SQLite

  int (*xColumnCount)(Fts5Context*);
  int (*xRowCount)(Fts5Context*, sqlite3_int64 *pnRow);
  int (*xColumnTotalSize)(Fts5Context*, int iCol, sqlite3_int64 *pnToken);

  int (*xTokenize)(Fts5Context*, 
    const char *pText, int nText, /* Text to tokenize */
    void *pCtx,                   /* Context passed to xToken() */
    int (*xToken)(void*, int, const char*, int, int, int)       /* Callback */
  );

  int (*xPhraseCount)(Fts5Context*);
  int (*xPhraseSize)(Fts5Context*, int iPhrase);

  int (*xInstCount)(Fts5Context*, int *pnInst);
  int (*xInst)(Fts5Context*, int iIdx, int *piPhrase, int *piCol, int *piOff);

  void (*xDelete)(Fts5Tokenizer*);
  int (*xTokenize)(Fts5Tokenizer*, 
      void *pCtx,
      int flags,
      const char *pText, int nText, 
      int (*xToken)(
        void *pCtx,         /* Copy of 2nd argument to xTokenize() */
        int tflags,         /* Mask of FTS5_TOKEN_* flags */
        const char *pToken, /* Pointer to buffer containing token */
        int nToken,         /* Size of token in bytes */
        int iStart,         /* Byte offset of token within input text */
        int iEnd            /* Byte offset of end of token within input text */

      )
  );
};

/* Flags that may be passed as the third argument to xTokenize() */
#define FTS5_TOKENIZE_QUERY     0x0001
#define FTS5_TOKENIZE_PREFIX    0x0002
#define FTS5_TOKENIZE_DOCUMENT  0x0004
#define FTS5_TOKENIZE_AUX       0x0008

/* Flags that may be passed by the tokenizer implementation back to FTS5
** as the third argument to the supplied xToken callback. */
#define FTS5_TOKEN_COLOCATED    0x0001      /* Same position as prev. token */

/*
** END OF CUSTOM TOKENIZERS
*************************************************************************/

/*************************************************************************
** FTS5 EXTENSION REGISTRATION API
*/

int sqlite3Fts5ConfigDeclareVtab(Fts5Config *pConfig);

int sqlite3Fts5Tokenize(
  Fts5Config *pConfig,            /* FTS5 Configuration object */
  int flags,                      /* FTS5_TOKENIZE_* flags */
  const char *pText, int nText,   /* Text to tokenize */
  void *pCtx,                     /* Context passed to xToken() */
  int (*xToken)(void*, int, const char*, int, int, int)    /* Callback */
);

void sqlite3Fts5Dequote(char *z);

/* Load the contents of the %_config table */
int sqlite3Fts5ConfigLoad(Fts5Config*, int);

}

/*
** Tokenizer callback used by implementation of highlight() function.
*/
static int fts5HighlightCb(
  void *pContext,                 /* Pointer to HighlightContext object */
  int tflags,                     /* Mask of FTS5_TOKEN_* flags */
  const char *pToken,             /* Buffer containing token */
  int nToken,                     /* Size of token in bytes */
  int iStartOff,                  /* Start offset of token */
  int iEndOff                     /* End offset of token */

){
  HighlightContext *p = (HighlightContext*)pContext;
  int rc = SQLITE_OK;
  int iPos;

  if( tflags & FTS5_TOKEN_COLOCATED ) return SQLITE_OK;
  iPos = p->iPos++;

  if( p->iRangeEnd>0 ){
    if( iPos<p->iRangeStart || iPos>p->iRangeEnd ) return SQLITE_OK;
    if( p->iRangeStart && iPos==p->iRangeStart ) p->iOff = iStartOff;
  }

  if( iPos==p->iter.iStart ){

** to be an SQLite error code and returned to the caller.
*/
int sqlite3Fts5Tokenize(
  Fts5Config *pConfig,            /* FTS5 Configuration object */
  int flags,                      /* FTS5_TOKENIZE_* flags */
  const char *pText, int nText,   /* Text to tokenize */
  void *pCtx,                     /* Context passed to xToken() */
  int (*xToken)(void*, int, const char*, int, int, int)    /* Callback */
){
  if( pText==0 ) return SQLITE_OK;
  return pConfig->pTokApi->xTokenize(
      pConfig->pTok, pCtx, flags, pText, nText, xToken
  );
}

};

/*
** Callback for tokenizing terms used by ParseTerm().
*/
static int fts5ParseTokenize(
  void *pContext,                 /* Pointer to Fts5InsertCtx object */
  int tflags,                     /* Mask of FTS5_TOKEN_* flags */
  const char *pToken,             /* Buffer containing token */
  int nToken,                     /* Size of token in bytes */
  int iStart,                     /* Start offset of token */
  int iEnd                        /* End offset of token */

){
  int rc = SQLITE_OK;
  const int SZALLOC = 8;
  TokenCtx *pCtx = (TokenCtx*)pContext;
  Fts5ExprPhrase *pPhrase = pCtx->pPhrase;
  Fts5ExprTerm *pTerm;

  if( tflags & FTS5_TOKEN_COLOCATED ) return rc;

  if( pPhrase==0 || (pPhrase->nTerm % SZALLOC)==0 ){
    Fts5ExprPhrase *pNew;
    int nNew = SZALLOC + (pPhrase ? pPhrase->nTerm : 0);

    pNew = (Fts5ExprPhrase*)sqlite3_realloc(pPhrase, 
        sizeof(Fts5ExprPhrase) + sizeof(Fts5ExprTerm) * nNew

  return sqlite3Fts5StorageRowCount(pTab->pStorage, pnRow);
}

static int fts5ApiTokenize(
  Fts5Context *pCtx, 
  const char *pText, int nText, 
  void *pUserData,
  int (*xToken)(void*, int, const char*, int, int, int)
){
  Fts5Cursor *pCsr = (Fts5Cursor*)pCtx;
  Fts5Table *pTab = (Fts5Table*)(pCsr->base.pVtab);
  return sqlite3Fts5Tokenize(
      pTab->pConfig, FTS5_TOKENIZE_AUX, pText, nText, pUserData, xToken
  );
}

    }
  }
  return rc;
}

static int fts5ColumnSizeCb(
  void *pContext,                 /* Pointer to int */
  int tflags,
  const char *pToken,             /* Buffer containing token */
  int nToken,                     /* Size of token in bytes */
  int iStart,                     /* Start offset of token */
  int iEnd                        /* End offset of token */

){
  int *pCnt = (int*)pContext;
  if( (tflags & FTS5_TOKEN_COLOCATED)==0 ){
    (*pCnt)++;
  }
  return SQLITE_OK;
}

static int fts5ApiColumnSize(Fts5Context *pCtx, int iCol, int *pnToken){
  Fts5Cursor *pCsr = (Fts5Cursor*)pCtx;
  Fts5Table *pTab = (Fts5Table*)(pCsr->base.pVtab);
  Fts5Config *pConfig = pTab->pConfig;

};

/*
** Tokenization callback used when inserting tokens into the FTS index.
*/
static int fts5StorageInsertCallback(
  void *pContext,                 /* Pointer to Fts5InsertCtx object */
  int tflags,
  const char *pToken,             /* Buffer containing token */
  int nToken,                     /* Size of token in bytes */
  int iStart,                     /* Start offset of token */
  int iEnd                        /* End offset of token */

){
  Fts5InsertCtx *pCtx = (Fts5InsertCtx*)pContext;
  Fts5Index *pIdx = pCtx->pStorage->pIndex;
  if( (tflags & FTS5_TOKEN_COLOCATED)==0 ){
    pCtx->szCol++;
  }
  return sqlite3Fts5IndexWrite(pIdx, pCtx->iCol, pCtx->szCol-1, pToken, nToken);
}

/*
** If a row with rowid iDel is present in the %_content table, add the
** delete-markers to the FTS index necessary to delete it. Do not actually
** remove the %_content row at this time though.
*/

};

/*
** Tokenization callback used by integrity check.
*/
static int fts5StorageIntegrityCallback(
  void *pContext,                 /* Pointer to Fts5InsertCtx object */
  int tflags,
  const char *pToken,             /* Buffer containing token */
  int nToken,                     /* Size of token in bytes */
  int iStart,                     /* Start offset of token */
  int iEnd                        /* End offset of token */

){
  Fts5IntegrityCtx *pCtx = (Fts5IntegrityCtx*)pContext;

  if( (tflags & FTS5_TOKEN_COLOCATED)==0 ){
    pCtx->szCol++;
  }
  pCtx->cksum ^= sqlite3Fts5IndexCksum(
      pCtx->pConfig, pCtx->iRowid, pCtx->iCol, pCtx->szCol-1, pToken, nToken
  );
  return SQLITE_OK;
}

/*
** Check that the contents of the FTS index match that of the %_content
** table. Return SQLITE_OK if they do, or SQLITE_CORRUPT if not. Return

typedef struct F5tAuxData F5tAuxData;
struct F5tAuxData {
  Tcl_Obj *pObj;
};

static int xTokenizeCb(
  void *pCtx, 
  int tflags,
  const char *zToken, int nToken, 
  int iStart, int iEnd
){
  F5tFunction *p = (F5tFunction*)pCtx;
  Tcl_Obj *pEval = Tcl_DuplicateObj(p->pScript);
  int rc;

  Tcl_IncrRefCount(pEval);
  Tcl_ListObjAppendElement(p->interp, pEval, Tcl_NewStringObj(zToken, nToken));

  Tcl_Obj *pRet;
  int bSubst;
  const char *zInput;
};

static int xTokenizeCb2(
  void *pCtx, 
  int tflags,
  const char *zToken, int nToken, 
  int iStart, int iEnd
){
  F5tTokenizeCtx *p = (F5tTokenizeCtx*)pCtx;
  if( p->bSubst ){
    Tcl_ListObjAppendElement(0, p->pRet, Tcl_NewStringObj(zToken, nToken));
    Tcl_ListObjAppendElement(
        0, p->pRet, Tcl_NewStringObj(&p->zInput[iStart], iEnd-iStart)
    );

typedef struct F5tTokenizerContext F5tTokenizerContext;
typedef struct F5tTokenizerCb F5tTokenizerCb;
typedef struct F5tTokenizerModule F5tTokenizerModule;
typedef struct F5tTokenizerModule F5tTokenizerInstance;

struct F5tTokenizerContext {
  void *pCtx;
  int (*xToken)(void*, int, const char*, int, int, int);
};

struct F5tTokenizerModule {
  Tcl_Interp *interp;
  Tcl_Obj *pScript;
  F5tTokenizerContext *pContext;
};

}

static int f5tTokenizerTokenize(
  Fts5Tokenizer *p, 
  void *pCtx,
  int flags,
  const char *pText, int nText, 
  int (*xToken)(void*, int, const char*, int, int, int)
){
  F5tTokenizerInstance *pInst = (F5tTokenizerInstance*)p;
  void *pOldCtx;
  int (*xOldToken)(void*, int, const char*, int, int, int);
  Tcl_Obj *pEval;
  int rc;

  pOldCtx = pInst->pContext->pCtx;
  xOldToken = pInst->pContext->xToken;

  pEval = Tcl_DuplicateObj(pInst->pScript);

  zToken = Tcl_GetStringFromObj(objv[1], &nToken);
  if( Tcl_GetIntFromObj(interp, objv[2], &iStart) 
   || Tcl_GetIntFromObj(interp, objv[3], &iEnd) 
  ){
    return TCL_ERROR;
  }

  rc = p->xToken(p->pCtx, 0, zToken, nToken, iStart, iEnd);
  Tcl_SetResult(interp, (char*)sqlite3ErrName(rc), TCL_VOLATILE);
  return TCL_OK;
}

static void f5tDelTokenizer(void *pCtx){
  F5tTokenizerModule *pMod = (F5tTokenizerModule*)pCtx;
  Tcl_DecrRefCount(pMod->pScript);

** Tokenize some text using the ascii tokenizer.
*/
static int fts5AsciiTokenize(
  Fts5Tokenizer *pTokenizer,
  void *pCtx,
  int flags,
  const char *pText, int nText,
  int (*xToken)(void*, int, const char*, int nToken, int iStart, int iEnd)
){
  AsciiTokenizer *p = (AsciiTokenizer*)pTokenizer;
  int rc = SQLITE_OK;
  int ie;
  int is = 0;


  char aFold[64];
  int nFold = sizeof(aFold);
  char *pFold = aFold;
  unsigned char *a = p->aTokenChar;

  while( is<nText && rc==SQLITE_OK ){

        break;
      }
      nFold = nByte*2;
    }
    asciiFold(pFold, &pText[is], nByte);

    /* Invoke the token callback */
    rc = xToken(pCtx, 0, pFold, nByte, is, ie);
    is = ie+1;
  }
  
  if( pFold!=aFold ) sqlite3_free(pFold);
  if( rc==SQLITE_DONE ) rc = SQLITE_OK;
  return rc;
}

}

static int fts5UnicodeTokenize(
  Fts5Tokenizer *pTokenizer,
  void *pCtx,
  int flags,
  const char *pText, int nText,
  int (*xToken)(void*, int, const char*, int nToken, int iStart, int iEnd)
){
  Unicode61Tokenizer *p = (Unicode61Tokenizer*)pTokenizer;
  int rc = SQLITE_OK;
  unsigned char *a = p->aTokenChar;


  unsigned char *zTerm = (unsigned char*)&pText[nText];
  unsigned char *zCsr = (unsigned char *)pText;

  /* Output buffer */
  char *aFold = p->aFold;
  int nFold = p->nFold;

        }
        zCsr++;
      }
      ie = zCsr - (unsigned char*)pText;
    }

    /* Invoke the token callback */
    rc = xToken(pCtx, 0, aFold, zOut-aFold, is, ie); 
  }
  
 tokenize_done:
  if( rc==SQLITE_DONE ) rc = SQLITE_OK;
  return rc;
}

  *ppOut = (Fts5Tokenizer*)pRet;
  return rc;
}

typedef struct PorterContext PorterContext;
struct PorterContext {
  void *pCtx;
  int (*xToken)(void*, int, const char*, int, int, int);
  char *aBuf;
};

typedef struct PorterRule PorterRule;
struct PorterRule {
  const char *zSuffix;
  int nSuffix;

      *pnBuf = nBuf-1;
    }
  }
}

static int fts5PorterCb(
  void *pCtx, 
  int tflags,
  const char *pToken, 
  int nToken, 
  int iStart, 
  int iEnd

){
  PorterContext *p = (PorterContext*)pCtx;

  char *aBuf;
  int nBuf;

  if( nToken>FTS5_PORTER_MAX_TOKEN || nToken<3 ) goto pass_through;

  /* Step 5b. */
  if( nBuf>1 && aBuf[nBuf-1]=='l' 
   && aBuf[nBuf-2]=='l' && fts5Porter_MGt1(aBuf, nBuf-1) 
  ){
    nBuf--;
  }

  return p->xToken(p->pCtx, tflags, aBuf, nBuf, iStart, iEnd);

 pass_through:
  return p->xToken(p->pCtx, tflags, pToken, nToken, iStart, iEnd);
}

/*
** Tokenize using the porter tokenizer.
*/
static int fts5PorterTokenize(
  Fts5Tokenizer *pTokenizer,
  void *pCtx,
  int flags,
  const char *pText, int nText,
  int (*xToken)(void*, int, const char*, int nToken, int iStart, int iEnd)
){
  PorterTokenizer *p = (PorterTokenizer*)pTokenizer;
  PorterContext sCtx;
  sCtx.xToken = xToken;
  sCtx.pCtx = pCtx;
  sCtx.aBuf = p->aBuf;
  return p->tokenizer.xTokenize(