6be2a4e6d3c1c0366314b0b980c4e78476f25b8c
3 * $Id: anag.c,v 1.5 2002/08/11 12:58:09 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 1.5 2002/08/11 12:58:09 mdw
33 * Added support for regular expression matching, if supported by the C
36 * Revision 1.4 2001/02/19 19:18:50 mdw
37 * Minor big fixes to parser.
39 * Revision 1.3 2001/02/16 21:45:19 mdw
40 * Be more helpful. Improve full help message. Special-case error for
41 * empty command strings.
43 * Revision 1.2 2001/02/07 09:09:11 mdw
44 * Fix spurious error when `-file' is used.
46 * Revision 1.1 2001/02/04 17:14:42 mdw
51 /*----- Header files ------------------------------------------------------*/
55 /*----- Static variables --------------------------------------------------*/
57 static const char *file
= DICTIONARY
;
59 /*----- Help text functions -----------------------------------------------*/
61 static void usage(FILE *fp
)
63 pquis(fp
, "Usage: $ [-f file] expression\n");
66 static void version(FILE *fp
)
68 pquis(fp
, "$, version " VERSION
"\n");
71 static void help(FILE *fp
)
77 Searches a wordlist, printing all of the words which match an expression.\n\
79 Options supported are:\n\
81 -h, --help display this help text\n\
82 -v, --version display the program's version number\n\
83 -u, --usage display a very brief usage message\n\
84 -f, --file FILE read wordlist from FILE, not `" DICTIONARY
"'\n\
86 The basic tests in the expression are:\n\
88 -anagram WORD matches a full-length anagram\n\
89 -subgram WORD matches words which only use letters in WORD\n\
90 -wildcard PATTERN matches with wildcards `*' and `?'\n\
91 -trackword WORD matches words which can be found in a trackword\n\
95 -regexp REGEXP matches with an (extended) regular expression\n\
100 These simple tests can be combined using the operators `-a', `-o' and `-n'\n\
101 (for `and', `or' and `not'; they may also be written `&', `|' and `!' if\n\
102 you like), and grouped using parentheses `(' and `)'.\n\
106 /*----- The options parser ------------------------------------------------*/
108 /* --- Options table structure --- */
118 O_HELP
, O_VERSION
, O_USAGE
,
120 O_AND
, O_OR
, O_NOT
, O_LPAREN
, O_RPAREN
,
121 O_ANAG
, O_SUBG
, O_WILD
, O_TRACK
, O_REGEXP
,
127 static const struct opt opttab
[] = {
129 /* --- Options -- don't form part of the language --- */
131 { "help", 0, OF_SHORT
, O_HELP
},
132 { "version", 0, OF_SHORT
, O_VERSION
},
133 { "usage", 0, OF_SHORT
, O_USAGE
},
134 { "file", 1, OF_SHORT
, O_FILE
},
136 /* --- Operators -- provide the basic structure of the language --- *
138 * These are also given magical names by the parser.
141 { "and", 0, OF_SHORT
, O_AND
},
142 { "or", 0, OF_SHORT
, O_OR
},
143 { "not", 0, OF_SHORT
, O_NOT
},
145 /* --- Actual matching oeprations -- do something useful --- */
147 { "anagram", 1, 0, O_ANAG
},
148 { "subgram", 1, 0, O_SUBG
},
149 { "wildcard", 1, 0, O_WILD
},
150 { "trackword", 1, 0, O_TRACK
},
152 { "regexp", 1, 0, O_REGEXP
},
155 /* --- End marker --- */
161 static const char *const *av
;
164 /* --- @nextopt@ --- *
166 * Arguments: @const char ***arg@ = where to store the arg pointer
168 * Returns: The tag of the next option.
170 * Use: Scans the next option off the command line. If the option
171 * doesn't form part of the language, it's processed internally,
172 * and you'll never see it from here. On exit, the @arg@
173 * pointer is set to contain the address of the option scanned,
174 * followed by its arguments if any. You're expected to know
175 * how many arguments there are for your option.
178 static unsigned nextopt(const char *const **arg
)
181 const struct opt
*o
, *oo
;
185 /* --- Pick the next option off the front --- */
192 /* --- Cope with various forms of magic --- */
195 if (!p
[1]) switch (*p
) {
196 case '&': return (O_AND
);
197 case '|': return (O_OR
);
198 case '!': return (O_NOT
);
199 case '(': return (O_LPAREN
);
200 case ')': return (O_RPAREN
);
205 /* --- Now cope with other sorts of weirdies --- *
207 * By the end of this, a leading `-' or `--' will have been stripped.
217 die("syntax error near `--': rubbish at end of line");
221 /* --- Now look the word up in my table --- */
225 for (o
= opttab
; o
->name
; o
++) {
226 if (strncmp(p
, o
->name
, sz
) == 0) {
227 if (strlen(o
->name
) == sz
|| ((o
->f
& OF_SHORT
) && sz
== 1)) {
232 die("ambiguous option name `-%s' (could match `-%s' or `-%s')",
233 p
, oo
->name
, o
->name
);
239 die("unrecognized option name `-%s'", p
);
241 /* --- Sort out the arguments --- */
243 if (ai
+ oo
->nargs
> ac
)
244 die("too few arguments for `-%s' (need %u)", oo
->name
, oo
->nargs
);
247 /* --- Now process the option --- */
267 die("syntax error near `%s': unknown token type", av
[ai
- 1]);
271 /*----- Node types for operators ------------------------------------------*/
273 /* --- Node structures --- */
275 typedef struct node_bin
{
281 typedef struct node_un
{
286 /* --- Node functions --- */
288 static int n_or(node
*nn
, const char *p
, size_t sz
)
290 node_bin
*n
= (node_bin
*)nn
;
291 return (n
->left
->func(n
->left
, p
, sz
) || n
->right
->func(n
->right
, p
, sz
));
294 static int n_and(node
*nn
, const char *p
, size_t sz
)
296 node_bin
*n
= (node_bin
*)nn
;
297 return (n
->left
->func(n
->left
, p
, sz
) && n
->right
->func(n
->right
, p
, sz
));
300 static int n_not(node
*nn
, const char *p
, size_t sz
)
302 node_un
*n
= (node_un
*)nn
;
303 return (!n
->arg
->func(n
->arg
, p
, sz
));
306 /*----- Parser for the expression syntax ----------------------------------*/
308 /* --- A parser context --- */
310 typedef struct p_ctx
{
312 const char *const *a
;
315 /* --- Parser structure --- *
317 * This is a simple recursive descent parser. The context retains
318 * information about the current token. Each function is passed the address
319 * of a node pointer to fill in. This simplifies the binary operator code
320 * somewhat, relative to returning pointers to node trees.
323 static void p_expr(p_ctx
*p
, node
**/
*nn*/
);
325 static void p_next(p_ctx
*p
)
327 static const char *const eof
[] = { "<end>", 0 };
328 p
->t
= nextopt(&p
->a
);
333 static void p_factor(p_ctx
*p
, node
**nn
)
336 if (p
->t
== O_LPAREN
) {
339 if (p
->t
!= O_RPAREN
)
340 die("syntax error near `%s': missing `)'", *p
->a
);
342 } else if (p
->t
== O_NOT
) {
343 n
= xmalloc(sizeof(node_un
));
347 p_factor(p
, &n
->arg
);
350 case O_ANAG
: *nn
= anagram(p
->a
+ 1); break;
351 case O_SUBG
: *nn
= subgram(p
->a
+ 1); break;
352 case O_WILD
: *nn
= wildcard(p
->a
+ 1); break;
353 case O_TRACK
: *nn
= trackword(p
->a
+ 1); break;
355 case O_REGEXP
: *nn
= regexp(p
->a
+ 1); break;
357 default: die("syntax error near `%s': unexpected token", *p
->a
);
363 static void p_term(p_ctx
*p
, node
**nn
)
378 n
= xmalloc(sizeof(node_bin
));
386 static void p_expr(p_ctx
*p
, node
**nn
)
394 n
= xmalloc(sizeof(node_bin
));
402 /* --- @p_argv@ --- *
404 * Arguments: @int argc@ = number of command-line arguments
405 * @const char *const argv[]@ = vectoor of arguments
407 * Returns: A compiled node, parsed from the arguments.
409 * Use: Does the donkey-work of parsing a command-line.
412 static node
*p_argv(int argc
, const char *const argv
[])
423 pquis(stderr
, "(Run `$ --help' for more detail.)\n");
428 die("syntax error near `%s': rubbish at end of line (too many `)'s?)",
434 /*----- Main code ---------------------------------------------------------*/
438 * Arguments: @int argc@ = number of command-line arguments
439 * @char *argv[]@ = vector of argument words
441 * Returns: Zero on success, nonzero on failure.
443 * Use: Picks entries from a word list which match particular
444 * expressions. This might be of assistance to word-game types.
447 int main(int argc
, char *argv
[])
455 n
= p_argv(argc
, (const char *const *)argv
);
457 if ((fp
= fopen(file
, "r")) == 0)
458 die("error opening `%s': %s", file
, strerror(errno
));
461 if (dstr_putline(&d
, fp
) < 0)
464 for (p
= q
= d
.buf
; p
< l
; p
++) {
465 if (!isalnum((unsigned char)*p
))
467 *q
++ = tolower((unsigned char)*p
);
471 if (n
->func(n
, d
.buf
, d
.len
)) {
472 fwrite(d
.buf
, 1, d
.len
, stdout
);
477 die("error reading `%s': %s", file
, strerror(errno
));
482 /*----- That's all, folks -------------------------------------------------*/