Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
Overview
Comment: | Update the spellfix virtual table to the latest development code. |
---|---|
Downloads: | Tarball | ZIP archive | SQL archive |
Timelines: | family | ancestors | descendants | both | trunk |
Files: | files | file ages | folders |
SHA1: |
6954fef006431d153de6e63e362b8d26 |
User & Date: | drh 2012-08-14 17:29:27 |
Context
2012-08-14
| ||
18:43 | Add an assert() to the btree rebalancer in order to silence a clang/scan-build warning. check-in: 6730579c user: drh tags: trunk | |
17:29 | Update the spellfix virtual table to the latest development code. check-in: 6954fef0 user: drh tags: trunk | |
01:45 | Refer to the file mapping Win32 API functions only when absolutely necessary. check-in: 1de2237d user: mistachkin tags: trunk | |
Changes
Changes to src/test_spellfix.c.
6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 ... 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 ... 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 ... 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 ... 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 ... 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 ... 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 ... 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 ... 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 .... 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 .... 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 .... 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 .... 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 .... 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 .... 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 .... 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 .... 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 .... 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 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 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 .... 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 .... 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 .... 2307 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 .... 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 .... 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 .... 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 .... 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 .... 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 .... 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 .... 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 .... 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 .... 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 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 .... 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 .... 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 .... 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 .... 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 .... 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 .... 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 |
** ** May you do good and not evil. ** May you find forgiveness for yourself and forgive others. ** May you share freely, never taking more than you give. ** ************************************************************************* ** ** This module implements a VIRTUAL TABLE that can be used to search ** a large vocabulary for close matches. For example, this virtual ** table can be used to suggest corrections to misspelled words. Or, ** it could be used with FTS4 to do full-text search using potentially ** misspelled words. ** ** Create an instance of the virtual table this way: ** ** CREATE VIRTUAL TABLE demo USING spellfix1; ** ** The "spellfix1" term is the name of this module. The "demo" is the ** name of the virtual table you will be creating. The table is initially ** empty. You have to populate it with your vocabulary. Suppose you ** have a list of words in a table named "big_vocabulary". Then do this: ** ** INSERT INTO demo(word) SELECT word FROM big_vocabulary; ** ** If you intend to use this virtual table in cooperation with an FTS4 ** table (for spelling correctly of search terms) then you can extract ** the vocabulary using an fts3aux table: ** ** INSERT INTO demo(word) SELECT term FROM search_aux WHERE col='*'; ** ** You can also provide the virtual table with a "rank" for each word. ** The "rank" is an estimate of how common the word is. Larger numbers ** mean the word is more common. If you omit the rank when populating ** the table, then a rank of 1 is assumed. But if you have rank ** information, you can supply it and the virtual table will show a ** slight preference for selecting more commonly used terms. To ** populate the rank from an fts4aux table "search_aux" do something ** like this: ** ** INSERT INTO demo(word,rank) ** SELECT term, documents FROM search_aux WHERE col='*'; ** ** To query the virtual table, include a MATCH operator in the WHERE ** clause. For example: ** ** SELECT word FROM demo WHERE word MATCH 'kennasaw'; ** ** Using a dataset of American place names (derived from ** http://geonames.usgs.gov/domestic/download_data.htm) the query above ** returns 20 results beginning with: ** ** kennesaw ** kenosha ** kenesaw ** kenaga ** keanak ** ** If you append the character '*' to the end of the pattern, then ** a prefix search is performed. For example: ** ** SELECT word FROM demo WHERE word MATCH 'kennes*'; ** ** Yields 20 results beginning with: ** ** kennesaw ** kennestone ** kenneson ** kenneys ** keanes ** keenes ** ** The virtual table actually has a unique rowid with five columns plus three ** extra hidden columns. The columns are as follows: ** ** rowid A unique integer number associated with each ** vocabulary item in the table. This can be used ** as a foreign key on other tables in the database. ** ** word The text of the word that matches the pattern. ** Both word and pattern can contains unicode characters ** and can be mixed case. ** ** rank This is the rank of the word, as specified in the ** original INSERT statement. ** ** distance This is an edit distance or Levensthein distance going ** from the pattern to the word. ** ** langid This is the language-id of the word. All queries are ** against a single language-id, which defaults to 0. ** For any given query this value is the same on all rows. ** ** score The score is a combination of rank and distance. The ** idea is that a lower score is better. The virtual table ** attempts to find words with the lowest score and ** by default (unless overridden by ORDER BY) returns ** results in order of increasing score. ** ** matchlen For prefix queries, the number of characters in the prefix ** of the returned value (word) that matched the query term. ** For non-prefix queries, the number of characters in the ** returned value. ** ** top (HIDDEN) For any query, this value is the same on all ** rows. It is an integer which is the maximum number of ** rows that will be output. The actually number of rows ** output might be less than this number, but it will never ** be greater. The default value for top is 20, but that ** can be changed for each query by including a term of ** the form "top=N" in the WHERE clause of the query. ** ** scope (HIDDEN) For any query, this value is the same on all ** rows. The scope is a measure of how widely the virtual ** table looks for matching words. Smaller values of ** scope cause a broader search. The scope is normally ** choosen automatically and is capped at 4. Applications ** can change the scope by including a term of the form ** "scope=N" in the WHERE clause of the query. Increasing ** the scope will make the query run faster, but will reduce ** the possible corrections. ** ** srchcnt (HIDDEN) For any query, this value is the same on all ** rows. This value is an integer which is the number of ** of words examined using the edit-distance algorithm to ** find the top matches that are ultimately displayed. This ** value is for diagnostic use only. ** ** soundslike (HIDDEN) When inserting vocabulary entries, this field ** can be set to an spelling that matches what the word ** sounds like. See the DEALING WITH UNUSUAL AND DIFFICULT ** SPELLINGS section below for details. ** ** When inserting into or updating the virtual table, only the rowid, word, ** rank, and langid may be changes. Any attempt to set or modify the values ** of distance, score, top, scope, or srchcnt is silently ignored. ** ** ALGORITHM ** ** A shadow table named "%_vocab" (where the % is replaced by the name of ** the virtual table; Ex: "demo_vocab" for the "demo" virtual table) is ** constructed with these columns: ** ** id The unique id (INTEGER PRIMARY KEY) ** ** rank The rank of word. ** ** langid The language id for this entry. ** ** word The original UTF8 text of the vocabulary word ** ** k1 The word transliterated into lower-case ASCII. ** There is a standard table of mappings from non-ASCII ** characters into ASCII. Examples: "æ" -> "ae", ** "þ" -> "th", "ß" -> "ss", "á" -> "a", ... The ** accessory function spellfix1_translit(X) will do ** the non-ASCII to ASCII mapping. The built-in lower(X) ** function will convert to lower-case. Thus: ** k1 = lower(spellfix1_translit(word)). ** ** k2 This field holds a phonetic code derived from k1. Letters ** 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 ** omitting the second character of a double-constant. Other transformations ** or more expensive. The idea is that the edit distance function returns ** a low cost of words that are similar and a higher cost for words ** that are futher apart. In this implementation, the maximum cost ** of any single-character edit (delete, insert, or substitute) is 100, ** with lower costs for some edits (such as transforming vowels). ** ** The "score" for a comparison is the edit distance between the pattern ** and the word, adjusted down by the base-2 logorithm of the word rank. ** For example, a match with distance 100 but rank 1000 would have a ** score of 122 (= 100 - log2(1000) + 32) where as a match with distance ** 100 with a rank of 1 would have a score of 131 (100 - log2(1) + 32). ** (NB: The constant 32 is added to each score to keep it from going ** negative in case the edit distance is zero.) In this way, frequently ** used words get a slightly lower cost which tends to move them toward ** the top of the list of alternative spellings. ** ** A straightforward implementation of a spelling corrector would be ** to compare the search term against every word in the vocabulary ** and select the 20 with the lowest scores. However, there will ** 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. ** ** For example, suppose the input word is "Paskagula". The phonetic ** key is "BACACALA" which is then truncated to 4 characters "BACA". ** The edit distance is then run on the 4980 entries (out of ** 272,597 entries total) of the vocabulary whose k2 values begin with ** BACA, yielding "Pascagoula" as the best match. ** ** Only terms of the vocabulary with a matching langid are searched. ** Hence, the same table can contain entries from multiple languages ** and only the requested language will be used. The default langid ** is 0. ** ** DEALING WITH UNUSUAL AND DIFFICULT SPELLINGS ** ** The algorithm above works quite well for most cases, but there are ** exceptions. These exceptions can be dealt with by making additional ** entries in the virtual table using the "soundslike" column. ** ** For example, many words of Greek origin begin with letters "ps" where ** the "p" is silent. Ex: psalm, pseudonym, psoriasis, psyche. In ** another example, many Scottish surnames can be spelled with an ** initial "Mac" or "Mc". Thus, "MacKay" and "McKay" are both pronounced ** the same. ** ** Accommodation can be made for words that are not spelled as they ** sound by making additional entries into the virtual table for the ** same word, but adding an alternative spelling in the "soundslike" ** column. For example, the canonical entry for "psalm" would be this: ** ** INSERT INTO demo(word) VALUES('psalm'); ** ** To enhance the ability to correct the spelling of "salm" into ** "psalm", make an addition entry like this: ** ** INSERT INTO demo(word,soundslike) VALUES('psalm','salm'); ** ** It is ok to make multiple entries for the same word as long as ** each entry has a different soundslike value. Note that if no ** soundslike value is specified, the soundslike defaults to the word ** itself. ** ** Listed below are some cases where it might make sense to add additional ** soundslike entries. The specific entries will depend on the application ** and the target language. ** ** * Silent "p" in words beginning with "ps": psalm, psyche ** ** * Silent "p" in words beginning with "pn": pneumonia, pneumatic ** ** * Silent "p" in words beginning with "pt": pterodactyl, ptolemaic ** ** * Silent "d" in words beginning with "dj": djinn, Djikarta ** ** * Silent "k" in words beginning with "kn": knight, Knuthson ** ** * Silent "g" in words beginning with "gn": gnarly, gnome, gnat ** ** * "Mac" versus "Mc" beginning Scottish surnames ** ** * "Tch" sounds in Slavic words: Tchaikovsky vs. Chaykovsky ** ** * The letter "j" pronounced like "h" in Spanish: LaJolla ** ** * Words beginning with "wr" versus "r": write vs. rite ** ** * Miscellanous problem words such as "debt", "tsetse", ** "Nguyen", "Van Nuyes". */ #if SQLITE_CORE # include "sqliteInt.h" #else # include <string.h> # include <stdio.h> # include <stdlib.h> ................................................................................ ** 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 ................................................................................ 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; ................................................................................ /* differ only in case */ return 0; } classFrom = characterClass(cPrev, cFrom); classTo = characterClass(cPrev, cTo); if( classFrom==classTo ){ /* Same character class */ return classFrom=='A' ? 25 : 40; } if( classFrom>=CCLASS_B && classFrom<=CCLASS_Y && classTo>=CCLASS_B && classTo<=CCLASS_Y ){ /* Convert from one consonant to another, but in a different class */ return 75; } /* Any other subsitution */ ................................................................................ ** ** If pnMatch is not NULL, then *pnMatch is set to the number of bytes ** of zB that matched the pattern in zA. If zA does not end with a '*', ** then this value is always the number of bytes in zB (i.e. strlen(zB)). ** If zA does end in a '*', then it is the number of bytes in the prefix ** of zB that was deemed to match zA. */ static int editdist1(const char *zA, const char *zB, int iLangId, int *pnMatch){ 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 */ ................................................................................ #if 0 printf("A=\"%s\" B=\"%s\" dc=%c\n", zA, zB, dc?dc:' '); #endif /* Verify input strings and measure their lengths */ for(nA=0; zA[nA]; nA++){ if( zA[nA]>127 ) return -2; } for(nB=0; zB[nB]; nB++){ 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; ................................................................................ if( m[xB]<res ){ res = m[xB]; if( pnMatch ) *pnMatch = xB+nMatch; } } }else{ res = m[nB]; if( pnMatch ) *pnMatch = -1; } 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, 0); 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); ................................................................................ */ 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; ................................................................................ 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. */ ................................................................................ } /* ** 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. ................................................................................ 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 = 0; break; } } return pStr; } #if 0 /* No longer used */ /* ** 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; } #endif /* 0 */ /* ** 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. */ ................................................................................ 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); ................................................................................ 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); ................................................................................ pFrom = editDist3FromStringNew(pLang, zA, nA); if( pFrom==0 ){ sqlite3_result_error_nomem(context); return; } dist = editDist3Core(pFrom, zB, nB, pLang, 0); editDist3FromStringDelete(pFrom); sqlite3_result_int(context, dist); } } /* ** Register the editDist3 function with SQLite */ static int editDist3Install(sqlite3 *db){ ................................................................................ /* ** Return the value of the first UTF-8 character in the string. */ static int utf8Read(const unsigned char *z, int n, int *pSize){ int c, i; if( n==0 ){ c = i = 0; }else{ c = z[0]; i = 1; if( c>=0xc0 ){ c = sqlite3Utf8Trans1[c-0xc0]; while( i<n && (z[i] & 0xc0)==0x80 ){ ................................................................................ ** The returned string might contain more characters than the input. ** ** Space to hold the returned string comes from sqlite3_malloc() and ** should be freed by the caller. */ static unsigned char *transliterate(const unsigned char *zIn, int nIn){ unsigned char *zOut = sqlite3_malloc( nIn*4 + 1 ); int i, c, sz, nOut; if( zOut==0 ) return 0; i = nOut = 0; while( i<nIn ){ c = utf8Read(zIn, nIn, &sz); zIn += sz; nIn -= sz; if( c<=127 ){ zOut[nOut++] = c; }else{ int xTop, xBtm, x; ................................................................................ } 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 */ ................................................................................ struct spellfix1_cursor { sqlite3_vtab_cursor base; /* Base class - must be first */ spellfix1_vtab *pVTab; /* The table to which this cursor belongs */ char *zPattern; /* rhs of MATCH clause */ 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 */ int iMatchlen; /* Value of matchlen column (or -1) */ ................................................................................ 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; ................................................................................ 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{ nDbName = strlen(zDbName); pNew = sqlite3_malloc( sizeof(*pNew) + nDbName + 1); if( pNew==0 ){ rc = SQLITE_NOMEM; }else{ memset(pNew, 0, sizeof(*pNew)); pNew->zDbName = (char*)&pNew[1]; memcpy(pNew->zDbName, zDbName, nDbName+1); 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, matchlen, 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_MATCHLEN 5 #define SPELLFIX_COL_PHONEHASH 6 #define SPELLFIX_COL_TOP 7 #define SPELLFIX_COL_SCOPE 8 #define SPELLFIX_COL_SRCHCNT 9 #define SPELLFIX_COL_SOUNDSLIKE 10 #define SPELLFIX_COL_COMMAND 11 } 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; } /* ** The xConnect and xCreate methods */ static int spellfix1Connect( ................................................................................ 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; ................................................................................ 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 ){ pIdxInfo->orderByConsumed = 1; /* Default order by iScore */ } if( iPlan&2 ){ pIdxInfo->aConstraintUsage[iLangTerm].argvIndex = idx++; pIdxInfo->aConstraintUsage[iLangTerm].omit = 1; ................................................................................ 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 ){ ................................................................................ } 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 ){ int iMatchlen = -1; 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, &iMatchlen); }else{ zK1 = (const char*)sqlite3_column_text(pStmt, 3); if( zK1==0 ) continue; iDist = editdist1(p->zPattern, zK1, pCur->iLang, 0); } 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); ................................................................................ }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; pCur->a[idx].iMatchlen = iMatchlen; memcpy(pCur->a[idx].zHash, zHash1, iScope+1); if( pCur->nRow<pCur->nAlloc ) pCur->nRow++; ................................................................................ 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( ................................................................................ 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 ){ ................................................................................ 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])); sqlite3_free(pCur->zPattern); pCur->zPattern = zPattern; 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.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); 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 ** prior to any spellfix1Column, spellfix1Rowid, or spellfix1Eof call. ................................................................................ /* ** Advance a cursor to its next row of output */ static int spellfix1Next(sqlite3_vtab_cursor *cur){ spellfix1_cursor *pCur = (spellfix1_cursor *)cur; if( pCur->iRow < pCur->nRow ) pCur->iRow++; return SQLITE_OK; } /* ** Return TRUE if we are at the end-of-file */ static int spellfix1Eof(sqlite3_vtab_cursor *cur){ ................................................................................ spellfix1_cursor *pCur = (spellfix1_cursor *)cur; return pCur->iRow>=pCur->nRow; } /* ** 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); ................................................................................ int nWord = strlen(zWord); if( nPattern>0 && pCur->zPattern[nPattern-1]=='*' ){ char *zTranslit; int res; zTranslit = (char *)transliterate((unsigned char *)zWord, nWord); if( !zTranslit ) return SQLITE_NOMEM; res = editdist1(pCur->zPattern, zTranslit, pCur->iLang, &iMatchlen); sqlite3_free(zTranslit); if( res<0 ) return SQLITE_NOMEM; iMatchlen = translen_to_charlen(zWord, nWord, iMatchlen); }else{ iMatchlen = utf8Charlen(zWord, nWord); } } ................................................................................ } /* ** The rowid. */ static int spellfix1Rowid(sqlite3_vtab_cursor *cur, sqlite_int64 *pRowid){ spellfix1_cursor *pCur = (spellfix1_cursor*)cur; *pRowid = pCur->a[pCur->iRow].iRowid; return SQLITE_OK; } /* ** The xUpdate() method. */ static int spellfix1Update( ................................................................................ iRank, iLang, zWord, zK1, zK2 ); *pRowid = sqlite3_last_insert_rowid(db); }else{ rowid = sqlite3_value_int64(argv[0]); newRowid = *pRowid = sqlite3_value_int64(argv[1]); spellfix1DbExec(&rc, db, "UPDATE \"%w\".\"%w_vocab\" SET id=%lld, rank=%d, lang=%d," " word=%Q, rank=%d, k1=%Q, k2=%Q WHERE id=%lld", p->zDbName, p->zTableName, newRowid, iRank, iLang, zWord, zK1, zK2, rowid ); } sqlite3_free(zK1); sqlite3_free(zK2); } ................................................................................ spellfix1DbExec(&rc, db, "ALTER TABLE \"%w\".\"%w_vocab\" RENAME TO \"%w_vocab\"", p->zDbName, p->zTableName, zNewName ); if( rc==SQLITE_OK ){ sqlite3_free(p->zTableName); p->zTableName = zNewName; } return rc; } /* ** A virtual table module that provides fuzzy search. ................................................................................ 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 ); } |
| | | < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < | < | | | | > > | < < | | | | > > | < | | > | | > | > > < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < | > > > | > | > > | | | < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < | | > > > | < < < < < < | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | > | < > > > > | > > > > > | > < > > > | | > > > > > > | > > > > > > > > | > | > > > > | > > > > > > > > < < < < < < < < < < < < < < < < < < < < > > > > < > > > > < > > > | > > > > > > > > > > > | > > > > > > > | > > > > > > > > > > > > | > > > | > | | > > < < < < < < |
6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 .. 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 ... 195 196 197 198 199 200 201 202 203 204 205 206 207 208 ... 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 ... 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 ... 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 ... 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 ... 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 ... 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 ... 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 ... 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 ... 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 ... 846 847 848 849 850 851 852 853 854 855 856 857 858 859 ... 894 895 896 897 898 899 900 901 902 903 904 905 906 907 ... 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 .... 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 .... 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 .... 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 .... 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 .... 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 .... 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 .... 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 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 .... 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 .... 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 .... 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 .... 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 2291 2292 2293 2294 2295 2296 2297 2298 .... 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 .... 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 .... 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 .... 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 .... 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 .... 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 .... 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 .... 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 .... 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 .... 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 .... 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 |
** ** May you do good and not evil. ** May you find forgiveness for yourself and forgive others. ** May you share freely, never taking more than you give. ** ************************************************************************* ** ** This module implements the spellfix1 VIRTUAL TABLE that can be used ** to search a large vocabulary for close matches. See separate ** documentation files (spellfix1.wiki and editdist3.wiki) for details. */ #if SQLITE_CORE # include "sqliteInt.h" #else # include <string.h> # include <stdio.h> # include <stdlib.h> ................................................................................ ** 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' Digits: 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 ................................................................................ case 'g': case 'k': { if( zIn[1]=='n' ){ zIn++; nIn--; } break; } } } 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; ................................................................................ /* differ only in case */ return 0; } classFrom = characterClass(cPrev, cFrom); classTo = characterClass(cPrev, cTo); if( classFrom==classTo ){ /* Same character class */ return 40; } if( classFrom>=CCLASS_B && classFrom<=CCLASS_Y && classTo>=CCLASS_B && classTo<=CCLASS_Y ){ /* Convert from one consonant to another, but in a different class */ return 75; } /* Any other subsitution */ ................................................................................ ** ** If pnMatch is not NULL, then *pnMatch is set to the number of bytes ** of zB that matched the pattern in zA. If zA does not end with a '*', ** then this value is always the number of bytes in zB (i.e. strlen(zB)). ** If zA does end in a '*', then it is the number of bytes in the prefix ** of zB that was deemed to match zA. */ static int editdist1(const char *zA, const char *zB, int *pnMatch){ 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 */ ................................................................................ #if 0 printf("A=\"%s\" B=\"%s\" dc=%c\n", zA, zB, dc?dc:' '); #endif /* Verify input strings and measure their lengths */ for(nA=0; zA[nA]; nA++){ if( zA[nA]&0x80 ) return -2; } for(nB=0; zB[nB]; nB++){ if( zB[nB]&0x80 ) 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; ................................................................................ if( m[xB]<res ){ res = m[xB]; if( pnMatch ) *pnMatch = xB+nMatch; } } }else{ res = m[nB]; /* In the current implementation, pnMatch is always NULL if zA does ** not end in "*" */ assert( pnMatch==0 ); } sqlite3_free(toFree); return res; } /* ** Function: editdist(A,B) ** ** 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 res = editdist1( (const char*)sqlite3_value_text(argv[0]), (const char*)sqlite3_value_text(argv[1]), 0); 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); ................................................................................ */ 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, rc2; 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 = zFrom ? sqlite3_column_bytes(pStmt, 1) : 0; const char *zTo = (const char*)sqlite3_column_text(pStmt, 2); int nTo = zTo ? sqlite3_column_bytes(pStmt, 2) : 0; int iCost = sqlite3_column_int(pStmt, 3); assert( zFrom!=0 || nFrom==0 ); assert( zTo!=0 || nTo==0 ); if( nFrom>100 || nTo>100 ) continue; if( iCost<0 ) continue; if( iLang!=iLangPrev ){ EditDist3Lang *pNew; pNew = sqlite3_realloc(p->a, (p->nLang+1)*sizeof(p->a[0])); if( pNew==0 ){ rc = SQLITE_NOMEM; break; } p->a = pNew; pLang = &p->a[p->nLang]; p->nLang++; pLang->iLang = iLang; pLang->iInsCost = 100; pLang->iDelCost = 100; pLang->iSubCost = 150; 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; ................................................................................ pCost->iCost = iCost; memcpy(pCost->a, zFrom, nFrom); memcpy(pCost->a + nFrom, zTo, nTo); pCost->pNext = pLang->pCost; pLang->pCost = pCost; } } rc2 = sqlite3_finalize(pStmt); if( rc==SQLITE_OK ) rc = rc2; return rc; } /* ** Return the length (in bytes) of a utf-8 character. Or return a maximum ** of N. */ ................................................................................ } /* ** 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){ assert( p->nFrom<=n ); 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. ................................................................................ const char *z, int n ){ EditDist3FromString *pStr; EditDist3Cost *p; int i; if( z==0 ) return 0; 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]; memset(pStr->a, 0, sizeof(pStr->a[0])*n); 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 = 0; break; } } return pStr; } /* ** 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. */ ................................................................................ EditDist3FromString f = *pFrom; EditDist3To *a2; unsigned int *m; int szRow; EditDist3Cost *p; int res; /* 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); ................................................................................ if( matchTo(p, z2+i2, n2-i2) ){ updateCost(m, cxd+p->nFrom+szRow*p->nTo, cxd, p->iCost); } } } } #if 0 /* Enable for debugging */ 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); ................................................................................ pFrom = editDist3FromStringNew(pLang, zA, nA); if( pFrom==0 ){ sqlite3_result_error_nomem(context); return; } dist = editDist3Core(pFrom, zB, nB, pLang, 0); editDist3FromStringDelete(pFrom); if( dist==(-1) ){ sqlite3_result_error_nomem(context); }else{ sqlite3_result_int(context, dist); } } } /* ** Register the editDist3 function with SQLite */ static int editDist3Install(sqlite3 *db){ ................................................................................ /* ** Return the value of the first UTF-8 character in the string. */ static int utf8Read(const unsigned char *z, int n, int *pSize){ int c, i; /* All callers to this routine (in the current implementation) ** always have n>0. */ if( NEVER(n==0) ){ c = i = 0; }else{ c = z[0]; i = 1; if( c>=0xc0 ){ c = sqlite3Utf8Trans1[c-0xc0]; while( i<n && (z[i] & 0xc0)==0x80 ){ ................................................................................ ** The returned string might contain more characters than the input. ** ** Space to hold the returned string comes from sqlite3_malloc() and ** should be freed by the caller. */ static unsigned char *transliterate(const unsigned char *zIn, int nIn){ unsigned char *zOut = sqlite3_malloc( nIn*4 + 1 ); int c, sz, nOut; if( zOut==0 ) return 0; nOut = 0; while( nIn>0 ){ c = utf8Read(zIn, nIn, &sz); zIn += sz; nIn -= sz; if( c<=127 ){ zOut[nOut++] = c; }else{ int xTop, xBtm, x; ................................................................................ } sqlite3_result_int(context, res); } /* End transliterate ****************************************************************************** ****************************************************************************** ** 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 1 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 */ ................................................................................ struct spellfix1_cursor { sqlite3_vtab_cursor base; /* Base class - must be first */ spellfix1_vtab *pVTab; /* The table to which this cursor belongs */ char *zPattern; /* rhs of MATCH clause */ int nRow; /* Number of rows of content */ int nAlloc; /* Number of allocated rows */ int iRow; /* Current row of content */ int iLang; /* Value of the langid= constraint */ int iTop; /* Value of the top= constraint */ int iScope; /* Value of the scope= constraint */ int nSearch; /* Number of vocabulary items checked */ sqlite3_stmt *pFullScan; /* Shadow query for a full table scan */ 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 */ int iMatchlen; /* Value of matchlen column (or -1) */ ................................................................................ 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); #if 0 /* The parser will never leave spaces at the end */ while( i>0 && isspace(zOut[i-1]) ){ i--; } #endif zOut[i] = 0; c = zOut[0]; if( c=='\'' || c=='"' ){ for(i=1, j=0; ALWAYS(zOut[i]); i++){ zOut[j++] = zOut[i]; if( zOut[i]==c ){ if( zOut[i+1]==c ){ i++; }else{ zOut[j-1] = 0; break; ................................................................................ const char *zModule = argv[0]; const char *zDbName = argv[1]; const char *zTableName = argv[2]; int nDbName; int rc = SQLITE_OK; int i; nDbName = strlen(zDbName); pNew = sqlite3_malloc( sizeof(*pNew) + nDbName + 1); if( pNew==0 ){ rc = SQLITE_NOMEM; }else{ memset(pNew, 0, sizeof(*pNew)); pNew->zDbName = (char*)&pNew[1]; memcpy(pNew->zDbName, zDbName, nDbName+1); 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, matchlen, phonehash HIDDEN, " "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_MATCHLEN 5 #define SPELLFIX_COL_PHONEHASH 6 #define SPELLFIX_COL_TOP 7 #define SPELLFIX_COL_SCOPE 8 #define SPELLFIX_COL_SRCHCNT 9 #define SPELLFIX_COL_SOUNDSLIKE 10 #define SPELLFIX_COL_COMMAND 11 } 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; } *pzErr = sqlite3_mprintf("bad argument to spellfix1(): \"%s\"", argv[i]); rc = SQLITE_ERROR; } } if( rc && pNew ){ *ppVTab = 0; spellfix1Uninit(0, &pNew->base); }else{ *ppVTab = (sqlite3_vtab *)pNew; } return rc; } /* ** The xConnect and xCreate methods */ static int spellfix1Connect( ................................................................................ int i; for(i=0; i<pCur->nRow; i++){ sqlite3_free(pCur->a[i].zWord); } pCur->nRow = 0; pCur->iRow = 0; pCur->nSearch = 0; if( pCur->pFullScan ){ sqlite3_finalize(pCur->pFullScan); pCur->pFullScan = 0; } } /* ** Resize the cursor to hold up to N rows of content */ static void spellfix1ResizeCursor(spellfix1_cursor *pCur, int N){ struct spellfix1_row *aNew; ................................................................................ iDistTerm = i; } } if( iPlan&1 ){ int idx = 2; pIdxInfo->idxNum = iPlan; if( pIdxInfo->nOrderBy==1 && pIdxInfo->aOrderBy[0].iColumn==SPELLFIX_COL_SCORE && pIdxInfo->aOrderBy[0].desc==0 ){ pIdxInfo->orderByConsumed = 1; /* Default order by iScore */ } if( iPlan&2 ){ pIdxInfo->aConstraintUsage[iLangTerm].argvIndex = idx++; pIdxInfo->aConstraintUsage[iLangTerm].omit = 1; ................................................................................ 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; int rc; if( pCur->a==0 || p->rc ) return; /* Prior memory allocation failure */ zClass = (char*)phoneticHash((unsigned char*)zQuery, nQuery); if( zClass==0 ){ p->rc = SQLITE_NOMEM; return; } nClass = strlen(zClass); if( nClass>SPELLFIX_MX_HASH-2 ){ ................................................................................ } memcpy(zHash1, zClass, iScope); sqlite3_free(zClass); zHash1[iScope] = 0; memcpy(zHash2, zHash1, iScope); zHash2[iScope] = 'Z'; zHash2[iScope+1] = 0; #if SPELLFIX_MX_RUN>1 for(i=0; i<p->nRun; i++){ if( strcmp(p->azPrior[i], zHash1)==0 ) return; } #endif assert( p->nRun<SPELLFIX_MX_RUN ); memcpy(p->azPrior[p->nRun++], zHash1, iScope+1); if( sqlite3_bind_text(pStmt, 1, zHash1, -1, SQLITE_STATIC)==SQLITE_NOMEM || sqlite3_bind_text(pStmt, 2, zHash2, -1, SQLITE_STATIC)==SQLITE_NOMEM ){ p->rc = SQLITE_NOMEM; return; } #if SPELLFIX_MX_RUN>1 for(i=0; i<pCur->nRow; i++){ if( pCur->a[i].iScore>iWorst ){ iWorst = pCur->a[i].iScore; idxWorst = i; } } #endif while( sqlite3_step(pStmt)==SQLITE_ROW ){ int iMatchlen = -1; 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, &iMatchlen); }else{ zK1 = (const char*)sqlite3_column_text(pStmt, 3); if( zK1==0 ) continue; iDist = editdist1(p->zPattern, zK1, 0); } if( iDist<0 ){ p->rc = SQLITE_NOMEM; break; } 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); ................................................................................ }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)); if( pCur->a[idx].zWord==0 ){ p->rc = SQLITE_NOMEM; break; } pCur->a[idx].iRowid = sqlite3_column_int64(pStmt, 0); pCur->a[idx].iRank = iRank; pCur->a[idx].iDistance = iDist; pCur->a[idx].iScore = iScore; pCur->a[idx].iMatchlen = iMatchlen; memcpy(pCur->a[idx].zHash, zHash1, iScope+1); if( pCur->nRow<pCur->nAlloc ) pCur->nRow++; ................................................................................ if( iWorst<iScore ){ iWorst = iScore; idxWorst = i; } } } } rc = sqlite3_reset(pStmt); if( rc ) p->rc = rc; } /* ** This version of the xFilter method work if the MATCH term is present ** and we are doing a scan. */ static int spellfix1FilterForMatch( ................................................................................ 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 = 0; /* 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 ){ ................................................................................ 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); if( pMatchStr3==0 ){ x.rc = SQLITE_NOMEM; goto filter_exit; } }else{ x.pLang = 0; } zPattern = (char*)transliterate(zMatchThis, sqlite3_value_bytes(argv[0])); sqlite3_free(pCur->zPattern); pCur->zPattern = zPattern; if( zPattern==0 ){ x.rc = SQLITE_NOMEM; goto filter_exit; } 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 ); if( zSql==0 ){ x.rc = SQLITE_NOMEM; pStmt = 0; goto filter_exit; } 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.iLang = iLang; x.rc = rc; x.pConfig3 = p->pConfig3; if( x.rc==SQLITE_OK ){ spellfix1RunQuery(&x, zPattern, nPattern); } if( pCur->a ){ qsort(pCur->a, pCur->nRow, sizeof(pCur->a[0]), spellfix1RowCompare); pCur->iTop = iLimit; pCur->iScope = iScope; }else{ x.rc = SQLITE_NOMEM; } filter_exit: sqlite3_finalize(pStmt); editDist3FromStringDelete(pMatchStr3); return x.rc; } /* ** This version of xFilter handles a full-table scan case */ static int spellfix1FilterForFullScan( spellfix1_cursor *pCur, int idxNum, int argc, sqlite3_value **argv ){ int rc; char *zSql; spellfix1_vtab *pVTab = pCur->pVTab; spellfix1ResetCursor(pCur); zSql = sqlite3_mprintf( "SELECT word, rank, NULL, langid, id FROM \"%w\".\"%w_vocab\"", pVTab->zDbName, pVTab->zTableName); if( zSql==0 ) return SQLITE_NOMEM; rc = sqlite3_prepare_v2(pVTab->db, zSql, -1, &pCur->pFullScan, 0); sqlite3_free(zSql); pCur->nRow = pCur->iRow = 0; if( rc==SQLITE_OK ){ rc = sqlite3_step(pCur->pFullScan); if( rc==SQLITE_ROW ){ pCur->iRow = -1; rc = SQLITE_OK; } if( rc==SQLITE_DONE ){ rc = SQLITE_OK; } }else{ pCur->iRow = 0; } return rc; } /* ** Called to "rewind" a cursor back to the beginning so that ** it starts its output over again. Always called at least once ** prior to any spellfix1Column, spellfix1Rowid, or spellfix1Eof call. ................................................................................ /* ** Advance a cursor to its next row of output */ static int spellfix1Next(sqlite3_vtab_cursor *cur){ spellfix1_cursor *pCur = (spellfix1_cursor *)cur; if( pCur->iRow < pCur->nRow ){ if( pCur->pFullScan ){ int rc = sqlite3_step(pCur->pFullScan); if( rc!=SQLITE_ROW ) pCur->iRow = pCur->nRow; }else{ pCur->iRow++; } } return SQLITE_OK; } /* ** Return TRUE if we are at the end-of-file */ static int spellfix1Eof(sqlite3_vtab_cursor *cur){ ................................................................................ spellfix1_cursor *pCur = (spellfix1_cursor *)cur; return pCur->iRow>=pCur->nRow; } /* ** Return columns from the current row. */ static int spellfix1Column( sqlite3_vtab_cursor *cur, sqlite3_context *ctx, int i ){ spellfix1_cursor *pCur = (spellfix1_cursor*)cur; if( pCur->pFullScan ){ if( i<=SPELLFIX_COL_LANGID ){ sqlite3_result_value(ctx, sqlite3_column_value(pCur->pFullScan, i)); }else{ sqlite3_result_null(ctx); } return SQLITE_OK; } 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); ................................................................................ int nWord = strlen(zWord); if( nPattern>0 && pCur->zPattern[nPattern-1]=='*' ){ char *zTranslit; int res; zTranslit = (char *)transliterate((unsigned char *)zWord, nWord); if( !zTranslit ) return SQLITE_NOMEM; res = editdist1(pCur->zPattern, zTranslit, &iMatchlen); sqlite3_free(zTranslit); if( res<0 ) return SQLITE_NOMEM; iMatchlen = translen_to_charlen(zWord, nWord, iMatchlen); }else{ iMatchlen = utf8Charlen(zWord, nWord); } } ................................................................................ } /* ** The rowid. */ static int spellfix1Rowid(sqlite3_vtab_cursor *cur, sqlite_int64 *pRowid){ spellfix1_cursor *pCur = (spellfix1_cursor*)cur; if( pCur->pFullScan ){ *pRowid = sqlite3_column_int64(pCur->pFullScan, 4); }else{ *pRowid = pCur->a[pCur->iRow].iRowid; } return SQLITE_OK; } /* ** The xUpdate() method. */ static int spellfix1Update( ................................................................................ iRank, iLang, zWord, zK1, zK2 ); *pRowid = sqlite3_last_insert_rowid(db); }else{ rowid = sqlite3_value_int64(argv[0]); newRowid = *pRowid = sqlite3_value_int64(argv[1]); spellfix1DbExec(&rc, db, "UPDATE \"%w\".\"%w_vocab\" SET id=%lld, rank=%d, langid=%d," " word=%Q, k1=%Q, k2=%Q WHERE id=%lld", p->zDbName, p->zTableName, newRowid, iRank, iLang, zWord, zK1, zK2, rowid ); } sqlite3_free(zK1); sqlite3_free(zK2); } ................................................................................ spellfix1DbExec(&rc, db, "ALTER TABLE \"%w\".\"%w_vocab\" RENAME TO \"%w_vocab\"", p->zDbName, p->zTableName, zNewName ); if( rc==SQLITE_OK ){ sqlite3_free(p->zTableName); p->zTableName = zNewName; }else{ sqlite3_free(zNewName); } return rc; } /* ** A virtual table module that provides fuzzy search. ................................................................................ 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_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_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 ); } |
Changes to test/spellfix.test.
133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 |
foreach w $vocab { execsql { INSERT INTO t3(word) VALUES($w) } } } {} breakpoint foreach {tn word res} { 1 kos* {kosher 3 kiosk 4 kudo 2 kappa 1 keypad 1} 2 kellj* {killjoy 5 killed 4 killingly 4 kill 4 killer 4} 3 kellj {kill 4 kills 5 killjoy 7 keel 4 killed 6} } { do_execsql_test 1.2.$tn { SELECT word, matchlen FROM t3 WHERE word MATCH $word LIMIT 5 } $res } finish_test |
| | | > |
133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 |
foreach w $vocab { execsql { INSERT INTO t3(word) VALUES($w) } } } {} breakpoint foreach {tn word res} { 1 kos* {kosher 3 kiosk 4 kudo 2 kiss 3 kissed 3} 2 kellj* {killjoy 5 kill 4 killed 4 killer 4 killers 4} 3 kellj {kill 4 kills 5 killjoy 7 keel 4 killed 6} } { do_execsql_test 1.2.$tn { SELECT word, matchlen FROM t3 WHERE word MATCH $word ORDER BY score, word LIMIT 5 } $res } finish_test |