1 /* |
|
2 * Portions Copyright 2000-2003 Sun Microsystems, Inc. All Rights Reserved. |
|
3 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. |
|
4 * |
|
5 * This code is free software; you can redistribute it and/or modify it |
|
6 * under the terms of the GNU General Public License version 2 only, as |
|
7 * published by the Free Software Foundation. Sun designates this |
|
8 * particular file as subject to the "Classpath" exception as provided |
|
9 * by Sun in the LICENSE file that accompanied this code. |
|
10 * |
|
11 * This code is distributed in the hope that it will be useful, but WITHOUT |
|
12 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or |
|
13 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License |
|
14 * version 2 for more details (a copy is included in the LICENSE file that |
|
15 * accompanied this code). |
|
16 * |
|
17 * You should have received a copy of the GNU General Public License version |
|
18 * 2 along with this work; if not, write to the Free Software Foundation, |
|
19 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. |
|
20 * |
|
21 * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara, |
|
22 * CA 95054 USA or visit www.sun.com if you need additional information or |
|
23 * have any questions. |
|
24 */ |
|
25 |
|
26 /* |
|
27 * (C) Copyright IBM Corp. 1999-2003 - All Rights Reserved |
|
28 * |
|
29 * The original version of this source code and documentation is |
|
30 * copyrighted and owned by IBM. These materials are provided |
|
31 * under terms of a License Agreement between IBM and Sun. |
|
32 * This technology is protected by multiple US and International |
|
33 * patents. This notice and attribution to IBM may not be removed. |
|
34 */ |
|
35 |
|
36 /* |
|
37 * file name: ubidi.h |
|
38 * encoding: US-ASCII |
|
39 * tab size: 8 (not used) |
|
40 * indentation:4 |
|
41 * |
|
42 * created on: 1999jul27 |
|
43 * created by: Markus W. Scherer |
|
44 */ |
|
45 |
|
46 #ifndef UBIDI_H |
|
47 #define UBIDI_H |
|
48 |
|
49 #include "utypes.h" |
|
50 #include "uchardir.h" |
|
51 |
|
52 /* |
|
53 * javadoc-style comments are intended to be transformed into HTML |
|
54 * using DOC++ - see |
|
55 * http://www.zib.de/Visual/software/doc++/index.html . |
|
56 * |
|
57 * The HTML documentation is created with |
|
58 * doc++ -H ubidi.h |
|
59 * |
|
60 * The following #define trick allows us to do it all in one file |
|
61 * and still be able to compile it. |
|
62 */ |
|
63 #define DOCXX_TAG |
|
64 #define BIDI_SAMPLE_CODE |
|
65 |
|
66 /** |
|
67 * @name BiDi algorithm for ICU |
|
68 * |
|
69 * <h2>BiDi algorithm for ICU</h2> |
|
70 * |
|
71 * This is an implementation of the Unicode Bidirectional algorithm. |
|
72 * The algorithm is defined in the |
|
73 * <a href="http://www.unicode.org/unicode/reports/tr9/">Unicode Technical Report 9</a>, |
|
74 * version 5, also described in The Unicode Standard, Version 3.0 .<p> |
|
75 * |
|
76 * <h3>General remarks about the API:</h3> |
|
77 * |
|
78 * In functions with an error code parameter, |
|
79 * the <code>pErrorCode</code> pointer must be valid |
|
80 * and the value that it points to must not indicate a failure before |
|
81 * the function call. Otherwise, the function returns immediately. |
|
82 * After the function call, the value indicates success or failure.<p> |
|
83 * |
|
84 * The <quote>limit</quote> of a sequence of characters is the position just after their |
|
85 * last character, i.e., one more than that position.<p> |
|
86 * |
|
87 * Some of the API functions provide access to <quote>runs</quote>. |
|
88 * Such a <quote>run</quote> is defined as a sequence of characters |
|
89 * that are at the same embedding level |
|
90 * after performing the BiDi algorithm.<p> |
|
91 * |
|
92 * @author Markus W. Scherer |
|
93 */ |
|
94 DOCXX_TAG |
|
95 /*@{*/ |
|
96 |
|
97 /** |
|
98 * UBiDiLevel is the type of the level values in this |
|
99 * BiDi implementation. |
|
100 * It holds an embedding level and indicates the visual direction |
|
101 * by its bit 0 (even/odd value).<p> |
|
102 * |
|
103 * It can also hold non-level values for the |
|
104 * <code>paraLevel</code> and <code>embeddingLevels</code> |
|
105 * arguments of <code>ubidi_setPara()</code>; there: |
|
106 * <ul> |
|
107 * <li>bit 7 of an <code>embeddingLevels[]</code> |
|
108 * value indicates whether the using application is |
|
109 * specifying the level of a character to <i>override</i> whatever the |
|
110 * BiDi implementation would resolve it to.</li> |
|
111 * <li><code>paraLevel</code> can be set to the |
|
112 * pesudo-level values <code>UBIDI_DEFAULT_LTR</code> |
|
113 * and <code>UBIDI_DEFAULT_RTL</code>.</li> |
|
114 * |
|
115 * @see ubidi_setPara |
|
116 * |
|
117 * <p>The related constants are not real, valid level values. |
|
118 * <code>UBIDI_DEFAULT_XXX</code> can be used to specify |
|
119 * a default for the paragraph level for |
|
120 * when the <code>ubidi_setPara()</code> function |
|
121 * shall determine it but there is no |
|
122 * strongly typed character in the input.<p> |
|
123 * |
|
124 * Note that the value for <code>UBIDI_DEFAULT_LTR</code> is even |
|
125 * and the one for <code>UBIDI_DEFAULT_RTL</code> is odd, |
|
126 * just like with normal LTR and RTL level values - |
|
127 * these special values are designed that way. Also, the implementation |
|
128 * assumes that UBIDI_MAX_EXPLICIT_LEVEL is odd. |
|
129 * |
|
130 * @see UBIDI_DEFAULT_LTR |
|
131 * @see UBIDI_DEFAULT_RTL |
|
132 * @see UBIDI_LEVEL_OVERRIDE |
|
133 * @see UBIDI_MAX_EXPLICIT_LEVEL |
|
134 */ |
|
135 typedef uint8_t UBiDiLevel; |
|
136 |
|
137 /** @memo If there is no strong character, then set the paragraph level to 0 (left-to-right). */ |
|
138 #define UBIDI_DEFAULT_LTR 0xfe |
|
139 |
|
140 /** @memo If there is no strong character, then set the paragraph level to 1 (right-to-left). */ |
|
141 #define UBIDI_DEFAULT_RTL 0xff |
|
142 |
|
143 /** |
|
144 * @memo Maximum explicit embedding level |
|
145 * (The maximum resolved level can be up to <code>UBIDI_MAX_EXPLICIT_LEVEL+1</code>). |
|
146 */ |
|
147 #define UBIDI_MAX_EXPLICIT_LEVEL 61 |
|
148 |
|
149 /** @memo Bit flag for level input: overrides directional properties. */ |
|
150 #define UBIDI_LEVEL_OVERRIDE 0x80 |
|
151 |
|
152 /** |
|
153 * @memo <code>UBiDiDirection</code> values indicate the text direction. |
|
154 */ |
|
155 enum UBiDiDirection { |
|
156 /** @memo All left-to-right text. This is a 0 value. */ |
|
157 UBIDI_LTR, |
|
158 /** @memo All right-to-left text. This is a 1 value. */ |
|
159 UBIDI_RTL, |
|
160 /** @memo Mixed-directional text. */ |
|
161 UBIDI_MIXED |
|
162 }; |
|
163 |
|
164 typedef enum UBiDiDirection UBiDiDirection; |
|
165 |
|
166 /** |
|
167 * Forward declaration of the <code>UBiDi</code> structure for the declaration of |
|
168 * the API functions. Its fields are implementation-specific.<p> |
|
169 * This structure holds information about a paragraph of text |
|
170 * with BiDi-algorithm-related details, or about one line of |
|
171 * such a paragraph.<p> |
|
172 * Reordering can be done on a line, or on a paragraph which is |
|
173 * then interpreted as one single line. |
|
174 */ |
|
175 struct UBiDi; |
|
176 |
|
177 typedef struct UBiDi UBiDi; |
|
178 |
|
179 /** |
|
180 * Allocate a <code>UBiDi</code> structure. |
|
181 * Such an object is initially empty. It is assigned |
|
182 * the BiDi properties of a paragraph by <code>ubidi_setPara()</code> |
|
183 * or the BiDi properties of a line of a paragraph by |
|
184 * <code>ubidi_getLine()</code>.<p> |
|
185 * This object can be reused for as long as it is not deallocated |
|
186 * by calling <code>ubidi_close()</code>.<p> |
|
187 * <code>ubidi_set()</code> will allocate additional memory for |
|
188 * internal structures as necessary. |
|
189 * |
|
190 * @return An empty <code>UBiDi</code> object. |
|
191 */ |
|
192 U_CAPI UBiDi * U_EXPORT2 |
|
193 ubidi_open(); |
|
194 |
|
195 /** |
|
196 * Allocate a <code>UBiDi</code> structure with preallocated memory |
|
197 * for internal structures. |
|
198 * This function provides a <code>UBiDi</code> object like <code>ubidi_open()</code> |
|
199 * with no arguments, but it also preallocates memory for internal structures |
|
200 * according to the sizings supplied by the caller.<p> |
|
201 * Subsequent functions will not allocate any more memory, and are thus |
|
202 * guaranteed not to fail because of lack of memory.<p> |
|
203 * The preallocation can be limited to some of the internal memory |
|
204 * by setting some values to 0 here. That means that if, e.g., |
|
205 * <code>maxRunCount</code> cannot be reasonably predetermined and should not |
|
206 * be set to <code>maxLength</code> (the only failproof value) to avoid |
|
207 * wasting memory, then <code>maxRunCount</code> could be set to 0 here |
|
208 * and the internal structures that are associated with it will be allocated |
|
209 * on demand, just like with <code>ubidi_open()</code>. |
|
210 * |
|
211 * @param maxLength is the maximum paragraph or line length that internal memory |
|
212 * will be preallocated for. An attempt to associate this object with a |
|
213 * longer text will fail, unless this value is 0, which leaves the allocation |
|
214 * up to the implementation. |
|
215 * |
|
216 * @param maxRunCount is the maximum anticipated number of same-level runs |
|
217 * that internal memory will be preallocated for. An attempt to access |
|
218 * visual runs on an object that was not preallocated for as many runs |
|
219 * as the text was actually resolved to will fail, |
|
220 * unless this value is 0, which leaves the allocation up to the implementation.<p> |
|
221 * The number of runs depends on the actual text and maybe anywhere between |
|
222 * 1 and <code>maxLength</code>. It is typically small.<p> |
|
223 * |
|
224 * @param pErrorCode must be a valid pointer to an error code value, |
|
225 * which must not indicate a failure before the function call. |
|
226 * |
|
227 * @return An empty <code>UBiDi</code> object with preallocated memory. |
|
228 */ |
|
229 U_CAPI UBiDi * U_EXPORT2 |
|
230 ubidi_openSized(int32_t maxLength, int32_t maxRunCount, UErrorCode *pErrorCode); |
|
231 |
|
232 /** |
|
233 * <code>ubidi_close()</code> must be called to free the memory |
|
234 * associated with a UBiDi object.<p> |
|
235 * |
|
236 * <strong>Important: </strong> |
|
237 * If a <code>UBiDi</code> object is the <quote>child</quote> |
|
238 * of another one (its <quote>parent</quote>), after calling |
|
239 * <code>ubidi_setLine()</code>, then the child object must |
|
240 * be destroyed (closed) or reused (by calling |
|
241 * <code>ubidi_setPara()</code> or <code>ubidi_setLine()</code>) |
|
242 * before the parent object. |
|
243 * |
|
244 * @param pBiDi is a <code>UBiDi</code> object. |
|
245 * |
|
246 * @see ubidi_setPara |
|
247 * @see ubidi_setLine |
|
248 */ |
|
249 U_CAPI void U_EXPORT2 |
|
250 ubidi_close(UBiDi *pBiDi); |
|
251 |
|
252 /** |
|
253 * Perform the Unicode BiDi algorithm. It is defined in the |
|
254 * <a href="http://www.unicode.org/unicode/reports/tr9/">Unicode Technical Report 9</a>, |
|
255 * version 5, |
|
256 * also described in The Unicode Standard, Version 3.0 .<p> |
|
257 * |
|
258 * This function takes a single plain text paragraph with or without |
|
259 * externally specified embedding levels from <quote>styled</quote> text |
|
260 * and computes the left-right-directionality of each character.<p> |
|
261 * |
|
262 * If the entire paragraph consists of text of only one direction, then |
|
263 * the function may not perform all the steps described by the algorithm, |
|
264 * i.e., some levels may not be the same as if all steps were performed. |
|
265 * This is not relevant for unidirectional text.<br> |
|
266 * For example, in pure LTR text with numbers the numbers would get |
|
267 * a resolved level of 2 higher than the surrounding text according to |
|
268 * the algorithm. This implementation may set all resolved levels to |
|
269 * the same value in such a case.<p> |
|
270 * |
|
271 * The text must be externally split into separate paragraphs (rule P1). |
|
272 * Paragraph separators (B) should appear at most at the very end. |
|
273 * |
|
274 * @param pBiDi A <code>UBiDi</code> object allocated with <code>ubidi_open()</code> |
|
275 * which will be set to contain the reordering information, |
|
276 * especially the resolved levels for all the characters in <code>text</code>. |
|
277 * |
|
278 * @param text is a pointer to the single-paragraph text that the |
|
279 * BiDi algorithm will be performed on |
|
280 * (step (P1) of the algorithm is performed externally). |
|
281 * <strong>The text must be (at least) <code>length</code> long.</strong> |
|
282 * |
|
283 * @param length is the length of the text; if <code>length==-1</code> then |
|
284 * the text must be zero-terminated. |
|
285 * |
|
286 * @param paraLevel specifies the default level for the paragraph; |
|
287 * it is typically 0 (LTR) or 1 (RTL). |
|
288 * If the function shall determine the paragraph level from the text, |
|
289 * then <code>paraLevel</code> can be set to |
|
290 * either <code>UBIDI_DEFAULT_LTR</code> |
|
291 * or <code>UBIDI_DEFAULT_RTL</code>; |
|
292 * if there is no strongly typed character, then |
|
293 * the desired default is used (0 for LTR or 1 for RTL). |
|
294 * Any other value between 0 and <code>UBIDI_MAX_EXPLICIT_LEVEL</code> is also valid, |
|
295 * with odd levels indicating RTL. |
|
296 * |
|
297 * @param embeddingLevels (in) may be used to preset the embedding and override levels, |
|
298 * ignoring characters like LRE and PDF in the text. |
|
299 * A level overrides the directional property of its corresponding |
|
300 * (same index) character if the level has the |
|
301 * <code>UBIDI_LEVEL_OVERRIDE</code> bit set.<p> |
|
302 * Except for that bit, it must be |
|
303 * <code>paraLevel<=embeddingLevels[]<=UBIDI_MAX_EXPLICIT_LEVEL</code>.<p> |
|
304 * <strong>Caution: </strong>A copy of this pointer, not of the levels, |
|
305 * will be stored in the <code>UBiDi</code> object; |
|
306 * the <code>embeddingLevels</code> array must not be |
|
307 * deallocated before the <code>UBiDi</code> structure is destroyed or reused, |
|
308 * and the <code>embeddingLevels</code> |
|
309 * should not be modified to avoid unexpected results on subsequent BiDi operations. |
|
310 * However, the <code>ubidi_setPara()</code> and |
|
311 * <code>ubidi_setLine()</code> functions may modify some or all of the levels.<p> |
|
312 * After the <code>UBiDi</code> object is reused or destroyed, the caller |
|
313 * must take care of the deallocation of the <code>embeddingLevels</code> array.<p> |
|
314 * <strong>The <code>embeddingLevels</code> array must be |
|
315 * at least <code>length</code> long.</strong> |
|
316 * |
|
317 * @param pErrorCode must be a valid pointer to an error code value, |
|
318 * which must not indicate a failure before the function call. |
|
319 */ |
|
320 U_CAPI void U_EXPORT2 |
|
321 ubidi_setPara(UBiDi *pBiDi, const UChar *text, int32_t length, |
|
322 UBiDiLevel paraLevel, UBiDiLevel *embeddingLevels, |
|
323 UErrorCode *pErrorCode); |
|
324 |
|
325 /** |
|
326 * <code>ubidi_getLine()</code> sets a <code>UBiDi</code> to |
|
327 * contain the reordering information, especially the resolved levels, |
|
328 * for all the characters in a line of text. This line of text is |
|
329 * specified by referring to a <code>UBiDi</code> object representing |
|
330 * this information for a paragraph of text, and by specifying |
|
331 * a range of indexes in this paragraph.<p> |
|
332 * In the new line object, the indexes will range from 0 to <code>limit-start</code>.<p> |
|
333 * |
|
334 * This is used after calling <code>ubidi_setPara()</code> |
|
335 * for a paragraph, and after line-breaking on that paragraph. |
|
336 * It is not necessary if the paragraph is treated as a single line.<p> |
|
337 * |
|
338 * After line-breaking, rules (L1) and (L2) for the treatment of |
|
339 * trailing WS and for reordering are performed on |
|
340 * a <code>UBiDi</code> object that represents a line.<p> |
|
341 * |
|
342 * <strong>Important: </strong><code>pLineBiDi</code> shares data with |
|
343 * <code>pParaBiDi</code>. |
|
344 * You must destroy or reuse <code>pLineBiDi</code> before <code>pParaBiDi</code>. |
|
345 * In other words, you must destroy or reuse the <code>UBiDi</code> object for a line |
|
346 * before the object for its parent paragraph. |
|
347 * |
|
348 * @param pParaBiDi is the parent paragraph object. |
|
349 * |
|
350 * @param start is the line's first index into the paragraph text. |
|
351 * |
|
352 * @param limit is just behind the line's last index into the paragraph text |
|
353 * (its last index +1).<br> |
|
354 * It must be <code>0<=start<=limit<=</code>paragraph length. |
|
355 * |
|
356 * @param pLineBiDi is the object that will now represent a line of the paragraph. |
|
357 * |
|
358 * @param pErrorCode must be a valid pointer to an error code value, |
|
359 * which must not indicate a failure before the function call. |
|
360 * |
|
361 * @see ubidi_setPara |
|
362 */ |
|
363 U_CAPI void U_EXPORT2 |
|
364 ubidi_setLine(const UBiDi *pParaBiDi, |
|
365 int32_t start, int32_t limit, |
|
366 UBiDi *pLineBiDi, |
|
367 UErrorCode *pErrorCode); |
|
368 |
|
369 /** |
|
370 * Get the directionality of the text. |
|
371 * |
|
372 * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
|
373 * |
|
374 * @return A <code>UBIDI_XXX</code> value that indicates if the entire text |
|
375 * represented by this object is unidirectional, |
|
376 * and which direction, or if it is mixed-directional. |
|
377 * |
|
378 * @see UBiDiDirection |
|
379 */ |
|
380 U_CAPI UBiDiDirection U_EXPORT2 |
|
381 ubidi_getDirection(const UBiDi *pBiDi); |
|
382 |
|
383 /** |
|
384 * Get the length of the text. |
|
385 * |
|
386 * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
|
387 * |
|
388 * @return The length of the text that the UBiDi object was created for. |
|
389 */ |
|
390 U_CAPI int32_t U_EXPORT2 |
|
391 ubidi_getLength(const UBiDi *pBiDi); |
|
392 |
|
393 /** |
|
394 * Get the paragraph level of the text. |
|
395 * |
|
396 * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
|
397 * |
|
398 * @return The paragraph level. |
|
399 * |
|
400 * @see UBiDiLevel |
|
401 */ |
|
402 U_CAPI UBiDiLevel U_EXPORT2 |
|
403 ubidi_getParaLevel(const UBiDi *pBiDi); |
|
404 |
|
405 /** |
|
406 * Get the level for one character. |
|
407 * |
|
408 * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
|
409 * |
|
410 * @param charIndex the index of a character. |
|
411 * |
|
412 * @return The level for the character at charIndex. |
|
413 * |
|
414 * @see UBiDiLevel |
|
415 */ |
|
416 U_CAPI UBiDiLevel U_EXPORT2 |
|
417 ubidi_getLevelAt(const UBiDi *pBiDi, int32_t charIndex); |
|
418 |
|
419 /** |
|
420 * Get an array of levels for each character.<p> |
|
421 * |
|
422 * Note that this function may allocate memory under some |
|
423 * circumstances, unlike <code>ubidi_getLevelAt()</code>. |
|
424 * |
|
425 * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
|
426 * |
|
427 * @param pErrorCode must be a valid pointer to an error code value, |
|
428 * which must not indicate a failure before the function call. |
|
429 * |
|
430 * @return The levels array for the text, |
|
431 * or <code>NULL</code> if an error occurs. |
|
432 * |
|
433 * @see UBiDiLevel |
|
434 */ |
|
435 U_CAPI const UBiDiLevel * U_EXPORT2 |
|
436 ubidi_getLevels(UBiDi *pBiDi, UErrorCode *pErrorCode); |
|
437 |
|
438 /** |
|
439 * Get a logical run. |
|
440 * This function returns information about a run and is used |
|
441 * to retrieve runs in logical order.<p> |
|
442 * This is especially useful for line-breaking on a paragraph. |
|
443 * |
|
444 * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
|
445 * |
|
446 * @param logicalStart is the first character of the run. |
|
447 * |
|
448 * @param pLogicalLimit will receive the limit of the run. |
|
449 * The l-value that you point to here may be the |
|
450 * same expression (variable) as the one for |
|
451 * <code>logicalStart</code>. |
|
452 * This pointer can be <code>NULL</code> if this |
|
453 * value is not necessary. |
|
454 * |
|
455 * @param pLevel will receive the level of the run. |
|
456 * This pointer can be <code>NULL</code> if this |
|
457 * value is not necessary. |
|
458 */ |
|
459 U_CAPI void U_EXPORT2 |
|
460 ubidi_getLogicalRun(const UBiDi *pBiDi, int32_t logicalStart, |
|
461 int32_t *pLogicalLimit, UBiDiLevel *pLevel); |
|
462 |
|
463 /** |
|
464 * Get the number of runs. |
|
465 * This function may invoke the actual reordering on the |
|
466 * <code>UBiDi</code> object, after <code>ubidi_setPara()</code> |
|
467 * may have resolved only the levels of the text. Therefore, |
|
468 * <code>ubidi_countRuns()</code> may have to allocate memory, |
|
469 * and may fail doing so. |
|
470 * |
|
471 * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
|
472 * |
|
473 * @param pErrorCode must be a valid pointer to an error code value, |
|
474 * which must not indicate a failure before the function call. |
|
475 * |
|
476 * @return The number of runs. |
|
477 */ |
|
478 U_CAPI int32_t U_EXPORT2 |
|
479 ubidi_countRuns(UBiDi *pBiDi, UErrorCode *pErrorCode); |
|
480 |
|
481 /** |
|
482 * Get one run's logical start, length, and directionality, |
|
483 * which can be 0 for LTR or 1 for RTL. |
|
484 * In an RTL run, the character at the logical start is |
|
485 * visually on the right of the displayed run. |
|
486 * The length is the number of characters in the run.<p> |
|
487 * <code>ubidi_countRuns()</code> should be called |
|
488 * before the runs are retrieved. |
|
489 * |
|
490 * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
|
491 * |
|
492 * @param runIndex is the number of the run in visual order, in the |
|
493 * range <code>[0..ubidi_countRuns(pBiDi)-1]</code>. |
|
494 * |
|
495 * @param pLogicalStart is the first logical character index in the text. |
|
496 * The pointer may be <code>NULL</code> if this index is not needed. |
|
497 * |
|
498 * @param pLength is the number of characters (at least one) in the run. |
|
499 * The pointer may be <code>NULL</code> if this is not needed. |
|
500 * |
|
501 * @return the directionality of the run, |
|
502 * <code>UBIDI_LTR==0</code> or <code>UBIDI_RTL==1</code>, |
|
503 * never <code>UBIDI_MIXED</code>. |
|
504 * |
|
505 * @see ubidi_countRuns |
|
506 * |
|
507 * Example: |
|
508 * <pre> |
|
509 * int32_t i, count=ubidi_countRuns(pBiDi), |
|
510 * logicalStart, visualIndex=0, length; |
|
511 * for(i=0; i<count; ++i) { |
|
512 * if(UBIDI_LTR==ubidi_getVisualRun(pBiDi, i, &logicalStart, &length)) { |
|
513 * do { // LTR |
|
514 * show_char(text[logicalStart++], visualIndex++); |
|
515 * } while(--length>0); |
|
516 * } else { |
|
517 * logicalStart+=length; // logicalLimit |
|
518 * do { // RTL |
|
519 * show_char(text[--logicalStart], visualIndex++); |
|
520 * } while(--length>0); |
|
521 * } |
|
522 * } |
|
523 * </pre> |
|
524 * |
|
525 * Note that in right-to-left runs, code like this places |
|
526 * modifier letters before base characters and second surrogates |
|
527 * before first ones. |
|
528 */ |
|
529 U_CAPI UBiDiDirection U_EXPORT2 |
|
530 ubidi_getVisualRun(UBiDi *pBiDi, int32_t runIndex, |
|
531 int32_t *pLogicalStart, int32_t *pLength); |
|
532 |
|
533 /** |
|
534 * Get the visual position from a logical text position. |
|
535 * If such a mapping is used many times on the same |
|
536 * <code>UBiDi</code> object, then calling |
|
537 * <code>ubidi_getLogicalMap()</code> is more efficient.<p> |
|
538 * |
|
539 * Note that in right-to-left runs, this mapping places |
|
540 * modifier letters before base characters and second surrogates |
|
541 * before first ones. |
|
542 * |
|
543 * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
|
544 * |
|
545 * @param logicalIndex is the index of a character in the text. |
|
546 * |
|
547 * @param pErrorCode must be a valid pointer to an error code value, |
|
548 * which must not indicate a failure before the function call. |
|
549 * |
|
550 * @return The visual position of this character. |
|
551 * |
|
552 * @see ubidi_getLogicalMap |
|
553 * @see ubidi_getLogicalIndex |
|
554 */ |
|
555 U_CAPI int32_t U_EXPORT2 |
|
556 ubidi_getVisualIndex(UBiDi *pBiDi, int32_t logicalIndex, UErrorCode *pErrorCode); |
|
557 |
|
558 /** |
|
559 * Get the logical text position from a visual position. |
|
560 * If such a mapping is used many times on the same |
|
561 * <code>UBiDi</code> object, then calling |
|
562 * <code>ubidi_getVisualMap()</code> is more efficient.<p> |
|
563 * |
|
564 * This is the inverse function to <code>ubidi_getVisualIndex()</code>. |
|
565 * |
|
566 * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
|
567 * |
|
568 * @param visualIndex is the visual position of a character. |
|
569 * |
|
570 * @param pErrorCode must be a valid pointer to an error code value, |
|
571 * which must not indicate a failure before the function call. |
|
572 * |
|
573 * @return The index of this character in the text. |
|
574 * |
|
575 * @see ubidi_getVisualMap |
|
576 * @see ubidi_getVisualIndex |
|
577 */ |
|
578 U_CAPI int32_t U_EXPORT2 |
|
579 ubidi_getLogicalIndex(UBiDi *pBiDi, int32_t visualIndex, UErrorCode *pErrorCode); |
|
580 |
|
581 /** |
|
582 * Get a logical-to-visual index map (array) for the characters in the UBiDi |
|
583 * (paragraph or line) object. |
|
584 * |
|
585 * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
|
586 * |
|
587 * @param indexMap is a pointer to an array of <code>ubidi_getLength()</code> |
|
588 * indexes which will reflect the reordering of the characters. |
|
589 * The array does not need to be initialized.<p> |
|
590 * The index map will result in <code>indexMap[logicalIndex]==visualIndex</code>.<p> |
|
591 * |
|
592 * @param pErrorCode must be a valid pointer to an error code value, |
|
593 * which must not indicate a failure before the function call. |
|
594 * |
|
595 * @see ubidi_getVisualMap |
|
596 * @see ubidi_getVisualIndex |
|
597 */ |
|
598 U_CAPI void U_EXPORT2 |
|
599 ubidi_getLogicalMap(UBiDi *pBiDi, int32_t *indexMap, UErrorCode *pErrorCode); |
|
600 |
|
601 /** |
|
602 * Get a visual-to-logical index map (array) for the characters in the UBiDi |
|
603 * (paragraph or line) object. |
|
604 * |
|
605 * @param pBiDi is the paragraph or line <code>UBiDi</code> object. |
|
606 * |
|
607 * @param indexMap is a pointer to an array of <code>ubidi_getLength()</code> |
|
608 * indexes which will reflect the reordering of the characters. |
|
609 * The array does not need to be initialized.<p> |
|
610 * The index map will result in <code>indexMap[visualIndex]==logicalIndex</code>.<p> |
|
611 * |
|
612 * @param pErrorCode must be a valid pointer to an error code value, |
|
613 * which must not indicate a failure before the function call. |
|
614 * |
|
615 * @see ubidi_getLogicalMap |
|
616 * @see ubidi_getLogicalIndex |
|
617 */ |
|
618 U_CAPI void U_EXPORT2 |
|
619 ubidi_getVisualMap(UBiDi *pBiDi, int32_t *indexMap, UErrorCode *pErrorCode); |
|
620 |
|
621 /** |
|
622 * This is a convenience function that does not use a UBiDi object. |
|
623 * It is intended to be used for when an application has determined the levels |
|
624 * of objects (character sequences) and just needs to have them reordered (L2). |
|
625 * This is equivalent to using <code>ubidi_getLogicalMap</code> on a |
|
626 * <code>UBiDi</code> object. |
|
627 * |
|
628 * @param levels is an array with <code>length</code> levels that have been determined by |
|
629 * the application. |
|
630 * |
|
631 * @param length is the number of levels in the array, or, semantically, |
|
632 * the number of objects to be reordered. |
|
633 * It must be <code>length>0</code>. |
|
634 * |
|
635 * @param indexMap is a pointer to an array of <code>length</code> |
|
636 * indexes which will reflect the reordering of the characters. |
|
637 * The array does not need to be initialized.<p> |
|
638 * The index map will result in <code>indexMap[logicalIndex]==visualIndex</code>. |
|
639 */ |
|
640 U_CAPI void U_EXPORT2 |
|
641 ubidi_reorderLogical(const UBiDiLevel *levels, int32_t length, int32_t *indexMap); |
|
642 |
|
643 /** |
|
644 * This is a convenience function that does not use a UBiDi object. |
|
645 * It is intended to be used for when an application has determined the levels |
|
646 * of objects (character sequences) and just needs to have them reordered (L2). |
|
647 * This is equivalent to using <code>ubidi_getVisualMap</code> on a |
|
648 * <code>UBiDi</code> object. |
|
649 * |
|
650 * @param levels is an array with <code>length</code> levels that have been determined by |
|
651 * the application. |
|
652 * |
|
653 * @param length is the number of levels in the array, or, semantically, |
|
654 * the number of objects to be reordered. |
|
655 * It must be <code>length>0</code>. |
|
656 * |
|
657 * @param indexMap is a pointer to an array of <code>length</code> |
|
658 * indexes which will reflect the reordering of the characters. |
|
659 * The array does not need to be initialized.<p> |
|
660 * The index map will result in <code>indexMap[visualIndex]==logicalIndex</code>. |
|
661 */ |
|
662 U_CAPI void U_EXPORT2 |
|
663 ubidi_reorderVisual(const UBiDiLevel *levels, int32_t length, int32_t *indexMap); |
|
664 |
|
665 /** |
|
666 * Invert an index map. |
|
667 * The one-to-one index mapping of the first map is inverted and written to |
|
668 * the second one. |
|
669 * |
|
670 * @param srcMap is an array with <code>length</code> indexes |
|
671 * which define the original mapping. |
|
672 * |
|
673 * @param destMap is an array with <code>length</code> indexes |
|
674 * which will be filled with the inverse mapping. |
|
675 * |
|
676 * @param length is the length of each array. |
|
677 */ |
|
678 U_CAPI void U_EXPORT2 |
|
679 ubidi_invertMap(const int32_t *srcMap, int32_t *destMap, int32_t length); |
|
680 |
|
681 /** |
|
682 * @name Sample code for the ICU BiDi API |
|
683 * |
|
684 * <h2>Rendering a paragraph with the ICU BiDi API</h2> |
|
685 * |
|
686 * This is (hypothetical) sample code that illustrates |
|
687 * how the ICU BiDi API could be used to render a paragraph of text. |
|
688 * Rendering code depends highly on the graphics system, |
|
689 * therefore this sample code must make a lot of assumptions, |
|
690 * which may or may not match any existing graphics system's properties. |
|
691 * |
|
692 * <p>The basic assumptions are:</p> |
|
693 * <ul> |
|
694 * <li>Rendering is done from left to right on a horizontal line.</li> |
|
695 * <li>A run of single-style, unidirectional text can be rendered at once.</li> |
|
696 * <li>Such a run of text is passed to the graphics system with |
|
697 * characters (code units) in logical order.</li> |
|
698 * <li>The line-breaking algorithm is very complicated |
|
699 * and Locale-dependent - |
|
700 * and therefore its implementation omitted from this sample code.</li> |
|
701 * </ul> |
|
702 * |
|
703 * <pre> |
|
704 * #include "ubidi.h" |
|
705 * |
|
706 * typedef enum { |
|
707 * styleNormal=0, styleSelected=1, |
|
708 * styleBold=2, styleItalics=4, |
|
709 * styleSuper=8, styleSub=16 |
|
710 * } Style; |
|
711 * |
|
712 * typedef struct { int32_t limit; Style style; } StyleRun; |
|
713 * |
|
714 * int getTextWidth(const UChar *text, int32_t start, int32_t limit, |
|
715 * const StyleRun *styleRuns, int styleRunCount); |
|
716 * |
|
717 * // set *pLimit and *pStyleRunLimit for a line |
|
718 * // from text[start] and from styleRuns[styleRunStart] |
|
719 * // using ubidi_getLogicalRun(para, ...) |
|
720 * void getLineBreak(const UChar *text, int32_t start, int32_t *pLimit, |
|
721 * UBiDi *para, |
|
722 * const StyleRun *styleRuns, int styleRunStart, int *pStyleRunLimit, |
|
723 * int *pLineWidth); |
|
724 * |
|
725 * // render runs on a line sequentially, always from left to right |
|
726 * |
|
727 * // prepare rendering a new line |
|
728 * void startLine(UBiDiDirection textDirection, int lineWidth); |
|
729 * |
|
730 * // render a run of text and advance to the right by the run width |
|
731 * // the text[start..limit-1] is always in logical order |
|
732 * void renderRun(const UChar *text, int32_t start, int32_t limit, |
|
733 * UBiDiDirection textDirection, Style style); |
|
734 * |
|
735 * // We could compute a cross-product |
|
736 * // from the style runs with the directional runs |
|
737 * // and then reorder it. |
|
738 * // Instead, here we iterate over each run type |
|
739 * // and render the intersections - |
|
740 * // with shortcuts in simple (and common) cases. |
|
741 * // renderParagraph() is the main function. |
|
742 * |
|
743 * // render a directional run with |
|
744 * // (possibly) multiple style runs intersecting with it |
|
745 * void renderDirectionalRun(const UChar *text, |
|
746 * int32_t start, int32_t limit, |
|
747 * UBiDiDirection direction, |
|
748 * const StyleRun *styleRuns, int styleRunCount) { |
|
749 * int i; |
|
750 * |
|
751 * // iterate over style runs |
|
752 * if(direction==UBIDI_LTR) { |
|
753 * int styleLimit; |
|
754 * |
|
755 * for(i=0; i<styleRunCount; ++i) { |
|
756 * styleLimit=styleRun[i].limit; |
|
757 * if(start<styleLimit) { |
|
758 * if(styleLimit>limit) { styleLimit=limit; } |
|
759 * renderRun(text, start, styleLimit, |
|
760 * direction, styleRun[i].style); |
|
761 * if(styleLimit==limit) { break; } |
|
762 * start=styleLimit; |
|
763 * } |
|
764 * } |
|
765 * } else { |
|
766 * int styleStart; |
|
767 * |
|
768 * for(i=styleRunCount-1; i>=0; --i) { |
|
769 * if(i>0) { |
|
770 * styleStart=styleRun[i-1].limit; |
|
771 * } else { |
|
772 * styleStart=0; |
|
773 * } |
|
774 * if(limit>=styleStart) { |
|
775 * if(styleStart<start) { styleStart=start; } |
|
776 * renderRun(text, styleStart, limit, |
|
777 * direction, styleRun[i].style); |
|
778 * if(styleStart==start) { break; } |
|
779 * limit=styleStart; |
|
780 * } |
|
781 * } |
|
782 * } |
|
783 * } |
|
784 * |
|
785 * // the line object represents text[start..limit-1] |
|
786 * void renderLine(UBiDi *line, const UChar *text, |
|
787 * int32_t start, int32_t limit, |
|
788 * const StyleRun *styleRuns, int styleRunCount) { |
|
789 * UBiDiDirection direction=ubidi_getDirection(line); |
|
790 * if(direction!=UBIDI_MIXED) { |
|
791 * // unidirectional |
|
792 * if(styleRunCount<=1) { |
|
793 * renderRun(text, start, limit, direction, styleRuns[0].style); |
|
794 * } else { |
|
795 * renderDirectionalRun(text, start, limit, |
|
796 * direction, styleRuns, styleRunCount); |
|
797 * } |
|
798 * } else { |
|
799 * // mixed-directional |
|
800 * int32_t count, i, length; |
|
801 * UBiDiLevel level; |
|
802 * |
|
803 * count=ubidi_countRuns(para, pErrorCode); |
|
804 * if(U_SUCCESS(*pErrorCode)) { |
|
805 * if(styleRunCount<=1) { |
|
806 * Style style=styleRuns[0].style; |
|
807 * |
|
808 * // iterate over directional runs |
|
809 * for(i=0; i<count; ++i) { |
|
810 * direction=ubidi_getVisualRun(para, i, &start, &length); |
|
811 * renderRun(text, start, start+length, direction, style); |
|
812 * } |
|
813 * } else { |
|
814 * int32_t j; |
|
815 * |
|
816 * // iterate over both directional and style runs |
|
817 * for(i=0; i<count; ++i) { |
|
818 * direction=ubidi_getVisualRun(line, i, &start, &length); |
|
819 * renderDirectionalRun(text, start, start+length, |
|
820 * direction, styleRuns, styleRunCount); |
|
821 * } |
|
822 * } |
|
823 * } |
|
824 * } |
|
825 * } |
|
826 * |
|
827 * void renderParagraph(const UChar *text, int32_t length, |
|
828 * UBiDiDirection textDirection, |
|
829 * const StyleRun *styleRuns, int styleRunCount, |
|
830 * int lineWidth, |
|
831 * UErrorCode *pErrorCode) { |
|
832 * UBiDi *para; |
|
833 * |
|
834 * if(pErrorCode==NULL || U_FAILURE(*pErrorCode) || length<=0) { |
|
835 * return; |
|
836 * } |
|
837 * |
|
838 * para=ubidi_openSized(length, 0, pErrorCode); |
|
839 * if(para==NULL) { return; } |
|
840 * |
|
841 * ubidi_setPara(para, text, length, |
|
842 * textDirection ? UBIDI_DEFAULT_RTL : UBIDI_DEFAULT_LTR, |
|
843 * NULL, pErrorCode); |
|
844 * if(U_SUCCESS(*pErrorCode)) { |
|
845 * UBiDiLevel paraLevel=1&ubidi_getParaLevel(para); |
|
846 * StyleRun styleRun={ length, styleNormal }; |
|
847 * int width; |
|
848 * |
|
849 * if(styleRuns==NULL || styleRunCount<=0) { |
|
850 * styleRunCount=1; |
|
851 * styleRuns=&styleRun; |
|
852 * } |
|
853 * |
|
854 * // assume styleRuns[styleRunCount-1].limit>=length |
|
855 * |
|
856 * width=getTextWidth(text, 0, length, styleRuns, styleRunCount); |
|
857 * if(width<=lineWidth) { |
|
858 * // everything fits onto one line |
|
859 * |
|
860 * // prepare rendering a new line from either left or right |
|
861 * startLine(paraLevel, width); |
|
862 * |
|
863 * renderLine(para, text, 0, length, |
|
864 * styleRuns, styleRunCount); |
|
865 * } else { |
|
866 * UBiDi *line; |
|
867 * |
|
868 * // we need to render several lines |
|
869 * line=ubidi_openSized(length, 0, pErrorCode); |
|
870 * if(line!=NULL) { |
|
871 * int32_t start=0, limit; |
|
872 * int styleRunStart=0, styleRunLimit; |
|
873 * |
|
874 * for(;;) { |
|
875 * limit=length; |
|
876 * styleRunLimit=styleRunCount; |
|
877 * getLineBreak(text, start, &limit, para, |
|
878 * styleRuns, styleRunStart, &styleRunLimit, |
|
879 * &width); |
|
880 * ubidi_setLine(para, start, limit, line, pErrorCode); |
|
881 * if(U_SUCCESS(*pErrorCode)) { |
|
882 * // prepare rendering a new line |
|
883 * // from either left or right |
|
884 * startLine(paraLevel, width); |
|
885 * |
|
886 * renderLine(line, text, start, limit, |
|
887 * styleRuns+styleRunStart, |
|
888 * styleRunLimit-styleRunStart); |
|
889 * } |
|
890 * if(limit==length) { break; } |
|
891 * start=limit; |
|
892 * styleRunStart=styleRunLimit-1; |
|
893 * if(start>=styleRuns[styleRunStart].limit) { |
|
894 * ++styleRunStart; |
|
895 * } |
|
896 * } |
|
897 * |
|
898 * ubidi_close(line); |
|
899 * } |
|
900 * } |
|
901 * } |
|
902 * |
|
903 * ubidi_close(para); |
|
904 * } |
|
905 * </pre> |
|
906 */ |
|
907 BIDI_SAMPLE_CODE |
|
908 /*@{*/ |
|
909 /*@}*/ |
|
910 |
|
911 /*@}*/ |
|
912 |
|
913 #endif |
|