3 * $Id: anag.c,v 1.4 2001/02/19 19:18:50 mdw Exp $
7 * (c) 2001 Mark Wooding
10 /*----- Licensing notice --------------------------------------------------*
12 * This file is part of Anag: a simple wordgame helper.
14 * Anag is free software; you can redistribute it and/or modify
15 * it under the terms of the GNU General Public License as published by
16 * the Free Software Foundation; either version 2 of the License, or
17 * (at your option) any later version.
19 * Anag is distributed in the hope that it will be useful,
20 * but WITHOUT ANY WARRANTY; without even the implied warranty of
21 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
22 * GNU General Public License for more details.
24 * You should have received a copy of the GNU General Public License
25 * along with Anag; if not, write to the Free Software Foundation,
26 * Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
29 /*----- Revision history --------------------------------------------------*
32 * Revision 1.4 2001/02/19 19:18:50 mdw
33 * Minor big fixes to parser.
35 * Revision 1.3 2001/02/16 21:45:19 mdw
36 * Be more helpful. Improve full help message. Special-case error for
37 * empty command strings.
39 * Revision 1.2 2001/02/07 09:09:11 mdw
40 * Fix spurious error when `-file' is used.
42 * Revision 1.1 2001/02/04 17:14:42 mdw
47 /*----- Header files ------------------------------------------------------*/
51 /*----- Static variables --------------------------------------------------*/
53 static const char *file = DICTIONARY;
55 /*----- Help text functions -----------------------------------------------*/
57 static void usage(FILE *fp)
59 pquis(fp, "Usage: $ [-f file] expression\n");
62 static void version(FILE *fp)
64 pquis(fp, "$, version " VERSION "\n");
67 static void help(FILE *fp)
73 Searches a wordlist, printing all of the words which match an expression.\n\
75 Options supported are:\n\
77 -h, --help display this help text\n\
78 -v, --version display the program's version number\n\
79 -u, --usage display a very brief usage message\n\
80 -f, --file FILE read wordlist from FILE, not `" DICTIONARY "'\n\
82 The basic tests in the expression are:\n\
84 -anagram WORD matches a full-length anagram\n\
85 -subgram WORD matches words which only use letters in WORD\n\
86 -wildcard PATTERN matches with wildcards `*' and `?'\n\
87 -trackword WORD matches words which can be found in a trackword\n\
89 These simple tests can be combined using the operators `-a', `-o' and `-n'\n\
90 (for `and', `or' and `not'; they may also be written `&', `|' and `!' if\n\
91 you like), and grouped using parentheses `(' and `)'.\n\
95 /*----- The options parser ------------------------------------------------*/
97 /* --- Options table structure --- */
107 O_HELP, O_VERSION, O_USAGE,
109 O_AND, O_OR, O_NOT, O_LPAREN, O_RPAREN,
110 O_ANAG, O_SUBG, O_WILD, O_TRACK,
116 static const struct opt opttab[] = {
118 /* --- Options -- don't form part of the language --- */
120 { "help", 0, OF_SHORT, O_HELP },
121 { "version", 0, OF_SHORT, O_VERSION },
122 { "usage", 0, OF_SHORT, O_USAGE },
123 { "file", 1, OF_SHORT, O_FILE },
125 /* --- Operators -- provide the basic structure of the language --- *
127 * These are also given magical names by the parser.
130 { "and", 0, OF_SHORT, O_AND },
131 { "or", 0, OF_SHORT, O_OR },
132 { "not", 0, OF_SHORT, O_NOT },
134 /* --- Actual matching oeprations -- do something useful --- */
136 { "anagram", 1, 0, O_ANAG },
137 { "subgram", 1, 0, O_SUBG },
138 { "wildcard", 1, 0, O_WILD },
139 { "trackword", 1, 0, O_TRACK },
141 /* --- End marker --- */
147 static const char *const *av;
150 /* --- @nextopt@ --- *
152 * Arguments: @const char ***arg@ = where to store the arg pointer
154 * Returns: The tag of the next option.
156 * Use: Scans the next option off the command line. If the option
157 * doesn't form part of the language, it's processed internally,
158 * and you'll never see it from here. On exit, the @arg@
159 * pointer is set to contain the address of the option scanned,
160 * followed by its arguments if any. You're expected to know
161 * how many arguments there are for your option.
164 static unsigned nextopt(const char *const **arg)
167 const struct opt *o, *oo;
171 /* --- Pick the next option off the front --- */
178 /* --- Cope with various forms of magic --- */
181 if (!p[1]) switch (*p) {
182 case '&': return (O_AND);
183 case '|': return (O_OR);
184 case '!': return (O_NOT);
185 case '(': return (O_LPAREN);
186 case ')': return (O_RPAREN);
191 /* --- Now cope with other sorts of weirdies --- *
193 * By the end of this, a leading `-' or `--' will have been stripped.
203 die("syntax error near `--': rubbish at end of line");
207 /* --- Now look the word up in my table --- */
211 for (o = opttab; o->name; o++) {
212 if (strncmp(p, o->name, sz) == 0) {
213 if (strlen(o->name) == sz || ((o->f & OF_SHORT) && sz == 1)) {
218 die("ambiguous option name `-%s' (could match `-%s' or `-%s')",
219 p, oo->name, o->name);
225 die("unrecognized option name `-%s'", p);
227 /* --- Sort out the arguments --- */
229 if (ai + oo->nargs > ac)
230 die("too few arguments for `-%s' (need %u)", oo->name, oo->nargs);
233 /* --- Now process the option --- */
253 die("syntax error near `%s': unknown token type", av[ai - 1]);
257 /*----- Node types for operators ------------------------------------------*/
259 /* --- Node structures --- */
261 typedef struct node_bin {
267 typedef struct node_un {
272 /* --- Node functions --- */
274 static int n_or(node *nn, const char *p, size_t sz)
276 node_bin *n = (node_bin *)nn;
277 return (n->left->func(n->left, p, sz) || n->right->func(n->right, p, sz));
280 static int n_and(node *nn, const char *p, size_t sz)
282 node_bin *n = (node_bin *)nn;
283 return (n->left->func(n->left, p, sz) && n->right->func(n->right, p, sz));
286 static int n_not(node *nn, const char *p, size_t sz)
288 node_un *n = (node_un *)nn;
289 return (!n->arg->func(n->arg, p, sz));
292 /*----- Parser for the expression syntax ----------------------------------*/
294 /* --- A parser context --- */
296 typedef struct p_ctx {
298 const char *const *a;
301 /* --- Parser structure --- *
303 * This is a simple recursive descent parser. The context retains
304 * information about the current token. Each function is passed the address
305 * of a node pointer to fill in. This simplifies the binary operator code
306 * somewhat, relative to returning pointers to node trees.
309 static void p_expr(p_ctx *p, node **/*nn*/);
311 static void p_next(p_ctx *p)
313 static const char *const eof[] = { "<end>", 0 };
314 p->t = nextopt(&p->a);
319 static void p_factor(p_ctx *p, node **nn)
322 if (p->t == O_LPAREN) {
325 if (p->t != O_RPAREN)
326 die("syntax error near `%s': missing `)'", *p->a);
328 } else if (p->t == O_NOT) {
329 n = xmalloc(sizeof(node_un));
333 p_factor(p, &n->arg);
336 case O_ANAG: *nn = anagram(p->a + 1); break;
337 case O_SUBG: *nn = subgram(p->a + 1); break;
338 case O_WILD: *nn = wildcard(p->a + 1); break;
339 case O_TRACK: *nn = trackword(p->a + 1); break;
340 default: die("syntax error near `%s': unexpected token", *p->a);
346 static void p_term(p_ctx *p, node **nn)
361 n = xmalloc(sizeof(node_bin));
369 static void p_expr(p_ctx *p, node **nn)
377 n = xmalloc(sizeof(node_bin));
385 /* --- @p_argv@ --- *
387 * Arguments: @int argc@ = number of command-line arguments
388 * @const char *const argv[]@ = vectoor of arguments
390 * Returns: A compiled node, parsed from the arguments.
392 * Use: Does the donkey-work of parsing a command-line.
395 static node *p_argv(int argc, const char *const argv[])
406 pquis(stderr, "(Run `$ --help' for more detail.)\n");
411 die("syntax error near `%s': rubbish at end of line (too many `)'s?)",
417 /*----- Main code ---------------------------------------------------------*/
421 * Arguments: @int argc@ = number of command-line arguments
422 * @char *argv[]@ = vector of argument words
424 * Returns: Zero on success, nonzero on failure.
426 * Use: Picks entries from a word list which match particular
427 * expressions. This might be of assistance to word-game types.
430 int main(int argc, char *argv[])
438 n = p_argv(argc, (const char *const *)argv);
440 if ((fp = fopen(file, "r")) == 0)
441 die("error opening `%s': %s", file, strerror(errno));
444 if (dstr_putline(&d, fp) < 0)
447 for (p = q = d.buf; p < l; p++) {
448 if (!isalnum((unsigned char)*p))
450 *q++ = tolower((unsigned char)*p);
454 if (n->func(n, d.buf, d.len)) {
455 fwrite(d.buf, 1, d.len, stdout);
460 die("error reading `%s': %s", file, strerror(errno));
465 /*----- That's all, folks -------------------------------------------------*/