidi_setPara |
560 | * @see ubidi_setLine |
561 | * @stable ICU 2.0 |
562 | */ |
563 | U_CAPI void U_EXPORT2 |
564 | ubidi_close(UBiDi *pBiDi); |
565 | |
566 | #if U_SHOW_CPLUSPLUS_API |
567 | |
568 | U_NAMESPACE_BEGIN |
569 | |
570 | /** |
571 | * \class LocalUBiDiPointer |
572 | * "Smart pointer" class, closes a UBiDi via ubidi_close(). |
573 | * For most methods see the LocalPointerBase base class. |
574 | * |
575 | * @see LocalPointerBase |
576 | * @see LocalPointer |
577 | * @stable ICU 4.4 |
578 | */ |
579 | U_DEFINE_LOCAL_OPEN_POINTER(LocalUBiDiPointer, UBiDi, ubidi_close); |
580 | |
581 | U_NAMESPACE_END |
582 | |
583 | #endif |
584 | |
585 | /** |
586 | * Modify the operation of the Bidi algorithm such that it |
587 | * approximates an "inverse Bidi" algorithm. This function |
588 | * must be called before <code>ubidi_setPara()</code>. |
589 | * |
590 | * <p>The normal operation of the Bidi algorithm as described |
591 | * in the Unicode Technical Report is to take text stored in logical |
592 | * (keyboard, typing) order and to determine the reordering of it for visual |
593 | * rendering. |
594 | * Some legacy systems store text in visual order, and for operations |
595 | * with standard, Unicode-based algorithms, the text needs to be transformed |
596 | * to logical order. This is effectively the inverse algorithm of the |
597 | * described Bidi algorithm. Note that there is no standard algorithm for |
598 | * this "inverse Bidi" and that the current implementation provides only an |
599 | * approximation of "inverse Bidi".</p> |
600 | * |
601 | * <p>With <code>isInverse</code> set to <code>true</code>, |
602 | * this function changes the behavior of some of the subsequent functions |
603 | * in a way that they can be used for the inverse Bidi algorithm. |
604 | * Specifically, runs of text with numeric characters will be treated in a |
605 | * special way and may need to be surrounded with LRM characters when they are |
606 | * written in reordered sequence.</p> |
607 | * |
608 | * <p>Output runs should be retrieved using <code>ubidi_getVisualRun()</code>. |
609 | * Since the actual input for "inverse Bidi" is visually ordered text and |
610 | * <code>ubidi_getVisualRun()</code> gets the reordered runs, these are actually |
611 | * the runs of the logically ordered output.</p> |
612 | * |
613 | * <p>Calling this function with argument <code>isInverse</code> set to |
614 | * <code>true</code> is equivalent to calling |
615 | * <code>ubidi_setReorderingMode</code> with argument |
616 | * <code>reorderingMode</code> |
617 | * set to <code>#UBIDI_REORDER_INVERSE_NUMBERS_AS_L</code>.<br> |
618 | * Calling this function with argument <code>isInverse</code> set to |
619 | * <code>false</code> is equivalent to calling |
620 | * <code>ubidi_setReorderingMode</code> with argument |
621 | * <code>reorderingMode</code> |
622 | * set to <code>#UBIDI_REORDER_DEFAULT</code>. |
623 | * |
624 | * @param pBiDi is a <code>UBiDi</code> object. |
625 | * |
626 | * @param isInverse specifies "forward" or "inverse" Bidi operation. |
627 | * |
628 | * @see ubidi_setPara |
629 | * @see ubidi_writeReordered |
630 | * @see ubidi_setReorderingMode |
631 | * @stable ICU 2.0 |
632 | */ |
633 | U_CAPI void U_EXPORT2 |
634 | ubidi_setInverse(UBiDi *pBiDi, UBool isInverse); |
635 | |
636 | /** |
637 | * Is this Bidi object set to perform the inverse Bidi algorithm? |
638 | * <p>Note: calling this function after setting the reordering mode with |
639 | * <code>ubidi_setReorderingMode</code> will return <code>true</code> if the |
640 | * reordering mode was set to <code>#UBIDI_REORDER_INVERSE_NUMBERS_AS_L</code>, |
641 | * <code>false</code> for all other values.</p> |
642 | * |
643 | * @param pBiDi is a <code>UBiDi</code> object. |
644 | * @return true if the Bidi object is set to perform the inverse Bidi algorithm |
645 | * by handling numbers as L. |
646 | * |
647 | * @see ubidi_setInverse |
648 | * @see ubidi_setReorderingMode |
649 | * @stable ICU 2.0 |
650 | */ |
651 | |
652 | U_CAPI UBool U_EXPORT2 |
653 | ubidi_isInverse(UBiDi *pBiDi); |
654 | |
655 | /** |
656 | * Specify whether block separators must be allocated level zero, |
657 | * so that successive paragraphs will progress from left to right. |
658 | * This function must be called before <code>ubidi_setPara()</code>. |
659 | * Paragraph separators (B) may appear in the text. Setting them to level zero |
660 | * means that all paragraph separators (including one possibly appearing |
661 | * in the last text position) are kept in the reordered text after the text |
662 | * that they follow in the source text. |
663 | * When this feature is not enabled, a paragraph separator at the last |
664 | * position of the text before reordering will go to the first position |
665 | * of the reordered text when the paragraph level is odd. |
666 | * |
667 | * @param pBiDi is a <code>UBiDi</code> object. |
668 | * |
669 | * @param orderParagraphsLTR specifies whether paragraph separators (B) must |
670 | * receive level 0, so that successive paragraphs progress from left to right. |
671 | * |
672 | * @see ubidi_setPara |
673 | * @stable ICU 3.4 |
674 | */ |
675 | U_CAPI void U_EXPORT2 |
676 | ubidi_orderParagraphsLTR(UBiDi *pBiDi, UBool orderParagraphsLTR); |
677 | |
678 | /** |
679 | * Is this Bidi object set to allocate level 0 to block separators so that |
680 | * successive paragraphs progress from left to right? |
681 | * |
682 | * @param pBiDi is a <code>UBiDi</code> object. |
683 | * @return true if the Bidi object is set to allocate level 0 to block |
684 | * separators. |
685 | * |
686 | * @see ubidi_orderParagraphsLTR |
687 | * @stable ICU 3.4 |
688 | */ |
689 | U_CAPI UBool U_EXPORT2 |
690 | ubidi_isOrderParagraphsLTR(UBiDi *pBiDi); |
691 | |
692 | /** |
693 | * <code>UBiDiReorderingMode</code> values indicate which variant of the Bidi |
694 | * algorithm to use. |
695 | * |
696 | * @see ubidi_setReorderingMode |
697 | * @stable ICU 3.6 |
698 | */ |
699 | typedef enum UBiDiReorderingMode { |
700 | /** Regular Logical to Visual Bidi algorithm according to Unicode. |
701 | * This is a 0 value. |
702 | * @stable ICU 3.6 */ |
703 | UBIDI_REORDER_DEFAULT = 0, |
704 | /** Logical to Visual algorithm which handles numbers in a way which |
705 | * mimics the behavior of Windows XP. |
706 | * @stable ICU 3.6 */ |
707 | UBIDI_REORDER_NUMBERS_SPECIAL, |
708 | /** Logical to Visual algorithm grouping numbers with adjacent R characters |
709 | * (reversible algorithm). |
710 | * @stable ICU 3.6 */ |
711 | UBIDI_REORDER_GROUP_NUMBERS_WITH_R, |
712 | /** Reorder runs only to transform a Logical LTR string to the Logical RTL |
713 | * string with the same display, or vice-versa.<br> |
714 | * If this mode is set together with option |
715 | * <code>#UBIDI_OPTION_INSERT_MARKS</code>, some Bidi controls in the source |
716 | * text may be removed and other controls may be added to produce the |
717 | * minimum combination which has the required display. |
718 | * @stable ICU 3.6 */ |
719 | UBIDI_REORDER_RUNS_ONLY, |
720 | /** Visual to Logical algorithm which handles numbers like L |
721 | * (same algorithm as selected by <code>ubidi_setInverse(true)</code>. |
722 | * @see ubidi_setInverse |
723 | * @stable ICU 3.6 */ |
724 | UBIDI_REORDER_INVERSE_NUMBERS_AS_L, |
725 | /** Visual to Logical algorithm equivalent to the regular Logical to Visual |
726 | * algorithm. |
727 | * @stable ICU 3.6 */ |
728 | UBIDI_REORDER_INVERSE_LIKE_DIRECT, |
729 | /** Inverse Bidi (Visual to Logical) algorithm for the |
730 | * <code>UBIDI_REORDER_NUMBERS_SPECIAL</code> Bidi algorithm. |
731 | * @stable ICU 3.6 */ |
732 | UBIDI_REORDER_INVERSE_FOR_NUMBERS_SPECIAL, |
733 | #ifndef U_HIDE_DEPRECATED_API |
734 | /** |
735 | * Number of values for reordering mode. |
736 | * @deprecated ICU 58 The numeric value may change over time, see ICU ticket #12420. |
737 | */ |
738 | UBIDI_REORDER_COUNT |
739 | #endif // U_HIDE_DEPRECATED_API |
740 | } UBiDiReorderingMode; |
741 | |
742 | /** |
743 | * Modify the operation of the Bidi algorithm such that it implements some |
744 | * variant to the basic Bidi algorithm or approximates an "inverse Bidi" |
745 | * algorithm, depending on different values of the "reordering mode". |
746 | * This function must be called before <code>ubidi_setPara()</code>, and stays |
747 | * in effect until called again with a different argument. |
748 | * |
749 | * <p>The normal operation of the Bidi algorithm as described |
750 | * in the Unicode Standard Annex #9 is to take text stored in logical |
751 | * (keyboard, typing) order and to determine how to reorder it for visual |
752 | * rendering.</p> |
753 | * |
754 | * <p>With the reordering mode set to a value other than |
755 | * <code>#UBIDI_REORDER_DEFAULT</code>, this function changes the behavior of |
756 | * some of the subsequent functions in a way such that they implement an |
757 | * inverse Bidi algorithm or some other algorithm variants.</p> |
758 | * |
759 | * <p>Some legacy systems store text in visual order, and for operations |
760 | * with standard, Unicode-based algorithms, the text needs to be transformed |
761 | * into logical order. This is effectively the inverse algorithm of the |
762 | * described Bidi algorithm. Note that there is no standard algorithm for |
763 | * this "inverse Bidi", so a number of variants are implemented here.</p> |
764 | * |
765 | * <p>In other cases, it may be desirable to emulate some variant of the |
766 | * Logical to Visual algorithm (e.g. one used in MS Windows), or perform a |
767 | * Logical to Logical transformation.</p> |
768 | * |
769 | * <ul> |
770 | * <li>When the reordering mode is set to <code>#UBIDI_REORDER_DEFAULT</code>, |
771 | * the standard Bidi Logical to Visual algorithm is applied.</li> |
772 | * |
773 | * <li>When the reordering mode is set to |
774 | * <code>#UBIDI_REORDER_NUMBERS_SPECIAL</code>, |
775 | * the algorithm used to perform Bidi transformations when calling |
776 | * <code>ubidi_setPara</code> should approximate the algorithm used in |
777 | * Microsoft Windows XP rather than strictly conform to the Unicode Bidi |
778 | * algorithm. |
779 | * <br> |
780 | * The differences between the basic algorithm and the algorithm addressed |
781 | * by this option are as follows: |
782 | * <ul> |
783 | * <li>Within text at an even embedding level, the sequence "123AB" |
784 | * (where AB represent R or AL letters) is transformed to "123BA" by the |
785 | * Unicode algorithm and to "BA123" by the Windows algorithm.</li> |
786 | * <li>Arabic-Indic numbers (AN) are handled by the Windows algorithm just |
787 | * like regular numbers (EN).</li> |
788 | * </ul></li> |
789 | * |
790 | * <li>When the reordering mode is set to |
791 | * <code>#UBIDI_REORDER_GROUP_NUMBERS_WITH_R</code>, |
792 | * numbers located between LTR text and RTL text are associated with the RTL |
793 | * text. For instance, an LTR paragraph with content "abc 123 DEF" (where |
794 | * upper case letters represent RTL characters) will be transformed to |
795 | * "abc FED 123" (and not "abc 123 FED"), "DEF 123 abc" will be transformed |
796 | * to "123 FED abc" and "123 FED abc" will be transformed to "DEF 123 abc". |
797 | * This makes the algorithm reversible and makes it useful when round trip |
798 | * (from visual to logical and back to visual) must be achieved without |
799 | * adding LRM characters. However, this is a variation from the standard |
800 | * Unicode Bidi algorithm.<br> |
801 | * The source text should not contain Bidi control characters other than LRM |
802 | * or RLM.</li> |
803 | * |
804 | * <li>When the reordering mode is set to |
805 | * <code>#UBIDI_REORDER_RUNS_ONLY</code>, |
806 | * a "Logical to Logical" transformation must be performed: |
807 | * <ul> |
808 | * <li>If the default text level of the source text (argument <code>paraLevel</code> |
809 | * in <code>ubidi_setPara</code>) is even, the source text will be handled as |
810 | * LTR logical text and will be transformed to the RTL logical text which has |
811 | * the same LTR visual display.</li> |
812 | * <li>If the default level of the source text is odd, the source text |
813 | * will be handled as RTL logical text and will be transformed to the |
814 | * LTR logical text which has the same LTR visual display.</li> |
815 | * </ul> |
816 | * This mode may be needed when logical text which is basically Arabic or |
817 | * Hebrew, with possible included numbers or phrases in English, has to be |
818 | * displayed as if it had an even embedding level (this can happen if the |
819 | * displaying application treats all text as if it was basically LTR). |
820 | * <br> |
821 | * This mode may also be needed in the reverse case, when logical text which is |
822 | * basically English, with possible included phrases in Arabic or Hebrew, has to |
823 | * be displayed as if it had an odd embedding level. |
824 | * <br> |
825 | * Both cases could be handled by adding LRE or RLE at the head of the text, |
826 | * if the display subsystem supports these formatting controls. If it does not, |
827 | * the problem may be handled by transforming the source text in this mode |
828 | * before displaying it, so that it will be displayed properly.<br> |
829 | * The source text should not contain Bidi control characters other than LRM |
830 | * or RLM.</li> |
831 | * |
832 | * <li>When the reordering mode is set to |
833 | * <code>#UBIDI_REORDER_INVERSE_NUMBERS_AS_L</code>, an "inverse Bidi" algorithm |
834 | * is applied. |
835 | * Runs of text with numeric characters will be treated like LTR letters and |
836 | * may need to be surrounded with LRM characters when they are written in |
837 | * reordered sequence (the option <code>#UBIDI_INSERT_LRM_FOR_NUMERIC</code> can |
838 | * be used with function <code>ubidi_writeReordered</code> to this end. This |
839 | * mode is equivalent to calling <code>ubidi_setInverse()</code> with |
840 | * argument <code>isInverse</code> set to <code>true</code>.</li> |
841 | * |
842 | * <li>When the reordering mode is set to |
843 | * <code>#UBIDI_REORDER_INVERSE_LIKE_DIRECT</code>, the "direct" Logical to Visual |
844 | * Bidi algorithm is used as an approximation of an "inverse Bidi" algorithm. |
845 | * This mode is similar to mode <code>#UBIDI_REORDER_INVERSE_NUMBERS_AS_L</code> |
846 | * but is closer to the regular Bidi algorithm. |
847 | * <br> |
848 | * For example, an LTR paragraph with the content "FED 123 456 CBA" (where |
849 | * upper case represents RTL characters) will be transformed to |
850 | * "ABC 456 123 DEF", as opposed to "DEF 123 456 ABC" |
851 | * with mode <code>UBIDI_REORDER_INVERSE_NUMBERS_AS_L</code>.<br> |
852 | * When used in conjunction with option |
853 | * <code>#UBIDI_OPTION_INSERT_MARKS</code>, this mode generally |
854 | * adds Bidi marks to the output significantly more sparingly than mode |
855 | * <code>#UBIDI_REORDER_INVERSE_NUMBERS_AS_L</code> with option |
856 | * <code>#UBIDI_INSERT_LRM_FOR_NUMERIC</code> in calls to |
857 | * <code>ubidi_writeReordered</code>.</li> |
858 | * |
859 | * <li>When the reordering mode is set to |
860 | * <code>#UBIDI_REORDER_INVERSE_FOR_NUMBERS_SPECIAL</code>, the Logical to Visual |
861 | * Bidi algorithm used in Windows XP is used as an approximation of an "inverse Bidi" algorithm. |
862 | * <br> |
863 | * For example, an LTR paragraph with the content "abc FED123" (where |
864 | * upper case represents RTL characters) will be transformed to "abc 123DEF."</li> |
865 | * </ul> |
866 | * |
867 | * <p>In all the reordering modes specifying an "inverse Bidi" algorithm |
868 | * (i.e. those with a name starting with <code>UBIDI_REORDER_INVERSE</code>), |
869 | * output runs should be retrieved using |
870 | * <code>ubidi_getVisualRun()</code>, and the output text with |
871 | * <code>ubidi_writeReordered()</code>. The caller should keep in mind that in |
872 | * "inverse Bidi" modes the input is actually visually ordered text and |
873 | * reordered output returned by <code>ubidi_getVisualRun()</code> or |
874 | * <code>ubidi_writeReordered()</code> are actually runs or character string |
875 | * of logically ordered output.<br> |
876 | * For all the "inverse Bidi" modes, the source text should not contain |
877 | * Bidi control characters other than LRM or RLM.</p> |
878 | * |
879 | * <p>Note that option <code>#UBIDI_OUTPUT_REVERSE</code> of |
880 | * <code>ubidi_writeReordered</code> has no useful meaning and should not be |
881 | * used in conjunction with any value of the reordering mode specifying |
882 | * "inverse Bidi" or with value <code>UBIDI_REORDER_RUNS_ONLY</code>. |
883 | * |
884 | * @param pBiDi is a <code>UBiDi</code> object. |
885 | * @param reorderingMode specifies the required variant of the Bidi algorithm. |
886 | * |
887 | * @see UBiDiReorderingMode |
888 | * @see ubidi_setInverse |
889 | * @see ubidi_setPara |
890 | * @see ubidi_writeReordered |
891 | * @stable ICU 3.6 |
892 | */ |
893 | U_CAPI void U_EXPORT2 |
894 | ubidi_setReorderingMode(UBiDi *pBiDi, UBiDiReorderingMode reorderingMode); |
895 | |
896 | /** |
897 | * What is the requested reordering mode for a given Bidi object? |
898 | * |
899 | * @param pBiDi is a <code>UBiDi</code> object. |
900 | * @return the current reordering mode of the Bidi object |
901 | * @see ubidi_setReorderingMode |
902 | * @stable ICU 3.6 |
903 | */ |
904 | U_CAPI UBiDiReorderingMode U_EXPORT2 |
905 | ubidi_getReorderingMode(UBiDi *pBiDi); |
906 | |
907 | /** |
908 | * <code>UBiDiReorderingOption</code> values indicate which options are |
909 | * specified to affect the Bidi algorithm. |
910 | * |
911 | * @see ubidi_setReorderingOptions |
912 | * @stable ICU 3.6 |
913 | */ |
914 | typedef enum UBiDiReorderingOption { |
915 | /** |
916 | * option value for <code>ubidi_setReorderingOptions</code>: |
917 | * disable all the options which can be set with this function |
918 | * @see ubidi_setReorderingOptions |
919 | * @stable ICU 3.6 |
920 | */ |
921 | UBIDI_OPTION_DEFAULT = 0, |
922 | |
923 | /** |
924 | * option bit for <code>ubidi_setReorderingOptions</code>: |
925 | * insert Bidi marks (LRM or RLM) when needed to ensure correct result of |
926 | * a reordering to a Logical order |
927 | * |
928 | * <p>This option must be set or reset before calling |
929 | * <code>ubidi_setPara</code>.</p> |
930 | * |
931 | * <p>This option is significant only with reordering modes which generate |
932 | * a result with Logical order, specifically:</p> |
933 | * <ul> |
934 | * <li><code>#UBIDI_REORDER_RUNS_ONLY</code></li> |
935 | * <li><code>#UBIDI_REORDER_INVERSE_NUMBERS_AS_L</code></li> |
936 | * <li><code>#UBIDI_REORDER_INVERSE_LIKE_DIRECT</code></li> |
937 | * <li><code>#UBIDI_REORDER_INVERSE_FOR_NUMBERS_SPECIAL</code></li> |
938 | * </ul> |
939 | * |
940 | * <p>If this option is set in conjunction with reordering mode |
941 | * <code>#UBIDI_REORDER_INVERSE_NUMBERS_AS_L</code> or with calling |
942 | * <code>ubidi_setInverse(true)</code>, it implies |
943 | * option <code>#UBIDI_INSERT_LRM_FOR_NUMERIC</code> |
944 | * in calls to function <code>ubidi_writeReordered()</code>.</p> |
945 | * |
946 | * <p>For other reordering modes, a minimum number of LRM or RLM characters |
947 | * will be added to the source text after reordering it so as to ensure |
948 | * round trip, i.e. when applying the inverse reordering mode on the |
949 | * resulting logical text with removal of Bidi marks |
950 | * (option <code>#UBIDI_OPTION_REMOVE_CONTROLS</code> set before calling |
951 | * <code>ubidi_setPara()</code> or option <code>#UBIDI_REMOVE_BIDI_CONTROLS</code> |
952 | * in <code>ubidi_writeReordered</code>), the result will be identical to the |
953 | * source text in the first transformation. |
954 | * |
955 | * <p>This option will be ignored if specified together with option |
956 | * <code>#UBIDI_OPTION_REMOVE_CONTROLS</code>. It inhibits option |
957 | * <code>UBIDI_REMOVE_BIDI_CONTROLS</code> in calls to function |
958 | * <code>ubidi_writeReordered()</code> and it implies option |
959 | * <code>#UBIDI_INSERT_LRM_FOR_NUMERIC</code> in calls to function |
960 | * <code>ubidi_writeReordered()</code> if the reordering mode is |
961 | * <code>#UBIDI_REORDER_INVERSE_NUMBERS_AS_L</code>.</p> |
962 | * |
963 | * @see ubidi_setReorderingMode |
964 | * @see ubidi_setReorderingOptions |
965 | * @stable ICU 3.6 |
966 | */ |
967 | UBIDI_OPTION_INSERT_MARKS = 1, |
968 | |
969 | /** |
970 | * option bit for <code>ubidi_setReorderingOptions</code>: |
971 | * remove Bidi control characters |
972 | * |
973 | * <p>This option must be set or reset before calling |
974 | * <code>ubidi_setPara</code>.</p> |
975 | * |
976 | * <p>This option nullifies option <code>#UBIDI_OPTION_INSERT_MARKS</code>. |
977 | * It inhibits option <code>#UBIDI_INSERT_LRM_FOR_NUMERIC</code> in calls |
978 | * to function <code>ubidi_writeReordered()</code> and it implies option |
979 | * <code>#UBIDI_REMOVE_BIDI_CONTROLS</code> in calls to that function.</p> |
980 | * |
981 | * @see ubidi_setReorderingMode |
982 | * @see ubidi_setReorderingOptions |
983 | * @stable ICU 3.6 |
984 | */ |
985 | UBIDI_OPTION_REMOVE_CONTROLS = 2, |
986 | |
987 | /** |
988 | * option bit for <code>ubidi_setReorderingOptions</code>: |
989 | * process the output as part of a stream to be continued |
990 | * |
991 | * <p>This option must be set or reset before calling |
992 | * <code>ubidi_setPara</code>.</p> |
993 | * |
994 | * <p>This option specifies that the caller is interested in processing large |
995 | * text object in parts. |
996 | * The results of the successive calls are expected to be concatenated by the |
997 | * caller. Only the call for the last part will have this option bit off.</p> |
998 | * |
999 | * <p>When this option bit is on, <code>ubidi_setPara()</code> may process |
1000 | * less than the full source text in order to truncate the text at a meaningful |
1001 | * boundary. The caller should call <code>ubidi_getProcessedLength()</code> |
1002 | * immediately after calling <code>ubidi_setPara()</code> in order to |
1003 | * determine how much of the source text has been processed. |
1004 | * Source text beyond that length should be resubmitted in following calls to |
1005 | * <code>ubidi_setPara</code>. The processed length may be less than |
1006 | * the length of the source text if a character preceding the last character of |
1007 | * the source text constitutes a reasonable boundary (like a block separator) |
1008 | * for text to be continued.<br> |
1009 | * If the last character of the source text constitutes a reasonable |
1010 | * boundary, the whole text will be processed at once.<br> |
1011 | * If nowhere in the source text there exists |
1012 | * such a reasonable boundary, the processed length will be zero.<br> |
1013 | * The caller should check for such an occurrence and do one of the following: |
1014 | * <ul><li>submit a larger amount of text with a better chance to include |
1015 | * a reasonable boundary.</li> |
1016 | * <li>resubmit the same text after turning off option |
1017 | * <code>UBIDI_OPTION_STREAMING</code>.</li></ul> |
1018 | * In all cases, this option should be turned off before processing the last |
1019 | * part of the text.</p> |
1020 | * |
1021 | * <p>When the <code>UBIDI_OPTION_STREAMING</code> option is used, |
1022 | * it is recommended to call <code>ubidi_orderParagraphsLTR()</code> with |
1023 | * argument <code>orderParagraphsLTR</code> set to <code>true</code> before |
1024 | * calling <code>ubidi_setPara</code> so that later paragraphs may be |
1025 | * concatenated to previous paragraphs on the right.</p> |
1026 | * |
1027 | * @see ubidi_setReorderingMode |
1028 | * @see ubidi_setReorderingOptions |
1029 | * @see ubidi_getProcessedLength |
1030 | * @see ubidi_orderParagraphsLTR |
1031 | * @stable ICU 3.6 |
1032 | */ |
1033 | UBIDI_OPTION_STREAMING = 4 |
1034 | } UBiDiReorderingOption; |
1035 | |
1036 | /** |
1037 | * Specify which of the reordering options |
1038 | * should be applied during Bidi transformations. |
1039 | * |
1040 | * @param pBiDi is a <code>UBiDi</code> object. |
1041 | * @param reorderingOptions is a combination of zero or more of the following |
1042 | * options: |
1043 | * <code>#UBIDI_OPTION_DEFAULT</code>, <code>#UBIDI_OPTION_INSERT_MARKS</code>, |
1044 | * <code>#UBIDI_OPTION_REMOVE_CONTROLS</code>, <code>#UBIDI_OPTION_STREAMING</code>. |
1045 | * |
1046 | * @see ubidi_getReorderingOptions |
1047 | * @stable ICU 3.6 |
1048 | */ |
1049 | U_CAPI void U_EXPORT2 |
1050 | ubidi_setReorderingOptions(UBiDi *pBiDi, uint32_t reorderingOptions); |
1051 | |
1052 | /** |
1053 | * What are the reordering options applied to a given Bidi object? |
1054 | * |
1055 | * @param pBiDi is a <code>UBiDi</code> object. |
1056 | * @return the current reordering options of the Bidi object |
1057 | * @see ubidi_setReorderingOptions |
1058 | * @stable ICU 3.6 |
1059 | */ |
1060 | U_CAPI uint32_t U_EXPORT2 |
1061 | ubidi_getReorderingOptions(UBiDi *pBiDi); |
1062 | |
1063 | /** |
1064 | * Set the context before a call to ubidi_setPara().<p> |
1065 | * |
1066 | * ubidi_setPara() computes the left-right directionality for a given piece |
1067 | * of text which is supplied as one of its arguments. Sometimes this piece |
1068 | * of text (the "main text") should be considered in context, because text |
1069 | * appearing before ("prologue") and/or after ("epilogue") the main text |
1070 | * may affect the result of this computation.<p> |
1071 | * |
1072 | * This function specifies the prologue and/or the epilogue for the next |
1073 | * call to ubidi_setPara(). The characters specified as prologue and |
1074 | * epilogue should not be modified by the calling program until the call |
1075 | * to ubidi_setPara() has returned. If successive calls to ubidi_setPara() |
1076 | * all need specification of a context, ubidi_setContext() must be called |
1077 | * before each call to ubidi_setPara(). In other words, a context is not |
1078 | * "remembered" after the following successful call to ubidi_setPara().<p> |
1079 | * |
1080 | * If a call to ubidi_setPara() specifies UBIDI_DEFAULT_LTR or |
1081 | * UBIDI_DEFAULT_RTL as paraLevel and is preceded by a call to |
1082 | * ubidi_setContext() which specifies a prologue, the paragraph level will |
1083 | * be computed taking in consideration the text in the prologue.<p> |
1084 | * |
1085 | * When ubidi_setPara() is called without a previous call to |
1086 | * ubidi_setContext, the main text is handled as if preceded and followed |
1087 | * by strong directional characters at the current paragraph level. |
1088 | * Calling ubidi_setContext() with specification of a prologue will change |
1089 | * this behavior by handling the main text as if preceded by the last |
1090 | * strong character appearing in the prologue, if any. |
1091 | * Calling ubidi_setContext() with specification of an epilogue will change |
1092 | * the behavior of ubidi_setPara() by handling the main text as if followed |
1093 | * by the first strong character or digit appearing in the epilogue, if any.<p> |
1094 | * |
1095 | * Note 1: if <code>ubidi_setContext</code> is called repeatedly without |
1096 | * calling <code>ubidi_setPara</code>, the earlier calls have no effect, |
1097 | * only the last call will be remembered for the next call to |
1098 | * <code>ubidi_setPara</code>.<p> |
1099 | * |
1100 | * Note 2: calling <code>ubidi_setContext(pBiDi, NULL, 0, NULL, 0, &errorCode)</code> |
1101 | * cancels any previous setting of non-empty prologue or epilogue. |
1102 | * The next call to <code>ubidi_setPara()</code> will process no |
1103 | * prologue or epilogue.<p> |
1104 | * |
1105 | * Note 3: users must be aware that even after setting the context |
1106 | * before a call to ubidi_setPara() to perform e.g. a logical to visual |
1107 | * transformation, the resulting string may not be identical to what it |
1108 | * would have been if all the text, including prologue and epilogue, had |
1109 | * been processed together.<br> |
1110 | * Example (upper case letters represent RTL characters):<br> |
1111 | * prologue = "<code>abc DE</code>"<br> |
1112 | * epilogue = none<br> |
1113 | * main text = "<code>FGH xyz</code>"<br> |
1114 | * paraLevel = UBIDI_LTR<br> |
1115 | * display without prologue = "<code>HGF xyz</code>" |
1116 | * ("HGF" is adjacent to "xyz")<br> |
1117 | * display with prologue = "<code>abc HGFED xyz</code>" |
1118 | * ("HGF" is not adjacent to "xyz")<br> |
1119 | * |
1120 | * @param pBiDi is a paragraph <code>UBiDi</code> object. |
1121 | * |
1122 | * @param prologue is a pointer to the text which precedes the text that |
1123 | * will be specified in a coming call to ubidi_setPara(). |
1124 | * If there is no prologue to consider, then <code>proLength</code> |
1125 | * must be zero and this pointer can be NULL. |
1126 | * |
1127 | * @param proLength is the length of the prologue; if <code>proLength==-1</code> |
1128 | * then the prologue must be zero-terminated. |
1129 | * Otherwise proLength must be >= 0. If <code>proLength==0</code>, it means |
1130 | * that there is no prologue to consider. |
1131 | * |
1132 | * @param epilogue is a pointer to the text which follows the text that |
1133 | * will be specified in a coming call to ubidi_setPara(). |
1134 | * If there is no epilogue to consider, then <code>epiLength</code> |
1135 | * must be zero and this pointer can be NULL. |
1136 | * |
1137 | * @param epiLength is the length of the epilogue; if <code>epiLength==-1</code> |
1138 | * then the epilogue must be zero-terminated. |
1139 | * Otherwise epiLength must be >= 0. If <code>epiLength==0</code>, it means |
1140 | * that there is no epilogue to consider. |
1141 | * |
1142 | * @param pErrorCode must be a valid pointer to an error code value. |
1143 | * |
1144 | * @see ubidi_setPara |
1145 | * @stable ICU 4.8 |
1146 | */ |
1147 | U_CAPI void U_EXPORT2 |
1148 | ubidi_setContext(UBiDi *pBiDi, |
1149 | const UChar *prologue, int32_t proLength, |
1150 | const UChar *epilogue, int32_t epiLength, |
1151 | UErrorCode *pErrorCode); |
1152 | |
1153 | /** |
1154 | * Perform the Unicode Bidi algorithm. It is defined in the |
1155 | * <a href="http://www.unicode.org/unicode/reports/tr9/">Unicode Standard Annex #9</a>, |
1156 | * version 13, |
1157 | * also described in The Unicode Standard, Version 4.0 .<p> |
1158 | * |
1159 | * This function takes a piece of plain text containing one or more paragraphs, |
1160 | * with or without externally specified embedding levels from <i>styled</i> |
1161 | * text and computes the left-right-directionality of each character.<p> |
1162 | * |
1163 | * If the entire text is all of the same directionality, then |
1164 | * the function may not perform all the steps described by the algorithm, |
1165 | * i.e., some levels may not be the same as if all steps were performed. |
1166 | * This is not relevant for unidirectional text.<br> |
1167 | * For example, in pure LTR text with numbers the numbers would get |
1168 | * a resolved level of 2 higher than the surrounding text according to |
1169 | * the algorithm. This implementation may set all resolved levels to |
1170 | * the same value in such a case.<p> |
1171 | * |
1172 | * The text can be composed of multiple paragraphs. Occurrence of a block |
1173 | * separator in the text terminates a paragraph, and whatever comes next starts |
1174 | * a new paragraph. The exception to this rule is when a Carriage Return (CR) |
1175 | * is followed by a Line Feed (LF). Both CR and LF are block separators, but |
1176 | * in that case, the pair of characters is considered as terminating the |
1177 | * preceding paragraph, and a new paragraph will be started by a character |
1178 | * coming after the LF. |
1179 | * |
1180 | * @param pBiDi A <code>UBiDi</code> object allocated with <code>ubidi_open()</code> |
1181 | * which will be set to contain the reordering information, |
1182 | * especially the resolved levels for all the characters in <code>text</code>. |
1183 | * |
1184 | * @param text is a pointer to the text that the Bidi algorithm will be performed on. |
1185 | * This pointer is stored in the UBiDi object and can be retrieved |
1186 | * with <code>ubidi_getText()</code>.<br> |
1187 | * <strong>Note:</strong> the text must be (at least) <code>length</code> long. |
1188 | * |
1189 | * @param length is the length of the text; if <code>length==-1</code> then |
1190 | * the text must be zero-terminated. |
1191 | * |
1192 | * @param paraLevel specifies the default level for the text; |
1193 | * it is typically 0 (LTR) or 1 (RTL). |
1194 | * If the function shall determine the paragraph level from the text, |
1195 | * then <code>paraLevel</code> can be set to |
1196 | * either <code>#UBIDI_DEFAULT_LTR</code> |
1197 | * or <code>#UBIDI_DEFAULT_RTL</code>; if the text contains multiple |
1198 | * paragraphs, the paragraph level shall be determined separately for |
1199 | * each paragraph; if a paragraph does not include any strongly typed |
1200 | * character, then the desired default is used (0 for LTR or 1 for RTL). |
1201 | * Any other value between 0 and <code>#UBIDI_MAX_EXPLICIT_LEVEL</code> |
1202 | * is also valid, with odd levels indicating RTL. |
1203 | * |
1204 | * @param embeddingLevels (in) may be used to preset the embedding and override levels, |
1205 | * ignoring characters like LRE and PDF in the text. |
1206 | * A level overrides the directional property of its corresponding |
1207 | * (same index) character if the level has the |
1208 | * <code>#UBIDI_LEVEL_OVERRIDE</code> bit set.<br><br> |
1209 | * Aside from that bit, it must be |
1210 | * <code>paraLevel<=embeddingLevels[]<=UBIDI_MAX_EXPLICIT_LEVEL</code>, |
1211 | * except that level 0 is always allowed. |
1212 | * Level 0 for a paragraph separator prevents reordering of paragraphs; |
1213 | * this only works reliably if <code>#UBIDI_LEVEL_OVERRIDE</code> |
1214 | * is also set for paragraph separators. |
1215 | * Level 0 for other characters is treated as a wildcard |
1216 | * and is lifted up to the resolved level of the surrounding paragraph.<br><br> |
1217 | * <strong>Caution: </strong>A copy of this pointer, not of the levels, |
1218 | * will be stored in the <code>UBiDi</code> object; |
1219 | * the <code>embeddingLevels</code> array must not be |
1220 | * deallocated before the <code>UBiDi</code> structure is destroyed or reused, |
1221 | * and the <code>embeddingLevels</code> |
1222 | * should not be modified to avoid unexpected results on subsequent Bidi operations. |
1223 | * However, the <code>ubidi_setPara()</code> and |
1224 | * <code>ubidi_setLine()</code> functions may modify some or all of the levels.<br><br> |
1225 | * After the <code>UBiDi</code> object is reused or destroyed, the caller |
1226 | * must take care of the deallocation of the <code>embeddingLevels</code> array.<br><br> |
1227 | * <strong>Note:</strong> the <code>embeddingLevels</code> array must be |
1228 | * at least <code>length</code> long. |
1229 | * This pointer can be <code>NULL</code> if this |
1230 | * value is not necessary. |
1231 | * |
1232 | * @param pErrorCode must be a valid pointer to an error code value. |
1233 | * @stable ICU 2.0 |
1234 | */ |
1235 | U_CAPI void U_EXPORT2 |
1236 | ubidi_setPara(UBiDi *pBiDi, const UChar *text, int32_t length, |
1237 | UBiDiLevel paraLevel, UBiDiLevel *embeddingLevels, |
1238 | UErrorCode *pErrorCode); |
1239 | |
1240 | /** |
1241 | * <code>ubidi_setLine()</code> sets a <code>UBiDi</code> to |
1242 | * contain the reordering information, especially the resolved levels, |
1243 | * for all the characters in a line of text. This line of text is |
1244 | * specified by referring to a <code>UBiDi</code> object representing |
1245 | * this information for a piece of text containing one or more paragraphs, |
1246 | * and by specifying a range of indexes in this text.<p> |
1247 | * In the new line object, the indexes will range from 0 to <code>limit-start-1</code>.<p> |
1248 | * |
1249 | * This is used after calling <code>ubidi_setPara()</code> |
1250 | * for a piece of text, and after line-breaking on that text. |
1251 | * It is not necessary if each paragraph is treated as a single line.<p> |
1252 | * |
1253 | * After line-breaking, rules (L1) and (L2) for the treatment of |
1254 | * trailing WS and for reordering are performed on |
1255 | * a <code>UBiDi</code> object that represents a line.<p> |
1256 | * |
1257 | * <strong>Important: </strong><code>pLineBiDi</code> shares data with |
1258 | * <code>pParaBiDi</code>. |
1259 | * You must destroy or reuse <code>pLineBiDi</code> before <code>pParaBiDi</code>. |
1260 | * In other words, you must destroy or reuse the <code>UBiDi</code> object for a line |
1261 | * before the object for its parent paragraph.<p> |
1262 | * |
1263 | * The text pointer that was stored in <code>pParaBiDi</code> is also copied, |
1264 | * and <code>start</code> is added to it so that it points to the beginning of the |
1265 | * line for this object. |
1266 | * |
1267 | * @param pParaBiDi is the parent paragraph object. It must have been set |
1268 | * by a successful call to ubidi_setPara. |
1269 | * |
1270 | * @param start is the line's first index into the text. |
1271 | * |
1272 | * @param limit is just behind the line's last index into the text |
1273 | * (its last index +1).<br> |
1274 | * It must be <code>0<=start<limit<=</code>containing paragraph limit. |
1275 | * If the specified line crosses a paragraph boundary, the function |
1276 | * will terminate with error code U_ILLEGAL_ARGUMENT_ERROR. |
1277 | * |
1278 | * @param pLineBiDi is the object that will now represent a line of the text. |
1279 | * |
1280 | * @param pErrorCode must be a valid pointer to an error code value. |
1281 | * |
1282 | * @see ubidi_setPara |
1283 | * @see ubidi_getProcessedLength |
1284 | * @stable ICU 2.0 |
1285 | */ |
1286 | U_CAPI void U_EXPORT2 |
1287 | ubidi_setLine(const UBiDi *pParaBiDi, |
1288 | int32_t start, int32_t limit, |
1289 | UBiDi *pLineBiDi, |
1290 | UErrorCode *pErrorCode); |
1291 | |
1292 | /** |
1293 | * Get the directionality of the text. |
1294 | * |
1295 | * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
1296 | * |
1297 | * @return a value of <code>UBIDI_LTR</code>, <code>UBIDI_RTL</code> |
1298 | * or <code>UBIDI_MIXED</code> |
1299 | * that indicates if the entire text |
1300 | * represented by this object is unidirectional, |
1301 | * and which direction, or if it is mixed-directional. |
1302 | * Note - The value <code>UBIDI_NEUTRAL</code> is never returned from this method. |
1303 | * |
1304 | * @see UBiDiDirection |
1305 | * @stable ICU 2.0 |
1306 | */ |
1307 | U_CAPI UBiDiDirection U_EXPORT2 |
1308 | ubidi_getDirection(const UBiDi *pBiDi); |
1309 | |
1310 | /** |
1311 | * Gets the base direction of the text provided according |
1312 | * to the Unicode Bidirectional Algorithm. The base direction |
1313 | * is derived from the first character in the string with bidirectional |
1314 | * character type L, R, or AL. If the first such character has type L, |
1315 | * <code>UBIDI_LTR</code> is returned. If the first such character has |
1316 | * type R or AL, <code>UBIDI_RTL</code> is returned. If the string does |
1317 | * not contain any character of these types, then |
1318 | * <code>UBIDI_NEUTRAL</code> is returned. |
1319 | * |
1320 | * This is a lightweight function for use when only the base direction |
1321 | * is needed and no further bidi processing of the text is needed. |
1322 | * |
1323 | * @param text is a pointer to the text whose base |
1324 | * direction is needed. |
1325 | * Note: the text must be (at least) @c length long. |
1326 | * |
1327 | * @param length is the length of the text; |
1328 | * if <code>length==-1</code> then the text |
1329 | * must be zero-terminated. |
1330 | * |
1331 | * @return <code>UBIDI_LTR</code>, <code>UBIDI_RTL</code>, |
1332 | * <code>UBIDI_NEUTRAL</code> |
1333 | * |
1334 | * @see UBiDiDirection |
1335 | * @stable ICU 4.6 |
1336 | */ |
1337 | U_CAPI UBiDiDirection U_EXPORT2 |
1338 | ubidi_getBaseDirection(const UChar *text, int32_t length ); |
1339 | |
1340 | /** |
1341 | * Get the pointer to the text. |
1342 | * |
1343 | * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
1344 | * |
1345 | * @return The pointer to the text that the UBiDi object was created for. |
1346 | * |
1347 | * @see ubidi_setPara |
1348 | * @see ubidi_setLine |
1349 | * @stable ICU 2.0 |
1350 | */ |
1351 | U_CAPI const UChar * U_EXPORT2 |
1352 | ubidi_getText(const UBiDi *pBiDi); |
1353 | |
1354 | /** |
1355 | * Get the length of the text. |
1356 | * |
1357 | * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
1358 | * |
1359 | * @return The length of the text that the UBiDi object was created for. |
1360 | * @stable ICU 2.0 |
1361 | */ |
1362 | U_CAPI int32_t U_EXPORT2 |
1363 | ubidi_getLength(const UBiDi *pBiDi); |
1364 | |
1365 | /** |
1366 | * Get the paragraph level of the text. |
1367 | * |
1368 | * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
1369 | * |
1370 | * @return The paragraph level. If there are multiple paragraphs, their |
1371 | * level may vary if the required paraLevel is UBIDI_DEFAULT_LTR or |
1372 | * UBIDI_DEFAULT_RTL. In that case, the level of the first paragraph |
1373 | * is returned. |
1374 | * |
1375 | * @see UBiDiLevel |
1376 | * @see ubidi_getParagraph |
1377 | * @see ubidi_getParagraphByIndex |
1378 | * @stable ICU 2.0 |
1379 | */ |
1380 | U_CAPI UBiDiLevel U_EXPORT2 |
1381 | ubidi_getParaLevel(const UBiDi *pBiDi); |
1382 | |
1383 | /** |
1384 | * Get the number of paragraphs. |
1385 | * |
1386 | * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
1387 | * |
1388 | * @return The number of paragraphs. |
1389 | * @stable ICU 3.4 |
1390 | */ |
1391 | U_CAPI int32_t U_EXPORT2 |
1392 | ubidi_countParagraphs(UBiDi *pBiDi); |
1393 | |
1394 | /** |
1395 | * Get a paragraph, given a position within the text. |
1396 | * This function returns information about a paragraph.<br> |
1397 | * Note: if the paragraph index is known, it is more efficient to |
1398 | * retrieve the paragraph information using ubidi_getParagraphByIndex().<p> |
1399 | * |
1400 | * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
1401 | * |
1402 | * @param charIndex is the index of a character within the text, in the |
1403 | * range <code>[0..ubidi_getProcessedLength(pBiDi)-1]</code>. |
1404 | * |
1405 | * @param pParaStart will receive the index of the first character of the |
1406 | * paragraph in the text. |
1407 | * This pointer can be <code>NULL</code> if this |
1408 | * value is not necessary. |
1409 | * |
1410 | * @param pParaLimit will receive the limit of the paragraph. |
1411 | * The l-value that you point to here may be the |
1412 | * same expression (variable) as the one for |
1413 | * <code>charIndex</code>. |
1414 | * This pointer can be <code>NULL</code> if this |
1415 | * value is not necessary. |
1416 | * |
1417 | * @param pParaLevel will receive the level of the paragraph. |
1418 | * This pointer can be <code>NULL</code> if this |
1419 | * value is not necessary. |
1420 | * |
1421 | * @param pErrorCode must be a valid pointer to an error code value. |
1422 | * |
1423 | * @return The index of the paragraph containing the specified position. |
1424 | * |
1425 | * @see ubidi_getProcessedLength |
1426 | * @stable ICU 3.4 |
1427 | */ |
1428 | U_CAPI int32_t U_EXPORT2 |
1429 | ubidi_getParagraph(const UBiDi *pBiDi, int32_t charIndex, int32_t *pParaStart, |
1430 | int32_t *pParaLimit, UBiDiLevel *pParaLevel, |
1431 | UErrorCode *pErrorCode); |
1432 | |
1433 | /** |
1434 | * Get a paragraph, given the index of this paragraph. |
1435 | * |
1436 | * This function returns information about a paragraph.<p> |
1437 | * |
1438 | * @param pBiDi is the paragraph <code>UBiDi</code> object. |
1439 | * |
1440 | * @param paraIndex is the number of the paragraph, in the |
1441 | * range <code>[0..ubidi_countParagraphs(pBiDi)-1]</code>. |
1442 | * |
1443 | * @param pParaStart will receive the index of the first character of the |
1444 | * paragraph in the text. |
1445 | * This pointer can be <code>NULL</code> if this |
1446 | * value is not necessary. |
1447 | * |
1448 | * @param pParaLimit will receive the limit of the paragraph. |
1449 | * This pointer can be <code>NULL</code> if this |
1450 | * value is not necessary. |
1451 | * |
1452 | * @param pParaLevel will receive the level of the paragraph. |
1453 | * This pointer can be <code>NULL</code> if this |
1454 | * value is not necessary. |
1455 | * |
1456 | * @param pErrorCode must be a valid pointer to an error code value. |
1457 | * |
1458 | * @stable ICU 3.4 |
1459 | */ |
1460 | U_CAPI void U_EXPORT2 |
1461 | ubidi_getParagraphByIndex(const UBiDi *pBiDi, int32_t paraIndex, |
1462 | int32_t *pParaStart, int32_t *pParaLimit, |
1463 | UBiDiLevel *pParaLevel, UErrorCode *pErrorCode); |
1464 | |
1465 | /** |
1466 | * Get the level for one character. |
1467 | * |
1468 | * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
1469 | * |
1470 | * @param charIndex the index of a character. It must be in the range |
1471 | * [0..ubidi_getProcessedLength(pBiDi)]. |
1472 | * |
1473 | * @return The level for the character at charIndex (0 if charIndex is not |
1474 | * in the valid range). |
1475 | * |
1476 | * @see UBiDiLevel |
1477 | * @see ubidi_getProcessedLength |
1478 | * @stable ICU 2.0 |
1479 | */ |
1480 | U_CAPI UBiDiLevel U_EXPORT2 |
1481 | ubidi_getLevelAt(const UBiDi *pBiDi, int32_t charIndex); |
1482 | |
1483 | /** |
1484 | * Get an array of levels for each character.<p> |
1485 | * |
1486 | * Note that this function may allocate memory under some |
1487 | * circumstances, unlike <code>ubidi_getLevelAt()</code>. |
1488 | * |
1489 | * @param pBiDi is the paragraph or line <code>UBiDi</code> object, whose |
1490 | * text length must be strictly positive. |
1491 | * |
1492 | * @param pErrorCode must be a valid pointer to an error code value. |
1493 | * |
1494 | * @return The levels array for the text, |
1495 | * or <code>NULL</code> if an error occurs. |
1496 | * |
1497 | * @see UBiDiLevel |
1498 | * @see ubidi_getProcessedLength |
1499 | * @stable ICU 2.0 |
1500 | */ |
1501 | U_CAPI const UBiDiLevel * U_EXPORT2 |
1502 | ubidi_getLevels(UBiDi *pBiDi, UErrorCode *pErrorCode); |
1503 | |
1504 | /** |
1505 | * Get a logical run. |
1506 | * This function returns information about a run and is used |
1507 | * to retrieve runs in logical order.<p> |
1508 | * This is especially useful for line-breaking on a paragraph. |
1509 | * |
1510 | * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
1511 | * |
1512 | * @param logicalPosition is a logical position within the source text. |
1513 | * |
1514 | * @param pLogicalLimit will receive the limit of the corresponding run. |
1515 | * The l-value that you point to here may be the |
1516 | * same expression (variable) as the one for |
1517 | * <code>logicalPosition</code>. |
1518 | * This pointer can be <code>NULL</code> if this |
1519 | * value is not necessary. |
1520 | * |
1521 | * @param pLevel will receive the level of the corresponding run. |
1522 | * This pointer can be <code>NULL</code> if this |
1523 | * value is not necessary. |
1524 | * |
1525 | * @see ubidi_getProcessedLength |
1526 | * @stable ICU 2.0 |
1527 | */ |
1528 | U_CAPI void U_EXPORT2 |
1529 | ubidi_getLogicalRun(const UBiDi *pBiDi, int32_t logicalPosition, |
1530 | int32_t *pLogicalLimit, UBiDiLevel *pLevel); |
1531 | |
1532 | /** |
1533 | * Get the number of runs. |
1534 | * This function may invoke the actual reordering on the |
1535 | * <code>UBiDi</code> object, after <code>ubidi_setPara()</code> |
1536 | * may have resolved only the levels of the text. Therefore, |
1537 | * <code>ubidi_countRuns()</code> may have to allocate memory, |
1538 | * and may fail doing so. |
1539 | * |
1540 | * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
1541 | * |
1542 | * @param pErrorCode must be a valid pointer to an error code value. |
1543 | * |
1544 | * @return The number of runs. |
1545 | * @stable ICU 2.0 |
1546 | */ |
1547 | U_CAPI int32_t U_EXPORT2 |
1548 | ubidi_countRuns(UBiDi *pBiDi, UErrorCode *pErrorCode); |
1549 | |
1550 | /** |
1551 | * Get one run's logical start, length, and directionality, |
1552 | * which can be 0 for LTR or 1 for RTL. |
1553 | * In an RTL run, the character at the logical start is |
1554 | * visually on the right of the displayed run. |
1555 | * The length is the number of characters in the run.<p> |
1556 | * <code>ubidi_countRuns()</code> should be called |
1557 | * before the runs are retrieved. |
1558 | * |
1559 | * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
1560 | * |
1561 | * @param runIndex is the number of the run in visual order, in the |
1562 | * range <code>[0..ubidi_countRuns(pBiDi)-1]</code>. |
1563 | * |
1564 | * @param pLogicalStart is the first logical character index in the text. |
1565 | * The pointer may be <code>NULL</code> if this index is not needed. |
1566 | * |
1567 | * @param pLength is the number of characters (at least one) in the run. |
1568 | * The pointer may be <code>NULL</code> if this is not needed. |
1569 | * |
1570 | * @return the directionality of the run, |
1571 | * <code>UBIDI_LTR==0</code> or <code>UBIDI_RTL==1</code>, |
1572 | * never <code>UBIDI_MIXED</code>, |
1573 | * never <code>UBIDI_NEUTRAL</code>. |
1574 | * |
1575 | * @see ubidi_countRuns |
1576 | * |
1577 | * Example: |
1578 | * <pre> |
1579 | * \code |
1580 | * int32_t i, count=ubidi_countRuns(pBiDi), |
1581 | * logicalStart, visualIndex=0, length; |
1582 | * for(i=0; i<count; ++i) { |
1583 | * if(UBIDI_LTR==ubidi_getVisualRun(pBiDi, i, &logicalStart, &length)) { |
1584 | * do { // LTR |
1585 | * show_char(text[logicalStart++], visualIndex++); |
1586 | * } while(--length>0); |
1587 | * } else { |
1588 | * logicalStart+=length; // logicalLimit |
1589 | * do { // RTL |
1590 | * show_char(text[--logicalStart], visualIndex++); |
1591 | * } while(--length>0); |
1592 | * } |
1593 | * } |
1594 | *\endcode |
1595 | * </pre> |
1596 | * |
1597 | * Note that in right-to-left runs, code like this places |
1598 | * second surrogates before first ones (which is generally a bad idea) |
1599 | * and combining characters before base characters. |
1600 | * <p> |
1601 | * Use of <code>ubidi_writeReordered()</code>, optionally with the |
1602 | * <code>#UBIDI_KEEP_BASE_COMBINING</code> option, can be considered in order |
1603 | * to avoid these issues. |
1604 | * @stable ICU 2.0 |
1605 | */ |
1606 | U_CAPI UBiDiDirection U_EXPORT2 |
1607 | ubidi_getVisualRun(UBiDi *pBiDi, int32_t runIndex, |
1608 | int32_t *pLogicalStart, int32_t *pLength); |
1609 | |
1610 | /** |
1611 | * Get the visual position from a logical text position. |
1612 | * If such a mapping is used many times on the same |
1613 | * <code>UBiDi</code> object, then calling |
1614 | * <code>ubidi_getLogicalMap()</code> is more efficient.<p> |
1615 | * |
1616 | * The value returned may be <code>#UBIDI_MAP_NOWHERE</code> if there is no |
1617 | * visual position because the corresponding text character is a Bidi control |
1618 | * removed from output by the option <code>#UBIDI_OPTION_REMOVE_CONTROLS</code>. |
1619 | * <p> |
1620 | * When the visual output is altered by using options of |
1621 | * <code>ubidi_writeReordered()</code> such as <code>UBIDI_INSERT_LRM_FOR_NUMERIC</code>, |
1622 | * <code>UBIDI_KEEP_BASE_COMBINING</code>, <code>UBIDI_OUTPUT_REVERSE</code>, |
1623 | * <code>UBIDI_REMOVE_BIDI_CONTROLS</code>, the visual position returned may not |
1624 | * be correct. It is advised to use, when possible, reordering options |
1625 | * such as <code>UBIDI_OPTION_INSERT_MARKS</code> and <code>UBIDI_OPTION_REMOVE_CONTROLS</code>. |
1626 | * <p> |
1627 | * Note that in right-to-left runs, this mapping places |
1628 | * second surrogates before first ones (which is generally a bad idea) |
1629 | * and combining characters before base characters. |
1630 | * Use of <code>ubidi_writeReordered()</code>, optionally with the |
1631 | * <code>#UBIDI_KEEP_BASE_COMBINING</code> option can be considered instead |
1632 | * of using the mapping, in order to avoid these issues. |
1633 | * |
1634 | * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
1635 | * |
1636 | * @param logicalIndex is the index of a character in the text. |
1637 | * |
1638 | * @param pErrorCode must be a valid pointer to an error code value. |
1639 | * |
1640 | * @return The visual position of this character. |
1641 | * |
1642 | * @see ubidi_getLogicalMap |
1643 | * @see ubidi_getLogicalIndex |
1644 | * @see ubidi_getProcessedLength |
1645 | * @stable ICU 2.0 |
1646 | */ |
1647 | U_CAPI int32_t U_EXPORT2 |
1648 | ubidi_getVisualIndex(UBiDi *pBiDi, int32_t logicalIndex, UErrorCode *pErrorCode); |
1649 | |
1650 | /** |
1651 | * Get the logical text position from a visual position. |
1652 | * If such a mapping is used many times on the same |
1653 | * <code>UBiDi</code> object, then calling |
1654 | * <code>ubidi_getVisualMap()</code> is more efficient.<p> |
1655 | * |
1656 | * The value returned may be <code>#UBIDI_MAP_NOWHERE</code> if there is no |
1657 | * logical position because the corresponding text character is a Bidi mark |
1658 | * inserted in the output by option <code>#UBIDI_OPTION_INSERT_MARKS</code>. |
1659 | * <p> |
1660 | * This is the inverse function to <code>ubidi_getVisualIndex()</code>. |
1661 | * <p> |
1662 | * When the visual output is altered by using options of |
1663 | * <code>ubidi_writeReordered()</code> such as <code>UBIDI_INSERT_LRM_FOR_NUMERIC</code>, |
1664 | * <code>UBIDI_KEEP_BASE_COMBINING</code>, <code>UBIDI_OUTPUT_REVERSE</code>, |
1665 | * <code>UBIDI_REMOVE_BIDI_CONTROLS</code>, the logical position returned may not |
1666 | * be correct. It is advised to use, when possible, reordering options |
1667 | * such as <code>UBIDI_OPTION_INSERT_MARKS</code> and <code>UBIDI_OPTION_REMOVE_CONTROLS</code>. |
1668 | * |
1669 | * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
1670 | * |
1671 | * @param visualIndex is the visual position of a character. |
1672 | * |
1673 | * @param pErrorCode must be a valid pointer to an error code value. |
1674 | * |
1675 | * @return The index of this character in the text. |
1676 | * |
1677 | * @see ubidi_getVisualMap |
1678 | * @see ubidi_getVisualIndex |
1679 | * @see ubidi_getResultLength |
1680 | * @stable ICU 2.0 |
1681 | */ |
1682 | U_CAPI int32_t U_EXPORT2 |
1683 | ubidi_getLogicalIndex(UBiDi *pBiDi, int32_t visualIndex, UErrorCode *pErrorCode); |
1684 | |
1685 | /** |
1686 | * Get a logical-to-visual index map (array) for the characters in the UBiDi |
1687 | * (paragraph or line) object. |
1688 | * <p> |
1689 | * Some values in the map may be <code>#UBIDI_MAP_NOWHERE</code> if the |
1690 | * corresponding text characters are Bidi controls removed from the visual |
1691 | * output by the option <code>#UBIDI_OPTION_REMOVE_CONTROLS</code>. |
1692 | * <p> |
1693 | * When the visual output is altered by using options of |
1694 | * <code>ubidi_writeReordered()</code> such as <code>UBIDI_INSERT_LRM_FOR_NUMERIC</code>, |
1695 | * <code>UBIDI_KEEP_BASE_COMBINING</code>, <code>UBIDI_OUTPUT_REVERSE</code>, |
1696 | * <code>UBIDI_REMOVE_BIDI_CONTROLS</code>, the visual positions returned may not |
1697 | * be correct. It is advised to use, when possible, reordering options |
1698 | * such as <code>UBIDI_OPTION_INSERT_MARKS</code> and <code>UBIDI_OPTION_REMOVE_CONTROLS</code>. |
1699 | * <p> |
1700 | * Note that in right-to-left runs, this mapping places |
1701 | * second surrogates before first ones (which is generally a bad idea) |
1702 | * and combining characters before base characters. |
1703 | * Use of <code>ubidi_writeReordered()</code>, optionally with the |
1704 | * <code>#UBIDI_KEEP_BASE_COMBINING</code> option can be considered instead |
1705 | * of using the mapping, in order to avoid these issues. |
1706 | * |
1707 | * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
1708 | * |
1709 | * @param indexMap is a pointer to an array of <code>ubidi_getProcessedLength()</code> |
1710 | * indexes which will reflect the reordering of the characters. |
1711 | * If option <code>#UBIDI_OPTION_INSERT_MARKS</code> is set, the number |
1712 | * of elements allocated in <code>indexMap</code> must be no less than |
1713 | * <code>ubidi_getResultLength()</code>. |
1714 | * The array does not need to be initialized.<br><br> |
1715 | * The index map will result in <code>indexMap[logicalIndex]==visualIndex</code>. |
1716 | * |
1717 | * @param pErrorCode must be a valid pointer to an error code value. |
1718 | * |
1719 | * @see ubidi_getVisualMap |
1720 | * @see ubidi_getVisualIndex |
1721 | * @see ubidi_getProcessedLength |
1722 | * @see ubidi_getResultLength |
1723 | * @stable ICU 2.0 |
1724 | */ |
1725 | U_CAPI void U_EXPORT2 |
1726 | ubidi_getLogicalMap(UBiDi *pBiDi, int32_t *indexMap, UErrorCode *pErrorCode); |
1727 | |
1728 | /** |
1729 | * Get a visual-to-logical index map (array) for the characters in the UBiDi |
1730 | * (paragraph or line) object. |
1731 | * <p> |
1732 | * Some values in the map may be <code>#UBIDI_MAP_NOWHERE</code> if the |
1733 | * corresponding text characters are Bidi marks inserted in the visual output |
1734 | * by the option <code>#UBIDI_OPTION_INSERT_MARKS</code>. |
1735 | * <p> |
1736 | * When the visual output is altered by using options of |
1737 | * <code>ubidi_writeReordered()</code> such as <code>UBIDI_INSERT_LRM_FOR_NUMERIC</code>, |
1738 | * <code>UBIDI_KEEP_BASE_COMBINING</code>, <code>UBIDI_OUTPUT_REVERSE</code>, |
1739 | * <code>UBIDI_REMOVE_BIDI_CONTROLS</code>, the logical positions returned may not |
1740 | * be correct. It is advised to use, when possible, reordering options |
1741 | * such as <code>UBIDI_OPTION_INSERT_MARKS</code> and <code>UBIDI_OPTION_REMOVE_CONTROLS</code>. |
1742 | * |
1743 | * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
1744 | * |
1745 | * @param indexMap is a pointer to an array of <code>ubidi_getResultLength()</code> |
1746 | * indexes which will reflect the reordering of the characters. |
1747 | * If option <code>#UBIDI_OPTION_REMOVE_CONTROLS</code> is set, the number |
1748 | * of elements allocated in <code>indexMap</code> must be no less than |
1749 | * <code>ubidi_getProcessedLength()</code>. |
1750 | * The array does not need to be initialized.<br><br> |
1751 | * The index map will result in <code>indexMap[visualIndex]==logicalIndex</code>. |
1752 | * |
1753 | * @param pErrorCode must be a valid pointer to an error code value. |
1754 | * |
1755 | * @see ubidi_getLogicalMap |
1756 | * @see ubidi_getLogicalIndex |
1757 | * @see ubidi_getProcessedLength |
1758 | * @see ubidi_getResultLength |
1759 | * @stable ICU 2.0 |
1760 | */ |
1761 | U_CAPI void U_EXPORT2 |
1762 | ubidi_getVisualMap(UBiDi *pBiDi, int32_t *indexMap, UErrorCode *pErrorCode); |
1763 | |
1764 | /** |
1765 | * This is a convenience function that does not use a UBiDi object. |
1766 | * It is intended to be used for when an application has determined the levels |
1767 | * of objects (character sequences) and just needs to have them reordered (L2). |
1768 | * This is equivalent to using <code>ubidi_getLogicalMap()</code> on a |
1769 | * <code>UBiDi</code> object. |
1770 | * |
1771 | * @param levels is an array with <code>length</code> levels that have been determined by |
1772 | * the application. |
1773 | * |
1774 | * @param length is the number of levels in the array, or, semantically, |
1775 | * the number of objects to be reordered. |
1776 | * It must be <code>length>0</code>. |
1777 | * |
1778 | * @param indexMap is a pointer to an array of <code>length</code> |
1779 | * indexes which will reflect the reordering of the characters. |
1780 | * The array does not need to be initialized.<p> |
1781 | * The index map will result in <code>indexMap[logicalIndex]==visualIndex</code>. |
1782 | * @stable ICU 2.0 |
1783 | */ |
1784 | U_CAPI void U_EXPORT2 |
1785 | ubidi_reorderLogical(const UBiDiLevel *levels, int32_t length, int32_t *indexMap); |
1786 | |
1787 | /** |
1788 | * This is a convenience function that does not use a UBiDi object. |
1789 | * It is intended to be used for when an application has determined the levels |
1790 | * of objects (character sequences) and just needs to have them reordered (L2). |
1791 | * This is equivalent to using <code>ubidi_getVisualMap()</code> on a |
1792 | * <code>UBiDi</code> object. |
1793 | * |
1794 | * @param levels is an array with <code>length</code> levels that have been determined by |
1795 | * the application. |
1796 | * |
1797 | * @param length is the number of levels in the array, or, semantically, |
1798 | * the number of objects to be reordered. |
1799 | * It must be <code>length>0</code>. |
1800 | * |
1801 | * @param indexMap is a pointer to an array of <code>length</code> |
1802 | * indexes which will reflect the reordering of the characters. |
1803 | * The array does not need to be initialized.<p> |
1804 | * The index map will result in <code>indexMap[visualIndex]==logicalIndex</code>. |
1805 | * @stable ICU 2.0 |
1806 | */ |
1807 | U_CAPI void U_EXPORT2 |
1808 | ubidi_reorderVisual(const UBiDiLevel *levels, int32_t length, int32_t *indexMap); |
1809 | |
1810 | /** |
1811 | * Invert an index map. |
1812 | * The index mapping of the first map is inverted and written to |
1813 | * the second one. |
1814 | * |
1815 | * @param srcMap is an array with <code>length</code> elements |
1816 | * which defines the original mapping from a source array containing |
1817 | * <code>length</code> elements to a destination array. |
1818 | * Some elements of the source array may have no mapping in the |
1819 | * destination array. In that case, their value will be |
1820 | * the special value <code>UBIDI_MAP_NOWHERE</code>. |
1821 | * All elements must be >=0 or equal to <code>UBIDI_MAP_NOWHERE</code>. |
1822 | * Some elements may have a value >= <code>length</code>, if the |
1823 | * destination array has more elements than the source array. |
1824 | * There must be no duplicate indexes (two or more elements with the |
1825 | * same value except <code>UBIDI_MAP_NOWHERE</code>). |
1826 | * |
1827 | * @param destMap is an array with a number of elements equal to 1 + the highest |
1828 | * value in <code>srcMap</code>. |
1829 | * <code>destMap</code> will be filled with the inverse mapping. |
1830 | * If element with index i in <code>srcMap</code> has a value k different |
1831 | * from <code>UBIDI_MAP_NOWHERE</code>, this means that element i of |
1832 | * the source array maps to element k in the destination array. |
1833 | * The inverse map will have value i in its k-th element. |
1834 | * For all elements of the destination array which do not map to |
1835 | * an element in the source array, the corresponding element in the |
1836 | * inverse map will have a value equal to <code>UBIDI_MAP_NOWHERE</code>. |
1837 | * |
1838 | * @param length is the length of each array. |
1839 | * @see UBIDI_MAP_NOWHERE |
1840 | * @stable ICU 2.0 |
1841 | */ |
1842 | U_CAPI void U_EXPORT2 |
1843 | ubidi_invertMap(const int32_t *srcMap, int32_t *destMap, int32_t length); |
1844 | |
1845 | /** option flags for ubidi_writeReordered() */ |
1846 | |
1847 | /** |
1848 | * option bit for ubidi_writeReordered(): |
1849 | * keep combining characters after their base characters in RTL runs |
1850 | * |
1851 | * @see ubidi_writeReordered |
1852 | * @stable ICU 2.0 |
1853 | */ |
1854 | #define UBIDI_KEEP_BASE_COMBINING 1 |
1855 | |
1856 | /** |
1857 | * option bit for ubidi_writeReordered(): |
1858 | * replace characters with the "mirrored" property in RTL runs |
1859 | * by their mirror-image mappings |
1860 | * |
1861 | * @see ubidi_writeReordered |
1862 | * @stable ICU 2.0 |
1863 | */ |
1864 | #define UBIDI_DO_MIRRORING 2 |
1865 | |
1866 | /** |
1867 | * option bit for ubidi_writeReordered(): |
1868 | * surround the run with LRMs if necessary; |
1869 | * this is part of the approximate "inverse Bidi" algorithm |
1870 | * |
1871 | * <p>This option does not imply corresponding adjustment of the index |
1872 | * mappings.</p> |
1873 | * |
1874 | * @see ubidi_setInverse |
1875 | * @see ubidi_writeReordered |
1876 | * @stable ICU 2.0 |
1877 | */ |
1878 | #define UBIDI_INSERT_LRM_FOR_NUMERIC 4 |
1879 | |
1880 | /** |
1881 | * option bit for ubidi_writeReordered(): |
1882 | * remove Bidi control characters |
1883 | * (this does not affect #UBIDI_INSERT_LRM_FOR_NUMERIC) |
1884 | * |
1885 | * <p>This option does not imply corresponding adjustment of the index |
1886 | * mappings.</p> |
1887 | * |
1888 | * @see ubidi_writeReordered |
1889 | * @stable ICU 2.0 |
1890 | */ |
1891 | #define UBIDI_REMOVE_BIDI_CONTROLS 8 |
1892 | |
1893 | /** |
1894 | * option bit for ubidi_writeReordered(): |
1895 | * write the output in reverse order |
1896 | * |
1897 | * <p>This has the same effect as calling <code>ubidi_writeReordered()</code> |
1898 | * first without this option, and then calling |
1899 | * <code>ubidi_writeReverse()</code> without mirroring. |
1900 | * Doing this in the same step is faster and avoids a temporary buffer. |
1901 | * An example for using this option is output to a character terminal that |
1902 | * is designed for RTL scripts and stores text in reverse order.</p> |
1903 | * |
1904 | * @see ubidi_writeReordered |
1905 | * @stable ICU 2.0 |
1906 | */ |
1907 | #define UBIDI_OUTPUT_REVERSE 16 |
1908 | |
1909 | /** |
1910 | * Get the length of the source text processed by the last call to |
1911 | * <code>ubidi_setPara()</code>. This length may be different from the length |
1912 | * of the source text if option <code>#UBIDI_OPTION_STREAMING</code> |
1913 | * has been set. |
1914 | * <br> |
1915 | * Note that whenever the length of the text affects the execution or the |
1916 | * result of a function, it is the processed length which must be considered, |
1917 | * except for <code>ubidi_setPara</code> (which receives unprocessed source |
1918 | * text) and <code>ubidi_getLength</code> (which returns the original length |
1919 | * of the source text).<br> |
1920 | * In particular, the processed length is the one to consider in the following |
1921 | * cases: |
1922 | * <ul> |
1923 | * <li>maximum value of the <code>limit</code> argument of |
1924 | * <code>ubidi_setLine</code></li> |
1925 | * <li>maximum value of the <code>charIndex</code> argument of |
1926 | * <code>ubidi_getParagraph</code></li> |
1927 | * <li>maximum value of the <code>charIndex</code> argument of |
1928 | * <code>ubidi_getLevelAt</code></li> |
1929 | * <li>number of elements in the array returned by <code>ubidi_getLevels</code></li> |
1930 | * <li>maximum value of the <code>logicalStart</code> argument of |
1931 | * <code>ubidi_getLogicalRun</code></li> |
1932 | * <li>maximum value of the <code>logicalIndex</code> argument of |
1933 | * <code>ubidi_getVisualIndex</code></li> |
1934 | * <li>number of elements filled in the <code>*indexMap</code> argument of |
1935 | * <code>ubidi_getLogicalMap</code></li> |
1936 | * <li>length of text processed by <code>ubidi_writeReordered</code></li> |
1937 | * </ul> |
1938 | * |
1939 | * @param pBiDi is the paragraph <code>UBiDi</code> object. |
1940 | * |
1941 | * @return The length of the part of the source text processed by |
1942 | * the last call to <code>ubidi_setPara</code>. |
1943 | * @see ubidi_setPara |
1944 | * @see UBIDI_OPTION_STREAMING |
1945 | * @stable ICU 3.6 |
1946 | */ |
1947 | U_CAPI int32_t U_EXPORT2 |
1948 | ubidi_getProcessedLength(const UBiDi *pBiDi); |
1949 | |
1950 | /** |
1951 | * Get the length of the reordered text resulting from the last call to |
1952 | * <code>ubidi_setPara()</code>. This length may be different from the length |
1953 | * of the source text if option <code>#UBIDI_OPTION_INSERT_MARKS</code> |
1954 | * or option <code>#UBIDI_OPTION_REMOVE_CONTROLS</code> has been set. |
1955 | * <br> |
1956 | * This resulting length is the one to consider in the following cases: |
1957 | * <ul> |
1958 | * <li>maximum value of the <code>visualIndex</code> argument of |
1959 | * <code>ubidi_getLogicalIndex</code></li> |
1960 | * <li>number of elements of the <code>*indexMap</code> argument of |
1961 | * <code>ubidi_getVisualMap</code></li> |
1962 | * </ul> |
1963 | * Note that this length stays identical to the source text length if |
1964 | * Bidi marks are inserted or removed using option bits of |
1965 | * <code>ubidi_writeReordered</code>, or if option |
1966 | * <code>#UBIDI_REORDER_INVERSE_NUMBERS_AS_L</code> has been set. |
1967 | * |
1968 | * @param pBiDi is the paragraph <code>UBiDi</code> object. |
1969 | * |
1970 | * @return The length of the reordered text resulting from |
1971 | * the last call to <code>ubidi_setPara</code>. |
1972 | * @see ubidi_setPara |
1973 | * @see UBIDI_OPTION_INSERT_MARKS |
1974 | * @see UBIDI_OPTION_REMOVE_CONTROLS |
1975 | * @stable ICU 3.6 |
1976 | */ |
1977 | U_CAPI int32_t U_EXPORT2 |
1978 | ubidi_getResultLength(const UBiDi *pBiDi); |
1979 | |
1980 | U_CDECL_BEGIN |
1981 | |
1982 | #ifndef U_HIDE_DEPRECATED_API |
1983 | /** |
1984 | * Value returned by <code>UBiDiClassCallback</code> callbacks when |
1985 | * there is no need to override the standard Bidi class for a given code point. |
1986 | * |
1987 | * This constant is deprecated; use u_getIntPropertyMaxValue(UCHAR_BIDI_CLASS)+1 instead. |
1988 | * |
1989 | * @see UBiDiClassCallback |
1990 | * @deprecated ICU 58 The numeric value may change over time, see ICU ticket #12420. |
1991 | */ |
1992 | #define U_BIDI_CLASS_DEFAULT U_CHAR_DIRECTION_COUNT |
1993 | #endif // U_HIDE_DEPRECATED_API |
1994 | |
1995 | /** |
1996 | * Callback type declaration for overriding default Bidi class values with |
1997 | * custom ones. |
1998 | * <p>Usually, the function pointer will be propagated to a <code>UBiDi</code> |
1999 | * object by calling the <code>ubidi_setClassCallback()</code> function; |
2000 | * then the callback will be invoked by the UBA implementation any time the |
2001 | * class of a character is to be determined.</p> |
2002 | * |
2003 | * @param context is a pointer to the callback private data. |
2004 | * |
2005 | * @param c is the code point to get a Bidi class for. |
2006 | * |
2007 | * @return The directional property / Bidi class for the given code point |
2008 | * <code>c</code> if the default class has been overridden, or |
2009 | * <code>u_getIntPropertyMaxValue(UCHAR_BIDI_CLASS)+1</code> |
2010 | * if the standard Bidi class value for <code>c</code> is to be used. |
2011 | * @see ubidi_setClassCallback |
2012 | * @see ubidi_getClassCallback |
2013 | * @stable ICU 3.6 |
2014 | */ |
2015 | typedef UCharDirection U_CALLCONV |
2016 | UBiDiClassCallback(const void *context, UChar32 c); |
2017 | |
2018 | U_CDECL_END |
2019 | |
2020 | /** |
2021 | * Retrieve the Bidi class for a given code point. |
2022 | * <p>If a <code>#UBiDiClassCallback</code> callback is defined and returns a |
2023 | * value other than <code>u_getIntPropertyMaxValue(UCHAR_BIDI_CLASS)+1</code>, |
2024 | * that value is used; otherwise the default class determination mechanism is invoked.</p> |
2025 | * |
2026 | * @param pBiDi is the paragraph <code>UBiDi</code> object. |
2027 | * |
2028 | * @param c is the code point whose Bidi class must be retrieved. |
2029 | * |
2030 | * @return The Bidi class for character <code>c</code> based |
2031 | * on the given <code>pBiDi</code> instance. |
2032 | * @see UBiDiClassCallback |
2033 | * @stable ICU 3.6 |
2034 | */ |
2035 | U_CAPI UCharDirection U_EXPORT2 |
2036 | ubidi_getCustomizedClass(UBiDi *pBiDi, UChar32 c); |
2037 | |
2038 | /** |
2039 | * Set the callback function and callback data used by the UBA |
2040 | * implementation for Bidi class determination. |
2041 | * <p>This may be useful for assigning Bidi classes to PUA characters, or |
2042 | * for special application needs. For instance, an application may want to |
2043 | * handle all spaces like L or R characters (according to the base direction) |
2044 | * when creating the visual ordering of logical lines which are part of a report |
2045 | * organized in columns: there should not be interaction between adjacent |
2046 | * cells.<p> |
2047 | * |
2048 | * @param pBiDi is the paragraph <code>UBiDi</code> object. |
2049 | * |
2050 | * @param newFn is the new callback function pointer. |
2051 | * |
2052 | * @param newContext is the new callback context pointer. This can be NULL. |
2053 | * |
2054 | * @param oldFn fillin: Returns the old callback function pointer. This can be |
2055 | * NULL. |
2056 | * |
2057 | * @param oldContext fillin: Returns the old callback's context. This can be |
2058 | * NULL. |
2059 | * |
2060 | * @param pErrorCode must be a valid pointer to an error code value. |
2061 | * |
2062 | * @see ubidi_getClassCallback |
2063 | * @stable ICU 3.6 |
2064 | */ |
2065 | U_CAPI void U_EXPORT2 |
2066 | ubidi_setClassCallback(UBiDi *pBiDi, UBiDiClassCallback *newFn, |
2067 | const void *newContext, UBiDiClassCallback **oldFn, |
2068 | const void **oldContext, UErrorCode *pErrorCode); |
2069 | |
2070 | /** |
2071 | * Get the current callback function used for Bidi class determination. |
2072 | * |
2073 | * @param pBiDi is the paragraph <code>UBiDi</code> object. |
2074 | * |
2075 | * @param fn fillin: Returns the callback function pointer. |
2076 | * |
2077 | * @param context fillin: Returns the callback's private context. |
2078 | * |
2079 | * @see ubidi_setClassCallback |
2080 | * @stable ICU 3.6 |
2081 | */ |
2082 | U_CAPI void U_EXPORT2 |
2083 | ubidi_getClassCallback(UBiDi *pBiDi, UBiDiClassCallback **fn, const void **context); |
2084 | |
2085 | /** |
2086 | * Take a <code>UBiDi</code> object containing the reordering |
2087 | * information for a piece of text (one or more paragraphs) set by |
2088 | * <code>ubidi_setPara()</code> or for a line of text set by |
2089 | * <code>ubidi_setLine()</code> and write a reordered string to the |
2090 | * destination buffer. |
2091 | * |
2092 | * This function preserves the integrity of characters with multiple |
2093 | * code units and (optionally) combining characters. |
2094 | * Characters in RTL runs can be replaced by mirror-image characters |
2095 | * in the destination buffer. Note that "real" mirroring has |
2096 | * to be done in a rendering engine by glyph selection |
2097 | * and that for many "mirrored" characters there are no |
2098 | * Unicode characters as mirror-image equivalents. |
2099 | * There are also options to insert or remove Bidi control |
2100 | * characters; see the description of the <code>destSize</code> |
2101 | * and <code>options</code> parameters and of the option bit flags. |
2102 | * |
2103 | * @param pBiDi A pointer to a <code>UBiDi</code> object that |
2104 | * is set by <code>ubidi_setPara()</code> or |
2105 | * <code>ubidi_setLine()</code> and contains the reordering |
2106 | * information for the text that it was defined for, |
2107 | * as well as a pointer to that text.<br><br> |
2108 | * The text was aliased (only the pointer was stored |
2109 | * without copying the contents) and must not have been modified |
2110 | * since the <code>ubidi_setPara()</code> call. |
2111 | * |
2112 | * @param dest A pointer to where the reordered text is to be copied. |
2113 | * The source text and <code>dest[destSize]</code> |
2114 | * must not overlap. |
2115 | * |
2116 | * @param destSize The size of the <code>dest</code> buffer, |
2117 | * in number of UChars. |
2118 | * If the <code>UBIDI_INSERT_LRM_FOR_NUMERIC</code> |
2119 | * option is set, then the destination length could be |
2120 | * as large as |
2121 | * <code>ubidi_getLength(pBiDi)+2*ubidi_countRuns(pBiDi)</code>. |
2122 | * If the <code>UBIDI_REMOVE_BIDI_CONTROLS</code> option |
2123 | * is set, then the destination length may be less than |
2124 | * <code>ubidi_getLength(pBiDi)</code>. |
2125 | * If none of these options is set, then the destination length |
2126 | * will be exactly <code>ubidi_getProcessedLength(pBiDi)</code>. |
2127 | * |
2128 | * @param options A bit set of options for the reordering that control |
2129 | * how the reordered text is written. |
2130 | * The options include mirroring the characters on a code |
2131 | * point basis and inserting LRM characters, which is used |
2132 | * especially for transforming visually stored text |
2133 | * to logically stored text (although this is still an |
2134 | * imperfect implementation of an "inverse Bidi" algorithm |
2135 | * because it uses the "forward Bidi" algorithm at its core). |
2136 | * The available options are: |
2137 | * <code>#UBIDI_DO_MIRRORING</code>, |
2138 | * <code>#UBIDI_INSERT_LRM_FOR_NUMERIC</code>, |
2139 | * <code>#UBIDI_KEEP_BASE_COMBINING</code>, |
2140 | * <code>#UBIDI_OUTPUT_REVERSE</code>, |
2141 | * <code>#UBIDI_REMOVE_BIDI_CONTROLS</code> |
2142 | * |
2143 | * @param pErrorCode must be a valid pointer to an error code value. |
2144 | * |
2145 | * @return The length of the output string. |
2146 | * |
2147 | * @see ubidi_getProcessedLength |
2148 | * @stable ICU 2.0 |
2149 | */ |
2150 | U_CAPI int32_t U_EXPORT2 |
2151 | ubidi_writeReordered(UBiDi *pBiDi, |
2152 | UChar *dest, int32_t destSize, |
2153 | uint16_t options, |
2154 | UErrorCode *pErrorCode); |
2155 | |
2156 | /** |
2157 | * Reverse a Right-To-Left run of Unicode text. |
2158 | * |
2159 | * This function preserves the integrity of characters with multiple |
2160 | * code units and (optionally) combining characters. |
2161 | * Characters can be replaced by mirror-image characters |
2162 | * in the destination buffer. Note that "real" mirroring has |
2163 | * to be done in a rendering engine by glyph selection |
2164 | * and that for many "mirrored" characters there are no |
2165 | * Unicode characters as mirror-image equivalents. |
2166 | * There are also options to insert or remove Bidi control |
2167 | * characters. |
2168 | * |
2169 | * This function is the implementation for reversing RTL runs as part |
2170 | * of <code>ubidi_writeReordered()</code>. For detailed descriptions |
2171 | * of the parameters, see there. |
2172 | * Since no Bidi controls are inserted here, the output string length |
2173 | * will never exceed <code>srcLength</code>. |
2174 | * |
2175 | * @see ubidi_writeReordered |
2176 | * |
2177 | * @param src A pointer to the RTL run text. |
2178 | * |
2179 | * @param srcLength The length of the RTL run. |
2180 | * |
2181 | * @param dest A pointer to where the reordered text is to be copied. |
2182 | * <code>src[srcLength]</code> and <code>dest[destSize]</code> |
2183 | * must not overlap. |
2184 | * |
2185 | * @param destSize The size of the <code>dest</code> buffer, |
2186 | * in number of UChars. |
2187 | * If the <code>UBIDI_REMOVE_BIDI_CONTROLS</code> option |
2188 | * is set, then the destination length may be less than |
2189 | * <code>srcLength</code>. |
2190 | * If this option is not set, then the destination length |
2191 | * will be exactly <code>srcLength</code>. |
2192 | * |
2193 | * @param options A bit set of options for the reordering that control |
2194 | * how the reordered text is written. |
2195 | * See the <code>options</code> parameter in <code>ubidi_writeReordered()</code>. |
2196 | * |
2197 | * @param pErrorCode must be a valid pointer to an error code value. |
2198 | * |
2199 | * @return The length of the output string. |
2200 | * @stable ICU 2.0 |
2201 | */ |
2202 | U_CAPI int32_t U_EXPORT2 |
2203 | ubidi_writeReverse(const UChar *src, int32_t srcLength, |
2204 | UChar *dest, int32_t destSize, |
2205 | uint16_t options, |
2206 | UErrorCode *pErrorCode); |
2207 | |
2208 | /*#define BIDI_SAMPLE_CODE*/ |
2209 | /*@}*/ |
2210 | |
2211 | #endif |
2212 | |