xref: /openbmc/linux/scripts/genksyms/lex.l (revision 9c1f8594)
1 /* Lexical analysis for genksyms.
2    Copyright 1996, 1997 Linux International.
3 
4    New implementation contributed by Richard Henderson <rth@tamu.edu>
5    Based on original work by Bjorn Ekwall <bj0rn@blox.se>
6 
7    Taken from Linux modutils 2.4.22.
8 
9    This program is free software; you can redistribute it and/or modify it
10    under the terms of the GNU General Public License as published by the
11    Free Software Foundation; either version 2 of the License, or (at your
12    option) any later version.
13 
14    This program is distributed in the hope that it will be useful, but
15    WITHOUT ANY WARRANTY; without even the implied warranty of
16    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
17    General Public License for more details.
18 
19    You should have received a copy of the GNU General Public License
20    along with this program; if not, write to the Free Software Foundation,
21    Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.  */
22 
23 
24 %{
25 
26 #include <limits.h>
27 #include <stdlib.h>
28 #include <string.h>
29 #include <ctype.h>
30 
31 #include "genksyms.h"
32 #include "parse.tab.h"
33 
34 /* We've got a two-level lexer here.  We let flex do basic tokenization
35    and then we categorize those basic tokens in the second stage.  */
36 #define YY_DECL		static int yylex1(void)
37 
38 %}
39 
40 IDENT			[A-Za-z_\$][A-Za-z0-9_\$]*
41 
42 O_INT			0[0-7]*
43 D_INT			[1-9][0-9]*
44 X_INT			0[Xx][0-9A-Fa-f]+
45 I_SUF			[Uu]|[Ll]|[Uu][Ll]|[Ll][Uu]
46 INT			({O_INT}|{D_INT}|{X_INT}){I_SUF}?
47 
48 FRAC			([0-9]*\.[0-9]+)|([0-9]+\.)
49 EXP			[Ee][+-]?[0-9]+
50 F_SUF			[FfLl]
51 REAL			({FRAC}{EXP}?{F_SUF}?)|([0-9]+{EXP}{F_SUF}?)
52 
53 STRING			L?\"([^\\\"]*\\.)*[^\\\"]*\"
54 CHAR			L?\'([^\\\']*\\.)*[^\\\']*\'
55 
56 MC_TOKEN		([~%^&*+=|<>/-]=)|(&&)|("||")|(->)|(<<)|(>>)
57 
58 /* We don't do multiple input files.  */
59 %option noyywrap
60 
61 %option noinput
62 
63 %%
64 
65 
66  /* Keep track of our location in the original source files.  */
67 ^#[ \t]+{INT}[ \t]+\"[^\"\n]+\".*\n	return FILENAME;
68 ^#.*\n					cur_line++;
69 \n					cur_line++;
70 
71  /* Ignore all other whitespace.  */
72 [ \t\f\v\r]+				;
73 
74 
75 {STRING}				return STRING;
76 {CHAR}					return CHAR;
77 {IDENT}					return IDENT;
78 
79  /* The Pedant requires that the other C multi-character tokens be
80     recognized as tokens.  We don't actually use them since we don't
81     parse expressions, but we do want whitespace to be arranged
82     around them properly.  */
83 {MC_TOKEN}				return OTHER;
84 {INT}					return INT;
85 {REAL}					return REAL;
86 
87 "..."					return DOTS;
88 
89  /* All other tokens are single characters.  */
90 .					return yytext[0];
91 
92 
93 %%
94 
95 /* Bring in the keyword recognizer.  */
96 
97 #include "keywords.hash.c"
98 
99 
100 /* Macros to append to our phrase collection list.  */
101 
102 /*
103  * We mark any token, that that equals to a known enumerator, as
104  * SYM_ENUM_CONST. The parser will change this for struct and union tags later,
105  * the only problem is struct and union members:
106  *    enum e { a, b }; struct s { int a, b; }
107  * but in this case, the only effect will be, that the ABI checksums become
108  * more volatile, which is acceptable. Also, such collisions are quite rare,
109  * so far it was only observed in include/linux/telephony.h.
110  */
111 #define _APP(T,L)	do {						   \
112 			  cur_node = next_node;				   \
113 			  next_node = xmalloc(sizeof(*next_node));	   \
114 			  next_node->next = cur_node;			   \
115 			  cur_node->string = memcpy(xmalloc(L+1), T, L+1); \
116 			  cur_node->tag =				   \
117 			    find_symbol(cur_node->string, SYM_ENUM_CONST, 1)?\
118 			    SYM_ENUM_CONST : SYM_NORMAL ;		   \
119 			} while (0)
120 
121 #define APP		_APP(yytext, yyleng)
122 
123 
124 /* The second stage lexer.  Here we incorporate knowledge of the state
125    of the parser to tailor the tokens that are returned.  */
126 
127 int
128 yylex(void)
129 {
130   static enum {
131     ST_NOTSTARTED, ST_NORMAL, ST_ATTRIBUTE, ST_ASM, ST_BRACKET, ST_BRACE,
132     ST_EXPRESSION, ST_TABLE_1, ST_TABLE_2, ST_TABLE_3, ST_TABLE_4,
133     ST_TABLE_5, ST_TABLE_6
134   } lexstate = ST_NOTSTARTED;
135 
136   static int suppress_type_lookup, dont_want_brace_phrase;
137   static struct string_list *next_node;
138 
139   int token, count = 0;
140   struct string_list *cur_node;
141 
142   if (lexstate == ST_NOTSTARTED)
143     {
144       next_node = xmalloc(sizeof(*next_node));
145       next_node->next = NULL;
146       lexstate = ST_NORMAL;
147     }
148 
149 repeat:
150   token = yylex1();
151 
152   if (token == 0)
153     return 0;
154   else if (token == FILENAME)
155     {
156       char *file, *e;
157 
158       /* Save the filename and line number for later error messages.  */
159 
160       if (cur_filename)
161 	free(cur_filename);
162 
163       file = strchr(yytext, '\"')+1;
164       e = strchr(file, '\"');
165       *e = '\0';
166       cur_filename = memcpy(xmalloc(e-file+1), file, e-file+1);
167       cur_line = atoi(yytext+2);
168 
169       goto repeat;
170     }
171 
172   switch (lexstate)
173     {
174     case ST_NORMAL:
175       switch (token)
176 	{
177 	case IDENT:
178 	  APP;
179 	  {
180 	    const struct resword *r = is_reserved_word(yytext, yyleng);
181 	    if (r)
182 	      {
183 		switch (token = r->token)
184 		  {
185 		  case ATTRIBUTE_KEYW:
186 		    lexstate = ST_ATTRIBUTE;
187 		    count = 0;
188 		    goto repeat;
189 		  case ASM_KEYW:
190 		    lexstate = ST_ASM;
191 		    count = 0;
192 		    goto repeat;
193 
194 		  case STRUCT_KEYW:
195 		  case UNION_KEYW:
196 		  case ENUM_KEYW:
197 		    dont_want_brace_phrase = 3;
198 		    suppress_type_lookup = 2;
199 		    goto fini;
200 
201 		  case EXPORT_SYMBOL_KEYW:
202 		      goto fini;
203 		  }
204 	      }
205 	    if (!suppress_type_lookup)
206 	      {
207 		if (find_symbol(yytext, SYM_TYPEDEF, 1))
208 		  token = TYPE;
209 	      }
210 	  }
211 	  break;
212 
213 	case '[':
214 	  APP;
215 	  lexstate = ST_BRACKET;
216 	  count = 1;
217 	  goto repeat;
218 
219 	case '{':
220 	  APP;
221 	  if (dont_want_brace_phrase)
222 	    break;
223 	  lexstate = ST_BRACE;
224 	  count = 1;
225 	  goto repeat;
226 
227 	case '=': case ':':
228 	  APP;
229 	  lexstate = ST_EXPRESSION;
230 	  break;
231 
232 	case DOTS:
233 	default:
234 	  APP;
235 	  break;
236 	}
237       break;
238 
239     case ST_ATTRIBUTE:
240       APP;
241       switch (token)
242 	{
243 	case '(':
244 	  ++count;
245 	  goto repeat;
246 	case ')':
247 	  if (--count == 0)
248 	    {
249 	      lexstate = ST_NORMAL;
250 	      token = ATTRIBUTE_PHRASE;
251 	      break;
252 	    }
253 	  goto repeat;
254 	default:
255 	  goto repeat;
256 	}
257       break;
258 
259     case ST_ASM:
260       APP;
261       switch (token)
262 	{
263 	case '(':
264 	  ++count;
265 	  goto repeat;
266 	case ')':
267 	  if (--count == 0)
268 	    {
269 	      lexstate = ST_NORMAL;
270 	      token = ASM_PHRASE;
271 	      break;
272 	    }
273 	  goto repeat;
274 	default:
275 	  goto repeat;
276 	}
277       break;
278 
279     case ST_BRACKET:
280       APP;
281       switch (token)
282 	{
283 	case '[':
284 	  ++count;
285 	  goto repeat;
286 	case ']':
287 	  if (--count == 0)
288 	    {
289 	      lexstate = ST_NORMAL;
290 	      token = BRACKET_PHRASE;
291 	      break;
292 	    }
293 	  goto repeat;
294 	default:
295 	  goto repeat;
296 	}
297       break;
298 
299     case ST_BRACE:
300       APP;
301       switch (token)
302 	{
303 	case '{':
304 	  ++count;
305 	  goto repeat;
306 	case '}':
307 	  if (--count == 0)
308 	    {
309 	      lexstate = ST_NORMAL;
310 	      token = BRACE_PHRASE;
311 	      break;
312 	    }
313 	  goto repeat;
314 	default:
315 	  goto repeat;
316 	}
317       break;
318 
319     case ST_EXPRESSION:
320       switch (token)
321 	{
322 	case '(': case '[': case '{':
323 	  ++count;
324 	  APP;
325 	  goto repeat;
326 	case '}':
327 	  /* is this the last line of an enum declaration? */
328 	  if (count == 0)
329 	    {
330 	      /* Put back the token we just read so's we can find it again
331 		 after registering the expression.  */
332 	      unput(token);
333 
334 	      lexstate = ST_NORMAL;
335 	      token = EXPRESSION_PHRASE;
336 	      break;
337 	    }
338 	  /* FALLTHRU */
339 	case ')': case ']':
340 	  --count;
341 	  APP;
342 	  goto repeat;
343 	case ',': case ';':
344 	  if (count == 0)
345 	    {
346 	      /* Put back the token we just read so's we can find it again
347 		 after registering the expression.  */
348 	      unput(token);
349 
350 	      lexstate = ST_NORMAL;
351 	      token = EXPRESSION_PHRASE;
352 	      break;
353 	    }
354 	  APP;
355 	  goto repeat;
356 	default:
357 	  APP;
358 	  goto repeat;
359 	}
360       break;
361 
362     case ST_TABLE_1:
363       goto repeat;
364 
365     case ST_TABLE_2:
366       if (token == IDENT && yyleng == 1 && yytext[0] == 'X')
367 	{
368 	  token = EXPORT_SYMBOL_KEYW;
369 	  lexstate = ST_TABLE_5;
370 	  APP;
371 	  break;
372 	}
373       lexstate = ST_TABLE_6;
374       /* FALLTHRU */
375 
376     case ST_TABLE_6:
377       switch (token)
378 	{
379 	case '{': case '[': case '(':
380 	  ++count;
381 	  break;
382 	case '}': case ']': case ')':
383 	  --count;
384 	  break;
385 	case ',':
386 	  if (count == 0)
387 	    lexstate = ST_TABLE_2;
388 	  break;
389 	};
390       goto repeat;
391 
392     case ST_TABLE_3:
393       goto repeat;
394 
395     case ST_TABLE_4:
396       if (token == ';')
397 	lexstate = ST_NORMAL;
398       goto repeat;
399 
400     case ST_TABLE_5:
401       switch (token)
402 	{
403 	case ',':
404 	  token = ';';
405 	  lexstate = ST_TABLE_2;
406 	  APP;
407 	  break;
408 	default:
409 	  APP;
410 	  break;
411 	}
412       break;
413 
414     default:
415       exit(1);
416     }
417 fini:
418 
419   if (suppress_type_lookup > 0)
420     --suppress_type_lookup;
421   if (dont_want_brace_phrase > 0)
422     --dont_want_brace_phrase;
423 
424   yylval = &next_node->next;
425 
426   return token;
427 }
428