|
|
1.1 root 1: /* Parse C expressions for CCCP.
2: Copyright (C) 1986 Free Software Foundation.
3:
4: NO WARRANTY
5:
6: BECAUSE THIS PROGRAM IS LICENSED FREE OF CHARGE, WE PROVIDE ABSOLUTELY
7: NO WARRANTY, TO THE EXTENT PERMITTED BY APPLICABLE STATE LAW. EXCEPT
8: WHEN OTHERWISE STATED IN WRITING, FREE SOFTWARE FOUNDATION, INC,
9: RICHARD M. STALLMAN AND/OR OTHER PARTIES PROVIDE THIS PROGRAM "AS IS"
10: WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING,
11: BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
12: FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY
13: AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE
14: DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR
15: CORRECTION.
16:
17: IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW WILL RICHARD M.
18: STALLMAN, THE FREE SOFTWARE FOUNDATION, INC., AND/OR ANY OTHER PARTY
19: WHO MAY MODIFY AND REDISTRIBUTE THIS PROGRAM AS PERMITTED BELOW, BE
20: LIABLE TO YOU FOR DAMAGES, INCLUDING ANY LOST PROFITS, LOST MONIES, OR
21: OTHER SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE
22: USE OR INABILITY TO USE (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR
23: DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY THIRD PARTIES OR
24: A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS) THIS
25: PROGRAM, EVEN IF YOU HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH
26: DAMAGES, OR FOR ANY CLAIM BY ANY OTHER PARTY.
27:
28: GENERAL PUBLIC LICENSE TO COPY
29:
30: 1. You may copy and distribute verbatim copies of this source file
31: as you receive it, in any medium, provided that you conspicuously
32: and appropriately publish on each copy a valid copyright notice
33: "Copyright (C) 1986 Free Software Foundation"; and include
34: following the copyright notice a verbatim copy of the above disclaimer
35: of warranty and of this License.
36:
37: 2. You may modify your copy or copies of this source file or
38: any portion of it, and copy and distribute such modifications under
39: the terms of Paragraph 1 above, provided that you also do the following:
40:
41: a) cause the modified files to carry prominent notices stating
42: that you changed the files and the date of any change; and
43:
44: b) cause the whole of any work that you distribute or publish,
45: that in whole or in part contains or is a derivative of this
46: program or any part thereof, to be licensed at no charge to all
47: third parties on terms identical to those contained in this
48: License Agreement (except that you may choose to grant more extensive
49: warranty protection to some or all third parties, at your option).
50:
51: c) You may charge a distribution fee for the physical act of
52: transferring a copy, and you may at your option offer warranty
53: protection in exchange for a fee.
54:
55: Mere aggregation of another unrelated program with this program (or its
56: derivative) on a volume of a storage or distribution medium does not bring
57: the other program under the scope of these terms.
58:
59: 3. You may copy and distribute this program (or a portion or derivative
60: of it, under Paragraph 2) in object code or executable form under the terms
61: of Paragraphs 1 and 2 above provided that you also do one of the following:
62:
63: a) accompany it with the complete corresponding machine-readable
64: source code, which must be distributed under the terms of
65: Paragraphs 1 and 2 above; or,
66:
67: b) accompany it with a written offer, valid for at least three
68: years, to give any third party free (except for a nominal
69: shipping charge) a complete machine-readable copy of the
70: corresponding source code, to be distributed under the terms of
71: Paragraphs 1 and 2 above; or,
72:
73: c) accompany it with the information you received as to where the
74: corresponding source code may be obtained. (This alternative is
75: allowed only for noncommercial distribution and only if you
76: received the program in object code or executable form alone.)
77:
78: For an executable file, complete source code means all the source code for
79: all modules it contains; but, as a special exception, it need not include
80: source code for modules which are standard libraries that accompany the
81: operating system on which the executable file runs.
82:
83: 4. You may not copy, sublicense, distribute or transfer this program
84: except as expressly provided under this License Agreement. Any attempt
85: otherwise to copy, sublicense, distribute or transfer this program is void and
86: your rights to use the program under this License agreement shall be
87: automatically terminated. However, parties who have received computer
88: software programs from you with this License Agreement will not have
89: their licenses terminated so long as such parties remain in full compliance.
90:
91: In other words, you are welcome to use, share and improve this program.
92: You are forbidden to forbid anyone else to use, share and improve
93: what you give them. Help stamp out software-hoarding!
94:
95: Adapted from expread.y of GDB by Paul Rubin, July 1986.
96:
97: /* Parse a C expression from text in a string */
98:
99: %{
100: #include <setjmp.h>
101: /* #define YYDEBUG 1 */
102:
103: static int yylex ();
104: static yyerror ();
105: int expression_value;
106:
107: static jmp_buf parse_return_error;
108:
109: /* some external tables of character types */
110: extern unsigned char is_idstart[], is_idchar[];
111:
112: %}
113:
114: %union {
115: long lval;
116: int voidval;
117: char *sval;
118: }
119:
120: %type <lval> exp exp1 start
121: %token <lval> INT CHAR
122: %token <sval> NAME
123: %token <lval> ERROR
124:
125: %left ','
126: %left OR
127: %left AND
128: %left '|'
129: %left '^'
130: %left '&'
131: %left EQUAL NOTEQUAL
132: %left '<' '>' LEQ GEQ
133: %left LSH RSH
134: %left '+' '-'
135: %left '*' '/' '%'
136: %right UNARY
137:
138: %%
139:
140: start : exp1
141: { expression_value = $1; }
142: ;
143:
144: /* Expressions, including the comma operator. */
145: exp1 : exp
146: | exp1 ',' exp
147: { $$ = $3; }
148: ;
149:
150: /* Expressions, not including the comma operator. */
151: exp : '-' exp %prec UNARY
152: { $$ = - $2; }
153: | '!' exp %prec UNARY
154: { $$ = ! $2; }
155: | '~' exp %prec UNARY
156: { $$ = ~ $2; }
157: | '(' exp1 ')'
158: { $$ = $2; }
159: ;
160:
161: /* Binary operators in order of decreasing precedence. */
162: exp : exp '*' exp
163: { $$ = $1 * $3; }
164: | exp '/' exp
165: { $$ = $1 / $3; }
166: | exp '%' exp
167: { $$ = $1 % $3; }
168: | exp '+' exp
169: { $$ = $1 + $3; }
170: | exp '-' exp
171: { $$ = $1 - $3; }
172: | exp LSH exp
173: { $$ = $1 << $3; }
174: | exp RSH exp
175: { $$ = $1 >> $3; }
176: | exp EQUAL exp
177: { $$ = ($1 == $3); }
178: | exp NOTEQUAL exp
179: { $$ = ($1 != $3); }
180: | exp LEQ exp
181: { $$ = ($1 <= $3); }
182: | exp GEQ exp
183: { $$ = ($1 >= $3); }
184: | exp '<' exp
185: { $$ = ($1 < $3); }
186: | exp '>' exp
187: { $$ = ($1 > $3); }
188: | exp '&' exp
189: { $$ = ($1 & $3); }
190: | exp '^' exp
191: { $$ = ($1 ^ $3); }
192: | exp '|' exp
193: { $$ = ($1 | $3); }
194: | exp AND exp
195: { $$ = ($1 && $3); }
196: | exp OR exp
197: { $$ = ($1 || $3); }
198: | exp '?' exp ':' exp
199: { $$ = $1 ? $3 : $5; }
200: | INT
201: { $$ = yylval.lval; }
202: | CHAR
203: { $$ = yylval.lval; }
204: | NAME
205: { $$ = 0; }
206: ;
207: %%
208:
209: /* During parsing of a C expression, the pointer to the next character
210: is in this variable. */
211:
212: static char *lexptr;
213:
214: /* Take care of parsing a number (anything that starts with a digit).
215: Set yylval and return the token type; update lexptr.
216: LEN is the number of characters in it. */
217:
218: /* maybe needs to actually deal with floating point numbers */
219:
220: static int
221: parse_number (olen)
222: int olen;
223: {
224: register char *p = lexptr;
225: register long n = 0;
226: register int c;
227: register int base = 10;
228: register len = olen;
229: char *err_copy;
230:
231: extern double atof ();
232:
233: for (c = 0; c < len; c++)
234: if (p[c] == '.') {
235: /* It's a float since it contains a point. */
236: yyerror ("floating point numbers not allowed in #if expressions");
237: return ERROR;
238:
239: /* ****************
240: yylval.dval = atof (p);
241: lexptr += len;
242: return FLOAT;
243: **************** */
244: }
245:
246: if (len >= 3 && (!strncmp (p, "0x", 2) || !strncmp (p, "0X", 2))) {
247: p += 2;
248: base = 16;
249: len -= 2;
250: }
251: else if (*p == '0')
252: base = 8;
253:
254: while (len-- > 0) {
255: c = *p++;
256: n *= base;
257: if (c >= '0' && c <= '9')
258: n += c - '0';
259: else {
260: if (c >= 'A' && c <= 'Z') c += 'a' - 'A';
261: if (base == 16 && c >= 'a' && c <= 'f')
262: n += c - 'a' + 10;
263: else if (len == 0 && c == 'l')
264: ;
265: else {
266: yyerror ("Invalid number in #if expression");
267: return ERROR;
268: }
269: }
270: }
271:
272: lexptr = p;
273: yylval.lval = n;
274: return INT;
275: }
276:
277: struct token {
278: char *operator;
279: int token;
280: };
281:
282: #define NULL 0
283:
284: static struct token tokentab2[] = {
285: {"&&", AND},
286: {"||", OR},
287: {"<<", LSH},
288: {">>", RSH},
289: {"==", EQUAL},
290: {"!=", NOTEQUAL},
291: {"<=", LEQ},
292: {">=", GEQ},
293: {NULL, ERROR}
294: };
295:
296: /* Read one token, getting characters through lexptr. */
297:
298: static int
299: yylex ()
300: {
301: register int c;
302: register int namelen;
303: register char *tokstart;
304: register struct token *toktab;
305:
306: retry:
307:
308: tokstart = lexptr;
309: c = *tokstart;
310: /* See if it is a special token of length 2. */
311: for (toktab = tokentab2; toktab->operator != NULL; toktab++)
312: if (c == *toktab->operator && tokstart[1] == toktab->operator[1]) {
313: lexptr += 2;
314: return toktab->token;
315: }
316:
317: switch (c) {
318: case 0:
319: return 0;
320:
321: case ' ':
322: case '\t':
323: case '\n':
324: lexptr++;
325: goto retry;
326:
327: case '\'':
328: lexptr++;
329: c = *lexptr++;
330: if (c == '\\')
331: c = parse_escape (&lexptr);
332: yylval.lval = c;
333: c = *lexptr++;
334: if (c != '\'') {
335: yyerror ("Invalid character constant in #if");
336: return ERROR;
337: }
338:
339: return CHAR;
340:
341: case '/': /* possible comment */
342: if (*lexptr != '*')
343: return c;
344: for (;;) {
345: while (*lexptr != '\0') {
346: if (*lexptr++ == '*' && *lexptr == '/') {
347: lexptr++;
348: goto retry;
349: }
350: }
351: }
352:
353: /* some of these chars are invalid in constant expressions;
354: maybe do something about them later */
355: case '+':
356: case '-':
357: case '*':
358: case '%':
359: case '|':
360: case '&':
361: case '^':
362: case '~':
363: case '!':
364: case '@':
365: case '<':
366: case '>':
367: case '(':
368: case ')':
369: case '[':
370: case ']':
371: case '.':
372: case '?':
373: case ':':
374: case '=':
375: case '{':
376: case '}':
377: case ',':
378: lexptr++;
379: return c;
380:
381: case '"':
382: yyerror ("double quoted strings not allowed in #if expressions");
383: return ERROR;
384: }
385: if (c >= '0' && c <= '9') {
386: /* It's a number */
387: for (namelen = 0;
388: c = tokstart[namelen], is_idchar[c] || c == '.';
389: namelen++)
390: ;
391: return parse_number (namelen);
392: }
393:
394: if (!is_idstart[c]) {
395: yyerror ("Invalid token in expression");
396: return ERROR;
397: }
398:
399: /* It is a name. See how long it is. */
400:
401: for (namelen = 0; is_idchar[tokstart[namelen]]; namelen++)
402: ;
403:
404: lexptr += namelen;
405: return NAME;
406: }
407:
408:
409: /* Parse a C escape sequence. STRING_PTR points to a variable
410: containing a pointer to the string to parse. That pointer
411: is updated past the characters we use. The value of the
412: escape sequence is returned.
413:
414: A negative value means the sequence \ newline was seen,
415: which is supposed to be equivalent to nothing at all.
416:
417: If \ is followed by a null character, we return a negative
418: value and leave the string pointer pointing at the null character.
419:
420: If \ is followed by 000, we return 0 and leave the string pointer
421: after the zeros. A value of 0 does not mean end of string. */
422:
423: static int
424: parse_escape (string_ptr)
425: char **string_ptr;
426: {
427: register int c = *(*string_ptr)++;
428: switch (c)
429: {
430: case 'a':
431: return '\a';
432: case 'b':
433: return '\b';
434: case 'e':
435: return 033;
436: case 'f':
437: return '\f';
438: case 'n':
439: return '\n';
440: case 'r':
441: return '\r';
442: case 't':
443: return '\t';
444: case 'v':
445: return '\v';
446: case '\n':
447: return -2;
448: case 0:
449: (*string_ptr)--;
450: return 0;
451: case '^':
452: c = *(*string_ptr)++;
453: if (c == '\\')
454: c = parse_escape (string_ptr);
455: if (c == '?')
456: return 0177;
457: return (c & 0200) | (c & 037);
458:
459: case '0':
460: case '1':
461: case '2':
462: case '3':
463: case '4':
464: case '5':
465: case '6':
466: case '7':
467: {
468: register int i = c - '0';
469: register int count = 0;
470: while (++count < 3)
471: {
472: if ((c = *(*string_ptr)++) >= '0' && c <= '7')
473: {
474: i *= 8;
475: i += c - '0';
476: }
477: else
478: {
479: (*string_ptr)--;
480: break;
481: }
482: }
483: return i;
484: }
485: default:
486: return c;
487: }
488: }
489:
490: static
491: yyerror (s)
492: char *s;
493: {
494: error (s);
495: longjmp (parse_return_error, 1);
496: }
497:
498: /* This page contains the entry point to this file. */
499:
500: /* Parse STRING as an expression, and complain if this fails
501: to use up all of the contents of STRING. */
502: int
503: parse_c_expression (string)
504: char *string;
505: {
506: lexptr = string;
507:
508: if (lexptr == 0 || *lexptr == 0) {
509: error ("empty #if expression");
510: return 0; /* don't include the #if group */
511: }
512:
513: /* if there is some sort of scanning error, just return 0 and assume
514: the parsing routine has printed an error message somewhere.
515: there is surely a better thing to do than this. */
516: if (setjmp(parse_return_error))
517: return 0;
518:
519: if (yyparse ())
520: return 0; /* actually this is never reached
521: the way things stand. */
522: if (*lexptr)
523: error ("Junk after end of expression.");
524:
525: return expression_value; /* set by yyparse() */
526: }
527:
528: #ifdef TEST_EXP_READER
529: /* main program, for testing purposes. */
530: main()
531: {
532: int n;
533: char buf[1024];
534: extern int yydebug;
535: /*
536: yydebug = 1;
537: */
538: initialize_random_junk ();
539:
540: for (;;) {
541: printf("enter expression: ");
542: n = 0;
543: while ((buf[n] = getchar()) != '\n')
544: n++;
545: buf[n] = '\0';
546: printf("parser returned %d\n", parse_c_expression(buf));
547: }
548: }
549:
550: /* table to tell if char can be part of a C identifier. */
551: char is_idchar[256];
552: /* table to tell if char can be first char of a c identifier. */
553: char is_idstart[256];
554: /* table to tell if c is horizontal space. isspace() thinks that
555: newline is space; this is not a good idea for this program. */
556: char is_hor_space[256];
557:
558: /*
559: * initialize random junk in the hash table and maybe other places
560: */
561: initialize_random_junk()
562: {
563: register int i;
564:
565: /*
566: * Set up is_idchar and is_idstart tables. These should be
567: * faster than saying (is_alpha(c) || c == '_'), etc.
568: * Must do set up these things before calling any routines tthat
569: * refer to them.
570: */
571: for (i = 'a'; i <= 'z'; i++) {
572: ++is_idchar[i - 'a' + 'A'];
573: ++is_idchar[i];
574: ++is_idstart[i - 'a' + 'A'];
575: ++is_idstart[i];
576: }
577: for (i = '0'; i <= '9'; i++)
578: ++is_idchar[i];
579: ++is_idchar['_'];
580: ++is_idstart['_'];
581:
582: /* horizontal space table */
583: ++is_hor_space[' '];
584: ++is_hor_space['\t'];
585: }
586:
587: error (msg)
588: {
589: printf("error: %s\n", msg);
590: }
591: #endif
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.