/ Check-in [76eca729]
Login

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

Overview
Comment:Modified statement machine in sqlite3_complete() to return 0 on empty string. Added/updated tests for same. Ticket [356c885b0b].
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1:76eca7295cf7df4bef013af6c8c37251300cd383
User & Date: shaneh 2009-12-17 22:17:39
Original Comment: Modified statement machine in sqlite3_complete() to return 0 on empty string. Added/updated tests for same. Ticket 356c885b0b.
References
2009-12-17
22:18 Ticket [356c885b] sqlite3_complete() should return 0 on empty string status still Fixed with 1 other change artifact: 9743fa03 user: shane
22:18 Fixed ticket [356c885b]. artifact: 2492e5e8 user: shane
Context
2009-12-20
15:00
Fix a problem in FTS3 phrase queries. Add tests to check that it really is fixed. check-in: 956de051 user: dan tags: trunk
2009-12-17
22:17
Modified statement machine in sqlite3_complete() to return 0 on empty string. Added/updated tests for same. Ticket [356c885b0b]. check-in: 76eca729 user: shaneh tags: trunk
22:12
Fixed some TCL test cases to work if SQLITE_OMIT_TRIGGER is defined. check-in: 85e3c734 user: shaneh tags: trunk
Changes
Hide Diffs Unified Diffs Ignore Whitespace Patch

Changes to src/complete.c.

36
37
38
39
40
41
42

43
44
45
46
47

48
49
50
51
52
53
54
55
56
57


58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93

94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116

117
118
119
120
121
122
123
124

125
126
127
128
129
130
131
132
133
...
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
...
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
/*
** 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

#define tkEXPLAIN 3
#define tkCREATE  4
#define tkTEMP    5
#define tkTRIGGER 6
#define tkEND     7


/*
** 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 7 states:
**


**   (0) 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.
**
**   (1) NORMAL    We are in the middle of statement which ends with a single
**                 semicolon.
**
**   (2) EXPLAIN   The keyword EXPLAIN has been seen at the beginning of 
**                 a statement.
**
**   (3) CREATE    The keyword CREATE has been seen at the beginning of a
**                 statement, possibly preceeded by EXPLAIN and/or followed by
**                 TEMP or TEMPORARY
**
**   (4) TRIGGER   We are in the middle of a trigger definition that must be
**                 ended by a semicolon, the keyword END, and another semicolon.
**
**   (5) SEMI      We've seen the first semicolon in the ";END;" that occurs at
**                 the end of a trigger definition.
**
**   (6) 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.

**
** 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[7][8] = {
                     /* Token:                                                */
     /* State:       **  SEMI  WS  OTHER EXPLAIN  CREATE  TEMP  TRIGGER  END  */
     /* 0   START: */ {    0,  0,     1,      2,      3,    1,       1,   1,  },
     /* 1  NORMAL: */ {    0,  1,     1,      1,      1,    1,       1,   1,  },
     /* 2 EXPLAIN: */ {    0,  2,     2,      1,      3,    1,       1,   1,  },
     /* 3  CREATE: */ {    0,  3,     1,      1,      1,    3,       4,   1,  },
     /* 4 TRIGGER: */ {    5,  4,     4,      4,      4,    4,       4,   4,  },
     /* 5    SEMI: */ {    5,  5,     4,      4,      4,    4,       4,   6,  },
     /* 6     END: */ {    0,  6,     4,      4,      4,    4,       4,   4,  },

  };
#else
  /* If triggers are not suppored by this compile then the statement machine
  ** used to detect the end of a statement is much simplier
  */
  static const u8 trans[2][3] = {
                     /* Token:           */
     /* State:       **  SEMI  WS  OTHER */

     /* 0   START: */ {    0,  0,     1, },
     /* 1  NORMAL: */ {    0,  1,     1, },
  };
#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==0;
        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==0;
}

#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.







>





>








|
|
>
>
|



|
|
|
|
|
|
|



|
|
|
|
|
|
|






|








>













|
|
|
|
|
|
|
|
|
|
>
|
|
|
|
|
|
|
|
>
|
|







 







|







 







|







36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
...
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
...
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
/*
** 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.

Changes to test/main.test.

21
22
23
24
25
26
27
28
29
30
31



32
33
34
35
36





37
38
39
40
41











42
43
44
45
46
47
48
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 {
  db complete {
  }
} {1}



do_test main-1.3 {
  db complete {
     -- a comment ;
  }
} {1}





do_test main-1.4 {
  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}







|


|
>
>
>
|



|
>
>
>
>
>
|




>
>
>
>
>
>
>
>
>
>
>







21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
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}