sqlite/ext/fts1/fts1_tokenizer1.c
drh 6da40bcd79 Add a Porter stemmer option to the FTS1 module. (CVS 3452)
FossilOrigin-Name: 936b06aaa8133e83104de87e03dc94e286a31f86
2006-10-01 18:41:19 +00:00

221 lines
6.0 KiB
C

/*
** The author disclaims copyright to this source code.
**
*************************************************************************
** Implementation of the "simple" full-text-search tokenizer.
*/
/*
** 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"
typedef struct simple_tokenizer {
sqlite3_tokenizer base;
char delim[128]; /* flag ASCII delimiters */
} simple_tokenizer;
typedef struct simple_tokenizer_cursor {
sqlite3_tokenizer_cursor base;
const char *pInput; /* input we are tokenizing */
int nBytes; /* size of the input */
int iOffset; /* current position in pInput */
int iToken; /* index of next token to be returned */
char *pToken; /* storage for current token */
int nTokenAllocated; /* space allocated to zToken buffer */
} simple_tokenizer_cursor;
/* Forward declaration */
static const sqlite3_tokenizer_module simpleTokenizerModule;
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
){
simple_tokenizer *t;
t = (simple_tokenizer *) calloc(sizeof(simple_tokenizer), 1);
/* TODO(shess) Delimiters need to remain the same from run to run,
** else we need to reindex. One solution would be a meta-table to
** track such information in the database, then we'd only want this
** information on the initial create.
*/
if( argc>1 ){
int i, n = strlen(argv[1]);
for(i=0; i<n; i++){
unsigned char ch = argv[1][i];
/* We explicitly don't support UTF-8 delimiters for now. */
if( ch>=0x80 ){
free(t);
return SQLITE_ERROR;
}
t->delim[ch] = 1;
}
} else {
/* Mark non-alphanumeric ASCII characters as delimiters */
int i;
for(i=1; i<0x80; i++){
t->delim[i] = !isalnum(i);
}
}
*ppTokenizer = &t->base;
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, /* The tokenizer */
const char *pInput, int nBytes, /* String to be tokenized */
sqlite3_tokenizer_cursor **ppCursor /* OUT: Tokenization cursor */
){
simple_tokenizer_cursor *c;
c = (simple_tokenizer_cursor *) malloc(sizeof(simple_tokenizer_cursor));
c->pInput = pInput;
if( pInput==0 ){
c->nBytes = 0;
}else if( nBytes<0 ){
c->nBytes = (int)strlen(pInput);
}else{
c->nBytes = nBytes;
}
c->iOffset = 0; /* start tokenizing at the beginning */
c->iToken = 0;
c->pToken = NULL; /* no space allocated, yet. */
c->nTokenAllocated = 0;
*ppCursor = &c->base;
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);
free(c);
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, /* 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;
unsigned char *p = (unsigned char *)c->pInput;
while( c->iOffset<c->nBytes ){
int iStartOffset;
/* Scan past delimiter characters */
while( c->iOffset<c->nBytes && isDelim(t, p[c->iOffset]) ){
c->iOffset++;
}
/* Count non-delimiter characters. */
iStartOffset = c->iOffset;
while( c->iOffset<c->nBytes && !isDelim(t, p[c->iOffset]) ){
c->iOffset++;
}
if( c->iOffset>iStartOffset ){
int i, n = c->iOffset-iStartOffset;
if( n>c->nTokenAllocated ){
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
** case-insensitivity.
*/
unsigned char ch = p[iStartOffset+i];
c->pToken[i] = ch<0x80 ? tolower(ch) : ch;
}
*ppToken = c->pToken;
*pnBytes = n;
*piStartOffset = iStartOffset;
*piEndOffset = c->iOffset;
*piPosition = c->iToken++;
return SQLITE_OK;
}
}
return SQLITE_DONE;
}
/*
** The set of routines that implement the simple tokenizer
*/
static const sqlite3_tokenizer_module simpleTokenizerModule = {
0,
simpleCreate,
simpleDestroy,
simpleOpen,
simpleClose,
simpleNext,
};
/*
** Allocate a new simple tokenizer. Return a pointer to the new
** tokenizer in *ppModule
*/
void sqlite3Fts1SimpleTokenizerModule(
sqlite3_tokenizer_module const**ppModule
){
*ppModule = &simpleTokenizerModule;
}
#endif /* !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS1) */