]>
Commit | Line | Data |
---|---|---|
73c04bcf A |
1 | /** |
2 | ******************************************************************************* | |
46f4442e A |
3 | * Copyright (C) 2006-2008, International Business Machines Corporation * |
4 | * and others. All Rights Reserved. * | |
73c04bcf A |
5 | ******************************************************************************* |
6 | */ | |
7 | ||
8 | #include "unicode/utypes.h" | |
9 | ||
10 | #if !UCONFIG_NO_BREAK_ITERATION | |
11 | ||
12 | #include "triedict.h" | |
13 | #include "unicode/chariter.h" | |
14 | #include "unicode/uchriter.h" | |
15 | #include "unicode/strenum.h" | |
16 | #include "unicode/uenum.h" | |
17 | #include "unicode/udata.h" | |
18 | #include "cmemory.h" | |
19 | #include "udataswp.h" | |
20 | #include "uvector.h" | |
21 | #include "uvectr32.h" | |
22 | #include "uarrsort.h" | |
23 | ||
24 | //#define DEBUG_TRIE_DICT 1 | |
25 | ||
26 | #ifdef DEBUG_TRIE_DICT | |
27 | #include <sys/times.h> | |
28 | #include <limits.h> | |
29 | #include <stdio.h> | |
30 | #endif | |
31 | ||
32 | U_NAMESPACE_BEGIN | |
33 | ||
34 | /******************************************************************* | |
35 | * TrieWordDictionary | |
36 | */ | |
37 | ||
38 | TrieWordDictionary::TrieWordDictionary() { | |
39 | } | |
40 | ||
41 | TrieWordDictionary::~TrieWordDictionary() { | |
42 | } | |
43 | ||
44 | /******************************************************************* | |
45 | * MutableTrieDictionary | |
46 | */ | |
47 | ||
48 | // Node structure for the ternary, uncompressed trie | |
49 | struct TernaryNode : public UMemory { | |
50 | UChar ch; // UTF-16 code unit | |
51 | uint16_t flags; // Flag word | |
52 | TernaryNode *low; // Less-than link | |
53 | TernaryNode *equal; // Equal link | |
54 | TernaryNode *high; // Greater-than link | |
55 | ||
56 | TernaryNode(UChar uc); | |
57 | ~TernaryNode(); | |
58 | }; | |
59 | ||
60 | enum MutableTrieNodeFlags { | |
61 | kEndsWord = 0x0001 // This node marks the end of a valid word | |
62 | }; | |
63 | ||
64 | inline | |
65 | TernaryNode::TernaryNode(UChar uc) { | |
66 | ch = uc; | |
67 | flags = 0; | |
68 | low = NULL; | |
69 | equal = NULL; | |
70 | high = NULL; | |
71 | } | |
72 | ||
73 | // Not inline since it's recursive | |
74 | TernaryNode::~TernaryNode() { | |
75 | delete low; | |
76 | delete equal; | |
77 | delete high; | |
78 | } | |
79 | ||
80 | MutableTrieDictionary::MutableTrieDictionary( UChar median, UErrorCode &status ) { | |
81 | // Start the trie off with something. Having the root node already present | |
82 | // cuts a special case out of the search/insertion functions. | |
83 | // Making it a median character cuts the worse case for searches from | |
84 | // 4x a balanced trie to 2x a balanced trie. It's best to choose something | |
85 | // that starts a word that is midway in the list. | |
86 | fTrie = new TernaryNode(median); | |
87 | if (fTrie == NULL) { | |
88 | status = U_MEMORY_ALLOCATION_ERROR; | |
89 | } | |
90 | fIter = utext_openUChars(NULL, NULL, 0, &status); | |
91 | if (U_SUCCESS(status) && fIter == NULL) { | |
92 | status = U_MEMORY_ALLOCATION_ERROR; | |
93 | } | |
94 | } | |
95 | ||
96 | MutableTrieDictionary::MutableTrieDictionary( UErrorCode &status ) { | |
97 | fTrie = NULL; | |
98 | fIter = utext_openUChars(NULL, NULL, 0, &status); | |
99 | if (U_SUCCESS(status) && fIter == NULL) { | |
100 | status = U_MEMORY_ALLOCATION_ERROR; | |
101 | } | |
102 | } | |
103 | ||
104 | MutableTrieDictionary::~MutableTrieDictionary() { | |
105 | delete fTrie; | |
106 | utext_close(fIter); | |
107 | } | |
108 | ||
109 | int32_t | |
110 | MutableTrieDictionary::search( UText *text, | |
111 | int32_t maxLength, | |
112 | int32_t *lengths, | |
113 | int &count, | |
114 | int limit, | |
115 | TernaryNode *&parent, | |
116 | UBool &pMatched ) const { | |
117 | // TODO: current implementation works in UTF-16 space | |
118 | const TernaryNode *up = NULL; | |
119 | const TernaryNode *p = fTrie; | |
120 | int mycount = 0; | |
121 | pMatched = TRUE; | |
122 | int i; | |
123 | ||
124 | UChar uc = utext_current32(text); | |
125 | for (i = 0; i < maxLength && p != NULL; ++i) { | |
126 | while (p != NULL) { | |
127 | if (uc < p->ch) { | |
128 | up = p; | |
129 | p = p->low; | |
130 | } | |
131 | else if (uc == p->ch) { | |
132 | break; | |
133 | } | |
134 | else { | |
135 | up = p; | |
136 | p = p->high; | |
137 | } | |
138 | } | |
139 | if (p == NULL) { | |
140 | pMatched = FALSE; | |
141 | break; | |
142 | } | |
143 | // Must be equal to get here | |
144 | if (limit > 0 && (p->flags & kEndsWord)) { | |
145 | lengths[mycount++] = i+1; | |
146 | --limit; | |
147 | } | |
148 | up = p; | |
149 | p = p->equal; | |
150 | uc = utext_next32(text); | |
151 | uc = utext_current32(text); | |
152 | } | |
153 | ||
154 | // Note that there is no way to reach here with up == 0 unless | |
155 | // maxLength is 0 coming in. | |
156 | parent = (TernaryNode *)up; | |
157 | count = mycount; | |
158 | return i; | |
159 | } | |
160 | ||
161 | void | |
162 | MutableTrieDictionary::addWord( const UChar *word, | |
163 | int32_t length, | |
164 | UErrorCode &status ) { | |
165 | #if 0 | |
166 | if (length <= 0) { | |
167 | status = U_ILLEGAL_ARGUMENT_ERROR; | |
168 | return; | |
169 | } | |
170 | #endif | |
171 | TernaryNode *parent; | |
172 | UBool pMatched; | |
173 | int count; | |
174 | fIter = utext_openUChars(fIter, word, length, &status); | |
175 | ||
176 | int matched; | |
177 | matched = search(fIter, length, NULL, count, 0, parent, pMatched); | |
178 | ||
179 | while (matched++ < length) { | |
180 | UChar32 uc = utext_next32(fIter); // TODO: supplemetary support? | |
181 | U_ASSERT(uc != U_SENTINEL); | |
182 | TernaryNode *newNode = new TernaryNode(uc); | |
183 | if (newNode == NULL) { | |
184 | status = U_MEMORY_ALLOCATION_ERROR; | |
185 | return; | |
186 | } | |
187 | if (pMatched) { | |
188 | parent->equal = newNode; | |
189 | } | |
190 | else { | |
191 | pMatched = TRUE; | |
192 | if (uc < parent->ch) { | |
193 | parent->low = newNode; | |
194 | } | |
195 | else { | |
196 | parent->high = newNode; | |
197 | } | |
198 | } | |
199 | parent = newNode; | |
200 | } | |
201 | ||
202 | parent->flags |= kEndsWord; | |
203 | } | |
204 | ||
205 | #if 0 | |
206 | void | |
207 | MutableTrieDictionary::addWords( UEnumeration *words, | |
208 | UErrorCode &status ) { | |
209 | int32_t length; | |
210 | const UChar *word; | |
211 | while ((word = uenum_unext(words, &length, &status)) && U_SUCCESS(status)) { | |
212 | addWord(word, length, status); | |
213 | } | |
214 | } | |
215 | #endif | |
216 | ||
217 | int32_t | |
218 | MutableTrieDictionary::matches( UText *text, | |
219 | int32_t maxLength, | |
220 | int32_t *lengths, | |
221 | int &count, | |
222 | int limit ) const { | |
223 | TernaryNode *parent; | |
224 | UBool pMatched; | |
225 | return search(text, maxLength, lengths, count, limit, parent, pMatched); | |
226 | } | |
227 | ||
228 | // Implementation of iteration for MutableTrieDictionary | |
229 | class MutableTrieEnumeration : public StringEnumeration { | |
230 | private: | |
231 | UStack fNodeStack; // Stack of nodes to process | |
232 | UVector32 fBranchStack; // Stack of which branch we are working on | |
233 | TernaryNode *fRoot; // Root node | |
73c04bcf A |
234 | enum StackBranch { |
235 | kLessThan, | |
236 | kEqual, | |
237 | kGreaterThan, | |
238 | kDone | |
239 | }; | |
240 | ||
241 | public: | |
46f4442e A |
242 | static UClassID U_EXPORT2 getStaticClassID(void); |
243 | virtual UClassID getDynamicClassID(void) const; | |
73c04bcf A |
244 | public: |
245 | MutableTrieEnumeration(TernaryNode *root, UErrorCode &status) | |
246 | : fNodeStack(status), fBranchStack(status) { | |
247 | fRoot = root; | |
248 | fNodeStack.push(root, status); | |
249 | fBranchStack.push(kLessThan, status); | |
250 | unistr.remove(); | |
251 | } | |
252 | ||
253 | virtual ~MutableTrieEnumeration() { | |
254 | } | |
255 | ||
256 | virtual StringEnumeration *clone() const { | |
257 | UErrorCode status = U_ZERO_ERROR; | |
258 | return new MutableTrieEnumeration(fRoot, status); | |
259 | } | |
260 | ||
261 | virtual const UnicodeString *snext(UErrorCode &status) { | |
262 | if (fNodeStack.empty() || U_FAILURE(status)) { | |
263 | return NULL; | |
264 | } | |
265 | TernaryNode *node = (TernaryNode *) fNodeStack.peek(); | |
266 | StackBranch where = (StackBranch) fBranchStack.peeki(); | |
267 | while (!fNodeStack.empty() && U_SUCCESS(status)) { | |
268 | UBool emit; | |
269 | UBool equal; | |
270 | ||
271 | switch (where) { | |
272 | case kLessThan: | |
273 | if (node->low != NULL) { | |
274 | fBranchStack.setElementAt(kEqual, fBranchStack.size()-1); | |
275 | node = (TernaryNode *) fNodeStack.push(node->low, status); | |
276 | where = (StackBranch) fBranchStack.push(kLessThan, status); | |
277 | break; | |
278 | } | |
279 | case kEqual: | |
280 | emit = (node->flags & kEndsWord) != 0; | |
281 | equal = (node->equal != NULL); | |
282 | // If this node should be part of the next emitted string, append | |
283 | // the UChar to the string, and make sure we pop it when we come | |
284 | // back to this node. The character should only be in the string | |
285 | // for as long as we're traversing the equal subtree of this node | |
286 | if (equal || emit) { | |
287 | unistr.append(node->ch); | |
288 | fBranchStack.setElementAt(kGreaterThan, fBranchStack.size()-1); | |
289 | } | |
290 | if (equal) { | |
291 | node = (TernaryNode *) fNodeStack.push(node->equal, status); | |
292 | where = (StackBranch) fBranchStack.push(kLessThan, status); | |
293 | } | |
294 | if (emit) { | |
295 | return &unistr; | |
296 | } | |
297 | if (equal) { | |
298 | break; | |
299 | } | |
300 | case kGreaterThan: | |
301 | // If this node's character is in the string, remove it. | |
302 | if (node->equal != NULL || (node->flags & kEndsWord)) { | |
303 | unistr.truncate(unistr.length()-1); | |
304 | } | |
305 | if (node->high != NULL) { | |
306 | fBranchStack.setElementAt(kDone, fBranchStack.size()-1); | |
307 | node = (TernaryNode *) fNodeStack.push(node->high, status); | |
308 | where = (StackBranch) fBranchStack.push(kLessThan, status); | |
309 | break; | |
310 | } | |
311 | case kDone: | |
312 | fNodeStack.pop(); | |
313 | fBranchStack.popi(); | |
314 | node = (TernaryNode *) fNodeStack.peek(); | |
315 | where = (StackBranch) fBranchStack.peeki(); | |
316 | break; | |
317 | default: | |
318 | return NULL; | |
319 | } | |
320 | } | |
321 | return NULL; | |
322 | } | |
323 | ||
324 | // Very expensive, but this should never be used. | |
325 | virtual int32_t count(UErrorCode &status) const { | |
326 | MutableTrieEnumeration counter(fRoot, status); | |
327 | int32_t result = 0; | |
328 | while (counter.snext(status) != NULL && U_SUCCESS(status)) { | |
329 | ++result; | |
330 | } | |
331 | return result; | |
332 | } | |
333 | ||
334 | virtual void reset(UErrorCode &status) { | |
335 | fNodeStack.removeAllElements(); | |
336 | fBranchStack.removeAllElements(); | |
337 | fNodeStack.push(fRoot, status); | |
338 | fBranchStack.push(kLessThan, status); | |
339 | unistr.remove(); | |
340 | } | |
341 | }; | |
342 | ||
46f4442e | 343 | UOBJECT_DEFINE_RTTI_IMPLEMENTATION(MutableTrieEnumeration) |
73c04bcf A |
344 | |
345 | StringEnumeration * | |
346 | MutableTrieDictionary::openWords( UErrorCode &status ) const { | |
347 | if (U_FAILURE(status)) { | |
348 | return NULL; | |
349 | } | |
350 | return new MutableTrieEnumeration(fTrie, status); | |
351 | } | |
352 | ||
353 | /******************************************************************* | |
354 | * CompactTrieDictionary | |
355 | */ | |
356 | ||
357 | struct CompactTrieHeader { | |
358 | uint32_t size; // Size of the data in bytes | |
359 | uint32_t magic; // Magic number (including version) | |
360 | uint16_t nodeCount; // Number of entries in offsets[] | |
361 | uint16_t root; // Node number of the root node | |
362 | uint32_t offsets[1]; // Offsets to nodes from start of data | |
363 | }; | |
364 | ||
365 | // Note that to avoid platform-specific alignment issues, all members of the node | |
366 | // structures should be the same size, or should contain explicit padding to | |
367 | // natural alignment boundaries. | |
368 | ||
369 | // We can't use a bitfield for the flags+count field, because the layout of those | |
370 | // is not portable. 12 bits of count allows for up to 4096 entries in a node. | |
371 | struct CompactTrieNode { | |
372 | uint16_t flagscount; // Count of sub-entries, plus flags | |
373 | }; | |
374 | ||
375 | enum CompactTrieNodeFlags { | |
376 | kVerticalNode = 0x1000, // This is a vertical node | |
377 | kParentEndsWord = 0x2000, // The node whose equal link points to this ends a word | |
378 | kReservedFlag1 = 0x4000, | |
379 | kReservedFlag2 = 0x8000, | |
380 | kCountMask = 0x0FFF, // The count portion of flagscount | |
381 | kFlagMask = 0xF000 // The flags portion of flagscount | |
382 | }; | |
383 | ||
384 | // The two node types are distinguished by the kVerticalNode flag. | |
385 | ||
386 | struct CompactTrieHorizontalEntry { | |
387 | uint16_t ch; // UChar | |
388 | uint16_t equal; // Equal link node index | |
389 | }; | |
390 | ||
391 | // We don't use inheritance here because C++ does not guarantee that the | |
392 | // base class comes first in memory!! | |
393 | ||
394 | struct CompactTrieHorizontalNode { | |
395 | uint16_t flagscount; // Count of sub-entries, plus flags | |
396 | CompactTrieHorizontalEntry entries[1]; | |
397 | }; | |
398 | ||
399 | struct CompactTrieVerticalNode { | |
400 | uint16_t flagscount; // Count of sub-entries, plus flags | |
401 | uint16_t equal; // Equal link node index | |
402 | uint16_t chars[1]; // Code units | |
403 | }; | |
404 | ||
405 | // {'Dic', 1}, version 1 | |
406 | #define COMPACT_TRIE_MAGIC_1 0x44696301 | |
407 | ||
408 | CompactTrieDictionary::CompactTrieDictionary(UDataMemory *dataObj, | |
409 | UErrorCode &status ) | |
410 | : fUData(dataObj) | |
411 | { | |
412 | fData = (const CompactTrieHeader *) udata_getMemory(dataObj); | |
413 | fOwnData = FALSE; | |
414 | if (fData->magic != COMPACT_TRIE_MAGIC_1) { | |
415 | status = U_ILLEGAL_ARGUMENT_ERROR; | |
416 | fData = NULL; | |
417 | } | |
418 | } | |
419 | CompactTrieDictionary::CompactTrieDictionary( const void *data, | |
420 | UErrorCode &status ) | |
421 | : fUData(NULL) | |
422 | { | |
423 | fData = (const CompactTrieHeader *) data; | |
424 | fOwnData = FALSE; | |
425 | if (fData->magic != COMPACT_TRIE_MAGIC_1) { | |
426 | status = U_ILLEGAL_ARGUMENT_ERROR; | |
427 | fData = NULL; | |
428 | } | |
429 | } | |
430 | ||
431 | CompactTrieDictionary::CompactTrieDictionary( const MutableTrieDictionary &dict, | |
432 | UErrorCode &status ) | |
433 | : fUData(NULL) | |
434 | { | |
435 | fData = compactMutableTrieDictionary(dict, status); | |
436 | fOwnData = !U_FAILURE(status); | |
437 | } | |
438 | ||
439 | CompactTrieDictionary::~CompactTrieDictionary() { | |
440 | if (fOwnData) { | |
441 | uprv_free((void *)fData); | |
442 | } | |
443 | if (fUData) { | |
444 | udata_close(fUData); | |
445 | } | |
446 | } | |
447 | ||
448 | uint32_t | |
449 | CompactTrieDictionary::dataSize() const { | |
450 | return fData->size; | |
451 | } | |
452 | ||
453 | const void * | |
454 | CompactTrieDictionary::data() const { | |
455 | return fData; | |
456 | } | |
457 | ||
458 | // This function finds the address of a node for us, given its node ID | |
459 | static inline const CompactTrieNode * | |
460 | getCompactNode(const CompactTrieHeader *header, uint16_t node) { | |
461 | return (const CompactTrieNode *)((const uint8_t *)header + header->offsets[node]); | |
462 | } | |
463 | ||
464 | int32_t | |
465 | CompactTrieDictionary::matches( UText *text, | |
466 | int32_t maxLength, | |
467 | int32_t *lengths, | |
468 | int &count, | |
469 | int limit ) const { | |
470 | // TODO: current implementation works in UTF-16 space | |
471 | const CompactTrieNode *node = getCompactNode(fData, fData->root); | |
472 | int mycount = 0; | |
473 | ||
474 | UChar uc = utext_current32(text); | |
475 | int i = 0; | |
476 | ||
477 | while (node != NULL) { | |
478 | // Check if the node we just exited ends a word | |
479 | if (limit > 0 && (node->flagscount & kParentEndsWord)) { | |
480 | lengths[mycount++] = i; | |
481 | --limit; | |
482 | } | |
483 | // Check that we haven't exceeded the maximum number of input characters. | |
484 | // We have to do that here rather than in the while condition so that | |
485 | // we can check for ending a word, above. | |
486 | if (i >= maxLength) { | |
487 | break; | |
488 | } | |
489 | ||
490 | int nodeCount = (node->flagscount & kCountMask); | |
491 | if (nodeCount == 0) { | |
492 | // Special terminal node; return now | |
493 | break; | |
494 | } | |
495 | if (node->flagscount & kVerticalNode) { | |
496 | // Vertical node; check all the characters in it | |
497 | const CompactTrieVerticalNode *vnode = (const CompactTrieVerticalNode *)node; | |
498 | for (int j = 0; j < nodeCount && i < maxLength; ++j) { | |
499 | if (uc != vnode->chars[j]) { | |
500 | // We hit a non-equal character; return | |
501 | goto exit; | |
502 | } | |
503 | utext_next32(text); | |
504 | uc = utext_current32(text); | |
505 | ++i; | |
506 | } | |
507 | // To get here we must have come through the whole list successfully; | |
508 | // go on to the next node. Note that a word cannot end in the middle | |
509 | // of a vertical node. | |
510 | node = getCompactNode(fData, vnode->equal); | |
511 | } | |
512 | else { | |
513 | // Horizontal node; do binary search | |
514 | const CompactTrieHorizontalNode *hnode = (const CompactTrieHorizontalNode *)node; | |
515 | int low = 0; | |
516 | int high = nodeCount-1; | |
517 | int middle; | |
518 | node = NULL; // If we don't find a match, we'll fall out of the loop | |
519 | while (high >= low) { | |
520 | middle = (high+low)/2; | |
521 | if (uc == hnode->entries[middle].ch) { | |
522 | // We hit a match; get the next node and next character | |
523 | node = getCompactNode(fData, hnode->entries[middle].equal); | |
524 | utext_next32(text); | |
525 | uc = utext_current32(text); | |
526 | ++i; | |
527 | break; | |
528 | } | |
529 | else if (uc < hnode->entries[middle].ch) { | |
530 | high = middle-1; | |
531 | } | |
532 | else { | |
533 | low = middle+1; | |
534 | } | |
535 | } | |
536 | } | |
537 | } | |
538 | exit: | |
539 | count = mycount; | |
540 | return i; | |
541 | } | |
542 | ||
543 | // Implementation of iteration for CompactTrieDictionary | |
544 | class CompactTrieEnumeration : public StringEnumeration { | |
545 | private: | |
546 | UVector32 fNodeStack; // Stack of nodes to process | |
547 | UVector32 fIndexStack; // Stack of where in node we are | |
548 | const CompactTrieHeader *fHeader; // Trie data | |
73c04bcf A |
549 | |
550 | public: | |
46f4442e A |
551 | static UClassID U_EXPORT2 getStaticClassID(void); |
552 | virtual UClassID getDynamicClassID(void) const; | |
73c04bcf A |
553 | public: |
554 | CompactTrieEnumeration(const CompactTrieHeader *header, UErrorCode &status) | |
555 | : fNodeStack(status), fIndexStack(status) { | |
556 | fHeader = header; | |
557 | fNodeStack.push(header->root, status); | |
558 | fIndexStack.push(0, status); | |
559 | unistr.remove(); | |
560 | } | |
561 | ||
562 | virtual ~CompactTrieEnumeration() { | |
563 | } | |
564 | ||
565 | virtual StringEnumeration *clone() const { | |
566 | UErrorCode status = U_ZERO_ERROR; | |
567 | return new CompactTrieEnumeration(fHeader, status); | |
568 | } | |
569 | ||
570 | virtual const UnicodeString * snext(UErrorCode &status); | |
571 | ||
572 | // Very expensive, but this should never be used. | |
573 | virtual int32_t count(UErrorCode &status) const { | |
574 | CompactTrieEnumeration counter(fHeader, status); | |
575 | int32_t result = 0; | |
576 | while (counter.snext(status) != NULL && U_SUCCESS(status)) { | |
577 | ++result; | |
578 | } | |
579 | return result; | |
580 | } | |
581 | ||
582 | virtual void reset(UErrorCode &status) { | |
583 | fNodeStack.removeAllElements(); | |
584 | fIndexStack.removeAllElements(); | |
585 | fNodeStack.push(fHeader->root, status); | |
586 | fIndexStack.push(0, status); | |
587 | unistr.remove(); | |
588 | } | |
589 | }; | |
590 | ||
46f4442e | 591 | UOBJECT_DEFINE_RTTI_IMPLEMENTATION(CompactTrieEnumeration) |
73c04bcf A |
592 | |
593 | const UnicodeString * | |
594 | CompactTrieEnumeration::snext(UErrorCode &status) { | |
595 | if (fNodeStack.empty() || U_FAILURE(status)) { | |
596 | return NULL; | |
597 | } | |
598 | const CompactTrieNode *node = getCompactNode(fHeader, fNodeStack.peeki()); | |
599 | int where = fIndexStack.peeki(); | |
600 | while (!fNodeStack.empty() && U_SUCCESS(status)) { | |
601 | int nodeCount = (node->flagscount & kCountMask); | |
602 | UBool goingDown = FALSE; | |
603 | if (nodeCount == 0) { | |
604 | // Terminal node; go up immediately | |
605 | fNodeStack.popi(); | |
606 | fIndexStack.popi(); | |
607 | node = getCompactNode(fHeader, fNodeStack.peeki()); | |
608 | where = fIndexStack.peeki(); | |
609 | } | |
610 | else if (node->flagscount & kVerticalNode) { | |
611 | // Vertical node | |
612 | const CompactTrieVerticalNode *vnode = (const CompactTrieVerticalNode *)node; | |
613 | if (where == 0) { | |
614 | // Going down | |
615 | unistr.append((const UChar *)vnode->chars, (int32_t) nodeCount); | |
616 | fIndexStack.setElementAt(1, fIndexStack.size()-1); | |
617 | node = getCompactNode(fHeader, fNodeStack.push(vnode->equal, status)); | |
618 | where = fIndexStack.push(0, status); | |
619 | goingDown = TRUE; | |
620 | } | |
621 | else { | |
622 | // Going up | |
623 | unistr.truncate(unistr.length()-nodeCount); | |
624 | fNodeStack.popi(); | |
625 | fIndexStack.popi(); | |
626 | node = getCompactNode(fHeader, fNodeStack.peeki()); | |
627 | where = fIndexStack.peeki(); | |
628 | } | |
629 | } | |
630 | else { | |
631 | // Horizontal node | |
632 | const CompactTrieHorizontalNode *hnode = (const CompactTrieHorizontalNode *)node; | |
633 | if (where > 0) { | |
634 | // Pop previous char | |
635 | unistr.truncate(unistr.length()-1); | |
636 | } | |
637 | if (where < nodeCount) { | |
638 | // Push on next node | |
639 | unistr.append((UChar)hnode->entries[where].ch); | |
640 | fIndexStack.setElementAt(where+1, fIndexStack.size()-1); | |
641 | node = getCompactNode(fHeader, fNodeStack.push(hnode->entries[where].equal, status)); | |
642 | where = fIndexStack.push(0, status); | |
643 | goingDown = TRUE; | |
644 | } | |
645 | else { | |
646 | // Going up | |
647 | fNodeStack.popi(); | |
648 | fIndexStack.popi(); | |
649 | node = getCompactNode(fHeader, fNodeStack.peeki()); | |
650 | where = fIndexStack.peeki(); | |
651 | } | |
652 | } | |
653 | // Check if the parent of the node we've just gone down to ends a | |
654 | // word. If so, return it. | |
655 | if (goingDown && (node->flagscount & kParentEndsWord)) { | |
656 | return &unistr; | |
657 | } | |
658 | } | |
659 | return NULL; | |
660 | } | |
661 | ||
662 | StringEnumeration * | |
663 | CompactTrieDictionary::openWords( UErrorCode &status ) const { | |
664 | if (U_FAILURE(status)) { | |
665 | return NULL; | |
666 | } | |
667 | return new CompactTrieEnumeration(fData, status); | |
668 | } | |
669 | ||
670 | // | |
671 | // Below here is all code related to converting a ternary trie to a compact trie | |
672 | // and back again | |
673 | // | |
674 | ||
675 | // Helper classes to construct the compact trie | |
676 | class BuildCompactTrieNode: public UMemory { | |
677 | public: | |
678 | UBool fParentEndsWord; | |
679 | UBool fVertical; | |
680 | UBool fHasDuplicate; | |
681 | int32_t fNodeID; | |
682 | UnicodeString fChars; | |
683 | ||
684 | public: | |
685 | BuildCompactTrieNode(UBool parentEndsWord, UBool vertical, UStack &nodes, UErrorCode &status) { | |
686 | fParentEndsWord = parentEndsWord; | |
687 | fHasDuplicate = FALSE; | |
688 | fVertical = vertical; | |
689 | fNodeID = nodes.size(); | |
690 | nodes.push(this, status); | |
691 | } | |
692 | ||
693 | virtual ~BuildCompactTrieNode() { | |
694 | } | |
695 | ||
696 | virtual uint32_t size() { | |
697 | return sizeof(uint16_t); | |
698 | } | |
699 | ||
700 | virtual void write(uint8_t *bytes, uint32_t &offset, const UVector32 &/*translate*/) { | |
701 | // Write flag/count | |
702 | *((uint16_t *)(bytes+offset)) = (fChars.length() & kCountMask) | |
703 | | (fVertical ? kVerticalNode : 0) | (fParentEndsWord ? kParentEndsWord : 0 ); | |
704 | offset += sizeof(uint16_t); | |
705 | } | |
706 | }; | |
707 | ||
708 | class BuildCompactTrieHorizontalNode: public BuildCompactTrieNode { | |
709 | public: | |
710 | UStack fLinks; | |
711 | ||
712 | public: | |
713 | BuildCompactTrieHorizontalNode(UBool parentEndsWord, UStack &nodes, UErrorCode &status) | |
714 | : BuildCompactTrieNode(parentEndsWord, FALSE, nodes, status), fLinks(status) { | |
715 | } | |
716 | ||
717 | virtual ~BuildCompactTrieHorizontalNode() { | |
718 | } | |
719 | ||
720 | virtual uint32_t size() { | |
721 | return offsetof(CompactTrieHorizontalNode,entries) + | |
722 | (fChars.length()*sizeof(CompactTrieHorizontalEntry)); | |
723 | } | |
724 | ||
725 | virtual void write(uint8_t *bytes, uint32_t &offset, const UVector32 &translate) { | |
726 | BuildCompactTrieNode::write(bytes, offset, translate); | |
727 | int32_t count = fChars.length(); | |
728 | for (int32_t i = 0; i < count; ++i) { | |
729 | CompactTrieHorizontalEntry *entry = (CompactTrieHorizontalEntry *)(bytes+offset); | |
730 | entry->ch = fChars[i]; | |
731 | entry->equal = translate.elementAti(((BuildCompactTrieNode *)fLinks[i])->fNodeID); | |
732 | #ifdef DEBUG_TRIE_DICT | |
733 | if (entry->equal == 0) { | |
734 | fprintf(stderr, "ERROR: horizontal link %d, logical node %d maps to physical node zero\n", | |
735 | i, ((BuildCompactTrieNode *)fLinks[i])->fNodeID); | |
736 | } | |
737 | #endif | |
738 | offset += sizeof(CompactTrieHorizontalEntry); | |
739 | } | |
740 | } | |
741 | ||
742 | void addNode(UChar ch, BuildCompactTrieNode *link, UErrorCode &status) { | |
743 | fChars.append(ch); | |
744 | fLinks.push(link, status); | |
745 | } | |
746 | }; | |
747 | ||
748 | class BuildCompactTrieVerticalNode: public BuildCompactTrieNode { | |
749 | public: | |
750 | BuildCompactTrieNode *fEqual; | |
751 | ||
752 | public: | |
753 | BuildCompactTrieVerticalNode(UBool parentEndsWord, UStack &nodes, UErrorCode &status) | |
754 | : BuildCompactTrieNode(parentEndsWord, TRUE, nodes, status) { | |
755 | fEqual = NULL; | |
756 | } | |
757 | ||
758 | virtual ~BuildCompactTrieVerticalNode() { | |
759 | } | |
760 | ||
761 | virtual uint32_t size() { | |
762 | return offsetof(CompactTrieVerticalNode,chars) + (fChars.length()*sizeof(uint16_t)); | |
763 | } | |
764 | ||
765 | virtual void write(uint8_t *bytes, uint32_t &offset, const UVector32 &translate) { | |
766 | CompactTrieVerticalNode *node = (CompactTrieVerticalNode *)(bytes+offset); | |
767 | BuildCompactTrieNode::write(bytes, offset, translate); | |
768 | node->equal = translate.elementAti(fEqual->fNodeID); | |
769 | offset += sizeof(node->equal); | |
770 | #ifdef DEBUG_TRIE_DICT | |
771 | if (node->equal == 0) { | |
772 | fprintf(stderr, "ERROR: vertical link, logical node %d maps to physical node zero\n", | |
773 | fEqual->fNodeID); | |
774 | } | |
775 | #endif | |
776 | fChars.extract(0, fChars.length(), (UChar *)node->chars); | |
777 | offset += sizeof(uint16_t)*fChars.length(); | |
778 | } | |
779 | ||
780 | void addChar(UChar ch) { | |
781 | fChars.append(ch); | |
782 | } | |
783 | ||
784 | void setLink(BuildCompactTrieNode *node) { | |
785 | fEqual = node; | |
786 | } | |
787 | }; | |
788 | ||
789 | // Forward declaration | |
790 | static void walkHorizontal(const TernaryNode *node, | |
791 | BuildCompactTrieHorizontalNode *building, | |
792 | UStack &nodes, | |
793 | UErrorCode &status); | |
794 | ||
795 | // Convert one node. Uses recursion. | |
796 | ||
797 | static BuildCompactTrieNode * | |
798 | compactOneNode(const TernaryNode *node, UBool parentEndsWord, UStack &nodes, UErrorCode &status) { | |
799 | if (U_FAILURE(status)) { | |
800 | return NULL; | |
801 | } | |
802 | BuildCompactTrieNode *result = NULL; | |
803 | UBool horizontal = (node->low != NULL || node->high != NULL); | |
804 | if (horizontal) { | |
805 | BuildCompactTrieHorizontalNode *hResult = | |
806 | new BuildCompactTrieHorizontalNode(parentEndsWord, nodes, status); | |
807 | if (hResult == NULL) { | |
808 | status = U_MEMORY_ALLOCATION_ERROR; | |
46f4442e | 809 | return NULL; |
73c04bcf A |
810 | } |
811 | if (U_SUCCESS(status)) { | |
812 | walkHorizontal(node, hResult, nodes, status); | |
813 | result = hResult; | |
814 | } | |
815 | } | |
816 | else { | |
817 | BuildCompactTrieVerticalNode *vResult = | |
818 | new BuildCompactTrieVerticalNode(parentEndsWord, nodes, status); | |
819 | if (vResult == NULL) { | |
820 | status = U_MEMORY_ALLOCATION_ERROR; | |
821 | } | |
46f4442e | 822 | else if (U_SUCCESS(status)) { |
73c04bcf A |
823 | UBool endsWord = FALSE; |
824 | // Take up nodes until we end a word, or hit a node with < or > links | |
825 | do { | |
826 | vResult->addChar(node->ch); | |
827 | endsWord = (node->flags & kEndsWord) != 0; | |
828 | node = node->equal; | |
829 | } | |
830 | while(node != NULL && !endsWord && node->low == NULL && node->high == NULL); | |
831 | if (node == NULL) { | |
832 | if (!endsWord) { | |
833 | status = U_ILLEGAL_ARGUMENT_ERROR; // Corrupt input trie | |
834 | } | |
835 | else { | |
836 | vResult->setLink((BuildCompactTrieNode *)nodes[1]); | |
837 | } | |
838 | } | |
839 | else { | |
840 | vResult->setLink(compactOneNode(node, endsWord, nodes, status)); | |
841 | } | |
842 | result = vResult; | |
843 | } | |
844 | } | |
845 | return result; | |
846 | } | |
847 | ||
848 | // Walk the set of peers at the same level, to build a horizontal node. | |
849 | // Uses recursion. | |
850 | ||
851 | static void walkHorizontal(const TernaryNode *node, | |
852 | BuildCompactTrieHorizontalNode *building, | |
853 | UStack &nodes, | |
854 | UErrorCode &status) { | |
855 | while (U_SUCCESS(status) && node != NULL) { | |
856 | if (node->low != NULL) { | |
857 | walkHorizontal(node->low, building, nodes, status); | |
858 | } | |
859 | BuildCompactTrieNode *link = NULL; | |
860 | if (node->equal != NULL) { | |
861 | link = compactOneNode(node->equal, (node->flags & kEndsWord) != 0, nodes, status); | |
862 | } | |
863 | else if (node->flags & kEndsWord) { | |
864 | link = (BuildCompactTrieNode *)nodes[1]; | |
865 | } | |
866 | if (U_SUCCESS(status) && link != NULL) { | |
867 | building->addNode(node->ch, link, status); | |
868 | } | |
869 | // Tail recurse manually instead of leaving it to the compiler. | |
870 | //if (node->high != NULL) { | |
871 | // walkHorizontal(node->high, building, nodes, status); | |
872 | //} | |
873 | node = node->high; | |
874 | } | |
875 | } | |
876 | ||
877 | U_NAMESPACE_END | |
46f4442e | 878 | U_NAMESPACE_USE |
73c04bcf A |
879 | U_CDECL_BEGIN |
880 | static int32_t U_CALLCONV | |
881 | _sortBuildNodes(const void * /*context*/, const void *voidl, const void *voidr) { | |
882 | BuildCompactTrieNode *left = *(BuildCompactTrieNode **)voidl; | |
883 | BuildCompactTrieNode *right = *(BuildCompactTrieNode **)voidr; | |
884 | // Check for comparing a node to itself, to avoid spurious duplicates | |
885 | if (left == right) { | |
886 | return 0; | |
887 | } | |
888 | // Most significant is type of node. Can never coalesce. | |
889 | if (left->fVertical != right->fVertical) { | |
890 | return left->fVertical - right->fVertical; | |
891 | } | |
892 | // Next, the "parent ends word" flag. If that differs, we cannot coalesce. | |
893 | if (left->fParentEndsWord != right->fParentEndsWord) { | |
894 | return left->fParentEndsWord - right->fParentEndsWord; | |
895 | } | |
896 | // Next, the string. If that differs, we can never coalesce. | |
897 | int32_t result = left->fChars.compare(right->fChars); | |
898 | if (result != 0) { | |
899 | return result; | |
900 | } | |
901 | // We know they're both the same node type, so branch for the two cases. | |
902 | if (left->fVertical) { | |
903 | result = ((BuildCompactTrieVerticalNode *)left)->fEqual->fNodeID | |
904 | - ((BuildCompactTrieVerticalNode *)right)->fEqual->fNodeID; | |
905 | } | |
906 | else { | |
907 | // We need to compare the links vectors. They should be the | |
908 | // same size because the strings were equal. | |
909 | // We compare the node IDs instead of the pointers, to handle | |
910 | // coalesced nodes. | |
911 | BuildCompactTrieHorizontalNode *hleft, *hright; | |
912 | hleft = (BuildCompactTrieHorizontalNode *)left; | |
913 | hright = (BuildCompactTrieHorizontalNode *)right; | |
914 | int32_t count = hleft->fLinks.size(); | |
915 | for (int32_t i = 0; i < count && result == 0; ++i) { | |
916 | result = ((BuildCompactTrieNode *)(hleft->fLinks[i]))->fNodeID - | |
917 | ((BuildCompactTrieNode *)(hright->fLinks[i]))->fNodeID; | |
918 | } | |
919 | } | |
920 | // If they are equal to each other, mark them (speeds coalescing) | |
921 | if (result == 0) { | |
922 | left->fHasDuplicate = TRUE; | |
923 | right->fHasDuplicate = TRUE; | |
924 | } | |
925 | return result; | |
926 | } | |
927 | U_CDECL_END | |
928 | U_NAMESPACE_BEGIN | |
929 | ||
930 | static void coalesceDuplicates(UStack &nodes, UErrorCode &status) { | |
931 | // We sort the array of nodes to place duplicates next to each other | |
932 | if (U_FAILURE(status)) { | |
933 | return; | |
934 | } | |
935 | int32_t size = nodes.size(); | |
936 | void **array = (void **)uprv_malloc(sizeof(void *)*size); | |
937 | if (array == NULL) { | |
938 | status = U_MEMORY_ALLOCATION_ERROR; | |
939 | return; | |
940 | } | |
941 | (void) nodes.toArray(array); | |
942 | ||
943 | // Now repeatedly identify duplicates until there are no more | |
944 | int32_t dupes = 0; | |
945 | long passCount = 0; | |
946 | #ifdef DEBUG_TRIE_DICT | |
947 | long totalDupes = 0; | |
948 | #endif | |
949 | do { | |
950 | BuildCompactTrieNode *node; | |
951 | BuildCompactTrieNode *first = NULL; | |
952 | BuildCompactTrieNode **p; | |
953 | BuildCompactTrieNode **pFirst = NULL; | |
954 | int32_t counter = size - 2; | |
955 | // Sort the array, skipping nodes 0 and 1. Use quicksort for the first | |
956 | // pass for speed. For the second and subsequent passes, we use stable | |
957 | // (insertion) sort for two reasons: | |
958 | // 1. The array is already mostly ordered, so we get better performance. | |
959 | // 2. The way we find one and only one instance of a set of duplicates is to | |
960 | // check that the node ID equals the array index. If we used an unstable | |
961 | // sort for the second or later passes, it's possible that none of the | |
962 | // duplicates would wind up with a node ID equal to its array index. | |
963 | // The sort stability guarantees that, because as we coalesce more and | |
964 | // more groups, the first element of the resultant group will be one of | |
965 | // the first elements of the groups being coalesced. | |
966 | // To use quicksort for the second and subsequent passes, we would have to | |
967 | // find the minimum of the node numbers in a group, and set all the nodes | |
968 | // in the group to that node number. | |
969 | uprv_sortArray(array+2, counter, sizeof(void *), _sortBuildNodes, NULL, (passCount > 0), &status); | |
970 | dupes = 0; | |
971 | for (p = (BuildCompactTrieNode **)array + 2; counter > 0; --counter, ++p) { | |
972 | node = *p; | |
973 | if (node->fHasDuplicate) { | |
974 | if (first == NULL) { | |
975 | first = node; | |
976 | pFirst = p; | |
977 | } | |
978 | else if (_sortBuildNodes(NULL, pFirst, p) != 0) { | |
979 | // Starting a new run of dupes | |
980 | first = node; | |
981 | pFirst = p; | |
982 | } | |
983 | else if (node->fNodeID != first->fNodeID) { | |
984 | // Slave one to the other, note duplicate | |
985 | node->fNodeID = first->fNodeID; | |
986 | dupes += 1; | |
987 | } | |
988 | } | |
989 | else { | |
990 | // This node has no dupes | |
991 | first = NULL; | |
992 | pFirst = NULL; | |
993 | } | |
994 | } | |
995 | passCount += 1; | |
996 | #ifdef DEBUG_TRIE_DICT | |
997 | totalDupes += dupes; | |
998 | fprintf(stderr, "Trie node dupe removal, pass %d: %d nodes tagged\n", passCount, dupes); | |
999 | #endif | |
1000 | } | |
1001 | while (dupes > 0); | |
1002 | #ifdef DEBUG_TRIE_DICT | |
1003 | fprintf(stderr, "Trie node dupe removal complete: %d tagged in %d passes\n", totalDupes, passCount); | |
1004 | #endif | |
1005 | ||
1006 | // We no longer need the temporary array, as the nodes have all been marked appropriately. | |
1007 | uprv_free(array); | |
1008 | } | |
1009 | ||
1010 | U_NAMESPACE_END | |
1011 | U_CDECL_BEGIN | |
1012 | static void U_CALLCONV _deleteBuildNode(void *obj) { | |
1013 | delete (BuildCompactTrieNode *) obj; | |
1014 | } | |
1015 | U_CDECL_END | |
1016 | U_NAMESPACE_BEGIN | |
1017 | ||
1018 | CompactTrieHeader * | |
1019 | CompactTrieDictionary::compactMutableTrieDictionary( const MutableTrieDictionary &dict, | |
1020 | UErrorCode &status ) { | |
1021 | if (U_FAILURE(status)) { | |
1022 | return NULL; | |
1023 | } | |
1024 | #ifdef DEBUG_TRIE_DICT | |
1025 | struct tms timing; | |
1026 | struct tms previous; | |
1027 | (void) ::times(&previous); | |
1028 | #endif | |
1029 | UStack nodes(_deleteBuildNode, NULL, status); // Index of nodes | |
1030 | ||
1031 | // Add node 0, used as the NULL pointer/sentinel. | |
1032 | nodes.addElement((int32_t)0, status); | |
1033 | ||
1034 | // Start by creating the special empty node we use to indicate that the parent | |
1035 | // terminates a word. This must be node 1, because the builder assumes | |
1036 | // that. | |
1037 | if (U_FAILURE(status)) { | |
1038 | return NULL; | |
1039 | } | |
1040 | BuildCompactTrieNode *terminal = new BuildCompactTrieNode(TRUE, FALSE, nodes, status); | |
1041 | if (terminal == NULL) { | |
1042 | status = U_MEMORY_ALLOCATION_ERROR; | |
1043 | } | |
1044 | ||
1045 | // This call does all the work of building the new trie structure. The root | |
1046 | // will be node 2. | |
1047 | BuildCompactTrieNode *root = compactOneNode(dict.fTrie, FALSE, nodes, status); | |
1048 | #ifdef DEBUG_TRIE_DICT | |
1049 | (void) ::times(&timing); | |
1050 | fprintf(stderr, "Compact trie built, %d nodes, time user %f system %f\n", | |
1051 | nodes.size(), (double)(timing.tms_utime-previous.tms_utime)/CLK_TCK, | |
1052 | (double)(timing.tms_stime-previous.tms_stime)/CLK_TCK); | |
1053 | previous = timing; | |
1054 | #endif | |
1055 | ||
1056 | // Now coalesce all duplicate nodes. | |
1057 | coalesceDuplicates(nodes, status); | |
1058 | #ifdef DEBUG_TRIE_DICT | |
1059 | (void) ::times(&timing); | |
1060 | fprintf(stderr, "Duplicates coalesced, time user %f system %f\n", | |
1061 | (double)(timing.tms_utime-previous.tms_utime)/CLK_TCK, | |
1062 | (double)(timing.tms_stime-previous.tms_stime)/CLK_TCK); | |
1063 | previous = timing; | |
1064 | #endif | |
1065 | ||
1066 | // Next, build the output trie. | |
1067 | // First we compute all the sizes and build the node ID translation table. | |
1068 | uint32_t totalSize = offsetof(CompactTrieHeader,offsets); | |
1069 | int32_t count = nodes.size(); | |
1070 | int32_t nodeCount = 1; // The sentinel node we already have | |
1071 | BuildCompactTrieNode *node; | |
46f4442e | 1072 | int32_t i; |
73c04bcf A |
1073 | UVector32 translate(count, status); // Should be no growth needed after this |
1074 | translate.push(0, status); // The sentinel node | |
1075 | ||
1076 | if (U_FAILURE(status)) { | |
1077 | return NULL; | |
1078 | } | |
1079 | ||
46f4442e | 1080 | for (i = 1; i < count; ++i) { |
73c04bcf A |
1081 | node = (BuildCompactTrieNode *)nodes[i]; |
1082 | if (node->fNodeID == i) { | |
1083 | // Only one node out of each duplicate set is used | |
1084 | if (i >= translate.size()) { | |
1085 | // Logically extend the mapping table | |
1086 | translate.setSize(i+1); | |
1087 | } | |
1088 | translate.setElementAt(nodeCount++, i); | |
1089 | totalSize += node->size(); | |
1090 | } | |
1091 | } | |
1092 | ||
1093 | // Check for overflowing 16 bits worth of nodes. | |
1094 | if (nodeCount > 0x10000) { | |
1095 | status = U_ILLEGAL_ARGUMENT_ERROR; | |
1096 | return NULL; | |
1097 | } | |
1098 | ||
1099 | // Add enough room for the offsets. | |
1100 | totalSize += nodeCount*sizeof(uint32_t); | |
1101 | #ifdef DEBUG_TRIE_DICT | |
1102 | (void) ::times(&timing); | |
1103 | fprintf(stderr, "Sizes/mapping done, time user %f system %f\n", | |
1104 | (double)(timing.tms_utime-previous.tms_utime)/CLK_TCK, | |
1105 | (double)(timing.tms_stime-previous.tms_stime)/CLK_TCK); | |
1106 | previous = timing; | |
1107 | fprintf(stderr, "%d nodes, %d unique, %d bytes\n", nodes.size(), nodeCount, totalSize); | |
1108 | #endif | |
1109 | uint8_t *bytes = (uint8_t *)uprv_malloc(totalSize); | |
1110 | if (bytes == NULL) { | |
1111 | status = U_MEMORY_ALLOCATION_ERROR; | |
1112 | return NULL; | |
1113 | } | |
1114 | ||
1115 | CompactTrieHeader *header = (CompactTrieHeader *)bytes; | |
1116 | header->size = totalSize; | |
1117 | header->nodeCount = nodeCount; | |
1118 | header->offsets[0] = 0; // Sentinel | |
1119 | header->root = translate.elementAti(root->fNodeID); | |
1120 | #ifdef DEBUG_TRIE_DICT | |
1121 | if (header->root == 0) { | |
1122 | fprintf(stderr, "ERROR: root node %d translate to physical zero\n", root->fNodeID); | |
1123 | } | |
1124 | #endif | |
1125 | uint32_t offset = offsetof(CompactTrieHeader,offsets)+(nodeCount*sizeof(uint32_t)); | |
1126 | nodeCount = 1; | |
1127 | // Now write the data | |
46f4442e | 1128 | for (i = 1; i < count; ++i) { |
73c04bcf A |
1129 | node = (BuildCompactTrieNode *)nodes[i]; |
1130 | if (node->fNodeID == i) { | |
1131 | header->offsets[nodeCount++] = offset; | |
1132 | node->write(bytes, offset, translate); | |
1133 | } | |
1134 | } | |
1135 | #ifdef DEBUG_TRIE_DICT | |
1136 | (void) ::times(&timing); | |
1137 | fprintf(stderr, "Trie built, time user %f system %f\n", | |
1138 | (double)(timing.tms_utime-previous.tms_utime)/CLK_TCK, | |
1139 | (double)(timing.tms_stime-previous.tms_stime)/CLK_TCK); | |
1140 | previous = timing; | |
1141 | fprintf(stderr, "Final offset is %d\n", offset); | |
1142 | ||
1143 | // Collect statistics on node types and sizes | |
1144 | int hCount = 0; | |
1145 | int vCount = 0; | |
1146 | size_t hSize = 0; | |
1147 | size_t vSize = 0; | |
1148 | size_t hItemCount = 0; | |
1149 | size_t vItemCount = 0; | |
1150 | uint32_t previousOff = offset; | |
46f4442e A |
1151 | for (uint16_t nodeIdx = nodeCount-1; nodeIdx >= 2; --nodeIdx) { |
1152 | const CompactTrieNode *node = getCompactNode(header, nodeIdx); | |
73c04bcf A |
1153 | if (node->flagscount & kVerticalNode) { |
1154 | vCount += 1; | |
1155 | vItemCount += (node->flagscount & kCountMask); | |
46f4442e | 1156 | vSize += previousOff-header->offsets[nodeIdx]; |
73c04bcf A |
1157 | } |
1158 | else { | |
1159 | hCount += 1; | |
1160 | hItemCount += (node->flagscount & kCountMask); | |
46f4442e | 1161 | hSize += previousOff-header->offsets[nodeIdx]; |
73c04bcf | 1162 | } |
46f4442e | 1163 | previousOff = header->offsets[nodeIdx]; |
73c04bcf A |
1164 | } |
1165 | fprintf(stderr, "Horizontal nodes: %d total, average %f bytes with %f items\n", hCount, | |
1166 | (double)hSize/hCount, (double)hItemCount/hCount); | |
1167 | fprintf(stderr, "Vertical nodes: %d total, average %f bytes with %f items\n", vCount, | |
1168 | (double)vSize/vCount, (double)vItemCount/vCount); | |
1169 | #endif | |
1170 | ||
1171 | if (U_FAILURE(status)) { | |
1172 | uprv_free(bytes); | |
1173 | header = NULL; | |
1174 | } | |
1175 | else { | |
1176 | header->magic = COMPACT_TRIE_MAGIC_1; | |
1177 | } | |
1178 | return header; | |
1179 | } | |
1180 | ||
1181 | // Forward declaration | |
1182 | static TernaryNode * | |
1183 | unpackOneNode( const CompactTrieHeader *header, const CompactTrieNode *node, UErrorCode &status ); | |
1184 | ||
1185 | ||
1186 | // Convert a horizontal node (or subarray thereof) into a ternary subtrie | |
1187 | static TernaryNode * | |
1188 | unpackHorizontalArray( const CompactTrieHeader *header, const CompactTrieHorizontalEntry *array, | |
1189 | int low, int high, UErrorCode &status ) { | |
1190 | if (U_FAILURE(status) || low > high) { | |
1191 | return NULL; | |
1192 | } | |
1193 | int middle = (low+high)/2; | |
1194 | TernaryNode *result = new TernaryNode(array[middle].ch); | |
1195 | if (result == NULL) { | |
1196 | status = U_MEMORY_ALLOCATION_ERROR; | |
1197 | return NULL; | |
1198 | } | |
1199 | const CompactTrieNode *equal = getCompactNode(header, array[middle].equal); | |
1200 | if (equal->flagscount & kParentEndsWord) { | |
1201 | result->flags |= kEndsWord; | |
1202 | } | |
1203 | result->low = unpackHorizontalArray(header, array, low, middle-1, status); | |
1204 | result->high = unpackHorizontalArray(header, array, middle+1, high, status); | |
1205 | result->equal = unpackOneNode(header, equal, status); | |
1206 | return result; | |
1207 | } | |
1208 | ||
1209 | // Convert one compact trie node into a ternary subtrie | |
1210 | static TernaryNode * | |
1211 | unpackOneNode( const CompactTrieHeader *header, const CompactTrieNode *node, UErrorCode &status ) { | |
1212 | int nodeCount = (node->flagscount & kCountMask); | |
1213 | if (nodeCount == 0 || U_FAILURE(status)) { | |
1214 | // Failure, or terminal node | |
1215 | return NULL; | |
1216 | } | |
1217 | if (node->flagscount & kVerticalNode) { | |
1218 | const CompactTrieVerticalNode *vnode = (const CompactTrieVerticalNode *)node; | |
1219 | TernaryNode *head = NULL; | |
1220 | TernaryNode *previous = NULL; | |
1221 | TernaryNode *latest = NULL; | |
1222 | for (int i = 0; i < nodeCount; ++i) { | |
1223 | latest = new TernaryNode(vnode->chars[i]); | |
1224 | if (latest == NULL) { | |
1225 | status = U_MEMORY_ALLOCATION_ERROR; | |
1226 | break; | |
1227 | } | |
1228 | if (head == NULL) { | |
1229 | head = latest; | |
1230 | } | |
1231 | if (previous != NULL) { | |
1232 | previous->equal = latest; | |
1233 | } | |
1234 | previous = latest; | |
1235 | } | |
1236 | if (latest != NULL) { | |
1237 | const CompactTrieNode *equal = getCompactNode(header, vnode->equal); | |
1238 | if (equal->flagscount & kParentEndsWord) { | |
1239 | latest->flags |= kEndsWord; | |
1240 | } | |
1241 | latest->equal = unpackOneNode(header, equal, status); | |
1242 | } | |
1243 | return head; | |
1244 | } | |
1245 | else { | |
1246 | // Horizontal node | |
1247 | const CompactTrieHorizontalNode *hnode = (const CompactTrieHorizontalNode *)node; | |
1248 | return unpackHorizontalArray(header, &hnode->entries[0], 0, nodeCount-1, status); | |
1249 | } | |
1250 | } | |
1251 | ||
1252 | MutableTrieDictionary * | |
1253 | CompactTrieDictionary::cloneMutable( UErrorCode &status ) const { | |
1254 | MutableTrieDictionary *result = new MutableTrieDictionary( status ); | |
1255 | if (result == NULL) { | |
1256 | status = U_MEMORY_ALLOCATION_ERROR; | |
1257 | return NULL; | |
1258 | } | |
1259 | TernaryNode *root = unpackOneNode(fData, getCompactNode(fData, fData->root), status); | |
1260 | if (U_FAILURE(status)) { | |
1261 | delete root; // Clean up | |
1262 | delete result; | |
1263 | return NULL; | |
1264 | } | |
1265 | result->fTrie = root; | |
1266 | return result; | |
1267 | } | |
1268 | ||
1269 | U_NAMESPACE_END | |
1270 | ||
1271 | U_CAPI int32_t U_EXPORT2 | |
1272 | triedict_swap(const UDataSwapper *ds, const void *inData, int32_t length, void *outData, | |
1273 | UErrorCode *status) { | |
1274 | ||
1275 | if (status == NULL || U_FAILURE(*status)) { | |
1276 | return 0; | |
1277 | } | |
1278 | if(ds==NULL || inData==NULL || length<-1 || (length>0 && outData==NULL)) { | |
1279 | *status=U_ILLEGAL_ARGUMENT_ERROR; | |
1280 | return 0; | |
1281 | } | |
1282 | ||
1283 | // | |
1284 | // Check that the data header is for for dictionary data. | |
1285 | // (Header contents are defined in genxxx.cpp) | |
1286 | // | |
1287 | const UDataInfo *pInfo = (const UDataInfo *)((const uint8_t *)inData+4); | |
1288 | if(!( pInfo->dataFormat[0]==0x54 && /* dataFormat="TrDc" */ | |
1289 | pInfo->dataFormat[1]==0x72 && | |
1290 | pInfo->dataFormat[2]==0x44 && | |
1291 | pInfo->dataFormat[3]==0x63 && | |
1292 | pInfo->formatVersion[0]==1 )) { | |
1293 | udata_printError(ds, "triedict_swap(): data format %02x.%02x.%02x.%02x (format version %02x) is not recognized\n", | |
1294 | pInfo->dataFormat[0], pInfo->dataFormat[1], | |
1295 | pInfo->dataFormat[2], pInfo->dataFormat[3], | |
1296 | pInfo->formatVersion[0]); | |
1297 | *status=U_UNSUPPORTED_ERROR; | |
1298 | return 0; | |
1299 | } | |
1300 | ||
1301 | // | |
1302 | // Swap the data header. (This is the generic ICU Data Header, not the | |
1303 | // CompactTrieHeader). This swap also conveniently gets us | |
1304 | // the size of the ICU d.h., which lets us locate the start | |
1305 | // of the RBBI specific data. | |
1306 | // | |
1307 | int32_t headerSize=udata_swapDataHeader(ds, inData, length, outData, status); | |
1308 | ||
1309 | // | |
1310 | // Get the CompactTrieHeader, and check that it appears to be OK. | |
1311 | // | |
1312 | const uint8_t *inBytes =(const uint8_t *)inData+headerSize; | |
1313 | const CompactTrieHeader *header = (const CompactTrieHeader *)inBytes; | |
1314 | if (ds->readUInt32(header->magic) != COMPACT_TRIE_MAGIC_1 | |
1315 | || ds->readUInt32(header->size) < sizeof(CompactTrieHeader)) | |
1316 | { | |
1317 | udata_printError(ds, "triedict_swap(): CompactTrieHeader is invalid.\n"); | |
1318 | *status=U_UNSUPPORTED_ERROR; | |
1319 | return 0; | |
1320 | } | |
1321 | ||
1322 | // | |
1323 | // Prefight operation? Just return the size | |
1324 | // | |
1325 | uint32_t totalSize = ds->readUInt32(header->size); | |
1326 | int32_t sizeWithUData = (int32_t)totalSize + headerSize; | |
1327 | if (length < 0) { | |
1328 | return sizeWithUData; | |
1329 | } | |
1330 | ||
1331 | // | |
1332 | // Check that length passed in is consistent with length from RBBI data header. | |
1333 | // | |
1334 | if (length < sizeWithUData) { | |
1335 | udata_printError(ds, "triedict_swap(): too few bytes (%d after ICU Data header) for trie data.\n", | |
1336 | totalSize); | |
1337 | *status=U_INDEX_OUTOFBOUNDS_ERROR; | |
1338 | return 0; | |
1339 | } | |
1340 | ||
1341 | // | |
1342 | // Swap the Data. Do the data itself first, then the CompactTrieHeader, because | |
1343 | // we need to reference the header to locate the data, and an | |
1344 | // inplace swap of the header leaves it unusable. | |
1345 | // | |
1346 | uint8_t *outBytes = (uint8_t *)outData + headerSize; | |
1347 | CompactTrieHeader *outputHeader = (CompactTrieHeader *)outBytes; | |
1348 | ||
1349 | #if 0 | |
1350 | // | |
1351 | // If not swapping in place, zero out the output buffer before starting. | |
1352 | // | |
1353 | if (inBytes != outBytes) { | |
1354 | uprv_memset(outBytes, 0, totalSize); | |
1355 | } | |
1356 | ||
1357 | // We need to loop through all the nodes in the offset table, and swap each one. | |
1358 | uint16_t nodeCount = ds->readUInt16(header->nodeCount); | |
1359 | // Skip node 0, which should always be 0. | |
1360 | for (int i = 1; i < nodeCount; ++i) { | |
1361 | uint32_t nodeOff = ds->readUInt32(header->offsets[i]); | |
1362 | const CompactTrieNode *inNode = (const CompactTrieNode *)(inBytes + nodeOff); | |
1363 | CompactTrieNode *outNode = (CompactTrieNode *)(outBytes + nodeOff); | |
1364 | uint16_t flagscount = ds->readUInt16(inNode->flagscount); | |
1365 | uint16_t itemCount = flagscount & kCountMask; | |
1366 | ds->writeUInt16(&outNode->flagscount, flagscount); | |
1367 | if (itemCount > 0) { | |
1368 | if (flagscount & kVerticalNode) { | |
1369 | ds->swapArray16(ds, inBytes+nodeOff+offsetof(CompactTrieVerticalNode,chars), | |
1370 | itemCount*sizeof(uint16_t), | |
1371 | outBytes+nodeOff+offsetof(CompactTrieVerticalNode,chars), status); | |
1372 | uint16_t equal = ds->readUInt16(inBytes+nodeOff+offsetof(CompactTrieVerticalNode,equal); | |
1373 | ds->writeUInt16(outBytes+nodeOff+offsetof(CompactTrieVerticalNode,equal)); | |
1374 | } | |
1375 | else { | |
1376 | const CompactTrieHorizontalNode *inHNode = (const CompactTrieHorizontalNode *)inNode; | |
1377 | CompactTrieHorizontalNode *outHNode = (CompactTrieHorizontalNode *)outNode; | |
1378 | for (int j = 0; j < itemCount; ++j) { | |
1379 | uint16_t word = ds->readUInt16(inHNode->entries[j].ch); | |
1380 | ds->writeUInt16(&outHNode->entries[j].ch, word); | |
1381 | word = ds->readUInt16(inHNode->entries[j].equal); | |
1382 | ds->writeUInt16(&outHNode->entries[j].equal, word); | |
1383 | } | |
1384 | } | |
1385 | } | |
1386 | } | |
1387 | #endif | |
1388 | ||
1389 | // All the data in all the nodes consist of 16 bit items. Swap them all at once. | |
1390 | uint16_t nodeCount = ds->readUInt16(header->nodeCount); | |
1391 | uint32_t nodesOff = offsetof(CompactTrieHeader,offsets)+((uint32_t)nodeCount*sizeof(uint32_t)); | |
1392 | ds->swapArray16(ds, inBytes+nodesOff, totalSize-nodesOff, outBytes+nodesOff, status); | |
1393 | ||
1394 | // Swap the header | |
1395 | ds->writeUInt32(&outputHeader->size, totalSize); | |
1396 | uint32_t magic = ds->readUInt32(header->magic); | |
1397 | ds->writeUInt32(&outputHeader->magic, magic); | |
1398 | ds->writeUInt16(&outputHeader->nodeCount, nodeCount); | |
1399 | uint16_t root = ds->readUInt16(header->root); | |
1400 | ds->writeUInt16(&outputHeader->root, root); | |
1401 | ds->swapArray32(ds, inBytes+offsetof(CompactTrieHeader,offsets), | |
1402 | sizeof(uint32_t)*(int32_t)nodeCount, | |
1403 | outBytes+offsetof(CompactTrieHeader,offsets), status); | |
1404 | ||
1405 | return sizeWithUData; | |
1406 | } | |
1407 | ||
1408 | #endif /* #if !UCONFIG_NO_BREAK_ITERATION */ |