mksyntax.c revision 20425 
1/*-
2 * Copyright (c) 1991, 1993
3 *	The Regents of the University of California.  All rights reserved.
4 *
5 * This code is derived from software contributed to Berkeley by
6 * Kenneth Almquist.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 * 1. Redistributions of source code must retain the above copyright
12 *    notice, this list of conditions and the following disclaimer.
13 * 2. Redistributions in binary form must reproduce the above copyright
14 *    notice, this list of conditions and the following disclaimer in the
15 *    documentation and/or other materials provided with the distribution.
16 * 3. All advertising materials mentioning features or use of this software
17 *    must display the following acknowledgement:
18 *	This product includes software developed by the University of
19 *	California, Berkeley and its contributors.
20 * 4. Neither the name of the University nor the names of its contributors
21 *    may be used to endorse or promote products derived from this software
22 *    without specific prior written permission.
23 *
24 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
25 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
26 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
27 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
28 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
29 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
30 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
31 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
32 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
33 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
34 * SUCH DAMAGE.
35 *
36 *	$Id: mksyntax.c,v 1.6 1996/09/01 10:20:56 peter Exp $
37 */
38
39#ifndef lint
40static char const copyright[] =
41"@(#) Copyright (c) 1991, 1993\n\
42	The Regents of the University of California.  All rights reserved.\n";
43#endif /* not lint */
44
45#ifndef lint
46static char const sccsid[] = "@(#)mksyntax.c	8.2 (Berkeley) 5/4/95";
47#endif /* not lint */
48
49/*
50 * This program creates syntax.h and syntax.c.
51 */
52
53#include <stdio.h>
54#include <string.h>
55#include "parser.h"
56
57
58struct synclass {
59	char *name;
60	char *comment;
61};
62
63/* Syntax classes */
64struct synclass synclass[] = {
65	{ "CWORD",	"character is nothing special" },
66	{ "CNL",	"newline character" },
67	{ "CBACK",	"a backslash character" },
68	{ "CSQUOTE",	"single quote" },
69	{ "CDQUOTE",	"double quote" },
70	{ "CENDQUOTE",	"a terminating quote" },
71	{ "CBQUOTE",	"backwards single quote" },
72	{ "CVAR",	"a dollar sign" },
73	{ "CENDVAR",	"a '}' character" },
74	{ "CLP",	"a left paren in arithmetic" },
75	{ "CRP",	"a right paren in arithmetic" },
76	{ "CEOF",	"end of file" },
77	{ "CCTL",	"like CWORD, except it must be escaped" },
78	{ "CSPCL",	"these terminate a word" },
79	{ NULL,		NULL }
80};
81
82
83/*
84 * Syntax classes for is_ functions.  Warning:  if you add new classes
85 * you may have to change the definition of the is_in_name macro.
86 */
87struct synclass is_entry[] = {
88	{ "ISDIGIT",	"a digit" },
89	{ "ISUPPER",	"an upper case letter" },
90	{ "ISLOWER",	"a lower case letter" },
91	{ "ISUNDER",	"an underscore" },
92	{ "ISSPECL",	"the name of a special parameter" },
93	{ NULL, 	NULL }
94};
95
96static char writer[] = "\
97/*\n\
98 * This file was generated by the mksyntax program.\n\
99 */\n\
100\n";
101
102
103static FILE *cfile;
104static FILE *hfile;
105static char *syntax[513];
106static int base;
107static int size;	/* number of values which a char variable can have */
108static int nbits;	/* number of bits in a character */
109static int digit_contig;/* true if digits are contiguous */
110
111static void filltable __P((char *));
112static void init __P((void));
113static void add __P((char *, char *));
114static void print __P((char *));
115static void output_type_macros __P((void));
116static void digit_convert __P((void));
117
118int
119main(argc, argv)
120	int argc;
121	char **argv;
122{
123	char c;
124	char d;
125	int sign;
126	int i;
127	char buf[80];
128	int pos;
129	static char digit[] = "0123456789";
130
131	/* Create output files */
132	if ((cfile = fopen("syntax.c", "w")) == NULL) {
133		perror("syntax.c");
134		exit(2);
135	}
136	if ((hfile = fopen("syntax.h", "w")) == NULL) {
137		perror("syntax.h");
138		exit(2);
139	}
140	fputs(writer, hfile);
141	fputs(writer, cfile);
142
143	/* Determine the characteristics of chars. */
144	c = -1;
145	if (c < 0)
146		sign = 1;
147	else
148		sign = 0;
149	for (nbits = 1 ; ; nbits++) {
150		d = (1 << nbits) - 1;
151		if (d == c)
152			break;
153	}
154#if 0
155	printf("%s %d bit chars\n", sign? "signed" : "unsigned", nbits);
156#endif
157	if (nbits > 9) {
158		fputs("Characters can't have more than 9 bits\n", stderr);
159		exit(2);
160	}
161	size = (1 << nbits) + 1;
162	base = 1;
163	if (sign)
164		base += 1 << (nbits - 1);
165	digit_contig = 1;
166	for (i = 0 ; i < 10 ; i++) {
167		if (digit[i] != '0' + i)
168			digit_contig = 0;
169	}
170
171	fputs("#include <sys/cdefs.h>\n", hfile);
172	fputs("#include <ctype.h>\n", hfile);
173
174	/* Generate the #define statements in the header file */
175	fputs("/* Syntax classes */\n", hfile);
176	for (i = 0 ; synclass[i].name ; i++) {
177		sprintf(buf, "#define %s %d", synclass[i].name, i);
178		fputs(buf, hfile);
179		for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07)
180			putc('\t', hfile);
181		fprintf(hfile, "/* %s */\n", synclass[i].comment);
182	}
183	putc('\n', hfile);
184	fputs("/* Syntax classes for is_ functions */\n", hfile);
185	for (i = 0 ; is_entry[i].name ; i++) {
186		sprintf(buf, "#define %s %#o", is_entry[i].name, 1 << i);
187		fputs(buf, hfile);
188		for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07)
189			putc('\t', hfile);
190		fprintf(hfile, "/* %s */\n", is_entry[i].comment);
191	}
192	putc('\n', hfile);
193	fprintf(hfile, "#define SYNBASE %d\n", base);
194	fprintf(hfile, "#define PEOF %d\n\n", -base);
195	putc('\n', hfile);
196	fputs("#define BASESYNTAX (basesyntax + SYNBASE)\n", hfile);
197	fputs("#define DQSYNTAX (dqsyntax + SYNBASE)\n", hfile);
198	fputs("#define SQSYNTAX (sqsyntax + SYNBASE)\n", hfile);
199	fputs("#define ARISYNTAX (arisyntax + SYNBASE)\n", hfile);
200	putc('\n', hfile);
201	output_type_macros();		/* is_digit, etc. */
202	putc('\n', hfile);
203
204	/* Generate the syntax tables. */
205	fputs("#include \"shell.h\"\n", cfile);
206	fputs("#include \"syntax.h\"\n\n", cfile);
207	init();
208	fputs("/* syntax table used when not in quotes */\n", cfile);
209	add("\n", "CNL");
210	add("\\", "CBACK");
211	add("'", "CSQUOTE");
212	add("\"", "CDQUOTE");
213	add("`", "CBQUOTE");
214	add("$", "CVAR");
215	add("}", "CENDVAR");
216	add("<>();&| \t", "CSPCL");
217	print("basesyntax");
218	init();
219	fputs("\n/* syntax table used when in double quotes */\n", cfile);
220	add("\n", "CNL");
221	add("\\", "CBACK");
222	add("\"", "CENDQUOTE");
223	add("`", "CBQUOTE");
224	add("$", "CVAR");
225	add("}", "CENDVAR");
226	/* ':/' for tilde expansion, '-' for [a\-x] pattern ranges */
227	add("!*?[=~:/-", "CCTL");
228	print("dqsyntax");
229	init();
230	fputs("\n/* syntax table used when in single quotes */\n", cfile);
231	add("\n", "CNL");
232	add("'", "CENDQUOTE");
233	/* ':/' for tilde expansion, '-' for [a\-x] pattern ranges */
234	add("!*?[=~:/-", "CCTL");
235	print("sqsyntax");
236	init();
237	fputs("\n/* syntax table used when in arithmetic */\n", cfile);
238	add("\n", "CNL");
239	add("\\", "CBACK");
240	add("`", "CBQUOTE");
241	add("'", "CSQUOTE");
242	add("\"", "CDQUOTE");
243	add("$", "CVAR");
244	add("}", "CENDVAR");
245	add("(", "CLP");
246	add(")", "CRP");
247	print("arisyntax");
248	filltable("0");
249	fputs("\n/* character classification table */\n", cfile);
250	add("0123456789", "ISDIGIT");
251	add("abcdefghijklmnopqrstucvwxyz", "ISLOWER");
252	add("ABCDEFGHIJKLMNOPQRSTUCVWXYZ", "ISUPPER");
253	add("_", "ISUNDER");
254	add("#?$!-*@", "ISSPECL");
255	print("is_type");
256	if (! digit_contig)
257		digit_convert();
258	exit(0);
259}
260
261
262
263/*
264 * Clear the syntax table.
265 */
266
267static void
268filltable(dftval)
269	char *dftval;
270{
271	int i;
272
273	for (i = 0 ; i < size ; i++)
274		syntax[i] = dftval;
275}
276
277
278/*
279 * Initialize the syntax table with default values.
280 */
281
282static void
283init()
284{
285	filltable("CWORD");
286	syntax[0] = "CEOF";
287	syntax[base + CTLESC] = "CCTL";
288	syntax[base + CTLVAR] = "CCTL";
289	syntax[base + CTLENDVAR] = "CCTL";
290	syntax[base + CTLBACKQ] = "CCTL";
291	syntax[base + CTLBACKQ + CTLQUOTE] = "CCTL";
292	syntax[base + CTLARI] = "CCTL";
293	syntax[base + CTLENDARI] = "CCTL";
294}
295
296
297/*
298 * Add entries to the syntax table.
299 */
300
301static void
302add(p, type)
303	char *p, *type;
304{
305	while (*p)
306		syntax[*p++ + base] = type;
307}
308
309
310
311/*
312 * Output the syntax table.
313 */
314
315static void
316print(name)
317	char *name;
318{
319	int i;
320	int col;
321
322	fprintf(hfile, "extern const char %s[];\n", name);
323	fprintf(cfile, "const char %s[%d] = {\n", name, size);
324	col = 0;
325	for (i = 0 ; i < size ; i++) {
326		if (i == 0) {
327			fputs("      ", cfile);
328		} else if ((i & 03) == 0) {
329			fputs(",\n      ", cfile);
330			col = 0;
331		} else {
332			putc(',', cfile);
333			while (++col < 9 * (i & 03))
334				putc(' ', cfile);
335		}
336		fputs(syntax[i], cfile);
337		col += strlen(syntax[i]);
338	}
339	fputs("\n};\n", cfile);
340}
341
342
343
344/*
345 * Output character classification macros (e.g. is_digit).  If digits are
346 * contiguous, we can test for them quickly.
347 */
348
349static char *macro[] = {
350	"#define is_digit(c)\t((is_type+SYNBASE)[c] & ISDIGIT)",
351	"#define is_alpha(c)\t((c) != PEOF && ((c) < CTLESC || (c) > CTLENDARI) && isalpha((unsigned char) (c)))",
352	"#define is_name(c)\t((c) != PEOF && ((c) < CTLESC || (c) > CTLENDARI) && ((c) == '_' || isalpha((unsigned char) (c))))",
353	"#define is_in_name(c)\t((c) != PEOF && ((c) < CTLESC || (c) > CTLENDARI) && ((c) == '_' || isalnum((unsigned char) (c))))",
354	"#define is_special(c)\t((is_type+SYNBASE)[c] & (ISSPECL|ISDIGIT))",
355	NULL
356};
357
358static void
359output_type_macros()
360{
361	char **pp;
362
363	if (digit_contig)
364		macro[0] = "#define is_digit(c)\t((unsigned)((c) - '0') <= 9)";
365	for (pp = macro ; *pp ; pp++)
366		fprintf(hfile, "%s\n", *pp);
367	if (digit_contig)
368		fputs("#define digit_val(c)\t((c) - '0')\n", hfile);
369	else
370		fputs("#define digit_val(c)\t(digit_value[c])\n", hfile);
371}
372
373
374
375/*
376 * Output digit conversion table (if digits are not contiguous).
377 */
378
379static void
380digit_convert()
381{
382	int maxdigit;
383	static char digit[] = "0123456789";
384	char *p;
385	int i;
386
387	maxdigit = 0;
388	for (p = digit ; *p ; p++)
389		if (*p > maxdigit)
390			maxdigit = *p;
391	fputs("extern const char digit_value[];\n", hfile);
392	fputs("\n\nconst char digit_value[] = {\n", cfile);
393	for (i = 0 ; i <= maxdigit ; i++) {
394		for (p = digit ; *p && *p != i ; p++);
395		if (*p == '\0')
396			p = digit;
397		fprintf(cfile, "      %d,\n", p - digit);
398	}
399	fputs("};\n", cfile);
400}
401