2 * Copyright (c) 1995, the EUROPAGATE consortium (see below).
4 * The EUROPAGATE consortium members are:
6 * University College Dublin
7 * Danmarks Teknologiske Videnscenter
8 * An Chomhairle Leabharlanna
9 * Consejo Superior de Investigaciones Cientificas
11 * Permission to use, copy, modify, distribute, and sell this software and
12 * its documentation, in whole or in part, for any purpose, is hereby granted,
15 * 1. This copyright and permission notice appear in all copies of the
16 * software and its documentation. Notices of copyright or attribution
17 * which appear at the beginning of any file must remain unchanged.
19 * 2. The names of EUROPAGATE or the project partners may not be used to
20 * endorse or promote products derived from this software without specific
21 * prior written permission.
23 * 3. Users of this software (implementors and gateway operators) agree to
24 * inform the EUROPAGATE consortium of their use of the software. This
25 * information will be used to evaluate the EUROPAGATE project and the
26 * software, and to plan further developments. The consortium may use
27 * the information in later publications.
29 * 4. Users of this software agree to make their best efforts, when
30 * documenting their use of the software, to acknowledge the EUROPAGATE
31 * consortium, and the role played by the software in their work.
33 * THIS SOFTWARE IS PROVIDED "AS IS" AND WITHOUT WARRANTY OF ANY KIND,
34 * EXPRESS, IMPLIED, OR OTHERWISE, INCLUDING WITHOUT LIMITATION, ANY
35 * WARRANTY OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
36 * IN NO EVENT SHALL THE EUROPAGATE CONSORTIUM OR ITS MEMBERS BE LIABLE
37 * FOR ANY SPECIAL, INCIDENTAL, INDIRECT OR CONSEQUENTIAL DAMAGES OF
38 * ANY KIND, OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA
39 * OR PROFITS, WHETHER OR NOT ADVISED OF THE POSSIBILITY OF DAMAGE, AND
40 * ON ANY THEORY OF LIABILITY, ARISING OUT OF OR IN CONNECTION WITH THE
41 * USE OR PERFORMANCE OF THIS SOFTWARE.
46 * \brief Implements CCL lexical analyzer (scanner)
48 /* CCL - lexical analysis
51 * $Id: ccltoken.c,v 1.5 2004-10-15 00:19:00 adam Exp $
55 * Revision 1.10 1995/07/11 12:28:31 adam
56 * New function: ccl_token_simple (split into simple tokens) and
57 * ccl_token_del (delete tokens).
59 * Revision 1.9 1995/05/16 09:39:28 adam
62 * Revision 1.8 1995/05/11 14:03:57 adam
63 * Changes in the reading of qualifier(s). New function: ccl_qual_fitem.
64 * New variable ccl_case_sensitive, which controls whether reserved
65 * words and field names are case sensitive or not.
67 * Revision 1.7 1995/04/19 12:11:24 adam
70 * Revision 1.6 1995/04/17 09:31:48 adam
71 * Improved handling of qualifiers. Aliases or reserved words.
73 * Revision 1.5 1995/02/23 08:32:00 adam
76 * Revision 1.3 1995/02/15 17:42:16 adam
77 * Minor changes of the api of this module. FILE* argument added
80 * Revision 1.2 1995/02/14 19:55:13 adam
81 * Header files ccl.h/cclp.h are gone! They have been merged an
82 * moved to ../include/ccl.h.
83 * Node kind(s) in ccl_rpn_node have changed names.
85 * Revision 1.1 1995/02/13 12:35:21 adam
86 * First version of CCL. Qualifiers aren't handled yet.
97 * token_cmp: Compare token with keyword(s)
98 * kw: Keyword list. Each keyword is separated by space.
100 * return: 1 if token string matches one of the keywords in list;
103 static int token_cmp (CCL_parser cclp, const char *kw, struct ccl_token *token)
105 const char *cp1 = kw;
108 int case_sensitive = cclp->ccl_case_sensitive;
110 aliases = ccl_qual_search_special(cclp->bibset, "case");
112 case_sensitive = atoi(aliases);
115 while ((cp2 = strchr (cp1, ' ')))
117 if (token->len == (size_t) (cp2-cp1))
121 if (!memcmp (cp1, token->name, token->len))
126 if (!ccl_memicmp (cp1, token->name, token->len))
133 return token->len == strlen(cp1)
134 && !memcmp (cp1, token->name, token->len);
135 return token->len == strlen(cp1) &&
136 !ccl_memicmp (cp1, token->name, token->len);
140 * ccl_token_simple: tokenize CCL raw tokens
142 struct ccl_token *ccl_token_simple (const char *command)
144 const char *cp = command;
145 struct ccl_token *first = NULL;
146 struct ccl_token *last = NULL;
150 while (*cp && strchr (" \t\r\n", *cp))
157 first = last = (struct ccl_token *)xmalloc (sizeof (*first));
163 last->next = (struct ccl_token *)xmalloc (sizeof(*first));
164 ccl_assert (last->next);
165 last->next->prev = last;
174 last->kind = CCL_TOK_EOL;
177 last->kind = CCL_TOK_TERM;
180 while (*cp && *cp != '\"')
189 while (*cp && !strchr (" \t\n\r", *cp))
194 last->kind = CCL_TOK_TERM;
202 * ccl_tokenize: tokenize CCL command string.
203 * return: CCL token list.
205 struct ccl_token *ccl_parser_tokenize (CCL_parser cclp, const char *command)
208 const unsigned char *cp = (const unsigned char *) command;
209 struct ccl_token *first = NULL;
210 struct ccl_token *last = NULL;
214 while (*cp && strchr (" \t\r\n", *cp))
221 first = last = (struct ccl_token *)xmalloc (sizeof (*first));
227 last->next = (struct ccl_token *)xmalloc (sizeof(*first));
228 ccl_assert (last->next);
229 last->next->prev = last;
233 last->name = (const char *) cp;
238 last->kind = CCL_TOK_EOL;
241 last->kind = CCL_TOK_LP;
244 last->kind = CCL_TOK_RP;
247 last->kind = CCL_TOK_COMMA;
251 last->kind = CCL_TOK_PROX;
261 if (*cp == '=' || *cp == '<' || *cp == '>')
264 last->kind = CCL_TOK_REL;
267 else if (cp[-1] == '=')
268 last->kind = CCL_TOK_EQ;
270 last->kind = CCL_TOK_REL;
273 last->kind = CCL_TOK_TERM;
274 last->name = (const char *) cp;
276 while (*cp && *cp != '\"')
285 if (!strchr ("(),%!><= \t\n\r", cp[-1]))
287 while (*cp && !strchr ("(),%!><= \t\n\r", *cp))
293 last->kind = CCL_TOK_TERM;
295 aliases = ccl_qual_search_special(cclp->bibset, "and");
297 aliases = cclp->ccl_token_and;
298 if (token_cmp (cclp, aliases, last))
299 last->kind = CCL_TOK_AND;
301 aliases = ccl_qual_search_special(cclp->bibset, "or");
303 aliases = cclp->ccl_token_or;
304 if (token_cmp (cclp, aliases, last))
305 last->kind = CCL_TOK_OR;
307 aliases = ccl_qual_search_special(cclp->bibset, "not");
309 aliases = cclp->ccl_token_not;
310 if (token_cmp (cclp, aliases, last))
311 last->kind = CCL_TOK_NOT;
313 aliases = ccl_qual_search_special(cclp->bibset, "set");
315 aliases = cclp->ccl_token_set;
317 if (token_cmp (cclp, aliases, last))
318 last->kind = CCL_TOK_SET;
324 struct ccl_token *ccl_token_add (struct ccl_token *at)
326 struct ccl_token *n = (struct ccl_token *)xmalloc (sizeof(*n));
334 n->kind = CCL_TOK_TERM;
340 struct ccl_token *ccl_tokenize (const char *command)
342 CCL_parser cclp = ccl_parser_create ();
343 struct ccl_token *list;
345 list = ccl_parser_tokenize (cclp, command);
347 ccl_parser_destroy (cclp);
352 * ccl_token_del: delete CCL tokens
354 void ccl_token_del (struct ccl_token *list)
356 struct ccl_token *list1;
366 char *ccl_strdup (const char *str)
368 int len = strlen(str);
369 char *p = (char*) xmalloc (len+1);
374 CCL_parser ccl_parser_create (void)
376 CCL_parser p = (CCL_parser)xmalloc (sizeof(*p));
379 p->look_token = NULL;
384 p->ccl_token_and = ccl_strdup("and");
385 p->ccl_token_or = ccl_strdup("or");
386 p->ccl_token_not = ccl_strdup("not andnot");
387 p->ccl_token_set = ccl_strdup("set");
388 p->ccl_case_sensitive = 1;
393 void ccl_parser_destroy (CCL_parser p)
397 xfree (p->ccl_token_and);
398 xfree (p->ccl_token_or);
399 xfree (p->ccl_token_not);
400 xfree (p->ccl_token_set);
404 void ccl_parser_set_op_and (CCL_parser p, const char *op)
408 if (p->ccl_token_and)
409 xfree (p->ccl_token_and);
410 p->ccl_token_and = ccl_strdup (op);
414 void ccl_parser_set_op_or (CCL_parser p, const char *op)
419 xfree (p->ccl_token_or);
420 p->ccl_token_or = ccl_strdup (op);
423 void ccl_parser_set_op_not (CCL_parser p, const char *op)
427 if (p->ccl_token_not)
428 xfree (p->ccl_token_not);
429 p->ccl_token_not = ccl_strdup (op);
432 void ccl_parser_set_op_set (CCL_parser p, const char *op)
436 if (p->ccl_token_set)
437 xfree (p->ccl_token_set);
438 p->ccl_token_set = ccl_strdup (op);
442 void ccl_parser_set_case (CCL_parser p, int case_sensitivity_flag)
445 p->ccl_case_sensitive = case_sensitivity_flag;