SQLite

/*
** Token types used by the sqlite3_complete() routine.  See the header
** comments on that procedure for additional information.
*/
#define tkSEMI    0
#define tkWS      1
#define tkOTHER   2
#ifndef SQLITE_OMIT_TRIGGER
#define tkEXPLAIN 3
#define tkCREATE  4
#define tkTEMP    5
#define tkTRIGGER 6
#define tkEND     7
#endif

/*
** Return TRUE if the given SQL string ends in a semicolon.
**
** Special handling is require for CREATE TRIGGER statements.
** Whenever the CREATE TRIGGER keywords are seen, the statement
** must end with ";END;".
**
** This implementation uses a state machine with 8 states:
**
**   (0) INVALID   We have not yet seen a non-whitespace character.
**
**   (1) START     At the beginning or end of an SQL statement.  This routine
**                 returns 1 if it ends in the START state and 0 if it ends
**                 in any other state.
**
**   (2) NORMAL    We are in the middle of statement which ends with a single
**                 semicolon.
**
**   (3) EXPLAIN   The keyword EXPLAIN has been seen at the beginning of 
**                 a statement.
**
**   (4) CREATE    The keyword CREATE has been seen at the beginning of a
**                 statement, possibly preceeded by EXPLAIN and/or followed by
**                 TEMP or TEMPORARY
**
**   (5) TRIGGER   We are in the middle of a trigger definition that must be
**                 ended by a semicolon, the keyword END, and another semicolon.
**
**   (6) SEMI      We've seen the first semicolon in the ";END;" that occurs at
**                 the end of a trigger definition.
**
**   (7) END       We've seen the ";END" of the ";END;" that occurs at the end
**                 of a trigger difinition.
**
** Transitions between states above are determined by tokens extracted
** from the input.  The following tokens are significant:
**
**   (0) tkSEMI      A semicolon.
**   (1) tkWS        Whitespace.
**   (2) tkOTHER     Any other SQL token.
**   (3) tkEXPLAIN   The "explain" keyword.
**   (4) tkCREATE    The "create" keyword.
**   (5) tkTEMP      The "temp" or "temporary" keyword.
**   (6) tkTRIGGER   The "trigger" keyword.
**   (7) tkEND       The "end" keyword.
**
** Whitespace never causes a state transition and is always ignored.
** This means that a SQL string of all whitespace is invalid.
**
** If we compile with SQLITE_OMIT_TRIGGER, all of the computation needed
** to recognize the end of a trigger can be omitted.  All we have to do
** is look for a semicolon that is not part of an string or comment.
*/
int sqlite3_complete(const char *zSql){
  u8 state = 0;   /* Current state, using numbers defined in header comment */
  u8 token;       /* Value of the next token */

#ifndef SQLITE_OMIT_TRIGGER
  /* A complex statement machine used to detect the end of a CREATE TRIGGER
  ** statement.  This is the normal case.
  */
  static const u8 trans[8][8] = {
                     /* Token:                                                */
     /* State:       **  SEMI  WS  OTHER  EXPLAIN  CREATE  TEMP  TRIGGER  END */
     /* 0 INVALID: */ {    1,  0,     2,       3,      4,    2,       2,   2, },
     /* 1   START: */ {    1,  1,     2,       3,      4,    2,       2,   2, },
     /* 2  NORMAL: */ {    1,  2,     2,       2,      2,    2,       2,   2, },
     /* 3 EXPLAIN: */ {    1,  3,     3,       2,      4,    2,       2,   2, },
     /* 4  CREATE: */ {    1,  4,     2,       2,      2,    4,       5,   2, },
     /* 5 TRIGGER: */ {    6,  5,     5,       5,      5,    5,       5,   5, },
     /* 6    SEMI: */ {    6,  6,     5,       5,      5,    5,       5,   7, },
     /* 7     END: */ {    1,  7,     5,       5,      5,    5,       5,   5, },
  };
#else
  /* If triggers are not supported by this compile then the statement machine
  ** used to detect the end of a statement is much simplier
  */
  static const u8 trans[3][3] = {
                     /* Token:           */
     /* State:       **  SEMI  WS  OTHER */
     /* 0 INVALID: */ {    1,  0,     2, },
     /* 1   START: */ {    1,  1,     2, },
     /* 2  NORMAL: */ {    1,  2,     2, },
  };
#endif /* SQLITE_OMIT_TRIGGER */

  while( *zSql ){
    switch( *zSql ){
      case ';': {  /* A semicolon */
        token = tkSEMI;

      }
      case '-': {   /* SQL-style comments from "--" to end of line */
        if( zSql[1]!='-' ){
          token = tkOTHER;
          break;
        }
        while( *zSql && *zSql!='\n' ){ zSql++; }
        if( *zSql==0 ) return state==1;
        token = tkWS;
        break;
      }
      case '[': {   /* Microsoft-style identifiers in [...] */
        zSql++;
        while( *zSql && *zSql!=']' ){ zSql++; }
        if( *zSql==0 ) return 0;

        }
        break;
      }
    }
    state = trans[state][token];
    zSql++;
  }
  return state==1;
}

#ifndef SQLITE_OMIT_UTF16
/*
** This routine is the same as the sqlite3_complete() routine described
** above, except that the parameter is required to be UTF-16 encoded, not
** UTF-8.

ifcapable {complete} {

# Tests of the sqlite_complete() function.
#
do_test main-1.1 {
  db complete {This is a test}
} {0}
do_test main-1.2.0 {
  db complete {
  }
} {0}
do_test main-1.2.1 {
  db complete {}
} {0}
do_test main-1.3.0 {
  db complete {
     -- a comment ;
  }
} {0}
do_test main-1.3.1 {
  db complete {
     /* a comment ; */
  }
} {0}
do_test main-1.4.0 {
  db complete {
     -- a comment ;
     ;
  }
} {1}
do_test main-1.4.1 {
  db complete {
     /* a comment ; */
     ;
  }
} {1}
do_test main-1.4.2 {
  db complete {
     /* a comment ; */ ;
  }
} {1}
do_test main-1.5 {
  db complete {DROP TABLE 'xyz;}
} {0}
do_test main-1.6 {
  db complete {DROP TABLE 'xyz';}
} {1}