|
sl@0
|
1 |
/*
|
|
sl@0
|
2 |
** 2001 September 15
|
|
sl@0
|
3 |
**
|
|
sl@0
|
4 |
** The author disclaims copyright to this source code. In place of
|
|
sl@0
|
5 |
** a legal notice, here is a blessing:
|
|
sl@0
|
6 |
**
|
|
sl@0
|
7 |
** May you do good and not evil.
|
|
sl@0
|
8 |
** May you find forgiveness for yourself and forgive others.
|
|
sl@0
|
9 |
** May you share freely, never taking more than you give.
|
|
sl@0
|
10 |
**
|
|
sl@0
|
11 |
*************************************************************************
|
|
sl@0
|
12 |
** An tokenizer for SQL
|
|
sl@0
|
13 |
**
|
|
sl@0
|
14 |
** This file contains C code that splits an SQL input string up into
|
|
sl@0
|
15 |
** individual tokens and sends those tokens one-by-one over to the
|
|
sl@0
|
16 |
** parser for analysis.
|
|
sl@0
|
17 |
**
|
|
sl@0
|
18 |
** $Id: tokenize.c,v 1.152 2008/09/01 15:52:11 drh Exp $
|
|
sl@0
|
19 |
*/
|
|
sl@0
|
20 |
#include "sqliteInt.h"
|
|
sl@0
|
21 |
#include <ctype.h>
|
|
sl@0
|
22 |
#include <stdlib.h>
|
|
sl@0
|
23 |
|
|
sl@0
|
24 |
/*
|
|
sl@0
|
25 |
** The charMap() macro maps alphabetic characters into their
|
|
sl@0
|
26 |
** lower-case ASCII equivalent. On ASCII machines, this is just
|
|
sl@0
|
27 |
** an upper-to-lower case map. On EBCDIC machines we also need
|
|
sl@0
|
28 |
** to adjust the encoding. Only alphabetic characters and underscores
|
|
sl@0
|
29 |
** need to be translated.
|
|
sl@0
|
30 |
*/
|
|
sl@0
|
31 |
#ifdef SQLITE_ASCII
|
|
sl@0
|
32 |
# define charMap(X) sqlite3UpperToLower[(unsigned char)X]
|
|
sl@0
|
33 |
#endif
|
|
sl@0
|
34 |
#ifdef SQLITE_EBCDIC
|
|
sl@0
|
35 |
# define charMap(X) ebcdicToAscii[(unsigned char)X]
|
|
sl@0
|
36 |
const unsigned char ebcdicToAscii[] = {
|
|
sl@0
|
37 |
/* 0 1 2 3 4 5 6 7 8 9 A B C D E F */
|
|
sl@0
|
38 |
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 0x */
|
|
sl@0
|
39 |
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 1x */
|
|
sl@0
|
40 |
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 2x */
|
|
sl@0
|
41 |
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 3x */
|
|
sl@0
|
42 |
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 4x */
|
|
sl@0
|
43 |
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 5x */
|
|
sl@0
|
44 |
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 95, 0, 0, /* 6x */
|
|
sl@0
|
45 |
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 7x */
|
|
sl@0
|
46 |
0, 97, 98, 99,100,101,102,103,104,105, 0, 0, 0, 0, 0, 0, /* 8x */
|
|
sl@0
|
47 |
0,106,107,108,109,110,111,112,113,114, 0, 0, 0, 0, 0, 0, /* 9x */
|
|
sl@0
|
48 |
0, 0,115,116,117,118,119,120,121,122, 0, 0, 0, 0, 0, 0, /* Ax */
|
|
sl@0
|
49 |
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* Bx */
|
|
sl@0
|
50 |
0, 97, 98, 99,100,101,102,103,104,105, 0, 0, 0, 0, 0, 0, /* Cx */
|
|
sl@0
|
51 |
0,106,107,108,109,110,111,112,113,114, 0, 0, 0, 0, 0, 0, /* Dx */
|
|
sl@0
|
52 |
0, 0,115,116,117,118,119,120,121,122, 0, 0, 0, 0, 0, 0, /* Ex */
|
|
sl@0
|
53 |
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* Fx */
|
|
sl@0
|
54 |
};
|
|
sl@0
|
55 |
#endif
|
|
sl@0
|
56 |
|
|
sl@0
|
57 |
/*
|
|
sl@0
|
58 |
** The sqlite3KeywordCode function looks up an identifier to determine if
|
|
sl@0
|
59 |
** it is a keyword. If it is a keyword, the token code of that keyword is
|
|
sl@0
|
60 |
** returned. If the input is not a keyword, TK_ID is returned.
|
|
sl@0
|
61 |
**
|
|
sl@0
|
62 |
** The implementation of this routine was generated by a program,
|
|
sl@0
|
63 |
** mkkeywordhash.h, located in the tool subdirectory of the distribution.
|
|
sl@0
|
64 |
** The output of the mkkeywordhash.c program is written into a file
|
|
sl@0
|
65 |
** named keywordhash.h and then included into this source file by
|
|
sl@0
|
66 |
** the #include below.
|
|
sl@0
|
67 |
*/
|
|
sl@0
|
68 |
#include "keywordhash.h"
|
|
sl@0
|
69 |
|
|
sl@0
|
70 |
|
|
sl@0
|
71 |
/*
|
|
sl@0
|
72 |
** If X is a character that can be used in an identifier then
|
|
sl@0
|
73 |
** IdChar(X) will be true. Otherwise it is false.
|
|
sl@0
|
74 |
**
|
|
sl@0
|
75 |
** For ASCII, any character with the high-order bit set is
|
|
sl@0
|
76 |
** allowed in an identifier. For 7-bit characters,
|
|
sl@0
|
77 |
** sqlite3IsIdChar[X] must be 1.
|
|
sl@0
|
78 |
**
|
|
sl@0
|
79 |
** For EBCDIC, the rules are more complex but have the same
|
|
sl@0
|
80 |
** end result.
|
|
sl@0
|
81 |
**
|
|
sl@0
|
82 |
** Ticket #1066. the SQL standard does not allow '$' in the
|
|
sl@0
|
83 |
** middle of identfiers. But many SQL implementations do.
|
|
sl@0
|
84 |
** SQLite will allow '$' in identifiers for compatibility.
|
|
sl@0
|
85 |
** But the feature is undocumented.
|
|
sl@0
|
86 |
*/
|
|
sl@0
|
87 |
#ifdef SQLITE_ASCII
|
|
sl@0
|
88 |
const char sqlite3IsAsciiIdChar[] = {
|
|
sl@0
|
89 |
/* x0 x1 x2 x3 x4 x5 x6 x7 x8 x9 xA xB xC xD xE xF */
|
|
sl@0
|
90 |
0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 2x */
|
|
sl@0
|
91 |
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, /* 3x */
|
|
sl@0
|
92 |
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 4x */
|
|
sl@0
|
93 |
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1, /* 5x */
|
|
sl@0
|
94 |
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 6x */
|
|
sl@0
|
95 |
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, /* 7x */
|
|
sl@0
|
96 |
};
|
|
sl@0
|
97 |
#define IdChar(C) (((c=C)&0x80)!=0 || (c>0x1f && sqlite3IsAsciiIdChar[c-0x20]))
|
|
sl@0
|
98 |
#endif
|
|
sl@0
|
99 |
#ifdef SQLITE_EBCDIC
|
|
sl@0
|
100 |
const char sqlite3IsEbcdicIdChar[] = {
|
|
sl@0
|
101 |
/* x0 x1 x2 x3 x4 x5 x6 x7 x8 x9 xA xB xC xD xE xF */
|
|
sl@0
|
102 |
0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, /* 4x */
|
|
sl@0
|
103 |
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 0, 0, 0, /* 5x */
|
|
sl@0
|
104 |
0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 0, 0, /* 6x */
|
|
sl@0
|
105 |
0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, /* 7x */
|
|
sl@0
|
106 |
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 0, /* 8x */
|
|
sl@0
|
107 |
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 0, 1, 0, /* 9x */
|
|
sl@0
|
108 |
1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 0, /* Ax */
|
|
sl@0
|
109 |
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* Bx */
|
|
sl@0
|
110 |
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, /* Cx */
|
|
sl@0
|
111 |
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, /* Dx */
|
|
sl@0
|
112 |
0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, /* Ex */
|
|
sl@0
|
113 |
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 0, /* Fx */
|
|
sl@0
|
114 |
};
|
|
sl@0
|
115 |
#define IdChar(C) (((c=C)>=0x42 && sqlite3IsEbcdicIdChar[c-0x40]))
|
|
sl@0
|
116 |
#endif
|
|
sl@0
|
117 |
|
|
sl@0
|
118 |
|
|
sl@0
|
119 |
/*
|
|
sl@0
|
120 |
** Return the length of the token that begins at z[0].
|
|
sl@0
|
121 |
** Store the token type in *tokenType before returning.
|
|
sl@0
|
122 |
*/
|
|
sl@0
|
123 |
int sqlite3GetToken(const unsigned char *z, int *tokenType){
|
|
sl@0
|
124 |
int i, c;
|
|
sl@0
|
125 |
switch( *z ){
|
|
sl@0
|
126 |
case ' ': case '\t': case '\n': case '\f': case '\r': {
|
|
sl@0
|
127 |
for(i=1; isspace(z[i]); i++){}
|
|
sl@0
|
128 |
*tokenType = TK_SPACE;
|
|
sl@0
|
129 |
return i;
|
|
sl@0
|
130 |
}
|
|
sl@0
|
131 |
case '-': {
|
|
sl@0
|
132 |
if( z[1]=='-' ){
|
|
sl@0
|
133 |
for(i=2; (c=z[i])!=0 && c!='\n'; i++){}
|
|
sl@0
|
134 |
*tokenType = TK_SPACE;
|
|
sl@0
|
135 |
return i;
|
|
sl@0
|
136 |
}
|
|
sl@0
|
137 |
*tokenType = TK_MINUS;
|
|
sl@0
|
138 |
return 1;
|
|
sl@0
|
139 |
}
|
|
sl@0
|
140 |
case '(': {
|
|
sl@0
|
141 |
*tokenType = TK_LP;
|
|
sl@0
|
142 |
return 1;
|
|
sl@0
|
143 |
}
|
|
sl@0
|
144 |
case ')': {
|
|
sl@0
|
145 |
*tokenType = TK_RP;
|
|
sl@0
|
146 |
return 1;
|
|
sl@0
|
147 |
}
|
|
sl@0
|
148 |
case ';': {
|
|
sl@0
|
149 |
*tokenType = TK_SEMI;
|
|
sl@0
|
150 |
return 1;
|
|
sl@0
|
151 |
}
|
|
sl@0
|
152 |
case '+': {
|
|
sl@0
|
153 |
*tokenType = TK_PLUS;
|
|
sl@0
|
154 |
return 1;
|
|
sl@0
|
155 |
}
|
|
sl@0
|
156 |
case '*': {
|
|
sl@0
|
157 |
*tokenType = TK_STAR;
|
|
sl@0
|
158 |
return 1;
|
|
sl@0
|
159 |
}
|
|
sl@0
|
160 |
case '/': {
|
|
sl@0
|
161 |
if( z[1]!='*' || z[2]==0 ){
|
|
sl@0
|
162 |
*tokenType = TK_SLASH;
|
|
sl@0
|
163 |
return 1;
|
|
sl@0
|
164 |
}
|
|
sl@0
|
165 |
for(i=3, c=z[2]; (c!='*' || z[i]!='/') && (c=z[i])!=0; i++){}
|
|
sl@0
|
166 |
if( c ) i++;
|
|
sl@0
|
167 |
*tokenType = TK_SPACE;
|
|
sl@0
|
168 |
return i;
|
|
sl@0
|
169 |
}
|
|
sl@0
|
170 |
case '%': {
|
|
sl@0
|
171 |
*tokenType = TK_REM;
|
|
sl@0
|
172 |
return 1;
|
|
sl@0
|
173 |
}
|
|
sl@0
|
174 |
case '=': {
|
|
sl@0
|
175 |
*tokenType = TK_EQ;
|
|
sl@0
|
176 |
return 1 + (z[1]=='=');
|
|
sl@0
|
177 |
}
|
|
sl@0
|
178 |
case '<': {
|
|
sl@0
|
179 |
if( (c=z[1])=='=' ){
|
|
sl@0
|
180 |
*tokenType = TK_LE;
|
|
sl@0
|
181 |
return 2;
|
|
sl@0
|
182 |
}else if( c=='>' ){
|
|
sl@0
|
183 |
*tokenType = TK_NE;
|
|
sl@0
|
184 |
return 2;
|
|
sl@0
|
185 |
}else if( c=='<' ){
|
|
sl@0
|
186 |
*tokenType = TK_LSHIFT;
|
|
sl@0
|
187 |
return 2;
|
|
sl@0
|
188 |
}else{
|
|
sl@0
|
189 |
*tokenType = TK_LT;
|
|
sl@0
|
190 |
return 1;
|
|
sl@0
|
191 |
}
|
|
sl@0
|
192 |
}
|
|
sl@0
|
193 |
case '>': {
|
|
sl@0
|
194 |
if( (c=z[1])=='=' ){
|
|
sl@0
|
195 |
*tokenType = TK_GE;
|
|
sl@0
|
196 |
return 2;
|
|
sl@0
|
197 |
}else if( c=='>' ){
|
|
sl@0
|
198 |
*tokenType = TK_RSHIFT;
|
|
sl@0
|
199 |
return 2;
|
|
sl@0
|
200 |
}else{
|
|
sl@0
|
201 |
*tokenType = TK_GT;
|
|
sl@0
|
202 |
return 1;
|
|
sl@0
|
203 |
}
|
|
sl@0
|
204 |
}
|
|
sl@0
|
205 |
case '!': {
|
|
sl@0
|
206 |
if( z[1]!='=' ){
|
|
sl@0
|
207 |
*tokenType = TK_ILLEGAL;
|
|
sl@0
|
208 |
return 2;
|
|
sl@0
|
209 |
}else{
|
|
sl@0
|
210 |
*tokenType = TK_NE;
|
|
sl@0
|
211 |
return 2;
|
|
sl@0
|
212 |
}
|
|
sl@0
|
213 |
}
|
|
sl@0
|
214 |
case '|': {
|
|
sl@0
|
215 |
if( z[1]!='|' ){
|
|
sl@0
|
216 |
*tokenType = TK_BITOR;
|
|
sl@0
|
217 |
return 1;
|
|
sl@0
|
218 |
}else{
|
|
sl@0
|
219 |
*tokenType = TK_CONCAT;
|
|
sl@0
|
220 |
return 2;
|
|
sl@0
|
221 |
}
|
|
sl@0
|
222 |
}
|
|
sl@0
|
223 |
case ',': {
|
|
sl@0
|
224 |
*tokenType = TK_COMMA;
|
|
sl@0
|
225 |
return 1;
|
|
sl@0
|
226 |
}
|
|
sl@0
|
227 |
case '&': {
|
|
sl@0
|
228 |
*tokenType = TK_BITAND;
|
|
sl@0
|
229 |
return 1;
|
|
sl@0
|
230 |
}
|
|
sl@0
|
231 |
case '~': {
|
|
sl@0
|
232 |
*tokenType = TK_BITNOT;
|
|
sl@0
|
233 |
return 1;
|
|
sl@0
|
234 |
}
|
|
sl@0
|
235 |
case '`':
|
|
sl@0
|
236 |
case '\'':
|
|
sl@0
|
237 |
case '"': {
|
|
sl@0
|
238 |
int delim = z[0];
|
|
sl@0
|
239 |
for(i=1; (c=z[i])!=0; i++){
|
|
sl@0
|
240 |
if( c==delim ){
|
|
sl@0
|
241 |
if( z[i+1]==delim ){
|
|
sl@0
|
242 |
i++;
|
|
sl@0
|
243 |
}else{
|
|
sl@0
|
244 |
break;
|
|
sl@0
|
245 |
}
|
|
sl@0
|
246 |
}
|
|
sl@0
|
247 |
}
|
|
sl@0
|
248 |
if( c=='\'' ){
|
|
sl@0
|
249 |
*tokenType = TK_STRING;
|
|
sl@0
|
250 |
return i+1;
|
|
sl@0
|
251 |
}else if( c!=0 ){
|
|
sl@0
|
252 |
*tokenType = TK_ID;
|
|
sl@0
|
253 |
return i+1;
|
|
sl@0
|
254 |
}else{
|
|
sl@0
|
255 |
*tokenType = TK_ILLEGAL;
|
|
sl@0
|
256 |
return i;
|
|
sl@0
|
257 |
}
|
|
sl@0
|
258 |
}
|
|
sl@0
|
259 |
case '.': {
|
|
sl@0
|
260 |
#ifndef SQLITE_OMIT_FLOATING_POINT
|
|
sl@0
|
261 |
if( !isdigit(z[1]) )
|
|
sl@0
|
262 |
#endif
|
|
sl@0
|
263 |
{
|
|
sl@0
|
264 |
*tokenType = TK_DOT;
|
|
sl@0
|
265 |
return 1;
|
|
sl@0
|
266 |
}
|
|
sl@0
|
267 |
/* If the next character is a digit, this is a floating point
|
|
sl@0
|
268 |
** number that begins with ".". Fall thru into the next case */
|
|
sl@0
|
269 |
}
|
|
sl@0
|
270 |
case '0': case '1': case '2': case '3': case '4':
|
|
sl@0
|
271 |
case '5': case '6': case '7': case '8': case '9': {
|
|
sl@0
|
272 |
*tokenType = TK_INTEGER;
|
|
sl@0
|
273 |
for(i=0; isdigit(z[i]); i++){}
|
|
sl@0
|
274 |
#ifndef SQLITE_OMIT_FLOATING_POINT
|
|
sl@0
|
275 |
if( z[i]=='.' ){
|
|
sl@0
|
276 |
i++;
|
|
sl@0
|
277 |
while( isdigit(z[i]) ){ i++; }
|
|
sl@0
|
278 |
*tokenType = TK_FLOAT;
|
|
sl@0
|
279 |
}
|
|
sl@0
|
280 |
if( (z[i]=='e' || z[i]=='E') &&
|
|
sl@0
|
281 |
( isdigit(z[i+1])
|
|
sl@0
|
282 |
|| ((z[i+1]=='+' || z[i+1]=='-') && isdigit(z[i+2]))
|
|
sl@0
|
283 |
)
|
|
sl@0
|
284 |
){
|
|
sl@0
|
285 |
i += 2;
|
|
sl@0
|
286 |
while( isdigit(z[i]) ){ i++; }
|
|
sl@0
|
287 |
*tokenType = TK_FLOAT;
|
|
sl@0
|
288 |
}
|
|
sl@0
|
289 |
#endif
|
|
sl@0
|
290 |
while( IdChar(z[i]) ){
|
|
sl@0
|
291 |
*tokenType = TK_ILLEGAL;
|
|
sl@0
|
292 |
i++;
|
|
sl@0
|
293 |
}
|
|
sl@0
|
294 |
return i;
|
|
sl@0
|
295 |
}
|
|
sl@0
|
296 |
case '[': {
|
|
sl@0
|
297 |
for(i=1, c=z[0]; c!=']' && (c=z[i])!=0; i++){}
|
|
sl@0
|
298 |
*tokenType = c==']' ? TK_ID : TK_ILLEGAL;
|
|
sl@0
|
299 |
return i;
|
|
sl@0
|
300 |
}
|
|
sl@0
|
301 |
case '?': {
|
|
sl@0
|
302 |
*tokenType = TK_VARIABLE;
|
|
sl@0
|
303 |
for(i=1; isdigit(z[i]); i++){}
|
|
sl@0
|
304 |
return i;
|
|
sl@0
|
305 |
}
|
|
sl@0
|
306 |
case '#': {
|
|
sl@0
|
307 |
for(i=1; isdigit(z[i]); i++){}
|
|
sl@0
|
308 |
if( i>1 ){
|
|
sl@0
|
309 |
/* Parameters of the form #NNN (where NNN is a number) are used
|
|
sl@0
|
310 |
** internally by sqlite3NestedParse. */
|
|
sl@0
|
311 |
*tokenType = TK_REGISTER;
|
|
sl@0
|
312 |
return i;
|
|
sl@0
|
313 |
}
|
|
sl@0
|
314 |
/* Fall through into the next case if the '#' is not followed by
|
|
sl@0
|
315 |
** a digit. Try to match #AAAA where AAAA is a parameter name. */
|
|
sl@0
|
316 |
}
|
|
sl@0
|
317 |
#ifndef SQLITE_OMIT_TCL_VARIABLE
|
|
sl@0
|
318 |
case '$':
|
|
sl@0
|
319 |
#endif
|
|
sl@0
|
320 |
case '@': /* For compatibility with MS SQL Server */
|
|
sl@0
|
321 |
case ':': {
|
|
sl@0
|
322 |
int n = 0;
|
|
sl@0
|
323 |
*tokenType = TK_VARIABLE;
|
|
sl@0
|
324 |
for(i=1; (c=z[i])!=0; i++){
|
|
sl@0
|
325 |
if( IdChar(c) ){
|
|
sl@0
|
326 |
n++;
|
|
sl@0
|
327 |
#ifndef SQLITE_OMIT_TCL_VARIABLE
|
|
sl@0
|
328 |
}else if( c=='(' && n>0 ){
|
|
sl@0
|
329 |
do{
|
|
sl@0
|
330 |
i++;
|
|
sl@0
|
331 |
}while( (c=z[i])!=0 && !isspace(c) && c!=')' );
|
|
sl@0
|
332 |
if( c==')' ){
|
|
sl@0
|
333 |
i++;
|
|
sl@0
|
334 |
}else{
|
|
sl@0
|
335 |
*tokenType = TK_ILLEGAL;
|
|
sl@0
|
336 |
}
|
|
sl@0
|
337 |
break;
|
|
sl@0
|
338 |
}else if( c==':' && z[i+1]==':' ){
|
|
sl@0
|
339 |
i++;
|
|
sl@0
|
340 |
#endif
|
|
sl@0
|
341 |
}else{
|
|
sl@0
|
342 |
break;
|
|
sl@0
|
343 |
}
|
|
sl@0
|
344 |
}
|
|
sl@0
|
345 |
if( n==0 ) *tokenType = TK_ILLEGAL;
|
|
sl@0
|
346 |
return i;
|
|
sl@0
|
347 |
}
|
|
sl@0
|
348 |
#ifndef SQLITE_OMIT_BLOB_LITERAL
|
|
sl@0
|
349 |
case 'x': case 'X': {
|
|
sl@0
|
350 |
if( z[1]=='\'' ){
|
|
sl@0
|
351 |
*tokenType = TK_BLOB;
|
|
sl@0
|
352 |
for(i=2; (c=z[i])!=0 && c!='\''; i++){
|
|
sl@0
|
353 |
if( !isxdigit(c) ){
|
|
sl@0
|
354 |
*tokenType = TK_ILLEGAL;
|
|
sl@0
|
355 |
}
|
|
sl@0
|
356 |
}
|
|
sl@0
|
357 |
if( i%2 || !c ) *tokenType = TK_ILLEGAL;
|
|
sl@0
|
358 |
if( c ) i++;
|
|
sl@0
|
359 |
return i;
|
|
sl@0
|
360 |
}
|
|
sl@0
|
361 |
/* Otherwise fall through to the next case */
|
|
sl@0
|
362 |
}
|
|
sl@0
|
363 |
#endif
|
|
sl@0
|
364 |
default: {
|
|
sl@0
|
365 |
if( !IdChar(*z) ){
|
|
sl@0
|
366 |
break;
|
|
sl@0
|
367 |
}
|
|
sl@0
|
368 |
for(i=1; IdChar(z[i]); i++){}
|
|
sl@0
|
369 |
*tokenType = keywordCode((char*)z, i);
|
|
sl@0
|
370 |
return i;
|
|
sl@0
|
371 |
}
|
|
sl@0
|
372 |
}
|
|
sl@0
|
373 |
*tokenType = TK_ILLEGAL;
|
|
sl@0
|
374 |
return 1;
|
|
sl@0
|
375 |
}
|
|
sl@0
|
376 |
|
|
sl@0
|
377 |
/*
|
|
sl@0
|
378 |
** Run the parser on the given SQL string. The parser structure is
|
|
sl@0
|
379 |
** passed in. An SQLITE_ status code is returned. If an error occurs
|
|
sl@0
|
380 |
** then an and attempt is made to write an error message into
|
|
sl@0
|
381 |
** memory obtained from sqlite3_malloc() and to make *pzErrMsg point to that
|
|
sl@0
|
382 |
** error message.
|
|
sl@0
|
383 |
*/
|
|
sl@0
|
384 |
int sqlite3RunParser(Parse *pParse, const char *zSql, char **pzErrMsg){
|
|
sl@0
|
385 |
int nErr = 0;
|
|
sl@0
|
386 |
int i;
|
|
sl@0
|
387 |
void *pEngine;
|
|
sl@0
|
388 |
int tokenType;
|
|
sl@0
|
389 |
int lastTokenParsed = -1;
|
|
sl@0
|
390 |
sqlite3 *db = pParse->db;
|
|
sl@0
|
391 |
int mxSqlLen = db->aLimit[SQLITE_LIMIT_SQL_LENGTH];
|
|
sl@0
|
392 |
|
|
sl@0
|
393 |
if( db->activeVdbeCnt==0 ){
|
|
sl@0
|
394 |
db->u1.isInterrupted = 0;
|
|
sl@0
|
395 |
}
|
|
sl@0
|
396 |
pParse->rc = SQLITE_OK;
|
|
sl@0
|
397 |
pParse->zTail = pParse->zSql = zSql;
|
|
sl@0
|
398 |
i = 0;
|
|
sl@0
|
399 |
assert( pzErrMsg!=0 );
|
|
sl@0
|
400 |
pEngine = sqlite3ParserAlloc((void*(*)(size_t))sqlite3Malloc);
|
|
sl@0
|
401 |
if( pEngine==0 ){
|
|
sl@0
|
402 |
db->mallocFailed = 1;
|
|
sl@0
|
403 |
return SQLITE_NOMEM;
|
|
sl@0
|
404 |
}
|
|
sl@0
|
405 |
assert( pParse->sLastToken.dyn==0 );
|
|
sl@0
|
406 |
assert( pParse->pNewTable==0 );
|
|
sl@0
|
407 |
assert( pParse->pNewTrigger==0 );
|
|
sl@0
|
408 |
assert( pParse->nVar==0 );
|
|
sl@0
|
409 |
assert( pParse->nVarExpr==0 );
|
|
sl@0
|
410 |
assert( pParse->nVarExprAlloc==0 );
|
|
sl@0
|
411 |
assert( pParse->apVarExpr==0 );
|
|
sl@0
|
412 |
while( !db->mallocFailed && zSql[i]!=0 ){
|
|
sl@0
|
413 |
assert( i>=0 );
|
|
sl@0
|
414 |
pParse->sLastToken.z = (u8*)&zSql[i];
|
|
sl@0
|
415 |
assert( pParse->sLastToken.dyn==0 );
|
|
sl@0
|
416 |
pParse->sLastToken.n = sqlite3GetToken((unsigned char*)&zSql[i],&tokenType);
|
|
sl@0
|
417 |
i += pParse->sLastToken.n;
|
|
sl@0
|
418 |
if( i>mxSqlLen ){
|
|
sl@0
|
419 |
pParse->rc = SQLITE_TOOBIG;
|
|
sl@0
|
420 |
break;
|
|
sl@0
|
421 |
}
|
|
sl@0
|
422 |
switch( tokenType ){
|
|
sl@0
|
423 |
case TK_SPACE: {
|
|
sl@0
|
424 |
if( db->u1.isInterrupted ){
|
|
sl@0
|
425 |
pParse->rc = SQLITE_INTERRUPT;
|
|
sl@0
|
426 |
sqlite3SetString(pzErrMsg, db, "interrupt");
|
|
sl@0
|
427 |
goto abort_parse;
|
|
sl@0
|
428 |
}
|
|
sl@0
|
429 |
break;
|
|
sl@0
|
430 |
}
|
|
sl@0
|
431 |
case TK_ILLEGAL: {
|
|
sl@0
|
432 |
sqlite3DbFree(db, *pzErrMsg);
|
|
sl@0
|
433 |
*pzErrMsg = sqlite3MPrintf(db, "unrecognized token: \"%T\"",
|
|
sl@0
|
434 |
&pParse->sLastToken);
|
|
sl@0
|
435 |
nErr++;
|
|
sl@0
|
436 |
goto abort_parse;
|
|
sl@0
|
437 |
}
|
|
sl@0
|
438 |
case TK_SEMI: {
|
|
sl@0
|
439 |
pParse->zTail = &zSql[i];
|
|
sl@0
|
440 |
/* Fall thru into the default case */
|
|
sl@0
|
441 |
}
|
|
sl@0
|
442 |
default: {
|
|
sl@0
|
443 |
sqlite3Parser(pEngine, tokenType, pParse->sLastToken, pParse);
|
|
sl@0
|
444 |
lastTokenParsed = tokenType;
|
|
sl@0
|
445 |
if( pParse->rc!=SQLITE_OK ){
|
|
sl@0
|
446 |
goto abort_parse;
|
|
sl@0
|
447 |
}
|
|
sl@0
|
448 |
break;
|
|
sl@0
|
449 |
}
|
|
sl@0
|
450 |
}
|
|
sl@0
|
451 |
}
|
|
sl@0
|
452 |
abort_parse:
|
|
sl@0
|
453 |
if( zSql[i]==0 && nErr==0 && pParse->rc==SQLITE_OK ){
|
|
sl@0
|
454 |
if( lastTokenParsed!=TK_SEMI ){
|
|
sl@0
|
455 |
sqlite3Parser(pEngine, TK_SEMI, pParse->sLastToken, pParse);
|
|
sl@0
|
456 |
pParse->zTail = &zSql[i];
|
|
sl@0
|
457 |
}
|
|
sl@0
|
458 |
sqlite3Parser(pEngine, 0, pParse->sLastToken, pParse);
|
|
sl@0
|
459 |
}
|
|
sl@0
|
460 |
#ifdef YYTRACKMAXSTACKDEPTH
|
|
sl@0
|
461 |
sqlite3StatusSet(SQLITE_STATUS_PARSER_STACK,
|
|
sl@0
|
462 |
sqlite3ParserStackPeak(pEngine)
|
|
sl@0
|
463 |
);
|
|
sl@0
|
464 |
#endif /* YYDEBUG */
|
|
sl@0
|
465 |
sqlite3ParserFree(pEngine, sqlite3_free);
|
|
sl@0
|
466 |
if( db->mallocFailed ){
|
|
sl@0
|
467 |
pParse->rc = SQLITE_NOMEM;
|
|
sl@0
|
468 |
}
|
|
sl@0
|
469 |
if( pParse->rc!=SQLITE_OK && pParse->rc!=SQLITE_DONE && pParse->zErrMsg==0 ){
|
|
sl@0
|
470 |
sqlite3SetString(&pParse->zErrMsg, db, "%s", sqlite3ErrStr(pParse->rc));
|
|
sl@0
|
471 |
}
|
|
sl@0
|
472 |
if( pParse->zErrMsg ){
|
|
sl@0
|
473 |
if( *pzErrMsg==0 ){
|
|
sl@0
|
474 |
*pzErrMsg = pParse->zErrMsg;
|
|
sl@0
|
475 |
}else{
|
|
sl@0
|
476 |
sqlite3DbFree(db, pParse->zErrMsg);
|
|
sl@0
|
477 |
}
|
|
sl@0
|
478 |
pParse->zErrMsg = 0;
|
|
sl@0
|
479 |
nErr++;
|
|
sl@0
|
480 |
}
|
|
sl@0
|
481 |
if( pParse->pVdbe && pParse->nErr>0 && pParse->nested==0 ){
|
|
sl@0
|
482 |
sqlite3VdbeDelete(pParse->pVdbe);
|
|
sl@0
|
483 |
pParse->pVdbe = 0;
|
|
sl@0
|
484 |
}
|
|
sl@0
|
485 |
#ifndef SQLITE_OMIT_SHARED_CACHE
|
|
sl@0
|
486 |
if( pParse->nested==0 ){
|
|
sl@0
|
487 |
sqlite3DbFree(db, pParse->aTableLock);
|
|
sl@0
|
488 |
pParse->aTableLock = 0;
|
|
sl@0
|
489 |
pParse->nTableLock = 0;
|
|
sl@0
|
490 |
}
|
|
sl@0
|
491 |
#endif
|
|
sl@0
|
492 |
#ifndef SQLITE_OMIT_VIRTUALTABLE
|
|
sl@0
|
493 |
sqlite3DbFree(db, pParse->apVtabLock);
|
|
sl@0
|
494 |
#endif
|
|
sl@0
|
495 |
|
|
sl@0
|
496 |
if( !IN_DECLARE_VTAB ){
|
|
sl@0
|
497 |
/* If the pParse->declareVtab flag is set, do not delete any table
|
|
sl@0
|
498 |
** structure built up in pParse->pNewTable. The calling code (see vtab.c)
|
|
sl@0
|
499 |
** will take responsibility for freeing the Table structure.
|
|
sl@0
|
500 |
*/
|
|
sl@0
|
501 |
sqlite3DeleteTable(pParse->pNewTable);
|
|
sl@0
|
502 |
}
|
|
sl@0
|
503 |
|
|
sl@0
|
504 |
sqlite3DeleteTrigger(db, pParse->pNewTrigger);
|
|
sl@0
|
505 |
sqlite3DbFree(db, pParse->apVarExpr);
|
|
sl@0
|
506 |
sqlite3DbFree(db, pParse->aAlias);
|
|
sl@0
|
507 |
while( pParse->pZombieTab ){
|
|
sl@0
|
508 |
Table *p = pParse->pZombieTab;
|
|
sl@0
|
509 |
pParse->pZombieTab = p->pNextZombie;
|
|
sl@0
|
510 |
sqlite3DeleteTable(p);
|
|
sl@0
|
511 |
}
|
|
sl@0
|
512 |
if( nErr>0 && (pParse->rc==SQLITE_OK || pParse->rc==SQLITE_DONE) ){
|
|
sl@0
|
513 |
pParse->rc = SQLITE_ERROR;
|
|
sl@0
|
514 |
}
|
|
sl@0
|
515 |
return nErr;
|
|
sl@0
|
516 |
}
|