fts1.c

sqlite-3.4.1,嵌入式数据库.是一个功能强大的开源数据库,给学习和研发以及小型公司的发展带来了全所未有的好处.

  if( rc!=SQLITE_OK ) return rc;

  rc = sql_step_statement(v, CONTENT_SELECT_STMT, &s);
  if( rc!=SQLITE_ROW ) return rc;

  values = (const char **) malloc(v->nColumn * sizeof(const char *));
  for(i=0; i<v->nColumn; ++i){
    if( sqlite3_column_type(s, i)==SQLITE_NULL ){
      values[i] = NULL;
    }else{
      values[i] = string_dup((char*)sqlite3_column_text(s, i));
    }
  }

  /* We expect only one row.  We must execute another sqlite3_step()
   * to complete the iteration; otherwise the table will remain locked. */
  rc = sqlite3_step(s);
  if( rc==SQLITE_DONE ){
    *pValues = values;
    return SQLITE_OK;
  }

  freeStringArray(v->nColumn, values);
  return rc;
}

/* delete from %_content where rowid = [iRow ] */
static int content_delete(fulltext_vtab *v, sqlite_int64 iRow){
  sqlite3_stmt *s;
  int rc = sql_get_statement(v, CONTENT_DELETE_STMT, &s);
  if( rc!=SQLITE_OK ) return rc;

  rc = sqlite3_bind_int64(s, 1, iRow);
  if( rc!=SQLITE_OK ) return rc;

  return sql_single_step_statement(v, CONTENT_DELETE_STMT, &s);
}

/* select rowid, doclist from %_term
 *  where term = [pTerm] and segment = [iSegment]
 * If found, returns SQLITE_ROW; the caller must free the
 * returned doclist.  If no rows found, returns SQLITE_DONE. */
static int term_select(fulltext_vtab *v, const char *pTerm, int nTerm,
                       int iSegment,
                       sqlite_int64 *rowid, DocList *out){
  sqlite3_stmt *s;
  int rc = sql_get_statement(v, TERM_SELECT_STMT, &s);
  if( rc!=SQLITE_OK ) return rc;

  rc = sqlite3_bind_text(s, 1, pTerm, nTerm, SQLITE_STATIC);
  if( rc!=SQLITE_OK ) return rc;

  rc = sqlite3_bind_int(s, 2, iSegment);
  if( rc!=SQLITE_OK ) return rc;

  rc = sql_step_statement(v, TERM_SELECT_STMT, &s);
  if( rc!=SQLITE_ROW ) return rc;

  *rowid = sqlite3_column_int64(s, 0);
  docListInit(out, DL_DEFAULT,
              sqlite3_column_blob(s, 1), sqlite3_column_bytes(s, 1));

  /* We expect only one row.  We must execute another sqlite3_step()
   * to complete the iteration; otherwise the table will remain locked. */
  rc = sqlite3_step(s);
  return rc==SQLITE_DONE ? SQLITE_ROW : rc;
}

/* Load the segment doclists for term pTerm and merge them in
** appropriate order into out.  Returns SQLITE_OK if successful.  If
** there are no segments for pTerm, successfully returns an empty
** doclist in out.
**
** Each document consists of 1 or more "columns".  The number of
** columns is v->nColumn.  If iColumn==v->nColumn, then return
** position information about all columns.  If iColumn<v->nColumn,
** then only return position information about the iColumn-th column
** (where the first column is 0).
*/
static int term_select_all(
  fulltext_vtab *v,     /* The fulltext index we are querying against */
  int iColumn,          /* If <nColumn, only look at the iColumn-th column */
  const char *pTerm,    /* The term whose posting lists we want */
  int nTerm,            /* Number of bytes in pTerm */
  DocList *out          /* Write the resulting doclist here */
){
  DocList doclist;
  sqlite3_stmt *s;
  int rc = sql_get_statement(v, TERM_SELECT_ALL_STMT, &s);
  if( rc!=SQLITE_OK ) return rc;

  rc = sqlite3_bind_text(s, 1, pTerm, nTerm, SQLITE_STATIC);
  if( rc!=SQLITE_OK ) return rc;

  docListInit(&doclist, DL_DEFAULT, 0, 0);

  /* TODO(shess) Handle schema and busy errors. */
  while( (rc=sql_step_statement(v, TERM_SELECT_ALL_STMT, &s))==SQLITE_ROW ){
    DocList old;

    /* TODO(shess) If we processed doclists from oldest to newest, we
    ** could skip the malloc() involved with the following call.  For
    ** now, I'd rather keep this logic similar to index_insert_term().
    ** We could additionally drop elements when we see deletes, but
    ** that would require a distinct version of docListAccumulate().
    */
    docListInit(&old, DL_DEFAULT,
                sqlite3_column_blob(s, 0), sqlite3_column_bytes(s, 0));

    if( iColumn<v->nColumn ){   /* querying a single column */
      docListRestrictColumn(&old, iColumn);
    }

    /* doclist contains the newer data, so write it over old.  Then
    ** steal accumulated result for doclist.
    */
    docListAccumulate(&old, &doclist);
    docListDestroy(&doclist);
    doclist = old;
  }
  if( rc!=SQLITE_DONE ){
    docListDestroy(&doclist);
    return rc;
  }

  docListDiscardEmpty(&doclist);
  *out = doclist;
  return SQLITE_OK;
}

/* insert into %_term (rowid, term, segment, doclist)
               values ([piRowid], [pTerm], [iSegment], [doclist])
** Lets sqlite select rowid if piRowid is NULL, else uses *piRowid.
**
** NOTE(shess) piRowid is IN, with values of "space of int64" plus
** null, it is not used to pass data back to the caller.
*/
static int term_insert(fulltext_vtab *v, sqlite_int64 *piRowid,
                       const char *pTerm, int nTerm,
                       int iSegment, DocList *doclist){
  sqlite3_stmt *s;
  int rc = sql_get_statement(v, TERM_INSERT_STMT, &s);
  if( rc!=SQLITE_OK ) return rc;

  if( piRowid==NULL ){
    rc = sqlite3_bind_null(s, 1);
  }else{
    rc = sqlite3_bind_int64(s, 1, *piRowid);
  }
  if( rc!=SQLITE_OK ) return rc;

  rc = sqlite3_bind_text(s, 2, pTerm, nTerm, SQLITE_STATIC);
  if( rc!=SQLITE_OK ) return rc;

  rc = sqlite3_bind_int(s, 3, iSegment);
  if( rc!=SQLITE_OK ) return rc;

  rc = sqlite3_bind_blob(s, 4, doclist->pData, doclist->nData, SQLITE_STATIC);
  if( rc!=SQLITE_OK ) return rc;

  return sql_single_step_statement(v, TERM_INSERT_STMT, &s);
}

/* update %_term set doclist = [doclist] where rowid = [rowid] */
static int term_update(fulltext_vtab *v, sqlite_int64 rowid,
                       DocList *doclist){
  sqlite3_stmt *s;
  int rc = sql_get_statement(v, TERM_UPDATE_STMT, &s);
  if( rc!=SQLITE_OK ) return rc;

  rc = sqlite3_bind_blob(s, 1, doclist->pData, doclist->nData, SQLITE_STATIC);
  if( rc!=SQLITE_OK ) return rc;

  rc = sqlite3_bind_int64(s, 2, rowid);
  if( rc!=SQLITE_OK ) return rc;

  return sql_single_step_statement(v, TERM_UPDATE_STMT, &s);
}

static int term_delete(fulltext_vtab *v, sqlite_int64 rowid){
  sqlite3_stmt *s;
  int rc = sql_get_statement(v, TERM_DELETE_STMT, &s);
  if( rc!=SQLITE_OK ) return rc;

  rc = sqlite3_bind_int64(s, 1, rowid);
  if( rc!=SQLITE_OK ) return rc;

  return sql_single_step_statement(v, TERM_DELETE_STMT, &s);
}

/*
** Free the memory used to contain a fulltext_vtab structure.
*/
static void fulltext_vtab_destroy(fulltext_vtab *v){
  int iStmt, i;

  TRACE(("FTS1 Destroy %p\n", v));
  for( iStmt=0; iStmt<MAX_STMT; iStmt++ ){
    if( v->pFulltextStatements[iStmt]!=NULL ){
      sqlite3_finalize(v->pFulltextStatements[iStmt]);
      v->pFulltextStatements[iStmt] = NULL;
    }
  }

  if( v->pTokenizer!=NULL ){
    v->pTokenizer->pModule->xDestroy(v->pTokenizer);
    v->pTokenizer = NULL;
  }
  
  free(v->azColumn);
  for(i = 0; i < v->nColumn; ++i) {
    sqlite3_free(v->azContentColumn[i]);
  }
  free(v->azContentColumn);
  free(v);
}

/*
** Token types for parsing the arguments to xConnect or xCreate.
*/
#define TOKEN_EOF         0    /* End of file */
#define TOKEN_SPACE       1    /* Any kind of whitespace */
#define TOKEN_ID          2    /* An identifier */
#define TOKEN_STRING      3    /* A string literal */
#define TOKEN_PUNCT       4    /* A single punctuation character */

/*
** If X is a character that can be used in an identifier then
** IdChar(X) will be true.  Otherwise it is false.
**
** For ASCII, any character with the high-order bit set is
** allowed in an identifier.  For 7-bit characters, 
** sqlite3IsIdChar[X] must be 1.
**
** Ticket #1066.  the SQL standard does not allow '$' in the
** middle of identfiers.  But many SQL implementations do. 
** SQLite will allow '$' in identifiers for compatibility.
** But the feature is undocumented.
*/
static const char isIdChar[] = {
/* x0 x1 x2 x3 x4 x5 x6 x7 x8 x9 xA xB xC xD xE xF */
    0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,  /* 2x */
    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)  (((c=C)&0x80)!=0 || (c>0x1f && isIdChar[c-0x20]))


/*
** Return the length of the token that begins at z[0]. 
** Store the token type in *tokenType before returning.
*/
static int getToken(const char *z, int *tokenType){
  int i, c;
  switch( *z ){
    case 0: {
      *tokenType = TOKEN_EOF;
      return 0;
    }
    case ' ': case '\t': case '\n': case '\f': case '\r': {
      for(i=1; safe_isspace(z[i]); i++){}
      *tokenType = TOKEN_SPACE;
      return i;
    }
    case '`':
    case '\'':
    case '"': {
      int delim = z[0];
      for(i=1; (c=z[i])!=0; i++){
        if( c==delim ){
          if( z[i+1]==delim ){
            i++;
          }else{
            break;
          }
        }
      }
      *tokenType = TOKEN_STRING;
      return i + (c!=0);
    }
    case '[': {
      for(i=1, c=z[0]; c!=']' && (c=z[i])!=0; i++){}
      *tokenType = TOKEN_ID;
      return i;
    }
    default: {
      if( !IdChar(*z) ){
        break;
      }
      for(i=1; IdChar(z[i]); i++){}
      *tokenType = TOKEN_ID;
      return i;
    }
  }
  *tokenType = TOKEN_PUNCT;
  return 1;
}

/*
** A token extracted from a string is an instance of the following
** structure.
*/
typedef struct Token {
  const char *z;       /* Pointer to token text.  Not '\000' terminated */
  short int n;         /* Length of the token text in bytes. */
} Token;

/*
** Given a input string (which is really one of the argv[] parameters
** passed into xConnect or xCreate) split the string up into tokens.
** Return an array of pointers to '\000' terminated strings, one string
** for each non-whitespace token.
**
** The returned array is terminated by a single NULL pointer.
**
** Space to hold the returned array is obtained from a single
** malloc and should be freed by passing the return value to free().
** The individual strings within the token list are all a part of
** the single memory allocation and will all be freed at once.
*/
static char **tokenizeString(const char *z, int *pnToken){
  int nToken = 0;
  Token *aToken = malloc( strlen(z) * sizeof(aToken[0]) );
  int n = 1;
  int e, i;
  int totalSize = 0;
  char **azToken;
  char *zCopy;
  while( n>0 ){
    n = getToken(z, &e);
    if( e!=TOKEN_SPACE ){
      aToken[nToken].z = z;
      aToken[nToken].n = n;
      nToken++;
      totalSize += n+1;
    }
    z += n;
  }
  azToken = (char**)malloc( nToken*sizeof(char*) + totalSize );
  zCopy = (char*)&azToken[nToken];
  nToken--;
  for(i=0; i<nToken; i++){
    azToken[i] = zCopy;
    n = aToken[i].n;
    memcpy(zCopy, aToken[i].z, n);
    zCopy[n] = 0;
    zCopy += n+1;
  }
  azToken[nToken] = 0;
  free(aToken);
  *pnToken = nToken;
  return azToken;
}

/*
** Convert an SQL-style quoted string into a normal string by removing
** the quote characters.  The conversion is done in-place.  If the
** input does not begin with a quote character, then this routine
** is a no-op.
**
** Examples:
**
**     "abc"   becomes   abc
**     'xyz'   becomes   xyz
**     [pqr]   becomes   pqr
**     `mno`   becomes   mno
*/
static void dequoteString(char *z){
  int quote;
  int i, j;
  if( z==0 ) return;
  quote = z[0];
  switch( quote ){
    case '\'':  break;
    case '"':   break;
    case '`':   break;                /* For MySQL compatibility */
    case '[':   quote = ']';  break;  /* For MS SqlServer compatibility */
    default:    return;
  }
  for(i=1, j=0; z[i]; i++){
    if( z[i]==quote ){
      if( z[i+1]==quote ){
        z[j++] = quote;
        i++;
      }else{
        z[j++] = 0;
        break;
      }
    }else{
      z[j++] = z[i];
    }
  }
}

/*
** The input azIn is a NULL-terminated list of tokens.  Remove the first
** token and all punctuation tokens.  Remove the quotes from
** around string literal tokens.
**
** Example:
**
**     input:      tokenize chinese ( 'simplifed' , 'mixed' )
**     output:     chinese simplifed mixed
**
** Another example:
**
**     input:      delimiters ( '[' , ']' , '...' )
**     output:     [ ] ...
*/
static void tokenListToIdList(char **azIn){
  int i, j;
  if( azIn ){
    for(i=0, j=-1; azIn[i]; i++){
      if( safe_isalnum(azIn[i][0]) || azIn[i][1] ){
        dequoteString(azIn[i]);
        if( j>=0 ){
          azIn[j] = azIn[i];
        }
        j++;
      }
    }
    azIn[j] = 0;
  }
}


/*
** Find the first alphanumeric token in the string zIn.  Null-terminate
** this token.  Remove any quotation marks.  And return a pointer to
** the result.