« back to all changes in this revision
Viewing changes to drizzled/lex_input_stream.h
- browse files at revision 1014.4.8
- compare with another revision
- compare with revision 1912
- download diff
- download tarball
- view history from revision 1014.4.8
- Committer: Jay Pipes
- Date: 2009-05-16 04:21:33 UTC
- mto: This revision was merged to the branch mainline in revision 1027.
- Revision ID: jpipes@serialcoder-20090516042133-b5vud3ivppeco3zp
Split Lex_input_stream out into its own header.
- files added:
- drizzled/lex_input_stream.h
- files modified:
- drizzled/Makefile.am
added
removed
drizzled/lex_input_stream.h
1
/* -*- mode: c++; c-basic-offset: 2; indent-tabs-mode: nil; -*-
2
* vim:expandtab:shiftwidth=2:tabstop=2:smarttab:
3
*
4
* Copyright (C) 2008 Sun Microsystems
5
*
6
* This program is free software; you can redistribute it and/or modify
7
* it under the terms of the GNU General Public License as published by
8
* the Free Software Foundation; version 2 of the License.
9
*
10
* This program is distributed in the hope that it will be useful,
11
* but WITHOUT ANY WARRANTY; without even the implied warranty of
12
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13
* GNU General Public License for more details.
14
*
15
* You should have received a copy of the GNU General Public License
16
* along with this program; if not, write to the Free Software
17
* Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
18
*/
19
20
#ifndef DRIZZLED_LEX_INPUT_STREAM_H
21
#define DRIZZLED_LEX_INPUT_STREAM_H
22
23
/**
24
@brief This class represents the character input stream consumed during
25
lexical analysis.
26
27
In addition to consuming the input stream, this class performs some
28
comment pre processing, by filtering out out of bound special text
29
from the query input stream.
30
Two buffers, with pointers inside each buffers, are maintained in
31
parallel. The 'raw' buffer is the original query text, which may
32
contain out-of-bound comments. The 'cpp' (for comments pre processor)
33
is the pre-processed buffer that contains only the query text that
34
should be seen once out-of-bound data is removed.
35
*/
36
37
class Lex_input_stream
38
{
39
public:
40
Lex_input_stream(Session *session, const char* buff, unsigned int length);
41
~Lex_input_stream();
42
43
/**
44
Set the echo mode.
45
46
When echo is true, characters parsed from the raw input stream are
47
preserved. When false, characters parsed are silently ignored.
48
@param echo the echo mode.
49
*/
50
void set_echo(bool echo)
51
{
52
m_echo= echo;
53
}
54
55
/**
56
Skip binary from the input stream.
57
@param n number of bytes to accept.
58
*/
59
void skip_binary(int n)
60
{
61
if (m_echo)
62
{
63
memcpy(m_cpp_ptr, m_ptr, n);
64
m_cpp_ptr += n;
65
}
66
m_ptr += n;
67
}
68
69
/**
70
Get a character, and advance in the stream.
71
@return the next character to parse.
72
*/
73
char yyGet()
74
{
75
char c= *m_ptr++;
76
if (m_echo)
77
*m_cpp_ptr++ = c;
78
return c;
79
}
80
81
/**
82
Get the last character accepted.
83
@return the last character accepted.
84
*/
85
char yyGetLast()
86
{
87
return m_ptr[-1];
88
}
89
90
/**
91
Look at the next character to parse, but do not accept it.
92
*/
93
char yyPeek()
94
{
95
return m_ptr[0];
96
}
97
98
/**
99
Look ahead at some character to parse.
100
@param n offset of the character to look up
101
*/
102
char yyPeekn(int n)
103
{
104
return m_ptr[n];
105
}
106
107
/**
108
Cancel the effect of the last yyGet() or yySkip().
109
Note that the echo mode should not change between calls to yyGet / yySkip
110
and yyUnget. The caller is responsible for ensuring that.
111
*/
112
void yyUnget()
113
{
114
m_ptr--;
115
if (m_echo)
116
m_cpp_ptr--;
117
}
118
119
/**
120
Accept a character, by advancing the input stream.
121
*/
122
void yySkip()
123
{
124
if (m_echo)
125
*m_cpp_ptr++ = *m_ptr++;
126
else
127
m_ptr++;
128
}
129
130
/**
131
Accept multiple characters at once.
132
@param n the number of characters to accept.
133
*/
134
void yySkipn(int n)
135
{
136
if (m_echo)
137
{
138
memcpy(m_cpp_ptr, m_ptr, n);
139
m_cpp_ptr += n;
140
}
141
m_ptr += n;
142
}
143
144
/**
145
End of file indicator for the query text to parse.
146
@return true if there are no more characters to parse
147
*/
148
bool eof()
149
{
150
return (m_ptr >= m_end_of_query);
151
}
152
153
/**
154
End of file indicator for the query text to parse.
155
@param n number of characters expected
156
@return true if there are less than n characters to parse
157
*/
158
bool eof(int n)
159
{
160
return ((m_ptr + n) >= m_end_of_query);
161
}
162
163
/** Get the raw query buffer. */
164
const char *get_buf()
165
{
166
return m_buf;
167
}
168
169
/** Get the pre-processed query buffer. */
170
const char *get_cpp_buf()
171
{
172
return m_cpp_buf;
173
}
174
175
/** Get the end of the raw query buffer. */
176
const char *get_end_of_query()
177
{
178
return m_end_of_query;
179
}
180
181
/** Mark the stream position as the start of a new token. */
182
void start_token()
183
{
184
m_tok_start_prev= m_tok_start;
185
m_tok_start= m_ptr;
186
m_tok_end= m_ptr;
187
188
m_cpp_tok_start_prev= m_cpp_tok_start;
189
m_cpp_tok_start= m_cpp_ptr;
190
m_cpp_tok_end= m_cpp_ptr;
191
}
192
193
/**
194
Adjust the starting position of the current token.
195
This is used to compensate for starting whitespace.
196
*/
197
void restart_token()
198
{
199
m_tok_start= m_ptr;
200
m_cpp_tok_start= m_cpp_ptr;
201
}
202
203
/** Get the token start position, in the raw buffer. */
204
const char *get_tok_start()
205
{
206
return m_tok_start;
207
}
208
209
/** Get the token start position, in the pre-processed buffer. */
210
const char *get_cpp_tok_start()
211
{
212
return m_cpp_tok_start;
213
}
214
215
/** Get the token end position, in the raw buffer. */
216
const char *get_tok_end()
217
{
218
return m_tok_end;
219
}
220
221
/** Get the token end position, in the pre-processed buffer. */
222
const char *get_cpp_tok_end()
223
{
224
return m_cpp_tok_end;
225
}
226
227
/** Get the previous token start position, in the raw buffer. */
228
const char *get_tok_start_prev()
229
{
230
return m_tok_start_prev;
231
}
232
233
/** Get the current stream pointer, in the raw buffer. */
234
const char *get_ptr()
235
{
236
return m_ptr;
237
}
238
239
/** Get the current stream pointer, in the pre-processed buffer. */
240
const char *get_cpp_ptr()
241
{
242
return m_cpp_ptr;
243
}
244
245
/** Get the length of the current token, in the raw buffer. */
246
uint32_t yyLength()
247
{
248
/*
249
The assumption is that the lexical analyser is always 1 character ahead,
250
which the -1 account for.
251
*/
252
assert(m_ptr > m_tok_start);
253
return (uint32_t) ((m_ptr - m_tok_start) - 1);
254
}
255
256
/** Get the utf8-body string. */
257
const char *get_body_utf8_str()
258
{
259
return m_body_utf8;
260
}
261
262
/** Get the utf8-body length. */
263
uint32_t get_body_utf8_length()
264
{
265
return m_body_utf8_ptr - m_body_utf8;
266
}
267
268
void body_utf8_start(Session *session, const char *begin_ptr);
269
void body_utf8_append(const char *ptr);
270
void body_utf8_append(const char *ptr, const char *end_ptr);
271
void body_utf8_append_literal(Session *session,
272
const LEX_STRING *txt,
273
const CHARSET_INFO * const txt_cs,
274
const char *end_ptr);
275
276
/** Current thread. */
277
Session *m_session;
278
279
/** Current line number. */
280
uint32_t yylineno;
281
282
/** Length of the last token parsed. */
283
uint32_t yytoklen;
284
285
/** Interface with bison, value of the last token parsed. */
286
LEX_YYSTYPE yylval;
287
288
/** LALR(2) resolution, look ahead token.*/
289
int lookahead_token;
290
291
/** LALR(2) resolution, value of the look ahead token.*/
292
LEX_YYSTYPE lookahead_yylval;
293
294
private:
295
/** Pointer to the current position in the raw input stream. */
296
const char *m_ptr;
297
298
/** Starting position of the last token parsed, in the raw buffer. */
299
const char *m_tok_start;
300
301
/** Ending position of the previous token parsed, in the raw buffer. */
302
const char *m_tok_end;
303
304
/** End of the query text in the input stream, in the raw buffer. */
305
const char *m_end_of_query;
306
307
/** Starting position of the previous token parsed, in the raw buffer. */
308
const char *m_tok_start_prev;
309
310
/** Begining of the query text in the input stream, in the raw buffer. */
311
const char *m_buf;
312
313
/** Length of the raw buffer. */
314
uint32_t m_buf_length;
315
316
/** Echo the parsed stream to the pre-processed buffer. */
317
bool m_echo;
318
319
/** Pre-processed buffer. */
320
char *m_cpp_buf;
321
322
/** Pointer to the current position in the pre-processed input stream. */
323
char *m_cpp_ptr;
324
325
/**
326
Starting position of the last token parsed,
327
in the pre-processed buffer.
328
*/
329
const char *m_cpp_tok_start;
330
331
/**
332
Starting position of the previous token parsed,
333
in the pre-procedded buffer.
334
*/
335
const char *m_cpp_tok_start_prev;
336
337
/**
338
Ending position of the previous token parsed,
339
in the pre-processed buffer.
340
*/
341
const char *m_cpp_tok_end;
342
343
/** UTF8-body buffer created during parsing. */
344
char *m_body_utf8;
345
346
/** Pointer to the current position in the UTF8-body buffer. */
347
char *m_body_utf8_ptr;
348
349
/**
350
Position in the pre-processed buffer. The query from m_cpp_buf to
351
m_cpp_utf_processed_ptr is converted to UTF8-body.
352
*/
353
const char *m_cpp_utf8_processed_ptr;
354
355
public:
356
357
/** Current state of the lexical analyser. */
358
enum my_lex_states next_state;
359
360
/**
361
Position of ';' in the stream, to delimit multiple queries.
362
This delimiter is in the raw buffer.
363
*/
364
const char *found_semicolon;
365
366
/** Token character bitmaps, to detect 7bit strings. */
367
unsigned char tok_bitmap;
368
369
/** SQL_MODE = IGNORE_SPACE. */
370
bool ignore_space;
371
372
/** State of the lexical analyser for comments. */
373
enum_comment_state in_comment;
374
375
/**
376
Starting position of the TEXT_STRING or IDENT in the pre-processed
377
buffer.
378
379
NOTE: this member must be used within DRIZZLElex() function only.
380
*/
381
const char *m_cpp_text_start;
382
383
/**
384
Ending position of the TEXT_STRING or IDENT in the pre-processed
385
buffer.
386
387
NOTE: this member must be used within DRIZZLElex() function only.
388
*/
389
const char *m_cpp_text_end;
390
391
/**
392
Character set specified by the character-set-introducer.
393
394
NOTE: this member must be used within DRIZZLElex() function only.
395
*/
396
const CHARSET_INFO *m_underscore_cs;
397
};
398
#endif /* DRIZZLED_LEX_INPUT_STREAM_H */
Loggerhead is a web-based interface for Breezy
Version: 2.0.1