3 * $Id: anag.c,v 1.3 2001/02/16 21:45:19 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 1.3 2001/02/16 21:45:19 mdw
33 * Be more helpful. Improve full help message. Special-case error for
34 * empty command strings.
36 * Revision 1.2 2001/02/07 09:09:11 mdw
37 * Fix spurious error when `-file' is used.
39 * Revision 1.1 2001/02/04 17:14:42 mdw
44 /*----- Header files ------------------------------------------------------*/
48 /*----- Static variables --------------------------------------------------*/
50 static const char *file
= DICTIONARY
;
52 /*----- Help text functions -----------------------------------------------*/
54 static void usage(FILE *fp
)
56 pquis(fp
, "Usage: $ [-f file] expression\n");
59 static void version(FILE *fp
)
61 pquis(fp
, "$, version " VERSION
"\n");
64 static void help(FILE *fp
)
70 Searches a wordlist, printing all of the words which match an expression.\n\
72 Options supported are:\n\
74 -h, --help display this help text\n\
75 -v, --version display the program's version number\n\
76 -u, --usage display a very brief usage message\n\
77 -f, --file FILE read wordlist from FILE, not `" DICTIONARY
"'\n\
79 The basic tests in the expression are:\n\
81 -anagram WORD matches a full-length anagram\n\
82 -subgram WORD matches words which only use letters in WORD\n\
83 -wildcard PATTERN matches with wildcards `*' and `?'\n\
84 -trackword WORD matches words which can be found in a trackword\n\
86 These simple tests can be combined using the operators `-a', `-o' and `-n'\n\
87 (for `and', `or' and `not'; they may also be written `&', `|' and `!' if\n\
88 you like), and grouped using parentheses `(' and `)'.\n\
92 /*----- The options parser ------------------------------------------------*/
94 /* --- Options table structure --- */
104 O_HELP
, O_VERSION
, O_USAGE
,
106 O_AND
, O_OR
, O_NOT
, O_LPAREN
, O_RPAREN
,
107 O_ANAG
, O_SUBG
, O_WILD
, O_TRACK
,
113 static const struct opt opttab
[] = {
115 /* --- Options -- don't form part of the language --- */
117 { "help", 0, OF_SHORT
, O_HELP
},
118 { "version", 0, OF_SHORT
, O_VERSION
},
119 { "usage", 0, OF_SHORT
, O_USAGE
},
120 { "file", 1, OF_SHORT
, O_FILE
},
122 /* --- Operators -- provide the basic structure of the language --- *
124 * These are also given magical names by the parser.
127 { "and", 0, OF_SHORT
, O_AND
},
128 { "or", 0, OF_SHORT
, O_OR
},
129 { "not", 0, OF_SHORT
, O_NOT
},
131 /* --- Actual matching oeprations -- do something useful --- */
133 { "anagram", 1, 0, O_ANAG
},
134 { "subgram", 1, 0, O_SUBG
},
135 { "wildcard", 1, 0, O_WILD
},
136 { "trackword", 1, 0, O_TRACK
},
138 /* --- End marker --- */
144 static const char *const *av
;
147 /* --- @nextopt@ --- *
149 * Arguments: @const char ***arg@ = where to store the arg pointer
151 * Returns: The tag of the next option.
153 * Use: Scans the next option off the command line. If the option
154 * doesn't form part of the language, it's processed internally,
155 * and you'll never see it from here. On exit, the @arg@
156 * pointer is set to contain the address of the option scanned,
157 * followed by its arguments if any. You're expected to know
158 * how many arguments there are for your option.
161 static unsigned nextopt(const char *const **arg
)
164 const struct opt
*o
, *oo
;
168 /* --- Pick the next option off the front --- */
175 /* --- Cope with various forms of magic --- */
178 if (!p
[1]) switch (*p
) {
179 case '&': return (O_AND
);
180 case '|': return (O_OR
);
181 case '!': return (O_NOT
);
182 case '(': return (O_LPAREN
);
183 case ')': return (O_RPAREN
);
188 /* --- Now cope with other sorts of weirdies --- *
190 * By the end of this, a leading `-' or `--' will have been stripped.
200 die("syntax error near `--': rubbish at end of line");
204 /* --- Now look the word up in my table --- */
208 for (o
= opttab
; o
->name
; o
++) {
209 if (strncmp(p
, o
->name
, sz
) == 0) {
210 if (strlen(o
->name
) == sz
|| ((o
->f
& OF_SHORT
) && sz
== 1)) {
215 die("ambiguous option name `-%s' (could match `-%s' or `-%s')",
216 p
, oo
->name
, o
->name
);
222 die("unrecognized option name `-%s'", p
);
224 /* --- Sort out the arguments --- */
226 if (ai
+ oo
->nargs
> ac
)
227 die("too few arguments for `-%s' (need %u)", oo
->name
, oo
->nargs
);
230 /* --- Now process the option --- */
250 die("syntax error near `%s': unknown token type", av
[ai
- 1]);
254 /*----- Node types for operators ------------------------------------------*/
256 /* --- Node structures --- */
258 typedef struct node_bin
{
264 typedef struct node_un
{
269 /* --- Node functions --- */
271 static int n_or(node
*nn
, const char *p
, size_t sz
)
273 node_bin
*n
= (node_bin
*)nn
;
274 return (n
->left
->func(n
->left
, p
, sz
) || n
->right
->func(n
->right
, p
, sz
));
277 static int n_and(node
*nn
, const char *p
, size_t sz
)
279 node_bin
*n
= (node_bin
*)nn
;
280 return (n
->left
->func(n
->left
, p
, sz
) && n
->right
->func(n
->right
, p
, sz
));
283 static int n_not(node
*nn
, const char *p
, size_t sz
)
285 node_un
*n
= (node_un
*)nn
;
286 return (!n
->arg
->func(n
->arg
, p
, sz
));
289 /*----- Parser for the expression syntax ----------------------------------*/
291 /* --- A parser context --- */
293 typedef struct p_ctx
{
295 const char *const *a
;
298 /* --- Parser structure --- *
300 * This is a simple recursive descent parser. The context retains
301 * information about the current token. Each function is passed the address
302 * of a node pointer to fill in. This simplifies the binary operator code
303 * somewhat, relative to returning pointers to node trees.
306 static void p_expr(p_ctx
*p
, node
**/
*nn*/
);
308 static void p_next(p_ctx
*p
)
310 static const char *const eof
[] = { "<end>", 0 };
311 p
->t
= nextopt(&p
->a
);
316 static void p_factor(p_ctx
*p
, node
**nn
)
319 if (p
->t
== O_LPAREN
) {
322 if (p
->t
!= O_RPAREN
)
323 die("syntax error near `%s': missing `('", *p
->a
);
325 } else if (p
->t
== O_NOT
) {
326 n
= xmalloc(sizeof(node_un
));
330 p_factor(p
, &n
->arg
);
333 case O_ANAG
: *nn
= anagram(p
->a
+ 1); break;
334 case O_SUBG
: *nn
= subgram(p
->a
+ 1); break;
335 case O_WILD
: *nn
= wildcard(p
->a
+ 1); break;
336 case O_TRACK
: *nn
= trackword(p
->a
+ 1); break;
337 default: die("syntax error near `%s': unexpected token", *p
->a
);
343 static void p_term(p_ctx
*p
, node
**nn
)
359 n
= xmalloc(sizeof(node_bin
));
367 static void p_expr(p_ctx
*p
, node
**nn
)
375 n
= xmalloc(sizeof(node_bin
));
383 /* --- @p_argv@ --- *
385 * Arguments: @int argc@ = number of command-line arguments
386 * @const char *const argv[]@ = vectoor of arguments
388 * Returns: A compiled node, parsed from the arguments.
390 * Use: Does the donkey-work of parsing a command-line.
393 static node
*p_argv(int argc
, const char *const argv
[])
404 pquis(stderr
, "(Run `$ --help' for more detail.)\n");
409 die("syntax error near `%s': rubbish at end of line (too many `)'s?)",
415 /*----- Main code ---------------------------------------------------------*/
419 * Arguments: @int argc@ = number of command-line arguments
420 * @char *argv[]@ = vector of argument words
422 * Returns: Zero on success, nonzero on failure.
424 * Use: Picks entries from a word list which match particular
425 * expressions. This might be of assistance to word-game types.
428 int main(int argc
, char *argv
[])
436 n
= p_argv(argc
, (const char *const *)argv
);
438 if ((fp
= fopen(file
, "r")) == 0)
439 die("error opening `%s': %s", file
, strerror(errno
));
442 if (dstr_putline(&d
, fp
) < 0)
445 for (p
= q
= d
.buf
; p
< l
; p
++) {
446 if (!isalnum((unsigned char)*p
))
448 *q
++ = tolower((unsigned char)*p
);
452 if (n
->func(n
, d
.buf
, d
.len
)) {
453 fwrite(d
.buf
, 1, d
.len
, stdout
);
458 die("error reading `%s': %s", file
, strerror(errno
));
463 /*----- That's all, folks -------------------------------------------------*/