From: drh Date: Sat, 21 Apr 2012 11:33:39 +0000 (+0000) Subject: If terminating interactive input to the command-line shell with ^D, issue X-Git-Tag: version-3.7.12~24 X-Git-Url: http://git.ipfire.org/?a=commitdiff_plain;h=9b8d3572a94969ca6e6aebc8dbe81c6213e4671a;p=thirdparty%2Fsqlite.git If terminating interactive input to the command-line shell with ^D, issue an extra \n to move the cursor to the next line before exiting. FossilOrigin-Name: feff1ef0b8f7b51ae80a9d34380b46a5103bf6cd --- diff --git a/manifest b/manifest index 164cba9eef..fd521769a8 100644 --- a/manifest +++ b/manifest @@ -1,5 +1,5 @@ -C Do\snot\sconsider\sa\sDISTINCT\sclause\sredundant\sunless\sa\ssubset\sof\sthe\sresult-set\sis\scollectively\ssubject\sto\sa\sUNIQUE\sconstraint\sand\sit\scan\sbe\sguaranteed\sthat\sall\scolumns\sof\sthe\ssubset\sare\sNOT\sNULL\s(either\sdue\sto\sNOT\sNULL\sconstraints\sWHERE\sclause\sterms).\sFix\sfor\s[385a5b56b9]. -D 2012-04-20T16:59:24.885 +C If\sterminating\sinteractive\sinput\sto\sthe\scommand-line\sshell\swith\s^D,\sissue\nan\sextra\s\\n\sto\smove\sthe\scursor\sto\sthe\snext\sline\sbefore\sexiting. +D 2012-04-21T11:33:39.769 F Makefile.arm-wince-mingw32ce-gcc d6df77f1f48d690bd73162294bbba7f59507c72f F Makefile.in 2f37e468503dbe79d35c9f6dffcf3fae1ae9ec20 F Makefile.linux-gcc 91d710bdc4998cb015f39edf3cb314ec4f4d7e23 @@ -174,7 +174,7 @@ F src/random.c cd4a67b3953b88019f8cd4ccd81394a8ddfaba50 F src/resolve.c 969ec2bc52db1b068054ecf5ddc74f244102a71d F src/rowset.c f6a49f3e9579428024662f6e2931832511f831a1 F src/select.c d7b9018b7dd2e821183d69477ab55c39b8272335 -F src/shell.c 11185a9a4574f363bd4268a2780d37480ae00040 +F src/shell.c dec1a1896ffa9eaedd6d9907cd43aca4b9d3295d F src/sqlite.h.in 4338f299fc83dada8407358d585c0e240ecb76a3 F src/sqlite3ext.h 6904f4aadf976f95241311fbffb00823075d9477 F src/sqliteInt.h c5e917c4f1453f3972b1fd0c81105dfe4f09cc32 @@ -218,6 +218,7 @@ F src/test_quota.h ee5da2ae7f84d1c8e0e0e2ab33f01d69f10259b5 F src/test_rtree.c aba603c949766c4193f1068b91c787f57274e0d9 F src/test_schema.c 8c06ef9ddb240c7a0fcd31bc221a6a2aade58bf0 F src/test_server.c 2f99eb2837dfa06a4aacf24af24c6affdf66a84f +F src/test_spellfix.c 495535f3eb57acdc384572da570e869bb1834bf4 F src/test_stat.c d7035cfcc0ff1f93c000b621f36524318e004e11 F src/test_superlock.c 2b97936ca127d13962c3605dbc9a4ef269c424cd F src/test_syscall.c a992d8c80ea91fbf21fb2dd570db40e77dd7e6ae @@ -994,7 +995,7 @@ F tool/tostr.awk e75472c2f98dd76e06b8c9c1367f4ab07e122d06 F tool/vdbe-compress.tcl d70ea6d8a19e3571d7ab8c9b75cba86d1173ff0f F tool/warnings-clang.sh a8a0a3babda96dfb1ff51adda3cbbf3dfb7266c2 F tool/warnings.sh fbc018d67fd7395f440c28f33ef0f94420226381 -P 9870e4c4fef10112c987c40cb1b95255a7214202 -R 7b0d4f32973f40703f4feb63496b017b -U dan -Z 293167d92a8f749c1bd20f8cdd573739 +P 7b8548b1872cc1225355ba8311e93dd08d6526e2 +R 6a28d821935f4caf8f00acbe95308ad8 +U drh +Z a4a6f8d23b8334a25a8498dbc291bb1a diff --git a/manifest.uuid b/manifest.uuid index 566056cbe1..7bec9efcd2 100644 --- a/manifest.uuid +++ b/manifest.uuid @@ -1 +1 @@ -7b8548b1872cc1225355ba8311e93dd08d6526e2 \ No newline at end of file +feff1ef0b8f7b51ae80a9d34380b46a5103bf6cd \ No newline at end of file diff --git a/src/shell.c b/src/shell.c index 2607e680de..7fc16ffe12 100644 --- a/src/shell.c +++ b/src/shell.c @@ -2572,7 +2572,9 @@ static int process_input(struct callback_data *p, FILE *in){ free(zLine); zLine = one_input_line(zSql, in); if( zLine==0 ){ - break; /* We have reached EOF */ + /* End of input */ + if( stdin_is_interactive ) printf("\n"); + break; } if( seenInterrupt ){ if( in!=0 ) break; diff --git a/src/test_spellfix.c b/src/test_spellfix.c new file mode 100644 index 0000000000..5a221e0b1b --- /dev/null +++ b/src/test_spellfix.c @@ -0,0 +1,1951 @@ +/* +** 2012 April 10 +** +** The author disclaims copyright to this source code. In place of +** a legal notice, here is a blessing: +** +** 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. +** +** 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_charclass(X). +** Hence: k2 = spellfix1_charclass(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_charclass(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 +# include +# include +# 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 +** 6 'L' Glides: L R +** 7 'M' Nasals: M N +** 8 'W' Letter W at the beginning of a word +** 9 'Y' Letter Y at the beginning of a word. +** 10 '9' A digit: 0 1 2 3 4 5 6 7 8 9 +** 11 ' ' White space +** 12 '?' 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_M 7 +#define CCLASS_W 8 +#define CCLASS_Y 9 +#define CCLASS_DIGIT 10 +#define CCLASS_SPACE 11 +#define CCLASS_OTHER 12 + +/* +** The following table gives the character class for non-initial ASCII +** characters. +*/ +static const unsigned char midClass[] = { + /* x0 x1 x2 x3 x4 x5 x6 x7 x8 x9 xa xb xc xd xe xf */ + /* 0x */ 12, 12, 12, 12, 12, 12, 12, 12, 12, 11, 11, 12, 11, 12, 12, 12, + /* 1x */ 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, + /* 2x */ 11, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, + /* 3x */ 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 12, 12, 12, 12, 12, 12, + /* 4x */ 12, 1, 2, 3, 4, 1, 2, 3, 0, 1, 3, 3, 6, 7, 7, 1, + /* 5x */ 2, 3, 6, 3, 4, 1, 2, 0, 3, 1, 3, 12, 12, 12, 12, 12, + /* 6x */ 12, 1, 2, 3, 4, 1, 2, 3, 0, 1, 3, 3, 6, 7, 7, 1, + /* 7x */ 2, 3, 6, 3, 4, 1, 2, 0, 3, 1, 3, 12, 12, 12, 12, 12, +}; + +/* +** 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[] = { + /* x0 x1 x2 x3 x4 x5 x6 x7 x8 x9 xa xb xc xd xe xf */ + /* 0x */ 12, 12, 12, 12, 12, 12, 12, 12, 12, 11, 11, 12, 11, 12, 12, 12, + /* 1x */ 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, + /* 2x */ 11, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, + /* 3x */ 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 12, 12, 12, 12, 12, 12, + /* 4x */ 12, 1, 2, 3, 4, 1, 2, 3, 5, 1, 3, 3, 6, 7, 7, 1, + /* 5x */ 2, 3, 6, 3, 4, 1, 2, 8, 3, 9, 3, 12, 12, 12, 12, 12, + /* 6x */ 12, 1, 2, 3, 4, 1, 2, 3, 5, 1, 3, 3, 6, 7, 7, 1, + /* 7x */ 2, 3, 6, 3, 4, 1, 2, 8, 3, 9, 3, 12, 12, 12, 12, 12, +}; + +/* +** Mapping from the character class number (0-12) 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[] = ".ABCDHLMWY9 ?"; + +/* +** Generate a string of character classes corresponding to the +** ASCII characters in the input string zIn. If the input is not +** ASCII then the behavior is undefined. +** +** Space to hold the result is obtained from sqlite3_malloc() +** +** Return NULL if memory allocation fails. +*/ +static unsigned char *characterClassString(const unsigned char *zIn, int nIn){ + unsigned char *zOut = sqlite3_malloc( nIn + 1 ); + int i; + int nOut = 0; + char cPrev = 0x77; + const unsigned char *aClass = initClass; + + if( zOut==0 ) return 0; + for(i=0; i='A' && cTo<='Z') || (cTo>='a' && cTo<='z')) ){ + /* 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 */ + return 100; +} + +/* +** Given two strings zA and zB which are pure ASCII, return the cost +** of transforming zA into zB. If zA ends with '*' assume that it is +** a prefix of zB and give only minimal penalty for extra characters +** on the end of zB. +** +** 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 editdist(const char *zA, const char *zB){ + 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 */ + 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 */ + + /* Early out if either input is NULL */ + if( zA==0 || zB==0 ) return -1; + + /* Skip any common prefix */ + while( zA[0] && zA[0]==zB[0] ){ dc = zA[0]; zA++; zB++; } + if( zA[0]==0 && zB[0]==0 ) return 0; + +#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)/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); + cAprev = cA; + } + return res; + } + + /* A is a prefix of B */ + if( zA[0]=='*' && zA[1]==0 ) return 0; + + /* Allocate and initialize the Wagner matrix */ + if( nB<(sizeof(mStack)*4)/(sizeof(mStack[0])*5) ){ + m = mStack; + }else{ + m = toFree = sqlite3_malloc( (nB+1)*5*sizeof(m[0])/4 ); + if( m==0 ) return -3; + } + 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; + m[xB] = m[xB-1] + insertOrDeleteCost(cBprev, cB); + cBprev = cB; + } + cAprev = dc; + for(xA=1; xA<=nA; xA++){ + int lastA = (xA==nA); + cA = zA[xA-1]; + if( cA=='*' && lastA ) break; + d = m[0]; + dc = cx[0]; + m[0] = d + insertOrDeleteCost(cAprev, cA); + cBprev = 0; + for(xB=1; xB<=nB; xB++){ + int totalCost, insCost, delCost, subCost, ncx; + cB = zB[xB-1]; + + /* Cost to insert cB */ + insCost = insertOrDeleteCost(cx[xB-1], cB); + if( lastA ) insCost /= FINAL_INS_COST_DIV; + + /* Cost to delete cA */ + delCost = insertOrDeleteCost(cx[xB], cA); + + /* Cost to substitute cA->cB */ + subCost = substituteCost(cx[xB-1], cA, cB); + + /* Best cost */ + totalCost = insCost + m[xB-1]; + ncx = cB; + if( (delCost + m[xB])nA ){ + res = m[nA]; + for(xB=nA+1; xB<=nB; xB++){ + if( m[xB]=0xc0 ){ + c = sqlite3Utf8Trans1[c-0xc0]; + while( i=xBtm ){ + x = (xTop + xBtm)/2; + if( translit[x].cFrom==c ){ + zOut[nOut++] = translit[x].cTo0; + if( translit[x].cTo1 ){ + zOut[nOut++] = translit[x].cTo1; + /* Add an extra "ch" after the "sh" for Щ and щ */ + if( c==0x0429 || c== 0x0449 ){ + zOut[nOut++] = 'c'; + zOut[nOut++] = 'h'; + } + } + c = 0; + break; + }else if( translit[x].cFrom>c ){ + xTop = x-1; + }else{ + xBtm = x+1; + } + } + if( c ) zOut[nOut++] = '?'; + } + } + zOut[nOut] = 0; + return zOut; +} + +/* +** spellfix1_translit(X) +** +** Convert a string that contains non-ASCII Roman characters into +** pure ASCII. +*/ +static void transliterateSqlFunc( + sqlite3_context *context, + int argc, + sqlite3_value **argv +){ + const unsigned char *zIn = sqlite3_value_text(argv[0]); + int nIn = sqlite3_value_bytes(argv[0]); + unsigned char *zOut = transliterate(zIn, nIn); + if( zOut==0 ){ + sqlite3_result_error_nomem(context); + }else{ + sqlite3_result_text(context, (char*)zOut, -1, sqlite3_free); + } +} + +/* +** spellfix1_scriptcode(X) +** +** Try to determine the dominant script used by the word X and return +** its ISO 15924 numeric code. +** +** The current implementation only understands the following scripts: +** +** 215 (Latin) +** 220 (Cyrillic) +** 200 (Greek) +** +** This routine will return 998 if the input X contains characters from +** two or more of the above scripts or 999 if X contains no characters +** from any of the above scripts. +*/ +static void scriptCodeSqlFunc( + sqlite3_context *context, + int argc, + sqlite3_value **argv +){ + const unsigned char *zIn = sqlite3_value_text(argv[0]); + int nIn = sqlite3_value_bytes(argv[0]); + int c, sz; + int scriptMask = 0; + int res; +# define SCRIPT_LATIN 0x0001 +# define SCRIPT_CYRILLIC 0x0002 +# define SCRIPT_GREEK 0x0004 + + while( nIn>0 ){ + c = utf8Read(zIn, nIn, &sz); + zIn += sz; + nIn -= sz; + if( c<0x02af ){ + scriptMask |= SCRIPT_LATIN; + }else if( c>=0x0400 && c<=0x04ff ){ + scriptMask |= SCRIPT_CYRILLIC; + }else if( c>=0x0386 && c<=0x03ce ){ + scriptMask |= SCRIPT_GREEK; + } + } + switch( scriptMask ){ + case 0: res = 999; break; + case SCRIPT_LATIN: res = 215; break; + case SCRIPT_CYRILLIC: res = 220; break; + case SCRIPT_GREEK: res = 200; break; + default: res = 998; break; + } + sqlite3_result_int(context, res); +} + +/***************************************************************************** +** Fuzzy-search virtual table +*****************************************************************************/ + +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 */ +}; + +/* 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 */ + } *a; +}; + +/* +** Construct one or more SQL statements from the format string given +** and then evaluate those statements. The success code is written +** into *pRc. +** +** If *pRc is initially non-zero then this routine is a no-op. +*/ +static void spellfix1DbExec( + int *pRc, /* Success code */ + sqlite3 *db, /* Database in which to run SQL */ + const char *zFormat, /* Format string for SQL */ + ... /* Arguments to the format string */ +){ + va_list ap; + char *zSql; + if( *pRc ) return; + va_start(ap, zFormat); + zSql = sqlite3_vmprintf(zFormat, ap); + va_end(ap); + if( zSql==0 ){ + *pRc = SQLITE_NOMEM; + }else{ + *pRc = sqlite3_exec(db, zSql, 0, 0, 0); + sqlite3_free(zSql); + } +} + +/* +** xDisconnect/xDestroy method for the fuzzy-search module. +*/ +static int spellfix1Uninit(int isDestroy, sqlite3_vtab *pVTab){ + spellfix1_vtab *p = (spellfix1_vtab*)pVTab; + int rc = SQLITE_OK; + 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 +** argv[3].. -> optional arguments (currently ignored) +*/ +static int spellfix1Init( + int isCreate, + sqlite3 *db, + void *pAux, + int argc, const char *const*argv, + sqlite3_vtab **ppVTab, + char **pzErr +){ + 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{ + 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,top HIDDEN,scope HIDDEN,srchcnt HIDDEN," + "soundslike HIDDEN)" + ); + } + 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 + ); + } + } + } + + *ppVTab = (sqlite3_vtab *)pNew; + return rc; +} + +/* +** The xConnect and xCreate methods +*/ +static int spellfix1Connect( + sqlite3 *db, + void *pAux, + int argc, const char *const*argv, + sqlite3_vtab **ppVTab, + char **pzErr +){ + return spellfix1Init(0, db, pAux, argc, argv, ppVTab, pzErr); +} +static int spellfix1Create( + sqlite3 *db, + void *pAux, + int argc, const char *const*argv, + sqlite3_vtab **ppVTab, + char **pzErr +){ + return spellfix1Init(1, db, pAux, argc, argv, ppVTab, pzErr); +} + +/* +** Reset a cursor so that it contains zero rows of content but holds +** space for N rows. +*/ +static void spellfix1ResetCursor(spellfix1_cursor *pCur, int N){ + int i; + for(i=0; inRow; i++){ + sqlite3_free(pCur->a[i].zWord); + } + pCur->a = sqlite3_realloc(pCur->a, sizeof(pCur->a[0])*N); + pCur->nAlloc = N; + pCur->nRow = 0; + pCur->iRow = 0; + pCur->nSearch = 0; +} + +/* +** Close a fuzzy-search cursor. +*/ +static int spellfix1Close(sqlite3_vtab_cursor *cur){ + spellfix1_cursor *pCur = (spellfix1_cursor *)cur; + spellfix1ResetCursor(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 +** +** 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 +** +** 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 i; + const struct sqlite3_index_constraint *pConstraint; + pConstraint = pIdxInfo->aConstraint; + for(i=0; inConstraint; i++, pConstraint++){ + if( pConstraint->usable==0 ) continue; + + /* Terms of the form: word MATCH $str */ + if( (iPlan & 1)==0 + && pConstraint->iColumn==0 + && 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==3 + && pConstraint->op==SQLITE_INDEX_CONSTRAINT_EQ + ){ + iPlan |= 2; + iLangTerm = i; + } + + /* Terms of the form: top = $top */ + if( (iPlan & 4)==0 + && pConstraint->iColumn==5 + && pConstraint->op==SQLITE_INDEX_CONSTRAINT_EQ + ){ + iPlan |= 4; + iTopTerm = i; + } + + /* Terms of the form: scope = $scope */ + if( (iPlan & 8)==0 + && pConstraint->iColumn==6 + && pConstraint->op==SQLITE_INDEX_CONSTRAINT_EQ + ){ + iPlan |= 8; + iScopeTerm = 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; + } + if( iPlan&4 ){ + 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; +} + +/* +** Open a new fuzzy-search cursor. +*/ +static int spellfix1Open(sqlite3_vtab *pVTab, sqlite3_vtab_cursor **ppCursor){ + spellfix1_vtab *p = (spellfix1_vtab*)pVTab; + spellfix1_cursor *pCur; + pCur = sqlite3_malloc( sizeof(*pCur) ); + if( pCur==0 ) return SQLITE_NOMEM; + memset(pCur, 0, sizeof(*pCur)); + pCur->pVTab = p; + *ppCursor = &pCur->base; + return SQLITE_OK; +} + +/* +** Adjust a distance measurement by the words rank in order to show +** preference to common words. +*/ +static int spellfix1Score(int iDistance, int iRank){ + int iLog2; + for(iLog2=0; iRank>0; iLog2++, iRank>>=1){} + return iDistance + 32 - iLog2; +} + +/* +** Compare two spellfix1_row objects for sorting purposes in qsort() such +** 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 +){ + const unsigned char *zPatternIn; + char *zPattern; + int nPattern; + char *zClass; + int nClass; + int iLimit = 20; + int iScope = 4; + int iLang = 0; + char *zSql; + int rc; + sqlite3_stmt *pStmt; + int idx = 1; + spellfix1_vtab *p = pCur->pVTab; + + 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 ){ + iScope = sqlite3_value_int(argv[idx++]); + if( iScope<1 ) iScope = 1; + } + spellfix1ResetCursor(pCur, iLimit); + zPatternIn = sqlite3_value_text(argv[0]); + if( zPatternIn==0 ) return SQLITE_OK; + zPattern = (char*)transliterate(zPatternIn, sqlite3_value_bytes(argv[0])); + if( zPattern==0 ) return SQLITE_NOMEM; + nPattern = strlen(zPattern); + if( zPattern[nPattern-1]=='*' ) nPattern--; + if( nPatterniScope ){ + zClass[iScope] = 0; + nClass = iScope; + } + zSql = sqlite3_mprintf( + "SELECT id, word, rank, k1" + " FROM \"%w\".\"%w_vocab\"" + " WHERE langid=%d AND k2 GLOB '%q*'", + p->zDbName, p->zTableName, iLang, zClass + ); + rc = sqlite3_prepare_v2(p->db, zSql, -1, &pStmt, 0); + sqlite3_free(zSql); + if( rc==SQLITE_OK ){ + const char *zK1; + int iDist; + int iRank; + int iScore; + int iWorst = 999999999; + int idx; + int idxWorst; + int i; + + while( sqlite3_step(pStmt)==SQLITE_ROW ){ + zK1 = (const char*)sqlite3_column_text(pStmt, 3); + if( zK1==0 ) continue; + pCur->nSearch++; + iRank = sqlite3_column_int(pStmt, 2); + iDist = editdist(zPattern, zK1); + iScore = spellfix1Score(iDist,iRank); + if( pCur->nRownAlloc ){ + idx = pCur->nRow; + }else if( iScorea[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; + if( pCur->nRownAlloc ) pCur->nRow++; + if( pCur->nRow==pCur->nAlloc ){ + iWorst = pCur->a[0].iScore; + idxWorst = 0; + for(i=1; inRow; i++){ + iScore = pCur->a[i].iScore; + if( iWorsta, pCur->nRow, sizeof(pCur->a[0]), spellfix1RowCompare); + pCur->iTop = iLimit; + pCur->iScope = iScope; + sqlite3_finalize(pStmt); + sqlite3_free(zPattern); + sqlite3_free(zClass); + return SQLITE_OK; +} + +/* +** 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, 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. +*/ +static int spellfix1Filter( + sqlite3_vtab_cursor *cur, + int idxNum, const char *idxStr, + int argc, sqlite3_value **argv +){ + spellfix1_cursor *pCur = (spellfix1_cursor *)cur; + int rc; + if( idxNum & 1 ){ + rc = spellfix1FilterForMatch(pCur, idxNum, argc, argv); + }else{ + rc = spellfix1FilterForFullScan(pCur, idxNum, argc, argv); + } + return rc; +} + + +/* +** 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 0: { + sqlite3_result_text(ctx, pCur->a[pCur->iRow].zWord, -1, SQLITE_STATIC); + break; + } + case 1: { + sqlite3_result_int(ctx, pCur->a[pCur->iRow].iRank); + break; + } + case 2: { + sqlite3_result_int(ctx, pCur->a[pCur->iRow].iDistance); + break; + } + case 3: { + sqlite3_result_int(ctx, pCur->iLang); + break; + } + case 4: { + sqlite3_result_int(ctx, pCur->a[pCur->iRow].iScore); + break; + } + case 5: { + sqlite3_result_int(ctx, pCur->iTop); + break; + } + case 6: { + sqlite3_result_int(ctx, pCur->iScope); + break; + } + case 7: { + sqlite3_result_int(ctx, pCur->nSearch); + break; + } + default: { + sqlite3_result_null(ctx); + break; + } + } + return SQLITE_OK; +} + +/* +** 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( + sqlite3_vtab *pVTab, + int argc, + sqlite3_value **argv, + sqlite_int64 *pRowid +){ + int rc = SQLITE_OK; + sqlite3_int64 rowid, newRowid; + spellfix1_vtab *p = (spellfix1_vtab*)pVTab; + sqlite3 *db = p->db; + + 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[2]); + int nWord = sqlite3_value_bytes(argv[2]); + int iLang = sqlite3_value_int(argv[5]); + int iRank = sqlite3_value_int(argv[3]); + const unsigned char *zSoundslike = sqlite3_value_text(argv[10]); + int nSoundslike = sqlite3_value_bytes(argv[10]); + char *zK1, *zK2; + int i; + char c; + + if( zWord==0 ){ + pVTab->zErrMsg = sqlite3_mprintf("%w.word may not be NULL", + p->zTableName); + return SQLITE_CONSTRAINT; + } + 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*)characterClassString((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) " + "VALUES(%d,%d,%Q,%Q,%Q)", + p->zDbName, p->zTableName, + 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); + } + return rc; +} + +/* +** Rename the spellfix1 table. +*/ +static int spellfix1Rename(sqlite3_vtab *pVTab, const char *zNew){ + spellfix1_vtab *p = (spellfix1_vtab*)pVTab; + sqlite3 *db = p->db; + int rc = SQLITE_OK; + char *zNewName = sqlite3_mprintf("%s", zNew); + if( zNewName==0 ){ + return SQLITE_NOMEM; + } + 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 sqlite3_module spellfix1Module = { + 0, /* iVersion */ + spellfix1Create, /* xCreate - handle CREATE VIRTUAL TABLE */ + spellfix1Connect, /* xConnect - reconnected to an existing table */ + spellfix1BestIndex, /* xBestIndex - figure out how to do a query */ + spellfix1Disconnect, /* xDisconnect - close a connection */ + spellfix1Destroy, /* xDestroy - handle DROP TABLE */ + spellfix1Open, /* xOpen - open a cursor */ + spellfix1Close, /* xClose - close a cursor */ + spellfix1Filter, /* xFilter - configure scan constraints */ + spellfix1Next, /* xNext - advance a cursor */ + spellfix1Eof, /* xEof - check for end of scan */ + spellfix1Column, /* xColumn - read data */ + spellfix1Rowid, /* xRowid - read data */ + spellfix1Update, /* xUpdate */ + 0, /* xBegin */ + 0, /* xSync */ + 0, /* xCommit */ + 0, /* xRollback */ + 0, /* xFindMethod */ + spellfix1Rename, /* xRename */ +}; + +/* +** Register the various functions and the virtual table. +*/ +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_charclass", 1, SQLITE_UTF8, 0, + characterClassSqlFunc, 0, 0); + nErr += sqlite3_create_function(db, "spellfix1_scriptcode", 1, SQLITE_UTF8, 0, + scriptCodeSqlFunc, 0, 0); + nErr += sqlite3_create_module(db, "spellfix1", &spellfix1Module, 0); + + /* Verify sanity of the translit[] table */ + for(i=0; i