Add a Porter stemmer option to the FTS1 module. (CVS 3452)
FossilOrigin-Name: 936b06aaa8133e83104de87e03dc94e286a31f86
This commit is contained in:
drh
parent
2564ef9781
commit
6da40bcd79
@ -1892,7 +1892,9 @@ static int constructVtab(
|
||||
/* TODO(shess) For now, add new tokenizers as else if clauses. */
|
||||
if( spec->azTokenizer[0]==0 || !strcmp(spec->azTokenizer[0], "simple") ){
|
||||
sqlite3Fts1SimpleTokenizerModule(&m);
|
||||
} else {
|
||||
}else if( strcmp(spec->azTokenizer[0], "porter")==0 ){
|
||||
sqlite3Fts1PorterTokenizerModule(&m);
|
||||
}else{
|
||||
*pzErr = sqlite3_mprintf("unknown tokenizer: %s", spec->azTokenizer[0]);
|
||||
rc = SQLITE_ERROR;
|
||||
goto err;
|
||||
|
||||
645
ext/fts1/fts1_porter.c
Normal file
645
ext/fts1/fts1_porter.c
Normal file
@ -0,0 +1,645 @@
|
||||
/*
|
||||
** 2006 September 30
|
||||
**
|
||||
** 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.
|
||||
**
|
||||
*************************************************************************
|
||||
** Implementation of the full-text-search tokenizer that implements
|
||||
** a Porter stemmer.
|
||||
*/
|
||||
|
||||
/*
|
||||
** The code in this file is only compiled if:
|
||||
**
|
||||
** * The FTS1 module is being built as an extension
|
||||
** (in which case SQLITE_CORE is not defined), or
|
||||
**
|
||||
** * The FTS1 module is being built into the core of
|
||||
** SQLite (in which case SQLITE_ENABLE_FTS1 is defined).
|
||||
*/
|
||||
#if !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS1)
|
||||
|
||||
|
||||
#include <assert.h>
|
||||
#if !defined(__APPLE__)
|
||||
#include <malloc.h>
|
||||
#else
|
||||
#include <stdlib.h>
|
||||
#endif
|
||||
#include <stdio.h>
|
||||
#include <string.h>
|
||||
#include <ctype.h>
|
||||
|
||||
#include "fts1_tokenizer.h"
|
||||
|
||||
/*
|
||||
** Class derived from sqlite3_tokenizer
|
||||
*/
|
||||
typedef struct porter_tokenizer {
|
||||
sqlite3_tokenizer base; /* Base class */
|
||||
} porter_tokenizer;
|
||||
|
||||
/*
|
||||
** Class derived from sqlit3_tokenizer_cursor
|
||||
*/
|
||||
typedef struct porter_tokenizer_cursor {
|
||||
sqlite3_tokenizer_cursor base;
|
||||
const char *zInput; /* input we are tokenizing */
|
||||
int nInput; /* size of the input */
|
||||
int iOffset; /* current position in zInput */
|
||||
int iToken; /* index of next token to be returned */
|
||||
char *zToken; /* storage for current token */
|
||||
int nAllocated; /* space allocated to zToken buffer */
|
||||
} porter_tokenizer_cursor;
|
||||
|
||||
|
||||
/* Forward declaration */
|
||||
static const sqlite3_tokenizer_module porterTokenizerModule;
|
||||
|
||||
|
||||
/*
|
||||
** Create a new tokenizer instance.
|
||||
*/
|
||||
static int porterCreate(
|
||||
int argc, const char * const *argv,
|
||||
sqlite3_tokenizer **ppTokenizer
|
||||
){
|
||||
porter_tokenizer *t;
|
||||
int i;
|
||||
|
||||
for(i=0; i<argc; i++) printf("argv[%d] = %s\n", i, argv[i]);
|
||||
t = (porter_tokenizer *) calloc(sizeof(porter_tokenizer), 1);
|
||||
*ppTokenizer = &t->base;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Destroy a tokenizer
|
||||
*/
|
||||
static int porterDestroy(sqlite3_tokenizer *pTokenizer){
|
||||
free(pTokenizer);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Prepare to begin tokenizing a particular string. The input
|
||||
** string to be tokenized is zInput[0..nInput-1]. A cursor
|
||||
** used to incrementally tokenize this string is returned in
|
||||
** *ppCursor.
|
||||
*/
|
||||
static int porterOpen(
|
||||
sqlite3_tokenizer *pTokenizer, /* The tokenizer */
|
||||
const char *zInput, int nInput, /* String to be tokenized */
|
||||
sqlite3_tokenizer_cursor **ppCursor /* OUT: Tokenization cursor */
|
||||
){
|
||||
porter_tokenizer_cursor *c;
|
||||
|
||||
c = (porter_tokenizer_cursor *) malloc(sizeof(porter_tokenizer_cursor));
|
||||
c->zInput = zInput;
|
||||
if( zInput==0 ){
|
||||
c->nInput = 0;
|
||||
}else if( nInput<0 ){
|
||||
c->nInput = (int)strlen(zInput);
|
||||
}else{
|
||||
c->nInput = nInput;
|
||||
}
|
||||
c->iOffset = 0; /* start tokenizing at the beginning */
|
||||
c->iToken = 0;
|
||||
c->zToken = NULL; /* no space allocated, yet. */
|
||||
c->nAllocated = 0;
|
||||
|
||||
*ppCursor = &c->base;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Close a tokenization cursor previously opened by a call to
|
||||
** porterOpen() above.
|
||||
*/
|
||||
static int porterClose(sqlite3_tokenizer_cursor *pCursor){
|
||||
porter_tokenizer_cursor *c = (porter_tokenizer_cursor *) pCursor;
|
||||
free(c->zToken);
|
||||
free(c);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
/*
|
||||
** Vowel or consonant
|
||||
*/
|
||||
static const char cType[] = {
|
||||
0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0,
|
||||
1, 1, 1, 2, 1
|
||||
};
|
||||
|
||||
/*
|
||||
** isConsonant() and isVowel() determine if their first character in
|
||||
** the string they point to is a consonant or a vowel, according
|
||||
** to Porter ruls.
|
||||
**
|
||||
** A consonate is any letter other than 'a', 'e', 'i', 'o', or 'u'.
|
||||
** 'Y' is a consonant unless it follows another consonant,
|
||||
** in which case it is a vowel.
|
||||
**
|
||||
** In these routine, the letters are in reverse order. So the 'y' rule
|
||||
** is that 'y' is a consonant unless it is followed by another
|
||||
** consonent.
|
||||
*/
|
||||
static int isVowel(const char*);
|
||||
static int isConsonant(const char *z){
|
||||
int j;
|
||||
char x = *z;
|
||||
if( x==0 ) return 0;
|
||||
assert( x>='a' && x<='z' );
|
||||
j = cType[x-'a'];
|
||||
if( j<2 ) return j;
|
||||
return z[1]==0 || isVowel(z + 1);
|
||||
}
|
||||
static int isVowel(const char *z){
|
||||
int j;
|
||||
char x = *z;
|
||||
if( x==0 ) return 0;
|
||||
assert( x>='a' && x<='z' );
|
||||
j = cType[x-'a'];
|
||||
if( j<2 ) return 1-j;
|
||||
return isConsonant(z + 1);
|
||||
}
|
||||
|
||||
/*
|
||||
** Let any sequence of one or more vowels be represented by V and let
|
||||
** C be sequence of one or more consonants. Then every word can be
|
||||
** represented as:
|
||||
**
|
||||
** [C] (VC){m} [V]
|
||||
**
|
||||
** In prose: A word is an optional consonant followed by zero or
|
||||
** vowel-consonant pairs followed by an optional vowel. "m" is the
|
||||
** number of vowel consonant pairs. This routine computes the value
|
||||
** of m for the first i bytes of a word.
|
||||
**
|
||||
** Return true if the m-value for z is 1 or more. In other words,
|
||||
** return true if z contains at least one vowel that is followed
|
||||
** by a consonant.
|
||||
**
|
||||
** In this routine z[] is in reverse order. So we are really looking
|
||||
** for an instance of of a consonant followed by a vowel.
|
||||
*/
|
||||
static int m_gt_0(const char *z){
|
||||
while( isVowel(z) ){ z++; }
|
||||
if( *z==0 ) return 0;
|
||||
while( isConsonant(z) ){ z++; }
|
||||
return *z!=0;
|
||||
}
|
||||
|
||||
/* Like mgt0 above except we are looking for a value of m which is
|
||||
** exactly 1
|
||||
*/
|
||||
static int m_eq_1(const char *z){
|
||||
while( isVowel(z) ){ z++; }
|
||||
if( *z==0 ) return 0;
|
||||
while( isConsonant(z) ){ z++; }
|
||||
if( *z==0 ) return 0;
|
||||
while( isVowel(z) ){ z++; }
|
||||
if( *z==0 ) return 1;
|
||||
while( isConsonant(z) ){ z++; }
|
||||
return *z==0;
|
||||
}
|
||||
|
||||
/* Like mgt0 above except we are looking for a value of m>1 instead
|
||||
** or m>0
|
||||
*/
|
||||
static int m_gt_1(const char *z){
|
||||
while( isVowel(z) ){ z++; }
|
||||
if( *z==0 ) return 0;
|
||||
while( isConsonant(z) ){ z++; }
|
||||
if( *z==0 ) return 0;
|
||||
while( isVowel(z) ){ z++; }
|
||||
if( *z==0 ) return 0;
|
||||
while( isConsonant(z) ){ z++; }
|
||||
return *z!=0;
|
||||
}
|
||||
|
||||
/*
|
||||
** Return TRUE if there is a vowel anywhere within z[0..n-1]
|
||||
*/
|
||||
static int hasVowel(const char *z){
|
||||
while( isConsonant(z) ){ z++; }
|
||||
return *z!=0;
|
||||
}
|
||||
|
||||
/*
|
||||
** Return TRUE if the word ends in a double consonant.
|
||||
**
|
||||
** The text is reversed here. So we are really looking at
|
||||
** the first two characters of z[].
|
||||
*/
|
||||
static int doubleConsonant(const char *z){
|
||||
return isConsonant(z) && z[0]==z[1] && isConsonant(z+1);
|
||||
}
|
||||
|
||||
/*
|
||||
** Return TRUE if the word ends with three letters which
|
||||
** are consonant-vowel-consonent and where the final consonant
|
||||
** is not 'w', 'x', or 'y'.
|
||||
**
|
||||
** The word is reversed here. So we are really checking the
|
||||
** first three letters and the first one cannot be in [wxy].
|
||||
*/
|
||||
static int star_oh(const char *z){
|
||||
return
|
||||
z[0]!=0 && isConsonant(z) &&
|
||||
z[0]!='w' && z[0]!='x' && z[0]!='y' &&
|
||||
z[1]!=0 && isVowel(z+1) &&
|
||||
z[2]!=0 && isConsonant(z+2);
|
||||
}
|
||||
|
||||
/*
|
||||
** If the word ends with zFrom and xCond() is true for the stem
|
||||
** of the word that preceeds the zFrom ending, then change the
|
||||
** ending to zTo.
|
||||
**
|
||||
** The input word *pz and zFrom are both in reverse order. zTo
|
||||
** is in normal order.
|
||||
**
|
||||
** Return TRUE if zFrom matches. Return FALSE if zFrom does not
|
||||
** match. Not that TRUE is returned even if xCond() fails and
|
||||
** no substitution occurs.
|
||||
*/
|
||||
static int stem(
|
||||
char **pz, /* The word being stemmed (Reversed) */
|
||||
const char *zFrom, /* If the ending matches this... (Reversed) */
|
||||
const char *zTo, /* ... change the ending to this (not reversed) */
|
||||
int (*xCond)(const char*) /* Condition that must be true */
|
||||
){
|
||||
char *z = *pz;
|
||||
while( *zFrom && *zFrom==*z ){ z++; zFrom++; }
|
||||
if( *zFrom!=0 ) return 0;
|
||||
if( xCond && !xCond(z) ) return 1;
|
||||
while( *zTo ){
|
||||
*(--z) = *(zTo++);
|
||||
}
|
||||
*pz = z;
|
||||
return 1;
|
||||
}
|
||||
|
||||
/*
|
||||
** This is the fallback stemmer used when the porter stemmer is
|
||||
** inappropriate. The input word is copied into the output with
|
||||
** US-ASCII case folding. If the input word is too long (more
|
||||
** than 20 bytes if it contains no digits or more than 6 bytes if
|
||||
** it contains digits) then word is truncated to 20 or 6 bytes
|
||||
** by taking 10 or 3 bytes from the beginning and end.
|
||||
*/
|
||||
static void copy_stemmer(const char *zIn, int nIn, char *zOut, int *pnOut){
|
||||
int i, mx, j;
|
||||
int hasDigit = 0;
|
||||
for(i=0; i<nIn; i++){
|
||||
int c = zIn[i];
|
||||
if( c>='A' && c<='Z' ){
|
||||
zOut[i] = c - 'A' + 'a';
|
||||
}else{
|
||||
if( c>='0' && c<='9' ) hasDigit = 1;
|
||||
zOut[i] = c;
|
||||
}
|
||||
}
|
||||
mx = hasDigit ? 3 : 10;
|
||||
if( nIn>mx*2 ){
|
||||
for(j=mx, i=nIn-mx; i<nIn; i++, j++){
|
||||
zOut[j] = zOut[i];
|
||||
}
|
||||
i = j;
|
||||
}
|
||||
zOut[i] = 0;
|
||||
*pnOut = i;
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
** Stem the input word zIn[0..nIn-1]. Store the output in zOut.
|
||||
** zOut is at least big enough to hold nIn bytes. Write the actual
|
||||
** size of the output word (exclusive of the '\0' terminator) into *pnOut.
|
||||
**
|
||||
** Any upper-case characters in the US-ASCII character set ([A-Z])
|
||||
** are converted to lower case. Upper-case UTF characters are
|
||||
** unchanged.
|
||||
**
|
||||
** Words that are longer than about 20 bytes are stemmed by retaining
|
||||
** a few bytes from the beginning and the end of the word. If the
|
||||
** word contains digits, 3 bytes are taken from the beginning and
|
||||
** 3 bytes from the end. For long words without digits, 10 bytes
|
||||
** are taken from each end. US-ASCII case folding still applies.
|
||||
**
|
||||
** If the input word contains not digits but does characters not
|
||||
** in [a-zA-Z] then no stemming is attempted and this routine just
|
||||
** copies the input into the input into the output with US-ASCII
|
||||
** case folding.
|
||||
**
|
||||
** Stemming never increases the length of the word. So there is
|
||||
** no chance of overflowing the zOut buffer.
|
||||
*/
|
||||
static void porter_stemmer(const char *zIn, int nIn, char *zOut, int *pnOut){
|
||||
int i, j, c;
|
||||
char zReverse[28];
|
||||
char *z, *z2;
|
||||
if( nIn<3 || nIn>=sizeof(zReverse)-7 ){
|
||||
/* The word is too big or too small for the porter stemmer.
|
||||
** Fallback to the copy stemmer */
|
||||
copy_stemmer(zIn, nIn, zOut, pnOut);
|
||||
return;
|
||||
}
|
||||
for(i=0, j=sizeof(zReverse)-6; i<nIn; i++, j--){
|
||||
c = zIn[i];
|
||||
if( c>='A' && c<='Z' ){
|
||||
zReverse[j] = c + 'a' - 'A';
|
||||
}else if( c>='a' && c<='z' ){
|
||||
zReverse[j] = c;
|
||||
}else{
|
||||
/* The use of a character not in [a-zA-Z] means that we fallback
|
||||
** to the copy stemmer */
|
||||
copy_stemmer(zIn, nIn, zOut, pnOut);
|
||||
return;
|
||||
}
|
||||
}
|
||||
memset(&zReverse[sizeof(zReverse)-5], 0, 5);
|
||||
z = &zReverse[j+1];
|
||||
|
||||
|
||||
/* Step 1a */
|
||||
if( z[0]=='s' ){
|
||||
if(
|
||||
!stem(&z, "sess", "ss", 0) &&
|
||||
!stem(&z, "sei", "i", 0) &&
|
||||
!stem(&z, "ss", "ss", 0)
|
||||
){
|
||||
z++;
|
||||
}
|
||||
}
|
||||
|
||||
/* Step 1b */
|
||||
z2 = z;
|
||||
if( stem(&z, "dee", "ee", m_gt_0) ){
|
||||
/* Do nothing. The work was all in the test */
|
||||
}else if(
|
||||
(stem(&z, "gni", "", hasVowel) || stem(&z, "de", "", hasVowel))
|
||||
&& z!=z2
|
||||
){
|
||||
if( stem(&z, "ta", "ate", 0) ||
|
||||
stem(&z, "lb", "ble", 0) ||
|
||||
stem(&z, "zi", "ize", 0) ){
|
||||
/* Do nothing. The work was all in the test */
|
||||
}else if( doubleConsonant(z) && (*z!='l' && *z!='s' && *z!='z') ){
|
||||
z++;
|
||||
}else if( m_eq_1(z) && star_oh(z) ){
|
||||
*(--z) = 'e';
|
||||
}
|
||||
}
|
||||
|
||||
/* Step 1c */
|
||||
if( z[0]=='y' && hasVowel(z+1) ){
|
||||
z[0] = 'i';
|
||||
}
|
||||
|
||||
/* Step 2 */
|
||||
switch( z[1] ){
|
||||
case 'a':
|
||||
stem(&z, "lanoita", "ate", m_gt_0) ||
|
||||
stem(&z, "lanoit", "tion", m_gt_0);
|
||||
break;
|
||||
case 'c':
|
||||
stem(&z, "icne", "ence", m_gt_0) ||
|
||||
stem(&z, "icna", "ance", m_gt_0);
|
||||
break;
|
||||
case 'e':
|
||||
stem(&z, "rezi", "ize", m_gt_0);
|
||||
break;
|
||||
case 'g':
|
||||
stem(&z, "igol", "log", m_gt_0);
|
||||
break;
|
||||
case 'l':
|
||||
stem(&z, "ilb", "ble", m_gt_0) ||
|
||||
stem(&z, "illa", "al", m_gt_0) ||
|
||||
stem(&z, "iltne", "ent", m_gt_0) ||
|
||||
stem(&z, "ile", "e", m_gt_0) ||
|
||||
stem(&z, "ilsuo", "ous", m_gt_0);
|
||||
break;
|
||||
case 'o':
|
||||
stem(&z, "noitazi", "ize", m_gt_0) ||
|
||||
stem(&z, "noita", "ate", m_gt_0) ||
|
||||
stem(&z, "rota", "ate", m_gt_0);
|
||||
break;
|
||||
case 's':
|
||||
stem(&z, "msila", "al", m_gt_0) ||
|
||||
stem(&z, "ssenevi", "ive", m_gt_0) ||
|
||||
stem(&z, "ssenluf", "ful", m_gt_0) ||
|
||||
stem(&z, "ssensuo", "ous", m_gt_0);
|
||||
break;
|
||||
case 't':
|
||||
stem(&z, "itila", "al", m_gt_0) ||
|
||||
stem(&z, "itivi", "ive", m_gt_0) ||
|
||||
stem(&z, "itilib", "ble", m_gt_0);
|
||||
break;
|
||||
}
|
||||
|
||||
/* Step 3 */
|
||||
switch( z[0] ){
|
||||
case 'e':
|
||||
stem(&z, "etaci", "ic", m_gt_0) ||
|
||||
stem(&z, "evita", "", m_gt_0) ||
|
||||
stem(&z, "ezila", "al", m_gt_0);
|
||||
break;
|
||||
case 'i':
|
||||
stem(&z, "itici", "ic", m_gt_0);
|
||||
break;
|
||||
case 'l':
|
||||
stem(&z, "laci", "ic", m_gt_0) ||
|
||||
stem(&z, "luf", "", m_gt_0);
|
||||
break;
|
||||
case 's':
|
||||
stem(&z, "ssen", "", m_gt_0);
|
||||
break;
|
||||
}
|
||||
|
||||
/* Step 4 */
|
||||
switch( z[1] ){
|
||||
case 'a':
|
||||
if( z[0]=='l' && m_gt_1(z+2) ){
|
||||
z += 2;
|
||||
}
|
||||
break;
|
||||
case 'c':
|
||||
if( z[0]=='e' && z[2]=='n' && (z[3]=='a' || z[3]=='e') && m_gt_1(z+4) ){
|
||||
z += 4;
|
||||
}
|
||||
break;
|
||||
case 'e':
|
||||
if( z[0]=='r' && m_gt_1(z+2) ){
|
||||
z += 2;
|
||||
}
|
||||
break;
|
||||
case 'i':
|
||||
if( z[0]=='c' && m_gt_1(z+2) ){
|
||||
z += 2;
|
||||
}
|
||||
break;
|
||||
case 'l':
|
||||
if( z[0]=='e' && z[2]=='b' && (z[3]=='a' || z[3]=='i') && m_gt_1(z+4) ){
|
||||
z += 4;
|
||||
}
|
||||
break;
|
||||
case 'n':
|
||||
if( z[0]=='t' ){
|
||||
if( z[2]=='a' ){
|
||||
if( m_gt_1(z+3) ){
|
||||
z += 3;
|
||||
}
|
||||
}else if( z[2]=='e' ){
|
||||
stem(&z, "tneme", "", m_gt_1) ||
|
||||
stem(&z, "tnem", "", m_gt_1) ||
|
||||
stem(&z, "tne", "", m_gt_1);
|
||||
}
|
||||
}
|
||||
break;
|
||||
case 'o':
|
||||
if( z[0]=='u' ){
|
||||
if( m_gt_1(z+2) ){
|
||||
z += 2;
|
||||
}
|
||||
}else if( z[3]=='s' || z[3]=='t' ){
|
||||
stem(&z, "noi", "", m_gt_1);
|
||||
}
|
||||
break;
|
||||
case 's':
|
||||
if( z[0]=='m' && z[2]=='i' && m_gt_1(z+3) ){
|
||||
z += 3;
|
||||
}
|
||||
break;
|
||||
case 't':
|
||||
stem(&z, "eta", "", m_gt_1) ||
|
||||
stem(&z, "iti", "", m_gt_1);
|
||||
break;
|
||||
case 'u':
|
||||
if( z[0]=='s' && z[2]=='o' && m_gt_1(z+3) ){
|
||||
z += 3;
|
||||
}
|
||||
break;
|
||||
case 'v':
|
||||
case 'z':
|
||||
if( z[0]=='e' && z[2]=='i' && m_gt_1(z+3) ){
|
||||
z += 3;
|
||||
}
|
||||
break;
|
||||
}
|
||||
|
||||
/* Step 5a */
|
||||
if( z[0]=='e' ){
|
||||
if( m_gt_1(z+1) ){
|
||||
z++;
|
||||
}else if( m_eq_1(z+1) && !star_oh(z+1) ){
|
||||
z++;
|
||||
}
|
||||
}
|
||||
|
||||
/* Step 5b */
|
||||
if( m_gt_1(z) && z[0]=='l' && z[1]=='l' ){
|
||||
z++;
|
||||
}
|
||||
|
||||
/* z[] is now the stemmed word in reverse order. Flip it back
|
||||
** around into forward order and return.
|
||||
*/
|
||||
*pnOut = i = strlen(z);
|
||||
zOut[i] = 0;
|
||||
while( *z ){
|
||||
zOut[--i] = *(z++);
|
||||
}
|
||||
}
|
||||
|
||||
/*
|
||||
** Characters that can be part of a token. We assume any character
|
||||
** whose value is greater than 0x80 (any UTF character) can be
|
||||
** part of a token. In other words, delimiters all must have
|
||||
** values of 0x7f or lower.
|
||||
*/
|
||||
const char isIdChar[] = {
|
||||
/* x0 x1 x2 x3 x4 x5 x6 x7 x8 x9 xA xB xC xD xE xF */
|
||||
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, /* 3x */
|
||||
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 4x */
|
||||
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1, /* 5x */
|
||||
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 6x */
|
||||
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, /* 7x */
|
||||
};
|
||||
#define idChar(C) (((ch=C)&0x80)!=0 || (ch>0x2f && isIdChar[ch-0x30]))
|
||||
#define isDelim(C) (((ch=C)&0x80)==0 && (ch<0x30 || !isIdChar[ch-0x30]))
|
||||
|
||||
/*
|
||||
** Extract the next token from a tokenization cursor. The cursor must
|
||||
** have been opened by a prior call to porterOpen().
|
||||
*/
|
||||
static int porterNext(
|
||||
sqlite3_tokenizer_cursor *pCursor, /* Cursor returned by porterOpen */
|
||||
const char **pzToken, /* OUT: *pzToken is the token text */
|
||||
int *pnBytes, /* OUT: Number of bytes in token */
|
||||
int *piStartOffset, /* OUT: Starting offset of token */
|
||||
int *piEndOffset, /* OUT: Ending offset of token */
|
||||
int *piPosition /* OUT: Position integer of token */
|
||||
){
|
||||
porter_tokenizer_cursor *c = (porter_tokenizer_cursor *) pCursor;
|
||||
const char *z = c->zInput;
|
||||
|
||||
while( c->iOffset<c->nInput ){
|
||||
int iStartOffset, ch;
|
||||
|
||||
/* Scan past delimiter characters */
|
||||
while( c->iOffset<c->nInput && isDelim(z[c->iOffset]) ){
|
||||
c->iOffset++;
|
||||
}
|
||||
|
||||
/* Count non-delimiter characters. */
|
||||
iStartOffset = c->iOffset;
|
||||
while( c->iOffset<c->nInput && !isDelim(z[c->iOffset]) ){
|
||||
c->iOffset++;
|
||||
}
|
||||
|
||||
if( c->iOffset>iStartOffset ){
|
||||
int n = c->iOffset-iStartOffset;
|
||||
if( n>c->nAllocated ){
|
||||
c->nAllocated = n+20;
|
||||
c->zToken = realloc(c->zToken, c->nAllocated);
|
||||
}
|
||||
porter_stemmer(&z[iStartOffset], n, c->zToken, pnBytes);
|
||||
*pzToken = c->zToken;
|
||||
*piStartOffset = iStartOffset;
|
||||
*piEndOffset = c->iOffset;
|
||||
*piPosition = c->iToken++;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
}
|
||||
return SQLITE_DONE;
|
||||
}
|
||||
|
||||
/*
|
||||
** The set of routines that implement the porter-stemmer tokenizer
|
||||
*/
|
||||
static const sqlite3_tokenizer_module porterTokenizerModule = {
|
||||
0,
|
||||
porterCreate,
|
||||
porterDestroy,
|
||||
porterOpen,
|
||||
porterClose,
|
||||
porterNext,
|
||||
};
|
||||
|
||||
/*
|
||||
** Allocate a new porter tokenizer. Return a pointer to the new
|
||||
** tokenizer in *ppModule
|
||||
*/
|
||||
void sqlite3Fts1PorterTokenizerModule(
|
||||
sqlite3_tokenizer_module const**ppModule
|
||||
){
|
||||
*ppModule = &porterTokenizerModule;
|
||||
}
|
||||
|
||||
#endif /* !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS1) */
|
||||
@ -85,5 +85,6 @@ struct sqlite3_tokenizer_cursor {
|
||||
** registration process.
|
||||
*/
|
||||
void sqlite3Fts1SimpleTokenizerModule(sqlite3_tokenizer_module const**ppModule);
|
||||
void sqlite3Fts1PorterTokenizerModule(sqlite3_tokenizer_module const**ppModule);
|
||||
|
||||
#endif /* _FTS1_TOKENIZER_H_ */
|
||||
|
||||
@ -52,6 +52,9 @@ static int isDelim(simple_tokenizer *t, unsigned char c){
|
||||
return c<0x80 && t->delim[c];
|
||||
}
|
||||
|
||||
/*
|
||||
** Create a new tokenizer instance.
|
||||
*/
|
||||
static int simpleCreate(
|
||||
int argc, const char * const *argv,
|
||||
sqlite3_tokenizer **ppTokenizer
|
||||
@ -87,15 +90,24 @@ static int simpleCreate(
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Destroy a tokenizer
|
||||
*/
|
||||
static int simpleDestroy(sqlite3_tokenizer *pTokenizer){
|
||||
free(pTokenizer);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Prepare to begin tokenizing a particular string. The input
|
||||
** string to be tokenized is pInput[0..nBytes-1]. A cursor
|
||||
** used to incrementally tokenize this string is returned in
|
||||
** *ppCursor.
|
||||
*/
|
||||
static int simpleOpen(
|
||||
sqlite3_tokenizer *pTokenizer,
|
||||
const char *pInput, int nBytes,
|
||||
sqlite3_tokenizer_cursor **ppCursor
|
||||
sqlite3_tokenizer *pTokenizer, /* The tokenizer */
|
||||
const char *pInput, int nBytes, /* String to be tokenized */
|
||||
sqlite3_tokenizer_cursor **ppCursor /* OUT: Tokenization cursor */
|
||||
){
|
||||
simple_tokenizer_cursor *c;
|
||||
|
||||
@ -117,6 +129,10 @@ static int simpleOpen(
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Close a tokenization cursor previously opened by a call to
|
||||
** simpleOpen() above.
|
||||
*/
|
||||
static int simpleClose(sqlite3_tokenizer_cursor *pCursor){
|
||||
simple_tokenizer_cursor *c = (simple_tokenizer_cursor *) pCursor;
|
||||
free(c->pToken);
|
||||
@ -124,10 +140,17 @@ static int simpleClose(sqlite3_tokenizer_cursor *pCursor){
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Extract the next token from a tokenization cursor. The cursor must
|
||||
** have been opened by a prior call to simpleOpen().
|
||||
*/
|
||||
static int simpleNext(
|
||||
sqlite3_tokenizer_cursor *pCursor,
|
||||
const char **ppToken, int *pnBytes,
|
||||
int *piStartOffset, int *piEndOffset, int *piPosition
|
||||
sqlite3_tokenizer_cursor *pCursor, /* Cursor returned by simpleOpen */
|
||||
const char **ppToken, /* OUT: *ppToken is the token text */
|
||||
int *pnBytes, /* OUT: Number of bytes in token */
|
||||
int *piStartOffset, /* OUT: Starting offset of token */
|
||||
int *piEndOffset, /* OUT: Ending offset of token */
|
||||
int *piPosition /* OUT: Position integer of token */
|
||||
){
|
||||
simple_tokenizer_cursor *c = (simple_tokenizer_cursor *) pCursor;
|
||||
simple_tokenizer *t = (simple_tokenizer *) pCursor->pTokenizer;
|
||||
@ -150,7 +173,8 @@ static int simpleNext(
|
||||
if( c->iOffset>iStartOffset ){
|
||||
int i, n = c->iOffset-iStartOffset;
|
||||
if( n>c->nTokenAllocated ){
|
||||
c->pToken = realloc(c->pToken, n);
|
||||
c->nTokenAllocated = n+20;
|
||||
c->pToken = realloc(c->pToken, c->nTokenAllocated);
|
||||
}
|
||||
for(i=0; i<n; i++){
|
||||
/* TODO(shess) This needs expansion to handle UTF-8
|
||||
@ -171,6 +195,9 @@ static int simpleNext(
|
||||
return SQLITE_DONE;
|
||||
}
|
||||
|
||||
/*
|
||||
** The set of routines that implement the simple tokenizer
|
||||
*/
|
||||
static const sqlite3_tokenizer_module simpleTokenizerModule = {
|
||||
0,
|
||||
simpleCreate,
|
||||
@ -180,6 +207,10 @@ static const sqlite3_tokenizer_module simpleTokenizerModule = {
|
||||
simpleNext,
|
||||
};
|
||||
|
||||
/*
|
||||
** Allocate a new simple tokenizer. Return a pointer to the new
|
||||
** tokenizer in *ppModule
|
||||
*/
|
||||
void sqlite3Fts1SimpleTokenizerModule(
|
||||
sqlite3_tokenizer_module const**ppModule
|
||||
){
|
||||
|
||||
19
manifest
19
manifest
@ -1,5 +1,5 @@
|
||||
C Make\ssure\smemory\sdoes\snot\sleak\swhen\spatching\sup\scolumn\snames\sso\sthat\nthey\sare\sunique\sin\sa\sjoin\sor\sview.\s\sTickets\s#1952\sand\s#2002.\s(CVS\s3451)
|
||||
D 2006-09-29T14:01:05
|
||||
C Add\sa\sPorter\sstemmer\soption\sto\sthe\sFTS1\smodule.\s(CVS\s3452)
|
||||
D 2006-10-01T18:41:20
|
||||
F Makefile.in cabd42d34340f49260bc2a7668c38eba8d4cfd99
|
||||
F Makefile.linux-gcc 2d8574d1ba75f129aba2019f0b959db380a90935
|
||||
F README 9c4e2d6706bdcc3efdd773ce752a8cdab4f90028
|
||||
@ -21,12 +21,13 @@ F ext/README.txt 913a7bd3f4837ab14d7e063304181787658b14e1
|
||||
F ext/fts1/README.txt 20ac73b006a70bcfd80069bdaf59214b6cf1db5e
|
||||
F ext/fts1/ft_hash.c 3927bd880e65329bdc6f506555b228b28924921b
|
||||
F ext/fts1/ft_hash.h 1a35e654a235c2c662d3ca0dfc3138ad60b8b7d5
|
||||
F ext/fts1/fts1.c bad8872dd51a51bc433a6b2d4d199c939e033cb9
|
||||
F ext/fts1/fts1.c 4f6278a61dd57a2ead405188eb847b3f682850e3
|
||||
F ext/fts1/fts1.h 6060b8f62c1d925ea8356cb1a6598073eb9159a6
|
||||
F ext/fts1/fts1_hash.c 3196cee866edbebb1c0521e21672e6d599965114
|
||||
F ext/fts1/fts1_hash.h 957d378355ed29f672cd5add012ce8b088a5e089
|
||||
F ext/fts1/fts1_tokenizer.h 12c0e7ad83120aff1f86ca848149f96f61da738b
|
||||
F ext/fts1/fts1_tokenizer1.c 3c96905873123f5bf76e01e6adb17177f4f2268d
|
||||
F ext/fts1/fts1_porter.c 18ec3c9f8e76598db135dbcc8692f1e5b0c1186b
|
||||
F ext/fts1/fts1_tokenizer.h fdea722c38a9f82ed921642981234f666e47919c
|
||||
F ext/fts1/fts1_tokenizer1.c 98c2bb9f1feb97294256850bd84baac6799168b8
|
||||
F ext/fts1/fulltext.c d935e600d87bc86b7d64f55c7520ea41d6034c5c
|
||||
F ext/fts1/fulltext.h 08525a47852d1d62a0be81d3fc3fe2d23b094efd
|
||||
F ext/fts1/simple_tokenizer.c 1844d72f7194c3fd3d7e4173053911bf0661b70d
|
||||
@ -193,6 +194,8 @@ F test/format4.test bf3bed3b13c63abfb3cfec232597a319a31d0bcc
|
||||
F test/fts1a.test 46090311f85da51bb33bd5ce84f7948359c6d8d7
|
||||
F test/fts1b.test 5d8a01aefbecc8b7442b36c94c05eb7a845462d5
|
||||
F test/fts1c.test 8790cc74bfc3141772f5cc0252dcdef20832f755
|
||||
F test/fts1d.test a73deace5c18df4a549b12908bade4f05dcf1a2f
|
||||
F test/fts1porter.test 947f842384721533eefe4bb67804f88f9e68f684
|
||||
F test/func.test 0ed54b5aeaad319f68016c033acfebef56f5874a
|
||||
F test/hook.test 7e7645fd9a033f79cce8fdff151e32715e7ec50a
|
||||
F test/in.test 369cb2aa1eab02296b4ec470732fe8c131260b1d
|
||||
@ -399,7 +402,7 @@ F www/tclsqlite.tcl bb0d1357328a42b1993d78573e587c6dcbc964b9
|
||||
F www/vdbe.tcl 87a31ace769f20d3627a64fa1fade7fed47b90d0
|
||||
F www/version3.tcl 890248cf7b70e60c383b0e84d77d5132b3ead42b
|
||||
F www/whentouse.tcl 97e2b5cd296f7d8057e11f44427dea8a4c2db513
|
||||
P 8cdf1d6ae018dfc93f8f0962b2530e31aa0bebff
|
||||
R e0824515a1c2aeb5e0fc636e6263bbc4
|
||||
P fcde639119c309c699ee9dd60ed60fd6e9c8c077
|
||||
R 47795ebdba00ff0307cc55aea533a6e6
|
||||
U drh
|
||||
Z 2aee5ccd57b1e344909056652e85b596
|
||||
Z b4c82aa43208ac6dbc1bc5d18d460b48
|
||||
|
||||
@ -1 +1 @@
|
||||
fcde639119c309c699ee9dd60ed60fd6e9c8c077
|
||||
936b06aaa8133e83104de87e03dc94e286a31f86
|
||||
65
test/fts1d.test
Normal file
65
test/fts1d.test
Normal file
@ -0,0 +1,65 @@
|
||||
# 2006 October 1
|
||||
#
|
||||
# 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 file implements regression tests for SQLite library. The
|
||||
# focus of this script is testing the FTS1 module, and in particular
|
||||
# the Porter stemmer.
|
||||
#
|
||||
# $Id: fts1d.test,v 1.1 2006/10/01 18:41:21 drh Exp $
|
||||
#
|
||||
|
||||
set testdir [file dirname $argv0]
|
||||
source $testdir/tester.tcl
|
||||
|
||||
# If SQLITE_ENABLE_FTS1 is defined, omit this file.
|
||||
ifcapable !fts1 {
|
||||
finish_test
|
||||
return
|
||||
}
|
||||
|
||||
do_test fts1d-1.1 {
|
||||
execsql {
|
||||
CREATE VIRTUAL TABLE t1 USING fts1(content, tokenize porter);
|
||||
INSERT INTO t1(rowid, content) VALUES(1, 'running and jumping');
|
||||
SELECT rowid FROM t1 WHERE content MATCH 'run jump';
|
||||
}
|
||||
} {1}
|
||||
do_test fts1d-1.2 {
|
||||
execsql {
|
||||
SELECT snippet(t1) FROM t1 WHERE t1 MATCH 'run jump';
|
||||
}
|
||||
} {{<b>running</b> and <b>jumping</b>}}
|
||||
do_test fts1d-1.3 {
|
||||
execsql {
|
||||
INSERT INTO t1(rowid, content)
|
||||
VALUES(2, 'abcdefghijklmnopqrstuvwyxz');
|
||||
SELECT rowid, snippet(t1) FROM t1 WHERE t1 MATCH 'abcdefghijqrstuvwyxz'
|
||||
}
|
||||
} {2 <b>abcdefghijklmnopqrstuvwyxz</b>}
|
||||
do_test fts1d-1.4 {
|
||||
execsql {
|
||||
SELECT rowid, snippet(t1) FROM t1 WHERE t1 MATCH 'abcdefghijXXXXqrstuvwyxz'
|
||||
}
|
||||
} {2 <b>abcdefghijklmnopqrstuvwyxz</b>}
|
||||
do_test fts1d-1.5 {
|
||||
execsql {
|
||||
INSERT INTO t1(rowid, content)
|
||||
VALUES(3, 'The value is 123456789');
|
||||
SELECT rowid, snippet(t1) FROM t1 WHERE t1 MATCH '123789'
|
||||
}
|
||||
} {3 {The value is <b>123456789</b>}}
|
||||
do_test fts1d-1.6 {
|
||||
execsql {
|
||||
SELECT rowid, snippet(t1) FROM t1 WHERE t1 MATCH '123000000789'
|
||||
}
|
||||
} {3 {The value is <b>123456789</b>}}
|
||||
|
||||
|
||||
finish_test
|
||||
23583
test/fts1porter.test
Normal file
23583
test/fts1porter.test
Normal file
File diff suppressed because it is too large
Load Diff
Loading…
Reference in New Issue
Block a user