1 | // Copyright (C) 2016 and later: Unicode, Inc. and others. |
2 | // License & terms of use: http://www.unicode.org/copyright.html |
3 | |
4 | // file: rbbi_cache.cpp |
5 | |
6 | #include "unicode/utypes.h" |
7 | |
8 | #if !UCONFIG_NO_BREAK_ITERATION |
9 | |
10 | #include "unicode/ubrk.h" |
11 | #include "unicode/rbbi.h" |
12 | |
13 | #include "rbbi_cache.h" |
14 | |
15 | #include "brkeng.h" |
16 | #include "cmemory.h" |
17 | #include "rbbidata.h" |
18 | #include "rbbirb.h" |
19 | #include "uassert.h" |
20 | #include "uvectr32.h" |
21 | |
22 | U_NAMESPACE_BEGIN |
23 | |
24 | /* |
25 | * DictionaryCache implementation |
26 | */ |
27 | |
28 | RuleBasedBreakIterator::DictionaryCache::DictionaryCache(RuleBasedBreakIterator *bi, UErrorCode &status) : |
29 | fBI(bi), fBreaks(status), fPositionInCache(-1), |
30 | fStart(0), fLimit(0), fFirstRuleStatusIndex(0), fOtherRuleStatusIndex(0) { |
31 | } |
32 | |
33 | RuleBasedBreakIterator::DictionaryCache::~DictionaryCache() { |
34 | } |
35 | |
36 | void RuleBasedBreakIterator::DictionaryCache::reset() { |
37 | fPositionInCache = -1; |
38 | fStart = 0; |
39 | fLimit = 0; |
40 | fFirstRuleStatusIndex = 0; |
41 | fOtherRuleStatusIndex = 0; |
42 | fBreaks.removeAllElements(); |
43 | } |
44 | |
45 | UBool RuleBasedBreakIterator::DictionaryCache::following(int32_t fromPos, int32_t *result, int32_t *statusIndex) { |
46 | if (fromPos >= fLimit || fromPos < fStart) { |
47 | fPositionInCache = -1; |
48 | return FALSE; |
49 | } |
50 | |
51 | // Sequential iteration, move from previous boundary to the following |
52 | |
53 | int32_t r = 0; |
54 | if (fPositionInCache >= 0 && fPositionInCache < fBreaks.size() && fBreaks.elementAti(fPositionInCache) == fromPos) { |
55 | ++fPositionInCache; |
56 | if (fPositionInCache >= fBreaks.size()) { |
57 | fPositionInCache = -1; |
58 | return FALSE; |
59 | } |
60 | r = fBreaks.elementAti(fPositionInCache); |
61 | U_ASSERT(r > fromPos); |
62 | *result = r; |
63 | *statusIndex = fOtherRuleStatusIndex; |
64 | return TRUE; |
65 | } |
66 | |
67 | // Random indexing. Linear search for the boundary following the given position. |
68 | |
69 | for (fPositionInCache = 0; fPositionInCache < fBreaks.size(); ++fPositionInCache) { |
70 | r= fBreaks.elementAti(fPositionInCache); |
71 | if (r > fromPos) { |
72 | *result = r; |
73 | *statusIndex = fOtherRuleStatusIndex; |
74 | return TRUE; |
75 | } |
76 | } |
77 | UPRV_UNREACHABLE; |
78 | } |
79 | |
80 | |
81 | UBool RuleBasedBreakIterator::DictionaryCache::preceding(int32_t fromPos, int32_t *result, int32_t *statusIndex) { |
82 | if (fromPos <= fStart || fromPos > fLimit) { |
83 | fPositionInCache = -1; |
84 | return FALSE; |
85 | } |
86 | |
87 | if (fromPos == fLimit) { |
88 | fPositionInCache = fBreaks.size() - 1; |
89 | if (fPositionInCache >= 0) { |
90 | U_ASSERT(fBreaks.elementAti(fPositionInCache) == fromPos); |
91 | } |
92 | } |
93 | |
94 | int32_t r; |
95 | if (fPositionInCache > 0 && fPositionInCache < fBreaks.size() && fBreaks.elementAti(fPositionInCache) == fromPos) { |
96 | --fPositionInCache; |
97 | r = fBreaks.elementAti(fPositionInCache); |
98 | U_ASSERT(r < fromPos); |
99 | *result = r; |
100 | *statusIndex = ( r== fStart) ? fFirstRuleStatusIndex : fOtherRuleStatusIndex; |
101 | return TRUE; |
102 | } |
103 | |
104 | if (fPositionInCache == 0) { |
105 | fPositionInCache = -1; |
106 | return FALSE; |
107 | } |
108 | |
109 | for (fPositionInCache = fBreaks.size()-1; fPositionInCache >= 0; --fPositionInCache) { |
110 | r = fBreaks.elementAti(fPositionInCache); |
111 | if (r < fromPos) { |
112 | *result = r; |
113 | *statusIndex = ( r == fStart) ? fFirstRuleStatusIndex : fOtherRuleStatusIndex; |
114 | return TRUE; |
115 | } |
116 | } |
117 | UPRV_UNREACHABLE; |
118 | } |
119 | |
120 | void RuleBasedBreakIterator::DictionaryCache::populateDictionary(int32_t startPos, int32_t endPos, |
121 | int32_t firstRuleStatus, int32_t otherRuleStatus) { |
122 | if ((endPos - startPos) <= 1) { |
123 | return; |
124 | } |
125 | |
126 | reset(); |
127 | fFirstRuleStatusIndex = firstRuleStatus; |
128 | fOtherRuleStatusIndex = otherRuleStatus; |
129 | |
130 | int32_t rangeStart = startPos; |
131 | int32_t rangeEnd = endPos; |
132 | |
133 | uint16_t category; |
134 | int32_t current; |
135 | UErrorCode status = U_ZERO_ERROR; |
136 | int32_t foundBreakCount = 0; |
137 | UText *text = &fBI->fText; |
138 | |
139 | // Loop through the text, looking for ranges of dictionary characters. |
140 | // For each span, find the appropriate break engine, and ask it to find |
141 | // any breaks within the span. |
142 | |
143 | utext_setNativeIndex(text, rangeStart); |
144 | UChar32 c = utext_current32(text); |
145 | category = UTRIE2_GET16(fBI->fData->fTrie, c); |
146 | |
147 | while(U_SUCCESS(status)) { |
148 | while((current = (int32_t)UTEXT_GETNATIVEINDEX(text)) < rangeEnd && (category & 0x4000) == 0) { |
149 | utext_next32(text); // TODO: cleaner loop structure. |
150 | c = utext_current32(text); |
151 | category = UTRIE2_GET16(fBI->fData->fTrie, c); |
152 | } |
153 | if (current >= rangeEnd) { |
154 | break; |
155 | } |
156 | |
157 | // We now have a dictionary character. Get the appropriate language object |
158 | // to deal with it. |
159 | const LanguageBreakEngine *lbe = fBI->getLanguageBreakEngine(c); |
160 | |
161 | // Ask the language object if there are any breaks. It will add them to the cache and |
162 | // leave the text pointer on the other side of its range, ready to search for the next one. |
163 | if (lbe != NULL) { |
164 | foundBreakCount += lbe->findBreaks(text, rangeStart, rangeEnd, fBreaks); |
165 | } |
166 | |
167 | // Reload the loop variables for the next go-round |
168 | c = utext_current32(text); |
169 | category = UTRIE2_GET16(fBI->fData->fTrie, c); |
170 | } |
171 | |
172 | // If we found breaks, ensure that the first and last entries are |
173 | // the original starting and ending position. And initialize the |
174 | // cache iteration position to the first entry. |
175 | |
176 | // printf("foundBreakCount = %d\n", foundBreakCount); |
177 | if (foundBreakCount > 0) { |
178 | U_ASSERT(foundBreakCount == fBreaks.size()); |
179 | if (startPos < fBreaks.elementAti(0)) { |
180 | // The dictionary did not place a boundary at the start of the segment of text. |
181 | // Add one now. This should not commonly happen, but it would be easy for interactions |
182 | // of the rules for dictionary segments and the break engine implementations to |
183 | // inadvertently cause it. Cover it here, just in case. |
184 | fBreaks.insertElementAt(startPos, 0, status); |
185 | } |
186 | if (endPos > fBreaks.peeki()) { |
187 | fBreaks.push(endPos, status); |
188 | } |
189 | fPositionInCache = 0; |
190 | // Note: Dictionary matching may extend beyond the original limit. |
191 | fStart = fBreaks.elementAti(0); |
192 | fLimit = fBreaks.peeki(); |
193 | } else { |
194 | // there were no language-based breaks, even though the segment contained |
195 | // dictionary characters. Subsequent attempts to fetch boundaries from the dictionary cache |
196 | // for this range will fail, and the calling code will fall back to the rule based boundaries. |
197 | } |
198 | } |
199 | |
200 | |
201 | /* |
202 | * BreakCache implemetation |
203 | */ |
204 | |
205 | RuleBasedBreakIterator::BreakCache::BreakCache(RuleBasedBreakIterator *bi, UErrorCode &status) : |
206 | fBI(bi), fSideBuffer(status) { |
207 | reset(); |
208 | } |
209 | |
210 | |
211 | RuleBasedBreakIterator::BreakCache::~BreakCache() { |
212 | } |
213 | |
214 | |
215 | void RuleBasedBreakIterator::BreakCache::reset(int32_t pos, int32_t ruleStatus) { |
216 | fStartBufIdx = 0; |
217 | fEndBufIdx = 0; |
218 | fTextIdx = pos; |
219 | fBufIdx = 0; |
220 | fBoundaries[0] = pos; |
221 | fStatuses[0] = (uint16_t)ruleStatus; |
222 | } |
223 | |
224 | |
225 | int32_t RuleBasedBreakIterator::BreakCache::current() { |
226 | fBI->fPosition = fTextIdx; |
227 | fBI->fRuleStatusIndex = fStatuses[fBufIdx]; |
228 | fBI->fDone = FALSE; |
229 | return fTextIdx; |
230 | } |
231 | |
232 | |
233 | void RuleBasedBreakIterator::BreakCache::following(int32_t startPos, UErrorCode &status) { |
234 | if (U_FAILURE(status)) { |
235 | return; |
236 | } |
237 | if (startPos == fTextIdx || seek(startPos) || populateNear(startPos, status)) { |
238 | // startPos is in the cache. Do a next() from that position. |
239 | // TODO: an awkward set of interactions with bi->fDone |
240 | // seek() does not clear it; it can't because of interactions with populateNear(). |
241 | // next() does not clear it in the fast-path case, where everything matters. Maybe it should. |
242 | // So clear it here, for the case where seek() succeeded on an iterator that had previously run off the end. |
243 | fBI->fDone = false; |
244 | next(); |
245 | } |
246 | return; |
247 | } |
248 | |
249 | |
250 | void RuleBasedBreakIterator::BreakCache::preceding(int32_t startPos, UErrorCode &status) { |
251 | if (U_FAILURE(status)) { |
252 | return; |
253 | } |
254 | if (startPos == fTextIdx || seek(startPos) || populateNear(startPos, status)) { |
255 | if (startPos == fTextIdx) { |
256 | previous(status); |
257 | } else { |
258 | // seek() leaves the BreakCache positioned at the preceding boundary |
259 | // if the requested position is between two bounaries. |
260 | // current() pushes the BreakCache position out to the BreakIterator itself. |
261 | U_ASSERT(startPos > fTextIdx); |
262 | current(); |
263 | } |
264 | } |
265 | return; |
266 | } |
267 | |
268 | |
269 | /* |
270 | * Out-of-line code for BreakCache::next(). |
271 | * Cache does not already contain the boundary |
272 | */ |
273 | void RuleBasedBreakIterator::BreakCache::nextOL() { |
274 | fBI->fDone = !populateFollowing(); |
275 | fBI->fPosition = fTextIdx; |
276 | fBI->fRuleStatusIndex = fStatuses[fBufIdx]; |
277 | return; |
278 | } |
279 | |
280 | |
281 | void RuleBasedBreakIterator::BreakCache::previous(UErrorCode &status) { |
282 | if (U_FAILURE(status)) { |
283 | return; |
284 | } |
285 | int32_t initialBufIdx = fBufIdx; |
286 | if (fBufIdx == fStartBufIdx) { |
287 | // At start of cache. Prepend to it. |
288 | populatePreceding(status); |
289 | } else { |
290 | // Cache already holds the next boundary |
291 | fBufIdx = modChunkSize(fBufIdx - 1); |
292 | fTextIdx = fBoundaries[fBufIdx]; |
293 | } |
294 | fBI->fDone = (fBufIdx == initialBufIdx); |
295 | fBI->fPosition = fTextIdx; |
296 | fBI->fRuleStatusIndex = fStatuses[fBufIdx]; |
297 | return; |
298 | } |
299 | |
300 | |
301 | UBool RuleBasedBreakIterator::BreakCache::seek(int32_t pos) { |
302 | if (pos < fBoundaries[fStartBufIdx] || pos > fBoundaries[fEndBufIdx]) { |
303 | return FALSE; |
304 | } |
305 | if (pos == fBoundaries[fStartBufIdx]) { |
306 | // Common case: seek(0), from BreakIterator::first() |
307 | fBufIdx = fStartBufIdx; |
308 | fTextIdx = fBoundaries[fBufIdx]; |
309 | return TRUE; |
310 | } |
311 | if (pos == fBoundaries[fEndBufIdx]) { |
312 | fBufIdx = fEndBufIdx; |
313 | fTextIdx = fBoundaries[fBufIdx]; |
314 | return TRUE; |
315 | } |
316 | |
317 | int32_t min = fStartBufIdx; |
318 | int32_t max = fEndBufIdx; |
319 | while (min != max) { |
320 | int32_t probe = (min + max + (min>max ? CACHE_SIZE : 0)) / 2; |
321 | probe = modChunkSize(probe); |
322 | if (fBoundaries[probe] > pos) { |
323 | max = probe; |
324 | } else { |
325 | min = modChunkSize(probe + 1); |
326 | } |
327 | } |
328 | U_ASSERT(fBoundaries[max] > pos); |
329 | fBufIdx = modChunkSize(max - 1); |
330 | fTextIdx = fBoundaries[fBufIdx]; |
331 | U_ASSERT(fTextIdx <= pos); |
332 | return TRUE; |
333 | } |
334 | |
335 | |
336 | UBool RuleBasedBreakIterator::BreakCache::populateNear(int32_t position, UErrorCode &status) { |
337 | if (U_FAILURE(status)) { |
338 | return FALSE; |
339 | } |
340 | U_ASSERT(position < fBoundaries[fStartBufIdx] || position > fBoundaries[fEndBufIdx]); |
341 | |
342 | // Find a boundary somewhere in the vicinity of the requested position. |
343 | // Depending on the safe rules and the text data, it could be either before, at, or after |
344 | // the requested position. |
345 | |
346 | |
347 | // If the requested position is not near already cached positions, clear the existing cache, |
348 | // find a near-by boundary and begin new cache contents there. |
349 | |
350 | if ((position < fBoundaries[fStartBufIdx] - 15) || position > (fBoundaries[fEndBufIdx] + 15)) { |
351 | int32_t aBoundary = 0; |
352 | int32_t ruleStatusIndex = 0; |
353 | if (position > 20) { |
354 | int32_t backupPos = fBI->handleSafePrevious(position); |
355 | |
356 | if (backupPos > 0) { |
357 | // Advance to the boundary following the backup position. |
358 | // There is a complication: the safe reverse rules identify pairs of code points |
359 | // that are safe. If advancing from the safe point moves forwards by less than |
360 | // two code points, we need to advance one more time to ensure that the boundary |
361 | // is good, including a correct rules status value. |
362 | // |
363 | fBI->fPosition = backupPos; |
364 | aBoundary = fBI->handleNext(); |
365 | if (aBoundary <= backupPos + 4) { |
366 | // +4 is a quick test for possibly having advanced only one codepoint. |
367 | // Four being the length of the longest potential code point, a supplementary in UTF-8 |
368 | utext_setNativeIndex(&fBI->fText, aBoundary); |
369 | if (backupPos == utext_getPreviousNativeIndex(&fBI->fText)) { |
370 | // The initial handleNext() only advanced by a single code point. Go again. |
371 | aBoundary = fBI->handleNext(); // Safe rules identify safe pairs. |
372 | } |
373 | } |
374 | ruleStatusIndex = fBI->fRuleStatusIndex; |
375 | } |
376 | } |
377 | reset(aBoundary, ruleStatusIndex); // Reset cache to hold aBoundary as a single starting point. |
378 | } |
379 | |
380 | // Fill in boundaries between existing cache content and the new requested position. |
381 | |
382 | if (fBoundaries[fEndBufIdx] < position) { |
383 | // The last position in the cache precedes the requested position. |
384 | // Add following position(s) to the cache. |
385 | while (fBoundaries[fEndBufIdx] < position) { |
386 | if (!populateFollowing()) { |
387 | UPRV_UNREACHABLE; |
388 | } |
389 | } |
390 | fBufIdx = fEndBufIdx; // Set iterator position to the end of the buffer. |
391 | fTextIdx = fBoundaries[fBufIdx]; // Required because populateFollowing may add extra boundaries. |
392 | while (fTextIdx > position) { // Move backwards to a position at or preceding the requested pos. |
393 | previous(status); |
394 | } |
395 | return true; |
396 | } |
397 | |
398 | if (fBoundaries[fStartBufIdx] > position) { |
399 | // The first position in the cache is beyond the requested position. |
400 | // back up more until we get a boundary <= the requested position. |
401 | while (fBoundaries[fStartBufIdx] > position) { |
402 | populatePreceding(status); |
403 | } |
404 | fBufIdx = fStartBufIdx; // Set iterator position to the start of the buffer. |
405 | fTextIdx = fBoundaries[fBufIdx]; // Required because populatePreceding may add extra boundaries. |
406 | while (fTextIdx < position) { // Move forwards to a position at or following the requested pos. |
407 | next(); |
408 | } |
409 | if (fTextIdx > position) { |
410 | // If position is not itself a boundary, the next() loop above will overshoot. |
411 | // Back up one, leaving cache position at the boundary preceding the requested position. |
412 | previous(status); |
413 | } |
414 | return true; |
415 | } |
416 | |
417 | U_ASSERT(fTextIdx == position); |
418 | return true; |
419 | } |
420 | |
421 | |
422 | |
423 | UBool RuleBasedBreakIterator::BreakCache::populateFollowing() { |
424 | int32_t fromPosition = fBoundaries[fEndBufIdx]; |
425 | int32_t fromRuleStatusIdx = fStatuses[fEndBufIdx]; |
426 | int32_t pos = 0; |
427 | int32_t ruleStatusIdx = 0; |
428 | |
429 | if (fBI->fDictionaryCache->following(fromPosition, &pos, &ruleStatusIdx)) { |
430 | addFollowing(pos, ruleStatusIdx, UpdateCachePosition); |
431 | return TRUE; |
432 | } |
433 | |
434 | fBI->fPosition = fromPosition; |
435 | pos = fBI->handleNext(); |
436 | if (pos == UBRK_DONE) { |
437 | return FALSE; |
438 | } |
439 | |
440 | ruleStatusIdx = fBI->fRuleStatusIndex; |
441 | if (fBI->fDictionaryCharCount > 0) { |
442 | // The text segment obtained from the rules includes dictionary characters. |
443 | // Subdivide it, with subdivided results going into the dictionary cache. |
444 | fBI->fDictionaryCache->populateDictionary(fromPosition, pos, fromRuleStatusIdx, ruleStatusIdx); |
445 | if (fBI->fDictionaryCache->following(fromPosition, &pos, &ruleStatusIdx)) { |
446 | addFollowing(pos, ruleStatusIdx, UpdateCachePosition); |
447 | return TRUE; |
448 | // TODO: may want to move a sizable chunk of dictionary cache to break cache at this point. |
449 | // But be careful with interactions with populateNear(). |
450 | } |
451 | } |
452 | |
453 | // Rule based segment did not include dictionary characters. |
454 | // Or, it did contain dictionary chars, but the dictionary segmenter didn't handle them, |
455 | // meaning that we didn't take the return, above. |
456 | // Add its end point to the cache. |
457 | addFollowing(pos, ruleStatusIdx, UpdateCachePosition); |
458 | |
459 | // Add several non-dictionary boundaries at this point, to optimize straight forward iteration. |
460 | // (subsequent calls to BreakIterator::next() will take the fast path, getting cached results. |
461 | // |
462 | for (int count=0; count<6; ++count) { |
463 | pos = fBI->handleNext(); |
464 | if (pos == UBRK_DONE || fBI->fDictionaryCharCount > 0) { |
465 | break; |
466 | } |
467 | addFollowing(pos, fBI->fRuleStatusIndex, RetainCachePosition); |
468 | } |
469 | |
470 | return TRUE; |
471 | } |
472 | |
473 | |
474 | UBool RuleBasedBreakIterator::BreakCache::populatePreceding(UErrorCode &status) { |
475 | if (U_FAILURE(status)) { |
476 | return FALSE; |
477 | } |
478 | |
479 | int32_t fromPosition = fBoundaries[fStartBufIdx]; |
480 | if (fromPosition == 0) { |
481 | return FALSE; |
482 | } |
483 | |
484 | int32_t position = 0; |
485 | int32_t positionStatusIdx = 0; |
486 | |
487 | if (fBI->fDictionaryCache->preceding(fromPosition, &position, &positionStatusIdx)) { |
488 | addPreceding(position, positionStatusIdx, UpdateCachePosition); |
489 | return TRUE; |
490 | } |
491 | |
492 | int32_t backupPosition = fromPosition; |
493 | |
494 | // Find a boundary somewhere preceding the first already-cached boundary |
495 | do { |
496 | backupPosition = backupPosition - 30; |
497 | if (backupPosition <= 0) { |
498 | backupPosition = 0; |
499 | } else { |
500 | backupPosition = fBI->handleSafePrevious(backupPosition); |
501 | } |
502 | if (backupPosition == UBRK_DONE || backupPosition == 0) { |
503 | position = 0; |
504 | positionStatusIdx = 0; |
505 | } else { |
506 | // Advance to the boundary following the backup position. |
507 | // There is a complication: the safe reverse rules identify pairs of code points |
508 | // that are safe. If advancing from the safe point moves forwards by less than |
509 | // two code points, we need to advance one more time to ensure that the boundary |
510 | // is good, including a correct rules status value. |
511 | // |
512 | fBI->fPosition = backupPosition; |
513 | position = fBI->handleNext(); |
514 | if (position <= backupPosition + 4) { |
515 | // +4 is a quick test for possibly having advanced only one codepoint. |
516 | // Four being the length of the longest potential code point, a supplementary in UTF-8 |
517 | utext_setNativeIndex(&fBI->fText, position); |
518 | if (backupPosition == utext_getPreviousNativeIndex(&fBI->fText)) { |
519 | // The initial handleNext() only advanced by a single code point. Go again. |
520 | position = fBI->handleNext(); // Safe rules identify safe pairs. |
521 | } |
522 | } |
523 | positionStatusIdx = fBI->fRuleStatusIndex; |
524 | } |
525 | } while (position >= fromPosition); |
526 | |
527 | // Find boundaries between the one we just located and the first already-cached boundary |
528 | // Put them in a side buffer, because we don't yet know where they will fall in the circular cache buffer.. |
529 | |
530 | fSideBuffer.removeAllElements(); |
531 | fSideBuffer.addElement(position, status); |
532 | fSideBuffer.addElement(positionStatusIdx, status); |
533 | |
534 | do { |
535 | int32_t prevPosition = fBI->fPosition = position; |
536 | int32_t prevStatusIdx = positionStatusIdx; |
537 | position = fBI->handleNext(); |
538 | positionStatusIdx = fBI->fRuleStatusIndex; |
539 | if (position == UBRK_DONE) { |
540 | break; |
541 | } |
542 | |
543 | UBool segmentHandledByDictionary = FALSE; |
544 | if (fBI->fDictionaryCharCount != 0) { |
545 | // Segment from the rules includes dictionary characters. |
546 | // Subdivide it, with subdivided results going into the dictionary cache. |
547 | int32_t dictSegEndPosition = position; |
548 | fBI->fDictionaryCache->populateDictionary(prevPosition, dictSegEndPosition, prevStatusIdx, positionStatusIdx); |
549 | while (fBI->fDictionaryCache->following(prevPosition, &position, &positionStatusIdx)) { |
550 | segmentHandledByDictionary = true; |
551 | U_ASSERT(position > prevPosition); |
552 | if (position >= fromPosition) { |
553 | break; |
554 | } |
555 | U_ASSERT(position <= dictSegEndPosition); |
556 | fSideBuffer.addElement(position, status); |
557 | fSideBuffer.addElement(positionStatusIdx, status); |
558 | prevPosition = position; |
559 | } |
560 | U_ASSERT(position==dictSegEndPosition || position>=fromPosition); |
561 | } |
562 | |
563 | if (!segmentHandledByDictionary && position < fromPosition) { |
564 | fSideBuffer.addElement(position, status); |
565 | fSideBuffer.addElement(positionStatusIdx, status); |
566 | } |
567 | } while (position < fromPosition); |
568 | |
569 | // Move boundaries from the side buffer to the main circular buffer. |
570 | UBool success = FALSE; |
571 | if (!fSideBuffer.isEmpty()) { |
572 | positionStatusIdx = fSideBuffer.popi(); |
573 | position = fSideBuffer.popi(); |
574 | addPreceding(position, positionStatusIdx, UpdateCachePosition); |
575 | success = TRUE; |
576 | } |
577 | |
578 | while (!fSideBuffer.isEmpty()) { |
579 | positionStatusIdx = fSideBuffer.popi(); |
580 | position = fSideBuffer.popi(); |
581 | if (!addPreceding(position, positionStatusIdx, RetainCachePosition)) { |
582 | // No space in circular buffer to hold a new preceding result while |
583 | // also retaining the current cache (iteration) position. |
584 | // Bailing out is safe; the cache will refill again if needed. |
585 | break; |
586 | } |
587 | } |
588 | |
589 | return success; |
590 | } |
591 | |
592 | |
593 | void RuleBasedBreakIterator::BreakCache::addFollowing(int32_t position, int32_t ruleStatusIdx, UpdatePositionValues update) { |
594 | U_ASSERT(position > fBoundaries[fEndBufIdx]); |
595 | U_ASSERT(ruleStatusIdx <= UINT16_MAX); |
596 | int32_t nextIdx = modChunkSize(fEndBufIdx + 1); |
597 | if (nextIdx == fStartBufIdx) { |
598 | fStartBufIdx = modChunkSize(fStartBufIdx + 6); // TODO: experiment. Probably revert to 1. |
599 | } |
600 | fBoundaries[nextIdx] = position; |
601 | fStatuses[nextIdx] = static_cast<uint16_t>(ruleStatusIdx); |
602 | fEndBufIdx = nextIdx; |
603 | if (update == UpdateCachePosition) { |
604 | // Set current position to the newly added boundary. |
605 | fBufIdx = nextIdx; |
606 | fTextIdx = position; |
607 | } else { |
608 | // Retaining the original cache position. |
609 | // Check if the added boundary wraps around the buffer, and would over-write the original position. |
610 | // It's the responsibility of callers of this function to not add too many. |
611 | U_ASSERT(nextIdx != fBufIdx); |
612 | } |
613 | } |
614 | |
615 | bool RuleBasedBreakIterator::BreakCache::addPreceding(int32_t position, int32_t ruleStatusIdx, UpdatePositionValues update) { |
616 | U_ASSERT(position < fBoundaries[fStartBufIdx]); |
617 | U_ASSERT(ruleStatusIdx <= UINT16_MAX); |
618 | int32_t nextIdx = modChunkSize(fStartBufIdx - 1); |
619 | if (nextIdx == fEndBufIdx) { |
620 | if (fBufIdx == fEndBufIdx && update == RetainCachePosition) { |
621 | // Failure. The insertion of the new boundary would claim the buffer position that is the |
622 | // current iteration position. And we also want to retain the current iteration position. |
623 | // (The buffer is already completely full of entries that precede the iteration position.) |
624 | return false; |
625 | } |
626 | fEndBufIdx = modChunkSize(fEndBufIdx - 1); |
627 | } |
628 | fBoundaries[nextIdx] = position; |
629 | fStatuses[nextIdx] = static_cast<uint16_t>(ruleStatusIdx); |
630 | fStartBufIdx = nextIdx; |
631 | if (update == UpdateCachePosition) { |
632 | fBufIdx = nextIdx; |
633 | fTextIdx = position; |
634 | } |
635 | return true; |
636 | } |
637 | |
638 | |
639 | void RuleBasedBreakIterator::BreakCache::dumpCache() { |
640 | #ifdef RBBI_DEBUG |
641 | RBBIDebugPrintf("fTextIdx:%d fBufIdx:%d\n" , fTextIdx, fBufIdx); |
642 | for (int32_t i=fStartBufIdx; ; i=modChunkSize(i+1)) { |
643 | RBBIDebugPrintf("%d %d\n" , i, fBoundaries[i]); |
644 | if (i == fEndBufIdx) { |
645 | break; |
646 | } |
647 | } |
648 | #endif |
649 | } |
650 | |
651 | U_NAMESPACE_END |
652 | |
653 | #endif // #if !UCONFIG_NO_BREAK_ITERATION |
654 | |