Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
Overview
Comment: | Update test_spellfix.c with latest changes. |
---|---|
Downloads: | Tarball | ZIP archive |
Timelines: | family | ancestors | descendants | both | trunk |
Files: | files | file ages | folders |
SHA1: |
cba2a65870481df213e006b07e74f0ca |
User & Date: | dan 2012-07-13 16:15:20.128 |
Context
2012-07-16
| ||
10:06 | If a specific database is nominated as part of a "PRAGMA integrity_check" or "PRAGMA quick_check" command, search for problems in the nominated database only. i.e. "PRAGMA main.quick_check" now only scans the main database, not all attached databases as before. (check-in: 4353e40b74 user: dan tags: trunk) | |
2012-07-13
| ||
19:26 | Add the "matchlen" column to the spellfix1 virtual table. (check-in: f24b9d87f6 user: dan tags: spellfix-matchlen) | |
16:15 | Update test_spellfix.c with latest changes. (check-in: cba2a65870 user: dan tags: trunk) | |
2012-06-30
| ||
22:22 | Setup the necessary library paths for cross-compilation with MSVC. (check-in: 7fac56ed9f user: mistachkin tags: trunk) | |
Changes
Changes to src/test8.c.
︙ | ︙ | |||
1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 | rc = sqlite3_declare_vtab(db, Tcl_GetString(objv[2])); if( rc!=SQLITE_OK ){ Tcl_SetResult(interp, (char *)sqlite3_errmsg(db), TCL_VOLATILE); return TCL_ERROR; } return TCL_OK; } #endif /* ifndef SQLITE_OMIT_VIRTUALTABLE */ /* ** Register commands with the TCL interpreter. */ int Sqlitetest8_Init(Tcl_Interp *interp){ #ifndef SQLITE_OMIT_VIRTUALTABLE static struct { char *zName; Tcl_ObjCmdProc *xProc; void *clientData; } aObjCmd[] = { | > > > > > > > > > > > > > > > > > > > > > > > > > | > | | 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 | rc = sqlite3_declare_vtab(db, Tcl_GetString(objv[2])); if( rc!=SQLITE_OK ){ Tcl_SetResult(interp, (char *)sqlite3_errmsg(db), TCL_VOLATILE); return TCL_ERROR; } return TCL_OK; } #include "test_spellfix.c" /* ** Register the spellfix virtual table module. */ static int register_spellfix_module( ClientData clientData, Tcl_Interp *interp, int objc, Tcl_Obj *CONST objv[] ){ static sqlite3_module aMod[3]; int iMod; sqlite3 *db; if( objc!=2 ){ Tcl_WrongNumArgs(interp, 1, objv, "DB"); return TCL_ERROR; } if( getDbPointer(interp, Tcl_GetString(objv[1]), &db) ) return TCL_ERROR; sqlite3Spellfix1Register(db); return TCL_OK; } #endif /* ifndef SQLITE_OMIT_VIRTUALTABLE */ /* ** Register commands with the TCL interpreter. */ int Sqlitetest8_Init(Tcl_Interp *interp){ #ifndef SQLITE_OMIT_VIRTUALTABLE static struct { char *zName; Tcl_ObjCmdProc *xProc; void *clientData; } aObjCmd[] = { { "register_echo_module", register_echo_module, 0 }, { "register_spellfix_module", register_spellfix_module, 0 }, { "sqlite3_declare_vtab", declare_vtab, 0 }, }; int i; for(i=0; i<sizeof(aObjCmd)/sizeof(aObjCmd[0]); i++){ Tcl_CreateObjCommand(interp, aObjCmd[i].zName, aObjCmd[i].xProc, aObjCmd[i].clientData, 0); } #endif |
︙ | ︙ |
Changes to src/test_spellfix.c.
︙ | ︙ | |||
161 162 163 164 165 166 167 | ** that have similar sounds are mapped into the same symbol. ** For example, all vowels and vowel clusters become the ** single symbol "A". And the letters "p", "b", "f", and ** "v" all become "B". All nasal sounds are represented ** as "N". And so forth. The mapping is base on ** ideas found in Soundex, Metaphone, and other ** long-standing phonetic matching systems. This key can | | | | 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 | ** that have similar sounds are mapped into the same symbol. ** For example, all vowels and vowel clusters become the ** single symbol "A". And the letters "p", "b", "f", and ** "v" all become "B". All nasal sounds are represented ** as "N". And so forth. The mapping is base on ** ideas found in Soundex, Metaphone, and other ** long-standing phonetic matching systems. This key can ** be generated by the function spellfix1_phonehash(X). ** Hence: k2 = spellfix1_phonehash(k1) ** ** There is also a function for computing the Wagner edit distance or the ** Levenshtein distance between a pattern and a word. This function ** is exposed as spellfix1_editdist(X,Y). The edit distance function ** returns the "cost" of converting X into Y. Some transformations ** cost more than others. Changing one vowel into a different vowel, ** for example is relatively cheap, as is doubling a constant, or |
︙ | ︙ | |||
197 198 199 200 201 202 203 | ** typically be hundreds of thousands or millions of words in the ** vocabulary, and so this approach is not fast enough. ** ** Suppose the term that is being spell-corrected is X. To limit ** the search space, X is converted to a k2-like key using the ** equivalent of: ** | | | 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 | ** typically be hundreds of thousands or millions of words in the ** vocabulary, and so this approach is not fast enough. ** ** Suppose the term that is being spell-corrected is X. To limit ** the search space, X is converted to a k2-like key using the ** equivalent of: ** ** key = spellfix1_phonehash(lower(spellfix1_translit(X))) ** ** This key is then limited to "scope" characters. The default scope ** value is 4, but an alternative scope can be specified using the ** "scope=N" term in the WHERE clause. After the key has been truncated, ** the edit distance is run against every term in the vocabulary that ** has a k2 value that begins with the abbreviated key. ** |
︙ | ︙ | |||
281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 | #else # include <string.h> # include <stdio.h> # include <stdlib.h> # include "sqlite3ext.h" SQLITE_EXTENSION_INIT1 #endif /* !SQLITE_CORE */ /* ** Character classes for ASCII characters: ** ** 0 '' Silent letters: H W ** 1 'A' Any vowel: A E I O U (Y) ** 2 'B' A bilabeal stop or fricative: B F P V ** 3 'C' Other fricatives or back stops: C G J K Q S X Z ** 4 'D' Alveolar stops: D T ** 5 'H' Letter H at the beginning of a word | > | > | | | | | | > | | | | | | < | | | | | | | | > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > < > | | | | | | | | | > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > | | | > > | > > | > > > | > > > > > > > > > > > > > > > > > > > > > > > > > > | | | | | > > > | 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 | #else # include <string.h> # include <stdio.h> # include <stdlib.h> # include "sqlite3ext.h" SQLITE_EXTENSION_INIT1 #endif /* !SQLITE_CORE */ #include <ctype.h> /* ** Character classes for ASCII characters: ** ** 0 '' Silent letters: H W ** 1 'A' Any vowel: A E I O U (Y) ** 2 'B' A bilabeal stop or fricative: B F P V ** 3 'C' Other fricatives or back stops: C G J K Q S X Z ** 4 'D' Alveolar stops: D T ** 5 'H' Letter H at the beginning of a word ** 6 'L' Glide: L ** 7 'R' Semivowel: R ** 8 'M' Nasals: M N ** 9 'W' Letter W at the beginning of a word ** 10 'Y' Letter Y at the beginning of a word. ** 11 '9' A digit: 0 1 2 3 4 5 6 7 8 9 ** 12 ' ' White space ** 13 '?' Other. */ #define CCLASS_SILENT 0 #define CCLASS_VOWEL 1 #define CCLASS_B 2 #define CCLASS_C 3 #define CCLASS_D 4 #define CCLASS_H 5 #define CCLASS_L 6 #define CCLASS_R 7 #define CCLASS_M 8 #define CCLASS_W 9 #define CCLASS_Y 10 #define CCLASS_DIGIT 11 #define CCLASS_SPACE 12 #define CCLASS_OTHER 13 /* ** The following table gives the character class for non-initial ASCII ** characters. */ static const unsigned char midClass[] = { /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_SPACE, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_SPACE, /* */ CCLASS_SPACE, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_SPACE, /* ! */ CCLASS_OTHER, /* " */ CCLASS_OTHER, /* # */ CCLASS_OTHER, /* $ */ CCLASS_OTHER, /* % */ CCLASS_OTHER, /* & */ CCLASS_OTHER, /* ' */ CCLASS_SILENT, /* ( */ CCLASS_OTHER, /* ) */ CCLASS_OTHER, /* * */ CCLASS_OTHER, /* + */ CCLASS_OTHER, /* , */ CCLASS_OTHER, /* - */ CCLASS_OTHER, /* . */ CCLASS_OTHER, /* / */ CCLASS_OTHER, /* 0 */ CCLASS_DIGIT, /* 1 */ CCLASS_DIGIT, /* 2 */ CCLASS_DIGIT, /* 3 */ CCLASS_DIGIT, /* 4 */ CCLASS_DIGIT, /* 5 */ CCLASS_DIGIT, /* 6 */ CCLASS_DIGIT, /* 7 */ CCLASS_DIGIT, /* 8 */ CCLASS_DIGIT, /* 9 */ CCLASS_DIGIT, /* : */ CCLASS_OTHER, /* ; */ CCLASS_OTHER, /* < */ CCLASS_OTHER, /* = */ CCLASS_OTHER, /* > */ CCLASS_OTHER, /* ? */ CCLASS_OTHER, /* @ */ CCLASS_OTHER, /* A */ CCLASS_VOWEL, /* B */ CCLASS_B, /* C */ CCLASS_C, /* D */ CCLASS_D, /* E */ CCLASS_VOWEL, /* F */ CCLASS_B, /* G */ CCLASS_C, /* H */ CCLASS_SILENT, /* I */ CCLASS_VOWEL, /* J */ CCLASS_C, /* K */ CCLASS_C, /* L */ CCLASS_L, /* M */ CCLASS_M, /* N */ CCLASS_M, /* O */ CCLASS_VOWEL, /* P */ CCLASS_B, /* Q */ CCLASS_C, /* R */ CCLASS_R, /* S */ CCLASS_C, /* T */ CCLASS_D, /* U */ CCLASS_VOWEL, /* V */ CCLASS_B, /* W */ CCLASS_SILENT, /* X */ CCLASS_C, /* Y */ CCLASS_VOWEL, /* Z */ CCLASS_C, /* [ */ CCLASS_OTHER, /* \ */ CCLASS_OTHER, /* ] */ CCLASS_OTHER, /* ^ */ CCLASS_OTHER, /* _ */ CCLASS_OTHER, /* ` */ CCLASS_OTHER, /* a */ CCLASS_VOWEL, /* b */ CCLASS_B, /* c */ CCLASS_C, /* d */ CCLASS_D, /* e */ CCLASS_VOWEL, /* f */ CCLASS_B, /* g */ CCLASS_C, /* h */ CCLASS_SILENT, /* i */ CCLASS_VOWEL, /* j */ CCLASS_C, /* k */ CCLASS_C, /* l */ CCLASS_L, /* m */ CCLASS_M, /* n */ CCLASS_M, /* o */ CCLASS_VOWEL, /* p */ CCLASS_B, /* q */ CCLASS_C, /* r */ CCLASS_R, /* s */ CCLASS_C, /* t */ CCLASS_D, /* u */ CCLASS_VOWEL, /* v */ CCLASS_B, /* w */ CCLASS_SILENT, /* x */ CCLASS_C, /* y */ CCLASS_VOWEL, /* z */ CCLASS_C, /* { */ CCLASS_OTHER, /* | */ CCLASS_OTHER, /* } */ CCLASS_OTHER, /* ~ */ CCLASS_OTHER, /* */ CCLASS_OTHER, }; /* ** This tables gives the character class for ASCII characters that form the ** initial character of a word. The only difference from midClass is with ** the letters H, W, and Y. */ static const unsigned char initClass[] = { /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_SPACE, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_SPACE, /* */ CCLASS_SPACE, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_OTHER, /* */ CCLASS_SPACE, /* ! */ CCLASS_OTHER, /* " */ CCLASS_OTHER, /* # */ CCLASS_OTHER, /* $ */ CCLASS_OTHER, /* % */ CCLASS_OTHER, /* & */ CCLASS_OTHER, /* ' */ CCLASS_OTHER, /* ( */ CCLASS_OTHER, /* ) */ CCLASS_OTHER, /* * */ CCLASS_OTHER, /* + */ CCLASS_OTHER, /* , */ CCLASS_OTHER, /* - */ CCLASS_OTHER, /* . */ CCLASS_OTHER, /* / */ CCLASS_OTHER, /* 0 */ CCLASS_DIGIT, /* 1 */ CCLASS_DIGIT, /* 2 */ CCLASS_DIGIT, /* 3 */ CCLASS_DIGIT, /* 4 */ CCLASS_DIGIT, /* 5 */ CCLASS_DIGIT, /* 6 */ CCLASS_DIGIT, /* 7 */ CCLASS_DIGIT, /* 8 */ CCLASS_DIGIT, /* 9 */ CCLASS_DIGIT, /* : */ CCLASS_OTHER, /* ; */ CCLASS_OTHER, /* < */ CCLASS_OTHER, /* = */ CCLASS_OTHER, /* > */ CCLASS_OTHER, /* ? */ CCLASS_OTHER, /* @ */ CCLASS_OTHER, /* A */ CCLASS_VOWEL, /* B */ CCLASS_B, /* C */ CCLASS_C, /* D */ CCLASS_D, /* E */ CCLASS_VOWEL, /* F */ CCLASS_B, /* G */ CCLASS_C, /* H */ CCLASS_SILENT, /* I */ CCLASS_VOWEL, /* J */ CCLASS_C, /* K */ CCLASS_C, /* L */ CCLASS_L, /* M */ CCLASS_M, /* N */ CCLASS_M, /* O */ CCLASS_VOWEL, /* P */ CCLASS_B, /* Q */ CCLASS_C, /* R */ CCLASS_R, /* S */ CCLASS_C, /* T */ CCLASS_D, /* U */ CCLASS_VOWEL, /* V */ CCLASS_B, /* W */ CCLASS_W, /* X */ CCLASS_C, /* Y */ CCLASS_Y, /* Z */ CCLASS_C, /* [ */ CCLASS_OTHER, /* \ */ CCLASS_OTHER, /* ] */ CCLASS_OTHER, /* ^ */ CCLASS_OTHER, /* _ */ CCLASS_OTHER, /* ` */ CCLASS_OTHER, /* a */ CCLASS_VOWEL, /* b */ CCLASS_B, /* c */ CCLASS_C, /* d */ CCLASS_D, /* e */ CCLASS_VOWEL, /* f */ CCLASS_B, /* g */ CCLASS_C, /* h */ CCLASS_SILENT, /* i */ CCLASS_VOWEL, /* j */ CCLASS_C, /* k */ CCLASS_C, /* l */ CCLASS_L, /* m */ CCLASS_M, /* n */ CCLASS_M, /* o */ CCLASS_VOWEL, /* p */ CCLASS_B, /* q */ CCLASS_C, /* r */ CCLASS_R, /* s */ CCLASS_C, /* t */ CCLASS_D, /* u */ CCLASS_VOWEL, /* v */ CCLASS_B, /* w */ CCLASS_W, /* x */ CCLASS_C, /* y */ CCLASS_Y, /* z */ CCLASS_C, /* { */ CCLASS_OTHER, /* | */ CCLASS_OTHER, /* } */ CCLASS_OTHER, /* ~ */ CCLASS_OTHER, /* */ CCLASS_OTHER, }; /* ** Mapping from the character class number (0-13) to a symbol for each ** character class. Note that initClass[] can be used to map the class ** symbol back into the class number. */ static const unsigned char className[] = ".ABCDHLRMWY9 ?"; /* ** Generate a "phonetic hash" from a string of ASCII characters ** in zIn[0..nIn-1]. ** ** * Map characters by character class as defined above. ** * Omit double-letters ** * Omit vowels beside R and L ** * Omit T when followed by CH ** * Omit W when followed by R ** * Omit D when followed by J or G ** * Omit K in KN or G in GN at the beginning of a word ** ** Space to hold the result is obtained from sqlite3_malloc() ** ** Return NULL if memory allocation fails. */ static unsigned char *phoneticHash(const unsigned char *zIn, int nIn){ unsigned char *zOut = sqlite3_malloc( nIn + 1 ); int i; int nOut = 0; char cPrev = 0x77; char cPrevX = 0x77; const unsigned char *aClass = initClass; if( zOut==0 ) return 0; if( nIn>2 ){ switch( zIn[0] ){ case 'g': case 'k': { if( zIn[1]=='n' ){ zIn++; nIn--; } break; } } } if( zIn[0]=='k' && zIn[1]=='n' ){ zIn++, nIn--; } for(i=0; i<nIn; i++){ unsigned char c = zIn[i]; if( i+1<nIn ){ if( c=='w' && zIn[i+1]=='r' ) continue; if( c=='d' && (zIn[i+1]=='j' || zIn[i+1]=='g') ) continue; if( i+2<nIn ){ if( c=='t' && zIn[i+1]=='c' && zIn[i+2]=='h' ) continue; } } c = aClass[c&0x7f]; if( c==CCLASS_SPACE ) continue; if( c==CCLASS_OTHER && cPrev!=CCLASS_DIGIT ) continue; if( c==CCLASS_VOWEL && (cPrevX==CCLASS_R || cPrevX==CCLASS_L) ){ continue; /* No vowels beside L or R */ } if( (c==CCLASS_R || c==CCLASS_L) && cPrevX==CCLASS_VOWEL ){ nOut--; /* No vowels beside L or R */ } cPrev = c; if( c==CCLASS_SILENT ) continue; cPrevX = c; if( c==CCLASS_SPACE ) continue; aClass = midClass; c = className[c]; if( c!=zOut[nOut-1] ) zOut[nOut++] = c; } zOut[nOut] = 0; return zOut; } /* ** This is an SQL function wrapper around phoneticHash(). See ** the description of phoneticHash() for additional information. */ static void phoneticHashSqlFunc( sqlite3_context *context, int argc, sqlite3_value **argv ){ const unsigned char *zIn; unsigned char *zOut; zIn = sqlite3_value_text(argv[0]); if( zIn==0 ) return; zOut = phoneticHash(zIn, sqlite3_value_bytes(argv[0])); if( zOut==0 ){ sqlite3_result_error_nomem(context); }else{ sqlite3_result_text(context, (char*)zOut, -1, sqlite3_free); } } /* ** Return the character class number for a character given its ** context. */ static char characterClass(char cPrev, char c){ return cPrev==0 ? initClass[c&0x7f] : midClass[c&0x7f]; } /* ** Return the cost of inserting or deleting character c immediately ** following character cPrev. If cPrev==0, that means c is the first ** character of the word. */ static int insertOrDeleteCost(char cPrev, char c, char cNext){ char classC = characterClass(cPrev, c); char classCprev; if( classC==CCLASS_SILENT ){ /* Insert or delete "silent" characters such as H or W */ return 1; } if( cPrev==c ){ /* Repeated characters, or miss a repeat */ return 10; } if( classC==CCLASS_VOWEL && (cPrev=='r' || cNext=='r') ){ return 20; /* Insert a vowel before or after 'r' */ } classCprev = characterClass(cPrev, cPrev); if( classC==classCprev ){ if( classC==CCLASS_VOWEL ){ /* Remove or add a new vowel to a vowel cluster */ return 15; }else{ |
︙ | ︙ | |||
496 497 498 499 500 501 502 | ** Smaller numbers mean a closer match. ** ** Negative values indicate an error: ** -1 One of the inputs is NULL ** -2 Non-ASCII characters on input ** -3 Unable to allocate memory */ | | > | 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 | ** Smaller numbers mean a closer match. ** ** Negative values indicate an error: ** -1 One of the inputs is NULL ** -2 Non-ASCII characters on input ** -3 Unable to allocate memory */ static int editdist1(const char *zA, const char *zB, int iLangId){ int nA, nB; /* Number of characters in zA[] and zB[] */ int xA, xB; /* Loop counters for zA[] and zB[] */ char cA, cB; /* Current character of zA and zB */ char cAprev, cBprev; /* Previous character of zA and zB */ char cAnext, cBnext; /* Next character in zA and zB */ int d; /* North-west cost value */ int dc = 0; /* North-west character value */ int res; /* Final result */ int *m; /* The cost matrix */ char *cx; /* Corresponding character values */ int *toFree = 0; /* Malloced space */ int mStack[60+15]; /* Stack space to use if not too much is needed */ |
︙ | ︙ | |||
532 533 534 535 536 537 538 | if( zB[nB]>127 ) return -2; } /* Special processing if either string is empty */ if( nA==0 ){ cBprev = dc; for(xB=res=0; (cB = zB[xB])!=0; xB++){ | | | | 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 | if( zB[nB]>127 ) return -2; } /* Special processing if either string is empty */ if( nA==0 ){ cBprev = dc; for(xB=res=0; (cB = zB[xB])!=0; xB++){ res += insertOrDeleteCost(cBprev, cB, zB[xB+1])/FINAL_INS_COST_DIV; cBprev = cB; } return res; } if( nB==0 ){ cAprev = dc; for(xA=res=0; (cA = zA[xA])!=0; xA++){ res += insertOrDeleteCost(cAprev, cA, zA[xA+1]); cAprev = cA; } return res; } /* A is a prefix of B */ if( zA[0]=='*' && zA[1]==0 ) return 0; |
︙ | ︙ | |||
563 564 565 566 567 568 569 570 571 | cx = (char*)&m[nB+1]; /* Compute the Wagner edit distance */ m[0] = 0; cx[0] = dc; cBprev = dc; for(xB=1; xB<=nB; xB++){ cB = zB[xB-1]; cx[xB] = cB; | > | > | > | | | 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 | cx = (char*)&m[nB+1]; /* Compute the Wagner edit distance */ m[0] = 0; cx[0] = dc; cBprev = dc; for(xB=1; xB<=nB; xB++){ cBnext = zB[xB]; cB = zB[xB-1]; cx[xB] = cB; m[xB] = m[xB-1] + insertOrDeleteCost(cBprev, cB, cBnext); cBprev = cB; } cAprev = dc; for(xA=1; xA<=nA; xA++){ int lastA = (xA==nA); cA = zA[xA-1]; cAnext = zA[xA]; if( cA=='*' && lastA ) break; d = m[0]; dc = cx[0]; m[0] = d + insertOrDeleteCost(cAprev, cA, cAnext); cBprev = 0; for(xB=1; xB<=nB; xB++){ int totalCost, insCost, delCost, subCost, ncx; cB = zB[xB-1]; cBnext = zB[xB]; /* Cost to insert cB */ insCost = insertOrDeleteCost(cx[xB-1], cB, cBnext); if( lastA ) insCost /= FINAL_INS_COST_DIV; /* Cost to delete cA */ delCost = insertOrDeleteCost(cx[xB], cA, cBnext); /* Cost to substitute cA->cB */ subCost = substituteCost(cx[xB-1], cA, cB); /* Best cost */ totalCost = insCost + m[xB-1]; ncx = cB; |
︙ | ︙ | |||
620 621 622 623 624 625 626 | cx[xB] = ncx; cBprev = cB; } cAprev = cA; } /* Free the wagner matrix and return the result */ | | | | > > > | | > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > | | 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 | cx[xB] = ncx; cBprev = cB; } cAprev = cA; } /* Free the wagner matrix and return the result */ if( cA=='*' ){ res = m[1]; for(xB=1; xB<=nB; xB++){ if( m[xB]<res ) res = m[xB]; } }else{ res = m[nB]; } sqlite3_free(toFree); return res; } /* ** Function: editdist(A,B) ** editdist(A,B,langid) ** ** Return the cost of transforming string A into string B. Both strings ** must be pure ASCII text. If A ends with '*' then it is assumed to be ** a prefix of B and extra characters on the end of B have minimal additional ** cost. */ static void editdistSqlFunc( sqlite3_context *context, int argc, sqlite3_value **argv ){ int langid = argc==2 ? 0 : sqlite3_value_int(argv[2]); int res = editdist1( (const char*)sqlite3_value_text(argv[0]), (const char*)sqlite3_value_text(argv[1]), langid); if( res<0 ){ if( res==(-3) ){ sqlite3_result_error_nomem(context); }else if( res==(-2) ){ sqlite3_result_error(context, "non-ASCII input to editdist()", -1); }else{ sqlite3_result_error(context, "NULL input to editdist()", -1); } }else{ sqlite3_result_int(context, res); } } /* End of the fixed-cost edit distance implementation ****************************************************************************** ***************************************************************************** ** Begin: Configurable cost unicode edit distance routines */ /* Forward declaration of structures */ typedef struct EditDist3Cost EditDist3Cost; typedef struct EditDist3Config EditDist3Config; typedef struct EditDist3Point EditDist3Point; typedef struct EditDist3From EditDist3From; typedef struct EditDist3FromString EditDist3FromString; typedef struct EditDist3To EditDist3To; typedef struct EditDist3ToString EditDist3ToString; typedef struct EditDist3Lang EditDist3Lang; /* ** An entry in the edit cost table */ struct EditDist3Cost { EditDist3Cost *pNext; /* Next cost element */ u8 nFrom; /* Number of bytes in aFrom */ u8 nTo; /* Number of bytes in aTo */ u16 iCost; /* Cost of this transformation */ char a[4] ; /* FROM string followed by TO string */ /* Additional TO and FROM string bytes appended as necessary */ }; /* ** Edit costs for a particular language ID */ struct EditDist3Lang { int iLang; /* Language ID */ int iInsCost; /* Default insertion cost */ int iDelCost; /* Default deletion cost */ int iSubCost; /* Default substitution cost */ EditDist3Cost *pCost; /* Costs */ }; /* ** The default EditDist3Lang object, with default costs. */ static const EditDist3Lang editDist3Lang = { 0, 100, 100, 150, 0 }; /* ** Complete configuration */ struct EditDist3Config { int nLang; /* Number of language IDs. Size of a[] */ EditDist3Lang *a; /* One for each distinct language ID */ }; /* ** Extra information about each character in the FROM string. */ struct EditDist3From { int nSubst; /* Number of substitution cost entries */ int nDel; /* Number of deletion cost entries */ int nByte; /* Number of bytes in this character */ EditDist3Cost **apSubst; /* Array of substitution costs for this element */ EditDist3Cost **apDel; /* Array of deletion cost entries */ }; /* ** A precompiled FROM string. * ** In the common case we expect the FROM string to be reused multiple times. ** In other words, the common case will be to measure the edit distance ** from a single origin string to multiple target strings. */ struct EditDist3FromString { char *z; /* The complete text of the FROM string */ int n; /* Number of characters in the FROM string */ int isPrefix; /* True if ends with '*' character */ EditDist3From *a; /* Extra info about each char of the FROM string */ }; /* ** Extra information about each character in the TO string. */ struct EditDist3To { int nIns; /* Number of insertion cost entries */ int nByte; /* Number of bytes in this character */ EditDist3Cost **apIns; /* Array of deletion cost entries */ }; /* ** A precompiled FROM string */ struct EditDist3ToString { char *z; /* The complete text of the TO string */ int n; /* Number of characters in the TO string */ EditDist3To *a; /* Extra info about each char of the TO string */ }; /* ** Clear or delete an instance of the object that records all edit-distance ** weights. */ static void editDist3ConfigClear(EditDist3Config *p){ int i; if( p==0 ) return; for(i=0; i<p->nLang; i++){ EditDist3Cost *pCost, *pNext; pCost = p->a[i].pCost; while( pCost ){ pNext = pCost->pNext; sqlite3_free(pCost); pCost = pNext; } } sqlite3_free(p->a); memset(p, 0, sizeof(*p)); } static void editDist3ConfigDelete(void *pIn){ EditDist3Config *p = (EditDist3Config*)pIn; editDist3ConfigClear(p); sqlite3_free(p); } /* ** Load all edit-distance weights from a table. */ static int editDist3ConfigLoad( EditDist3Config *p, /* The edit distance configuration to load */ sqlite3 *db, /* Load from this database */ const char *zTable /* Name of the table from which to load */ ){ sqlite3_stmt *pStmt; int rc; char *zSql; int iLangPrev = -9999; EditDist3Lang *pLang; zSql = sqlite3_mprintf("SELECT iLang, cFrom, cTo, iCost" " FROM \"%w\" WHERE iLang>=0 ORDER BY iLang", zTable); if( zSql==0 ) return SQLITE_NOMEM; rc = sqlite3_prepare(db, zSql, -1, &pStmt, 0); sqlite3_free(zSql); if( rc ) return rc; editDist3ConfigClear(p); while( sqlite3_step(pStmt)==SQLITE_ROW ){ int iLang = sqlite3_column_int(pStmt, 0); const char *zFrom = (const char*)sqlite3_column_text(pStmt, 1); int nFrom = sqlite3_column_bytes(pStmt, 1); const char *zTo = (const char*)sqlite3_column_text(pStmt, 2); int nTo = sqlite3_column_bytes(pStmt, 2); int iCost = sqlite3_column_int(pStmt, 3); if( nFrom>100 || nFrom<0 || nTo>100 || nTo<0 ) continue; if( iCost<0 ) continue; if( iLang!=iLangPrev ){ EditDist3Lang *pNew; p->nLang++; pNew = sqlite3_realloc(p->a, p->nLang*sizeof(p->a[0])); if( pNew==0 ){ rc = SQLITE_NOMEM; break; } p->a = pNew; pLang = &p->a[p->nLang-1]; pLang->iLang = iLang; pLang->iInsCost = 100; pLang->iDelCost = 100; pLang->iSubCost = 200; pLang->pCost = 0; iLangPrev = iLang; } if( nFrom==1 && zFrom[0]=='?' && nTo==0 ){ pLang->iDelCost = iCost; }else if( nFrom==0 && nTo==1 && zTo[0]=='?' ){ pLang->iInsCost = iCost; }else if( nFrom==1 && nTo==1 && zFrom[0]=='?' && zTo[0]=='?' ){ pLang->iSubCost = iCost; }else{ EditDist3Cost *pCost; int nExtra = nFrom + nTo - 4; if( nExtra<0 ) nExtra = 0; pCost = sqlite3_malloc( sizeof(*pCost) + nExtra ); if( pCost==0 ){ rc = SQLITE_NOMEM; break; } pCost->nFrom = nFrom; pCost->nTo = nTo; pCost->iCost = iCost; memcpy(pCost->a, zFrom, nFrom); memcpy(pCost->a + nFrom, zTo, nTo); pCost->pNext = pLang->pCost; pLang->pCost = pCost; } } sqlite3_finalize(pStmt); return rc; } /* ** Return the length (in bytes) of a utf-8 character. Or return a maximum ** of N. */ static int utf8Len(unsigned char c, int N){ int len = 1; if( c>0x7f ){ if( (c&0xe0)==0xc0 ){ len = 2; }else if( (c&0xf0)==0xe0 ){ len = 3; }else{ len = 4; } } if( len>N ) len = N; return len; } /* ** Return TRUE (non-zero) of the To side of the given cost matches ** the given string. */ static int matchTo(EditDist3Cost *p, const char *z, int n){ if( p->nTo>n ) return 0; if( memcmp(p->a+p->nFrom, z, p->nTo)!=0 ) return 0; return 1; } /* ** Return TRUE (non-zero) of the To side of the given cost matches ** the given string. */ static int matchFrom(EditDist3Cost *p, const char *z, int n){ if( p->nFrom>n ) return 0; if( memcmp(p->a, z, p->nFrom)!=0 ) return 0; return 1; } /* ** Return TRUE (non-zero) of the next FROM character and the next TO ** character are the same. */ static int matchFromTo( EditDist3FromString *pStr, /* Left hand string */ int n1, /* Index of comparison character on the left */ const char *z2, /* Right-handl comparison character */ int n2 /* Bytes remaining in z2[] */ ){ int b1 = pStr->a[n1].nByte; if( b1>n2 ) return 0; if( memcmp(pStr->z+n1, z2, b1)!=0 ) return 0; return 1; } /* ** Delete an EditDist3FromString objecct */ static void editDist3FromStringDelete(EditDist3FromString *p){ int i; if( p ){ for(i=0; i<p->n; i++){ sqlite3_free(p->a[i].apDel); sqlite3_free(p->a[i].apSubst); } sqlite3_free(p); } } /* ** Create a EditDist3FromString object. */ static EditDist3FromString *editDist3FromStringNew( const EditDist3Lang *pLang, const char *z, int n ){ EditDist3FromString *pStr; EditDist3Cost *p; int i; if( n<0 ) n = (int)strlen(z); pStr = sqlite3_malloc( sizeof(*pStr) + sizeof(pStr->a[0])*n + n + 1 ); if( pStr==0 ) return 0; pStr->a = (EditDist3From*)&pStr[1]; pStr->n = n; pStr->z = (char*)&pStr->a[n]; memcpy(pStr->z, z, n+1); if( n && z[n-1]=='*' ){ pStr->isPrefix = 1; n--; pStr->n--; pStr->z[n] = 0; }else{ pStr->isPrefix = 0; } for(i=0; i<n; i++){ EditDist3From *pFrom = &pStr->a[i]; memset(pFrom, 0, sizeof(*pFrom)); pFrom->nByte = utf8Len((unsigned char)z[i], n-i); for(p=pLang->pCost; p; p=p->pNext){ EditDist3Cost **apNew; if( i+p->nFrom>n ) continue; if( matchFrom(p, z+i, n-i)==0 ) continue; if( p->nTo==0 ){ apNew = sqlite3_realloc(pFrom->apDel, sizeof(*apNew)*(pFrom->nDel+1)); if( apNew==0 ) break; pFrom->apDel = apNew; apNew[pFrom->nDel++] = p; }else{ apNew = sqlite3_realloc(pFrom->apSubst, sizeof(*apNew)*(pFrom->nSubst+1)); if( apNew==0 ) break; pFrom->apSubst = apNew; apNew[pFrom->nSubst++] = p; } } if( p ){ editDist3FromStringDelete(pStr); pStr = 0; break; } } return pStr; } /* ** Return the number of bytes in the common prefix of two UTF8 strings. ** Only complete characters are considered. */ static int editDist3PrefixLen(const char *z1, const char *z2){ int n = 0; while( z1[n] && z1[n]==z2[n] ){ n++; } while( n && (z1[n]&0xc0)==0x80 ){ n--; } return n; } /* ** Return the number of bytes in the common suffix of two UTF8 strings. ** Only complete characters are considered. */ static int editDist3SuffixLen(const char *z1, int n1, const char *z2, int n2){ int origN1 = n1; while( n1>0 && n2>0 && z1[n1-1]==z2[n2-1] ){ n1--; n2--; } while( n1<origN1 && (z1[n1]&0xc0)==0x80 ){ n1++; n2++; } return origN1 - n1; } /* ** Update entry m[i] such that it is the minimum of its current value ** and m[j]+iCost. ** ** If the iCost is 1,000,000 or greater, then consider the cost to be ** infinite and skip the update. */ static void updateCost( unsigned int *m, int i, int j, int iCost ){ int b; if( iCost<10000 ){ b = m[j] + iCost; if( b<m[i] ) m[i] = b; } } /* Compute the edit distance between two strings. ** ** If an error occurs, return a negative number which is the error code. */ static int editDist3Core( EditDist3FromString *pFrom, /* The FROM string */ const char *z2, /* The TO string */ int n2, /* Length of the TO string */ const EditDist3Lang *pLang /* Edit weights for a particular language ID */ ){ int k, n; int i1, b1; int i2, b2; EditDist3FromString f = *pFrom; EditDist3To *a2; unsigned int *m; int szRow; EditDist3Cost *p; int res; #if 0 /* Remove comment prefix and suffix */ n = editDist3PrefixLen(f.z, z2); if( f.n==n2 && n2==n ) return 0; /* Identical strings */ f.n -= n; f.z += n; f.a += n; n2 -= n; z2 += n; if( f.isPrefix==0 ){ n = editDist3SuffixLen(f.z, f.n, z2, n2); f.n -= n; n2 -= n; } #endif /* allocate the Wagner matrix and the aTo[] array for the TO string */ n = (f.n+1)*(n2+1); n = (n+1)&~1; m = sqlite3_malloc( n*sizeof(m[0]) + sizeof(a2[0])*n2 ); if( m==0 ) return -1; /* Out of memory */ a2 = (EditDist3To*)&m[n]; memset(a2, 0, sizeof(a2[0])*n2); /* Fill in the a1[] matrix for all characters of the TO string */ for(i2=0; i2<n2; i2++){ a2[i2].nByte = utf8Len((unsigned char)z2[i2], n2-i2); for(p=pLang->pCost; p; p=p->pNext){ EditDist3Cost **apNew; if( p->nFrom>0 ) continue; if( i2+p->nTo>n2 ) continue; if( matchTo(p, z2+i2, n2-i2)==0 ) continue; a2[i2].nIns++; apNew = sqlite3_realloc(a2[i2].apIns, sizeof(*apNew)*a2[i2].nIns); if( apNew==0 ){ res = -1; /* Out of memory */ goto editDist3Abort; } a2[i2].apIns = apNew; a2[i2].apIns[a2[i2].nIns-1] = p; } } /* Prepare to compute the minimum edit distance */ szRow = f.n+1; memset(m, 0x01, (n2+1)*szRow*sizeof(m[0])); m[0] = 0; /* First fill in the top-row of the matrix with FROM deletion costs */ for(i1=0; i1<f.n; i1 += b1){ b1 = f.a[i1].nByte; updateCost(m, i1+b1, i1, pLang->iDelCost); for(k=0; k<f.a[i1].nDel; k++){ p = f.a[i1].apDel[k]; updateCost(m, i1+p->nFrom, i1, p->iCost); } } /* Fill in all subsequent rows, top-to-bottom, left-to-right */ for(i2=0; i2<n2; i2 += b2){ int rx; /* Starting index for current row */ int rxp; /* Starting index for previous row */ b2 = a2[i2].nByte; rx = szRow*(i2+b2); rxp = szRow*i2; updateCost(m, rx, rxp, pLang->iInsCost); for(k=0; k<a2[i2].nIns; k++){ p = a2[i2].apIns[k]; updateCost(m, szRow*(i2+p->nTo), rxp, p->iCost); } for(i1=0; i1<f.n; i1+=b1){ int cx; /* Index of current cell */ int cxp; /* Index of cell immediately to the left */ int cxd; /* Index of cell to the left and one row above */ int cxu; /* Index of cell immediately above */ b1 = f.a[i1].nByte; cxp = rx + i1; cx = cxp + b1; cxd = rxp + i1; cxu = cxd + b1; updateCost(m, cx, cxp, pLang->iDelCost); for(k=0; k<f.a[i1].nDel; k++){ p = f.a[i1].apDel[k]; updateCost(m, cxp+p->nFrom, cxp, p->iCost); } updateCost(m, cx, cxu, pLang->iInsCost); if( matchFromTo(&f, i1, z2+i2, n2-i2) ){ updateCost(m, cx, cxd, 0); } updateCost(m, cx, cxd, pLang->iSubCost); for(k=0; k<f.a[i1].nSubst; k++){ p = f.a[i1].apSubst[k]; if( matchTo(p, z2+i2, n2-i2) ){ updateCost(m, cxd+p->nFrom+szRow*p->nTo, cxd, p->iCost); } } } } #if 0 printf(" ^"); for(i1=0; i1<f.n; i1++) printf(" %c-%2x", f.z[i1], f.z[i1]&0xff); printf("\n ^:"); for(i1=0; i1<szRow; i1++){ int v = m[i1]; if( v>9999 ) printf(" ****"); else printf(" %4d", v); } printf("\n"); for(i2=0; i2<n2; i2++){ printf("%c-%02x:", z2[i2], z2[i2]&0xff); for(i1=0; i1<szRow; i1++){ int v = m[(i2+1)*szRow+i1]; if( v>9999 ) printf(" ****"); else printf(" %4d", v); } printf("\n"); } #endif /* Free memory allocations and return the result */ res = (int)m[szRow*(n2+1)-1]; if( f.isPrefix ){ for(i2=f.n; i2<n2; i2++){ int b = m[szRow*i2-1]; if( b<res ) res = b; } } editDist3Abort: for(i2=0; i2<n2; i2++) sqlite3_free(a2[i2].apIns); sqlite3_free(m); return res; } /* ** Get an appropriate EditDist3Lang object. */ static const EditDist3Lang *editDist3FindLang( EditDist3Config *pConfig, int iLang ){ int i; for(i=0; i<pConfig->nLang; i++){ if( pConfig->a[i].iLang==iLang ) return &pConfig->a[i]; } return &editDist3Lang; } /* ** Function: editdist3(A,B,iLang) ** editdist3(tablename) ** ** Return the cost of transforming string A into string B using edit ** weights for iLang. ** ** The second form loads edit weights into memory from a table. */ static void editDist3SqlFunc( sqlite3_context *context, int argc, sqlite3_value **argv ){ EditDist3Config *pConfig = (EditDist3Config*)sqlite3_user_data(context); sqlite3 *db = sqlite3_context_db_handle(context); int rc; if( argc==1 ){ const char *zTable = (const char*)sqlite3_value_text(argv[0]); rc = editDist3ConfigLoad(pConfig, db, zTable); if( rc ) sqlite3_result_error_code(context, rc); }else{ const char *zA = (const char*)sqlite3_value_text(argv[0]); const char *zB = (const char*)sqlite3_value_text(argv[1]); int nA = sqlite3_value_bytes(argv[0]); int nB = sqlite3_value_bytes(argv[1]); int iLang = argc==3 ? sqlite3_value_int(argv[2]) : 0; const EditDist3Lang *pLang = editDist3FindLang(pConfig, iLang); EditDist3FromString *pFrom; int dist; pFrom = editDist3FromStringNew(pLang, zA, nA); if( pFrom==0 ){ sqlite3_result_error_nomem(context); return; } dist = editDist3Core(pFrom, zB, nB, pLang); editDist3FromStringDelete(pFrom); sqlite3_result_int(context, dist); } } /* ** Register the editDist3 function with SQLite */ static int editDist3Install(sqlite3 *db){ int rc; EditDist3Config *pConfig = sqlite3_malloc( sizeof(*pConfig) ); if( pConfig==0 ) return SQLITE_NOMEM; memset(pConfig, 0, sizeof(*pConfig)); rc = sqlite3_create_function_v2(db, "editdist3", 2, SQLITE_UTF8, pConfig, editDist3SqlFunc, 0, 0, 0); if( rc==SQLITE_OK ){ rc = sqlite3_create_function_v2(db, "editdist3", 3, SQLITE_UTF8, pConfig, editDist3SqlFunc, 0, 0, 0); } if( rc==SQLITE_OK ){ rc = sqlite3_create_function_v2(db, "editdist3", 1, SQLITE_UTF8, pConfig, editDist3SqlFunc, 0, 0, editDist3ConfigDelete); }else{ sqlite3_free(pConfig); } return rc; } /* End configurable cost unicode edit distance routines ****************************************************************************** ****************************************************************************** ** Begin transliterate unicode-to-ascii implementation */ #if !SQLITE_AMALGAMATION /* ** This lookup table is used to help decode the first byte of ** a multi-byte UTF8 character. */ static const unsigned char sqlite3Utf8Trans1[] = { 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f, |
︙ | ︙ | |||
1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 | { 0x0423, 0x55, 0x00 }, /* У to U */ { 0x0424, 0x46, 0x00 }, /* Ф to F */ { 0x0425, 0x4b, 0x68 }, /* Х to Kh */ { 0x0426, 0x54, 0x63 }, /* Ц to Tc */ { 0x0427, 0x43, 0x68 }, /* Ч to Ch */ { 0x0428, 0x53, 0x68 }, /* Ш to Sh */ { 0x0429, 0x53, 0x68 }, /* Щ to Shch */ { 0x042B, 0x59, 0x00 }, /* Ы to Y */ { 0x042D, 0x45, 0x00 }, /* Э to E */ { 0x042E, 0x49, 0x75 }, /* Ю to Iu */ { 0x042F, 0x49, 0x61 }, /* Я to Ia */ { 0x0430, 0x61, 0x00 }, /* а to a */ { 0x0431, 0x62, 0x00 }, /* б to b */ { 0x0432, 0x76, 0x00 }, /* в to v */ { 0x0433, 0x67, 0x00 }, /* г to g */ | > > | 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 | { 0x0423, 0x55, 0x00 }, /* У to U */ { 0x0424, 0x46, 0x00 }, /* Ф to F */ { 0x0425, 0x4b, 0x68 }, /* Х to Kh */ { 0x0426, 0x54, 0x63 }, /* Ц to Tc */ { 0x0427, 0x43, 0x68 }, /* Ч to Ch */ { 0x0428, 0x53, 0x68 }, /* Ш to Sh */ { 0x0429, 0x53, 0x68 }, /* Щ to Shch */ { 0x042A, 0x61, 0x00 }, /* to A */ { 0x042B, 0x59, 0x00 }, /* Ы to Y */ { 0x042C, 0x59, 0x00 }, /* to Y */ { 0x042D, 0x45, 0x00 }, /* Э to E */ { 0x042E, 0x49, 0x75 }, /* Ю to Iu */ { 0x042F, 0x49, 0x61 }, /* Я to Ia */ { 0x0430, 0x61, 0x00 }, /* а to a */ { 0x0431, 0x62, 0x00 }, /* б to b */ { 0x0432, 0x76, 0x00 }, /* в to v */ { 0x0433, 0x67, 0x00 }, /* г to g */ |
︙ | ︙ | |||
1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 | { 0x0443, 0x75, 0x00 }, /* у to u */ { 0x0444, 0x66, 0x00 }, /* ф to f */ { 0x0445, 0x6b, 0x68 }, /* х to kh */ { 0x0446, 0x74, 0x63 }, /* ц to tc */ { 0x0447, 0x63, 0x68 }, /* ч to ch */ { 0x0448, 0x73, 0x68 }, /* ш to sh */ { 0x0449, 0x73, 0x68 }, /* щ to shch */ { 0x044B, 0x79, 0x00 }, /* ы to y */ { 0x044D, 0x65, 0x00 }, /* э to e */ { 0x044E, 0x69, 0x75 }, /* ю to iu */ { 0x044F, 0x69, 0x61 }, /* я to ia */ { 0x0450, 0x65, 0x00 }, /* ѐ to e */ { 0x0451, 0x65, 0x00 }, /* ё to e */ { 0x0452, 0x64, 0x00 }, /* ђ to d */ { 0x0453, 0x67, 0x00 }, /* ѓ to g */ | > > | 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 | { 0x0443, 0x75, 0x00 }, /* у to u */ { 0x0444, 0x66, 0x00 }, /* ф to f */ { 0x0445, 0x6b, 0x68 }, /* х to kh */ { 0x0446, 0x74, 0x63 }, /* ц to tc */ { 0x0447, 0x63, 0x68 }, /* ч to ch */ { 0x0448, 0x73, 0x68 }, /* ш to sh */ { 0x0449, 0x73, 0x68 }, /* щ to shch */ { 0x044A, 0x61, 0x00 }, /* to a */ { 0x044B, 0x79, 0x00 }, /* ы to y */ { 0x044C, 0x79, 0x00 }, /* to y */ { 0x044D, 0x65, 0x00 }, /* э to e */ { 0x044E, 0x69, 0x75 }, /* ю to iu */ { 0x044F, 0x69, 0x61 }, /* я to ia */ { 0x0450, 0x65, 0x00 }, /* ѐ to e */ { 0x0451, 0x65, 0x00 }, /* ё to e */ { 0x0452, 0x64, 0x00 }, /* ђ to d */ { 0x0453, 0x67, 0x00 }, /* ѓ to g */ |
︙ | ︙ | |||
1218 1219 1220 1221 1222 1223 1224 | case SCRIPT_CYRILLIC: res = 220; break; case SCRIPT_GREEK: res = 200; break; default: res = 998; break; } sqlite3_result_int(context, res); } | > | > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > | > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > | > > > > > > > > > | | | | > > | | > | 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 | case SCRIPT_CYRILLIC: res = 220; break; case SCRIPT_GREEK: res = 200; break; default: res = 998; break; } sqlite3_result_int(context, res); } /* End transliterate ****************************************************************************** ****************************************************************************** ** Begin Polloc & Zamora SPEEDCOP style keying functions. */ /* ** The Pollock & Zamora skeleton function. Move all consonants to the ** front and all vowels to the end, removing duplicates. Except if the ** first letter is a vowel then it remains as the first letter. */ static void pollockSkeletonKey(const char *zIn, char *zOut){ int i, j; unsigned char c; char seen[26]; static const unsigned char isVowel[] = { 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0 }; memset(seen, 0, sizeof(seen)); for(i=j=0; (c = (unsigned char)zIn[i])!=0; i++){ if( c<'a' || c>'z' ) continue; if( j>0 || isVowel[c-'a'] ) continue; if( seen[c-'a'] ) continue; seen[c-'a'] = 1; zOut[j++] = c; } for(i=0; (c = (unsigned char)zIn[i])!=0; i++){ if( c<'a' || c>'z' ) continue; if( seen[c-'a'] ) continue; if( !isVowel[c-'a'] ) continue; seen[c-'a'] = 1; zOut[j++] = c; } zOut[j] = 0; } /* ** Function: pollock_skeleton(X) ** ** Return the Pollock and Zamora skeleton key for a string X of all ** lower-case letters. */ static void pollockSkeletonSqlFunc( sqlite3_context *context, int argc, sqlite3_value **argv ){ const char *zIn = (const char*)sqlite3_value_text(argv[0]); int nIn = sqlite3_value_bytes(argv[0]); char *zOut; if( zIn ){ zOut = sqlite3_malloc( nIn + 1 ); if( zOut==0 ){ sqlite3_result_error_nomem(context); }else{ pollockSkeletonKey(zIn, zOut); sqlite3_result_text(context, (char*)zOut, -1, sqlite3_free); } } } /* ** The Pollock & Zamora omission key. ** ** The key consists of unique consonants in the following order: ** ** jkqxzvwybfmgpdhclntsr ** ** These are followed by unique vowels in input order. */ static void pollockOmissionKey(const char *zIn, char *zOut){ int i, j; unsigned char c; char seen[26]; static const unsigned char isVowel[] = { 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0 }; static const unsigned char constOrder[] = "jkqxzvwybfmgpdhclntsr"; memset(seen, 0, sizeof(seen)); for(i=j=0; (c = (unsigned char)zIn[i])!=0; i++){ if( c<'a' || c>'z' ) continue; if( isVowel[c-'a'] ) continue; if( seen[c-'a'] ) continue; seen[c-'a'] = 1; } for(i=0; (c = constOrder[i])!=0; i++){ if( seen[c-'a'] ) zOut[j++] = c; } for(i=0; (c = (unsigned char)zIn[i])!=0; i++){ if( c<'a' || c>'z' ) continue; if( seen[c-'a'] ) continue; if( !isVowel[c-'a'] ) continue; seen[c-'a'] = 1; zOut[j++] = c; } zOut[j] = 0; } /* ** Function: pollock_omission(X) ** ** Return the Pollock and Zamora omission key for a string X of all ** lower-case letters. */ static void pollockOmissionSqlFunc( sqlite3_context *context, int argc, sqlite3_value **argv ){ const char *zIn = (const char*)sqlite3_value_text(argv[0]); int nIn = sqlite3_value_bytes(argv[0]); char *zOut; if( zIn ){ zOut = sqlite3_malloc( nIn + 1 ); if( zOut==0 ){ sqlite3_result_error_nomem(context); }else{ pollockOmissionKey(zIn, zOut); sqlite3_result_text(context, (char*)zOut, -1, sqlite3_free); } } } /* End SPEEDCOP keying functions ****************************************************************************** ****************************************************************************** ** Begin spellfix1 virtual table. */ /* Maximum length of a phonehash used for querying the shadow table */ #define SPELLFIX_MX_HASH 8 /* Maximum number of hash strings to examine per query */ #define SPELLFIX_MX_RUN 8 typedef struct spellfix1_vtab spellfix1_vtab; typedef struct spellfix1_cursor spellfix1_cursor; /* Fuzzy-search virtual table object */ struct spellfix1_vtab { sqlite3_vtab base; /* Base class - must be first */ sqlite3 *db; /* Database connection */ char *zDbName; /* Name of database holding this table */ char *zTableName; /* Name of the virtual table */ char *zCostTable; /* Table holding edit-distance cost numbers */ EditDist3Config *pConfig3; /* Parsed edit distance costs */ }; /* Fuzzy-search cursor object */ struct spellfix1_cursor { sqlite3_vtab_cursor base; /* Base class - must be first */ spellfix1_vtab *pVTab; /* The table to which this cursor belongs */ int nRow; /* Number of rows of content */ int nAlloc; /* Number of allocated rows */ int iRow; /* Current row of content */ int iLang; /* Value of the lang= constraint */ int iTop; /* Value of the top= constraint */ int iScope; /* Value of the scope= constraint */ int nSearch; /* Number of vocabulary items checked */ struct spellfix1_row { /* For each row of content */ sqlite3_int64 iRowid; /* Rowid for this row */ char *zWord; /* Text for this row */ int iRank; /* Rank for this row */ int iDistance; /* Distance from pattern for this row */ int iScore; /* Score for sorting */ char zHash[SPELLFIX_MX_HASH]; /* the phonehash used for this match */ } *a; }; /* ** Construct one or more SQL statements from the format string given ** and then evaluate those statements. The success code is written ** into *pRc. |
︙ | ︙ | |||
1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 | if( isDestroy ){ sqlite3 *db = p->db; spellfix1DbExec(&rc, db, "DROP TABLE IF EXISTS \"%w\".\"%w_vocab\"", p->zDbName, p->zTableName); } if( rc==SQLITE_OK ){ sqlite3_free(p->zTableName); sqlite3_free(p); } return rc; } static int spellfix1Disconnect(sqlite3_vtab *pVTab){ return spellfix1Uninit(0, pVTab); } static int spellfix1Destroy(sqlite3_vtab *pVTab){ return spellfix1Uninit(1, pVTab); } /* ** xConnect/xCreate method for the spellfix1 module. Arguments are: ** ** argv[0] -> module name ("spellfix1") ** argv[1] -> database name ** argv[2] -> table name | > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > | 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 | if( isDestroy ){ sqlite3 *db = p->db; spellfix1DbExec(&rc, db, "DROP TABLE IF EXISTS \"%w\".\"%w_vocab\"", p->zDbName, p->zTableName); } if( rc==SQLITE_OK ){ sqlite3_free(p->zTableName); editDist3ConfigDelete(p->pConfig3); sqlite3_free(p->zCostTable); sqlite3_free(p); } return rc; } static int spellfix1Disconnect(sqlite3_vtab *pVTab){ return spellfix1Uninit(0, pVTab); } static int spellfix1Destroy(sqlite3_vtab *pVTab){ return spellfix1Uninit(1, pVTab); } /* ** Make a copy of a string. Remove leading and trailing whitespace ** and dequote it. */ static char *spellfix1Dequote(const char *zIn){ char *zOut; int i, j; char c; while( isspace(zIn[0]) ) zIn++; zOut = sqlite3_mprintf("%s", zIn); if( zOut==0 ) return 0; i = (int)strlen(zOut); while( i>0 && isspace(zOut[i-1]) ){ i--; } zOut[i] = 0; c = zOut[0]; if( c=='\'' || c=='"' ){ for(i=1, j=0; zOut[i]; i++){ zOut[j++] = zOut[i]; if( zOut[i]==c ){ if( zOut[i+1]==c ){ i++; }else{ zOut[j-1] = 0; break; } } } } return zOut; } /* ** xConnect/xCreate method for the spellfix1 module. Arguments are: ** ** argv[0] -> module name ("spellfix1") ** argv[1] -> database name ** argv[2] -> table name |
︙ | ︙ | |||
1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 | ){ spellfix1_vtab *pNew = 0; const char *zModule = argv[0]; const char *zDbName = argv[1]; const char *zTableName = argv[2]; int nDbName; int rc = SQLITE_OK; if( argc<3 ){ *pzErr = sqlite3_mprintf( "%s: wrong number of CREATE VIRTUAL TABLE arguments", argv[0] ); rc = SQLITE_ERROR; }else{ | > | 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 | ){ spellfix1_vtab *pNew = 0; const char *zModule = argv[0]; const char *zDbName = argv[1]; const char *zTableName = argv[2]; int nDbName; int rc = SQLITE_OK; int i; if( argc<3 ){ *pzErr = sqlite3_mprintf( "%s: wrong number of CREATE VIRTUAL TABLE arguments", argv[0] ); rc = SQLITE_ERROR; }else{ |
︙ | ︙ | |||
1348 1349 1350 1351 1352 1353 1354 | pNew->zTableName = sqlite3_mprintf("%s", zTableName); pNew->db = db; if( pNew->zTableName==0 ){ rc = SQLITE_NOMEM; }else{ rc = sqlite3_declare_vtab(db, "CREATE TABLE x(word,rank,distance,langid," | | | > > > > > > > > > > > > > > > > > > > | 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 | pNew->zTableName = sqlite3_mprintf("%s", zTableName); pNew->db = db; if( pNew->zTableName==0 ){ rc = SQLITE_NOMEM; }else{ rc = sqlite3_declare_vtab(db, "CREATE TABLE x(word,rank,distance,langid," "score, phonehash,top HIDDEN,scope HIDDEN,srchcnt HIDDEN," "soundslike HIDDEN,command HIDDEN)" ); #define SPELLFIX_COL_WORD 0 #define SPELLFIX_COL_RANK 1 #define SPELLFIX_COL_DISTANCE 2 #define SPELLFIX_COL_LANGID 3 #define SPELLFIX_COL_SCORE 4 #define SPELLFIX_COL_PHONEHASH 5 #define SPELLFIX_COL_TOP 6 #define SPELLFIX_COL_SCOPE 7 #define SPELLFIX_COL_SRCHCNT 8 #define SPELLFIX_COL_SOUNDSLIKE 9 #define SPELLFIX_COL_COMMAND 10 } if( rc==SQLITE_OK && isCreate ){ sqlite3_uint64 r; spellfix1DbExec(&rc, db, "CREATE TABLE IF NOT EXISTS \"%w\".\"%w_vocab\"(\n" " id INTEGER PRIMARY KEY,\n" " rank INT,\n" " langid INT,\n" " word TEXT,\n" " k1 TEXT,\n" " k2 TEXT\n" ");\n", zDbName, zTableName ); sqlite3_randomness(sizeof(r), &r); spellfix1DbExec(&rc, db, "CREATE INDEX IF NOT EXISTS \"%w\".\"%w_index_%llx\" " "ON \"%w_vocab\"(langid,k2);", zDbName, zModule, r, zTableName ); } for(i=3; rc==SQLITE_OK && i<argc; i++){ if( memcmp(argv[i],"edit_cost_table=",16)==0 && pNew->zCostTable==0 ){ pNew->zCostTable = spellfix1Dequote(&argv[i][16]); if( pNew->zCostTable==0 ) rc = SQLITE_NOMEM; continue; } rc = SQLITE_ERROR; } } } *ppVTab = (sqlite3_vtab *)pNew; return rc; } |
︙ | ︙ | |||
1402 1403 1404 1405 1406 1407 1408 | sqlite3_vtab **ppVTab, char **pzErr ){ return spellfix1Init(1, db, pAux, argc, argv, ppVTab, pzErr); } /* | | < | < < > > > > > > > > > > > > > > > > > > > | > > > > > > | | | | > > > > > > > > > > | 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 | sqlite3_vtab **ppVTab, char **pzErr ){ return spellfix1Init(1, db, pAux, argc, argv, ppVTab, pzErr); } /* ** Clear all of the content from a cursor. */ static void spellfix1ResetCursor(spellfix1_cursor *pCur){ int i; for(i=0; i<pCur->nRow; i++){ sqlite3_free(pCur->a[i].zWord); } pCur->nRow = 0; pCur->iRow = 0; pCur->nSearch = 0; } /* ** Resize the cursor to hold up to N rows of content */ static void spellfix1ResizeCursor(spellfix1_cursor *pCur, int N){ struct spellfix1_row *aNew; assert( N>=pCur->nRow ); aNew = sqlite3_realloc(pCur->a, sizeof(pCur->a[0])*N); if( aNew==0 && N>0 ){ spellfix1ResetCursor(pCur); sqlite3_free(pCur->a); pCur->nAlloc = 0; pCur->a = 0; }else{ pCur->nAlloc = N; pCur->a = aNew; } } /* ** Close a fuzzy-search cursor. */ static int spellfix1Close(sqlite3_vtab_cursor *cur){ spellfix1_cursor *pCur = (spellfix1_cursor *)cur; spellfix1ResetCursor(pCur); spellfix1ResizeCursor(pCur, 0); sqlite3_free(pCur); return SQLITE_OK; } /* ** Search for terms of these forms: ** ** (A) word MATCH $str ** (B) langid == $langid ** (C) top = $top ** (D) scope = $scope ** (E) distance < $distance ** (F) distance <= $distance ** ** The plan number is a bit mask formed with these bits: ** ** 0x01 (A) is found ** 0x02 (B) is found ** 0x04 (C) is found ** 0x08 (D) is found ** 0x10 (E) is found ** 0x20 (F) is found ** ** filter.argv[*] values contains $str, $langid, $top, and $scope, ** if specified and in that order. */ static int spellfix1BestIndex(sqlite3_vtab *tab, sqlite3_index_info *pIdxInfo){ int iPlan = 0; int iLangTerm = -1; int iTopTerm = -1; int iScopeTerm = -1; int iDistTerm = -1; int i; const struct sqlite3_index_constraint *pConstraint; pConstraint = pIdxInfo->aConstraint; for(i=0; i<pIdxInfo->nConstraint; i++, pConstraint++){ if( pConstraint->usable==0 ) continue; /* Terms of the form: word MATCH $str */ if( (iPlan & 1)==0 && pConstraint->iColumn==SPELLFIX_COL_WORD && pConstraint->op==SQLITE_INDEX_CONSTRAINT_MATCH ){ iPlan |= 1; pIdxInfo->aConstraintUsage[i].argvIndex = 1; pIdxInfo->aConstraintUsage[i].omit = 1; } /* Terms of the form: langid = $langid */ if( (iPlan & 2)==0 && pConstraint->iColumn==SPELLFIX_COL_LANGID && pConstraint->op==SQLITE_INDEX_CONSTRAINT_EQ ){ iPlan |= 2; iLangTerm = i; } /* Terms of the form: top = $top */ if( (iPlan & 4)==0 && pConstraint->iColumn==SPELLFIX_COL_TOP && pConstraint->op==SQLITE_INDEX_CONSTRAINT_EQ ){ iPlan |= 4; iTopTerm = i; } /* Terms of the form: scope = $scope */ if( (iPlan & 8)==0 && pConstraint->iColumn==SPELLFIX_COL_SCOPE && pConstraint->op==SQLITE_INDEX_CONSTRAINT_EQ ){ iPlan |= 8; iScopeTerm = i; } /* Terms of the form: distance < $dist or distance <= $dist */ if( (iPlan & (16|32))==0 && pConstraint->iColumn==SPELLFIX_COL_DISTANCE && (pConstraint->op==SQLITE_INDEX_CONSTRAINT_LT || pConstraint->op==SQLITE_INDEX_CONSTRAINT_LE) ){ iPlan |= pConstraint->op==SQLITE_INDEX_CONSTRAINT_LT ? 16 : 32; iDistTerm = i; } } if( iPlan&1 ){ int idx = 2; pIdxInfo->idxNum = iPlan; if( pIdxInfo->nOrderBy==1 && pIdxInfo->aOrderBy[0].iColumn==4 && pIdxInfo->aOrderBy[0].desc==0 |
︙ | ︙ | |||
1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 | pIdxInfo->aConstraintUsage[iTopTerm].argvIndex = idx++; pIdxInfo->aConstraintUsage[iTopTerm].omit = 1; } if( iPlan&8 ){ pIdxInfo->aConstraintUsage[iScopeTerm].argvIndex = idx++; pIdxInfo->aConstraintUsage[iScopeTerm].omit = 1; } pIdxInfo->estimatedCost = (double)10000; }else{ pIdxInfo->idxNum = 0; pIdxInfo->estimatedCost = (double)10000000; } return SQLITE_OK; } | > > > > | 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 | pIdxInfo->aConstraintUsage[iTopTerm].argvIndex = idx++; pIdxInfo->aConstraintUsage[iTopTerm].omit = 1; } if( iPlan&8 ){ pIdxInfo->aConstraintUsage[iScopeTerm].argvIndex = idx++; pIdxInfo->aConstraintUsage[iScopeTerm].omit = 1; } if( iPlan&(16|32) ){ pIdxInfo->aConstraintUsage[iDistTerm].argvIndex = idx++; pIdxInfo->aConstraintUsage[iDistTerm].omit = 1; } pIdxInfo->estimatedCost = (double)10000; }else{ pIdxInfo->idxNum = 0; pIdxInfo->estimatedCost = (double)10000000; } return SQLITE_OK; } |
︙ | ︙ | |||
1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 | ** that they sort in order of increasing distance. */ static int spellfix1RowCompare(const void *A, const void *B){ const struct spellfix1_row *a = (const struct spellfix1_row*)A; const struct spellfix1_row *b = (const struct spellfix1_row*)B; return a->iScore - b->iScore; } /* ** This version of the xFilter method work if the MATCH term is present ** and we are doing a scan. */ static int spellfix1FilterForMatch( spellfix1_cursor *pCur, int idxNum, int argc, sqlite3_value **argv ){ | > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > | > | | < < | | | | < | > | | > > > > > > > > > > > > > | | > > > > > > | > | | > > > > > > | < < < < < < < < | | | | | | | | | | | > > | < < < < < < < < < < < < < < | > | < < < < < | < < > > | | < > | | | > | > > > | | > | > | | | > | | | > | 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 | ** that they sort in order of increasing distance. */ static int spellfix1RowCompare(const void *A, const void *B){ const struct spellfix1_row *a = (const struct spellfix1_row*)A; const struct spellfix1_row *b = (const struct spellfix1_row*)B; return a->iScore - b->iScore; } /* ** A structure used to pass information from spellfix1FilterForMatch() ** into spellfix1RunQuery(). */ typedef struct MatchQuery { spellfix1_cursor *pCur; /* The cursor being queried */ sqlite3_stmt *pStmt; /* shadow table query statment */ char zHash[SPELLFIX_MX_HASH]; /* The current phonehash for zPattern */ const char *zPattern; /* Transliterated input string */ int nPattern; /* Length of zPattern */ EditDist3FromString *pMatchStr3; /* Original unicode string */ EditDist3Config *pConfig3; /* Edit-distance cost coefficients */ const EditDist3Lang *pLang; /* The selected language coefficients */ int iLang; /* The language id */ int iScope; /* Default scope */ int iMaxDist; /* Maximum allowed edit distance, or -1 */ int rc; /* Error code */ int nRun; /* Number of prior runs for the same zPattern */ char azPrior[SPELLFIX_MX_RUN][SPELLFIX_MX_HASH]; /* Prior hashes */ } MatchQuery; /* ** Run a query looking for the best matches against zPattern using ** zHash as the character class seed hash. */ static void spellfix1RunQuery(MatchQuery *p, const char *zQuery, int nQuery){ const char *zK1; const char *zWord; int iDist; int iRank; int iScore; int iWorst = 0; int idx; int idxWorst = -1; int i; int iScope = p->iScope; spellfix1_cursor *pCur = p->pCur; sqlite3_stmt *pStmt = p->pStmt; char zHash1[SPELLFIX_MX_HASH]; char zHash2[SPELLFIX_MX_HASH]; char *zClass; int nClass; if( pCur->a==0 || p->rc ) return; /* Prior memory allocation failure */ if( p->nRun>=SPELLFIX_MX_RUN ) return; zClass = (char*)phoneticHash((unsigned char*)zQuery, nQuery); if( zClass==0 ){ p->rc = SQLITE_NOMEM; return; } nClass = strlen(zClass); if( nClass>SPELLFIX_MX_HASH-2 ){ nClass = SPELLFIX_MX_HASH-2; zClass[nClass] = 0; } if( nClass<=iScope ){ if( nClass>2 ){ iScope = nClass-1; }else{ iScope = nClass; } } memcpy(zHash1, zClass, iScope); sqlite3_free(zClass); zHash1[iScope] = 0; memcpy(zHash2, zHash1, iScope); zHash2[iScope] = 'Z'; zHash2[iScope+1] = 0; for(i=0; i<p->nRun; i++){ if( strcmp(p->azPrior[i], zHash1)==0 ) return; } memcpy(p->azPrior[p->nRun++], zHash1, iScope+1); sqlite3_bind_text(pStmt, 1, zHash1, -1, SQLITE_STATIC); sqlite3_bind_text(pStmt, 2, zHash2, -1, SQLITE_STATIC); for(i=0; i<pCur->nRow; i++){ if( pCur->a[i].iScore>iWorst ){ iWorst = pCur->a[i].iScore; idxWorst = i; } } while( sqlite3_step(pStmt)==SQLITE_ROW ){ iRank = sqlite3_column_int(pStmt, 2); if( p->pMatchStr3 ){ int nWord = sqlite3_column_bytes(pStmt, 1); zWord = (const char*)sqlite3_column_text(pStmt, 1); iDist = editDist3Core(p->pMatchStr3, zWord, nWord, p->pLang); }else{ zK1 = (const char*)sqlite3_column_text(pStmt, 3); if( zK1==0 ) continue; iDist = editdist1(p->zPattern, zK1, pCur->iLang); } pCur->nSearch++; iScore = spellfix1Score(iDist,iRank); if( p->iMaxDist>=0 ){ if( iDist>p->iMaxDist ) continue; if( pCur->nRow>=pCur->nAlloc-1 ){ spellfix1ResizeCursor(pCur, pCur->nAlloc*2 + 10); if( pCur->a==0 ) break; } idx = pCur->nRow; }else if( pCur->nRow<pCur->nAlloc ){ idx = pCur->nRow; }else if( iScore<iWorst ){ idx = idxWorst; sqlite3_free(pCur->a[idx].zWord); }else{ continue; } pCur->a[idx].zWord = sqlite3_mprintf("%s", sqlite3_column_text(pStmt, 1)); pCur->a[idx].iRowid = sqlite3_column_int64(pStmt, 0); pCur->a[idx].iRank = iRank; pCur->a[idx].iDistance = iDist; pCur->a[idx].iScore = iScore; memcpy(pCur->a[idx].zHash, zHash1, iScope+1); if( pCur->nRow<pCur->nAlloc ) pCur->nRow++; if( pCur->nRow==pCur->nAlloc ){ iWorst = pCur->a[0].iScore; idxWorst = 0; for(i=1; i<pCur->nRow; i++){ iScore = pCur->a[i].iScore; if( iWorst<iScore ){ iWorst = iScore; idxWorst = i; } } } } sqlite3_reset(pStmt); } /* ** This version of the xFilter method work if the MATCH term is present ** and we are doing a scan. */ static int spellfix1FilterForMatch( spellfix1_cursor *pCur, int idxNum, int argc, sqlite3_value **argv ){ const unsigned char *zMatchThis; /* RHS of the MATCH operator */ EditDist3FromString *pMatchStr3 = 0; /* zMatchThis as an editdist string */ char *zPattern; /* Transliteration of zMatchThis */ int nPattern; /* Length of zPattern */ int iLimit = 20; /* Max number of rows of output */ int iScope = 3; /* Use this many characters of zClass */ int iLang = 0; /* Language code */ char *zSql; /* SQL of shadow table query */ sqlite3_stmt *pStmt; /* Shadow table query */ int rc; /* Result code */ int idx = 1; /* Next available filter parameter */ spellfix1_vtab *p = pCur->pVTab; /* The virtual table that owns pCur */ MatchQuery x; /* For passing info to RunQuery() */ /* Load the cost table if we have not already done so */ if( p->zCostTable!=0 && p->pConfig3==0 ){ p->pConfig3 = sqlite3_malloc( sizeof(p->pConfig3[0]) ); if( p->pConfig3==0 ) return SQLITE_NOMEM; memset(p->pConfig3, 0, sizeof(p->pConfig3[0])); rc = editDist3ConfigLoad(p->pConfig3, p->db, p->zCostTable); if( rc ) return rc; } memset(&x, 0, sizeof(x)); x.iScope = 3; /* Default scope if none specified by "WHERE scope=N" */ x.iMaxDist = -1; /* Maximum allowed edit distance */ if( idxNum&2 ){ iLang = sqlite3_value_int(argv[idx++]); } if( idxNum&4 ){ iLimit = sqlite3_value_int(argv[idx++]); if( iLimit<1 ) iLimit = 1; } if( idxNum&8 ){ x.iScope = sqlite3_value_int(argv[idx++]); if( x.iScope<1 ) x.iScope = 1; if( x.iScope>SPELLFIX_MX_HASH-2 ) x.iScope = SPELLFIX_MX_HASH-2; } if( idxNum&(16|32) ){ x.iMaxDist = sqlite3_value_int(argv[idx++]); if( idxNum&16 ) x.iMaxDist--; if( x.iMaxDist<0 ) x.iMaxDist = 0; } spellfix1ResetCursor(pCur); spellfix1ResizeCursor(pCur, iLimit); zMatchThis = sqlite3_value_text(argv[0]); if( zMatchThis==0 ) return SQLITE_OK; if( p->pConfig3 ){ x.pLang = editDist3FindLang(p->pConfig3, iLang); pMatchStr3 = editDist3FromStringNew(x.pLang, (const char*)zMatchThis, -1); }else{ x.pLang = 0; } zPattern = (char*)transliterate(zMatchThis, sqlite3_value_bytes(argv[0])); if( zPattern==0 ) return SQLITE_NOMEM; nPattern = strlen(zPattern); if( zPattern[nPattern-1]=='*' ) nPattern--; zSql = sqlite3_mprintf( "SELECT id, word, rank, k1" " FROM \"%w\".\"%w_vocab\"" " WHERE langid=%d AND k2>=?1 AND k2<?2", p->zDbName, p->zTableName, iLang ); rc = sqlite3_prepare_v2(p->db, zSql, -1, &pStmt, 0); sqlite3_free(zSql); pCur->iLang = iLang; x.pCur = pCur; x.pStmt = pStmt; x.zPattern = zPattern; x.nPattern = nPattern; x.pMatchStr3 = pMatchStr3; x.iLang = iLang; x.rc = rc; x.pConfig3 = p->pConfig3; if( x.rc==SQLITE_OK ){ spellfix1RunQuery(&x, zPattern, nPattern); } #if 0 /* Convert "ght" to "t" in the original pattern and try again */ if( x.rc==SQLITE_OK ){ int i, j; /* Loop counters */ char zQuery[50]; /* Space for alternative query string */ for(i=j=0; i<nPattern && i<sizeof(zQuery)-1; i++){ char c = zPattern[i]; if( c=='g' && i<nPattern-2 && zPattern[i+1]=='h' && zPattern[i+2]=='t' ){ i += 2; c= 't'; } zQuery[j++] = c; } zQuery[j] = 0; if( j<i ){ spellfix1RunQuery(&x, zQuery, j); } } #endif if( pCur->a ){ qsort(pCur->a, pCur->nRow, sizeof(pCur->a[0]), spellfix1RowCompare); pCur->iTop = iLimit; pCur->iScope = iScope; } sqlite3_finalize(pStmt); sqlite3_free(zPattern); editDist3FromStringDelete(pMatchStr3); return pCur->a ? x.rc : SQLITE_NOMEM; } /* ** This version of xFilter handles a full-table scan case */ static int spellfix1FilterForFullScan( spellfix1_cursor *pCur, int idxNum, int argc, sqlite3_value **argv ){ spellfix1ResetCursor(pCur); spellfix1ResizeCursor(pCur, 0); return SQLITE_OK; } /* ** Called to "rewind" a cursor back to the beginning so that ** it starts its output over again. Always called at least once |
︙ | ︙ | |||
1725 1726 1727 1728 1729 1730 1731 | /* ** Return columns from the current row. */ static int spellfix1Column(sqlite3_vtab_cursor *cur, sqlite3_context *ctx, int i){ spellfix1_cursor *pCur = (spellfix1_cursor*)cur; switch( i ){ | | | | | | > > > > | | | | 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 | /* ** Return columns from the current row. */ static int spellfix1Column(sqlite3_vtab_cursor *cur, sqlite3_context *ctx, int i){ spellfix1_cursor *pCur = (spellfix1_cursor*)cur; switch( i ){ case SPELLFIX_COL_WORD: { sqlite3_result_text(ctx, pCur->a[pCur->iRow].zWord, -1, SQLITE_STATIC); break; } case SPELLFIX_COL_RANK: { sqlite3_result_int(ctx, pCur->a[pCur->iRow].iRank); break; } case SPELLFIX_COL_DISTANCE: { sqlite3_result_int(ctx, pCur->a[pCur->iRow].iDistance); break; } case SPELLFIX_COL_LANGID: { sqlite3_result_int(ctx, pCur->iLang); break; } case SPELLFIX_COL_SCORE: { sqlite3_result_int(ctx, pCur->a[pCur->iRow].iScore); break; } case SPELLFIX_COL_PHONEHASH: { sqlite3_result_text(ctx, pCur->a[pCur->iRow].zHash, -1, SQLITE_STATIC); break; } case SPELLFIX_COL_TOP: { sqlite3_result_int(ctx, pCur->iTop); break; } case SPELLFIX_COL_SCOPE: { sqlite3_result_int(ctx, pCur->iScope); break; } case SPELLFIX_COL_SRCHCNT: { sqlite3_result_int(ctx, pCur->nSearch); break; } default: { sqlite3_result_null(ctx); break; } |
︙ | ︙ | |||
1795 1796 1797 1798 1799 1800 1801 | if( argc==1 ){ /* A delete operation on the rowid given by argv[0] */ rowid = *pRowid = sqlite3_value_int64(argv[0]); spellfix1DbExec(&rc, db, "DELETE FROM \"%w\".\"%w_vocab\" " " WHERE id=%lld", p->zDbName, p->zTableName, rowid); }else{ | | | | | | > | > > > > > > | | | > > > > > > > > > > | | 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 | if( argc==1 ){ /* A delete operation on the rowid given by argv[0] */ rowid = *pRowid = sqlite3_value_int64(argv[0]); spellfix1DbExec(&rc, db, "DELETE FROM \"%w\".\"%w_vocab\" " " WHERE id=%lld", p->zDbName, p->zTableName, rowid); }else{ const unsigned char *zWord = sqlite3_value_text(argv[SPELLFIX_COL_WORD+2]); int nWord = sqlite3_value_bytes(argv[SPELLFIX_COL_WORD+2]); int iLang = sqlite3_value_int(argv[SPELLFIX_COL_LANGID+2]); int iRank = sqlite3_value_int(argv[SPELLFIX_COL_RANK+2]); const unsigned char *zSoundslike = sqlite3_value_text(argv[SPELLFIX_COL_SOUNDSLIKE+2]); int nSoundslike = sqlite3_value_bytes(argv[SPELLFIX_COL_SOUNDSLIKE+2]); char *zK1, *zK2; int i; char c; if( zWord==0 ){ /* Inserts of the form: INSERT INTO table(command) VALUES('xyzzy'); ** cause zWord to be NULL, so we look at the "command" column to see ** what special actions to take */ const char *zCmd = (const char*)sqlite3_value_text(argv[SPELLFIX_COL_COMMAND+2]); if( zCmd==0 ){ pVTab->zErrMsg = sqlite3_mprintf("%s.word may not be NULL", p->zTableName); return SQLITE_CONSTRAINT; } if( strcmp(zCmd,"reset")==0 ){ /* Reset the edit cost table (if there is one). */ editDist3ConfigDelete(p->pConfig3); p->pConfig3 = 0; return SQLITE_OK; } pVTab->zErrMsg = sqlite3_mprintf("unknown value for %s.command: \"%w\"", p->zTableName, zCmd); return SQLITE_ERROR; } if( iRank<1 ) iRank = 1; if( zSoundslike ){ zK1 = (char*)transliterate(zSoundslike, nSoundslike); }else{ zK1 = (char*)transliterate(zWord, nWord); } if( zK1==0 ) return SQLITE_NOMEM; for(i=0; (c = zK1[i])!=0; i++){ if( c>='A' && c<='Z' ) zK1[i] += 'a' - 'A'; } zK2 = (char*)phoneticHash((const unsigned char*)zK1, i); if( zK2==0 ){ sqlite3_free(zK1); return SQLITE_NOMEM; } if( sqlite3_value_type(argv[0])==SQLITE_NULL ){ spellfix1DbExec(&rc, db, "INSERT INTO \"%w\".\"%w_vocab\"(rank,langid,word,k1,k2) " |
︙ | ︙ | |||
1908 1909 1910 1911 1912 1913 1914 | static int spellfix1Register(sqlite3 *db){ int nErr = 0; int i; nErr += sqlite3_create_function(db, "spellfix1_translit", 1, SQLITE_UTF8, 0, transliterateSqlFunc, 0, 0); nErr += sqlite3_create_function(db, "spellfix1_editdist", 2, SQLITE_UTF8, 0, editdistSqlFunc, 0, 0); | | | > > > > > > > | | 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 | static int spellfix1Register(sqlite3 *db){ int nErr = 0; int i; nErr += sqlite3_create_function(db, "spellfix1_translit", 1, SQLITE_UTF8, 0, transliterateSqlFunc, 0, 0); nErr += sqlite3_create_function(db, "spellfix1_editdist", 2, SQLITE_UTF8, 0, editdistSqlFunc, 0, 0); nErr += sqlite3_create_function(db, "spellfix1_editdist", 3, SQLITE_UTF8, 0, editdistSqlFunc, 0, 0); nErr += sqlite3_create_function(db, "spellfix1_phonehash", 1, SQLITE_UTF8, 0, phoneticHashSqlFunc, 0, 0); nErr += sqlite3_create_function(db, "spellfix1_scriptcode", 1, SQLITE_UTF8, 0, scriptCodeSqlFunc, 0, 0); nErr += sqlite3_create_function(db, "pollock_skeleton", 1, SQLITE_UTF8, 0, pollockSkeletonSqlFunc, 0, 0); nErr += sqlite3_create_function(db, "pollock_omission", 1, SQLITE_UTF8, 0, pollockOmissionSqlFunc, 0, 0); nErr += sqlite3_create_module(db, "spellfix1", &spellfix1Module, 0); nErr += editDist3Install(db); /* Verify sanity of the translit[] table */ for(i=0; i<sizeof(translit)/sizeof(translit[0])-1; i++){ assert( translit[i].cFrom<translit[i+1].cFrom ); } return nErr ? SQLITE_ERROR : SQLITE_OK; } #if SQLITE_CORE || defined(SQLITE_TEST) /* ** Register the spellfix1 virtual table and its associated functions. |
︙ | ︙ |