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.
44 /* CCL - lexical analysis
47 * $Id: ccltoken.c,v 1.1 2003-10-27 12:21:30 adam Exp $
51 * Revision 1.10 1995/07/11 12:28:31 adam
52 * New function: ccl_token_simple (split into simple tokens) and
53 * ccl_token_del (delete tokens).
55 * Revision 1.9 1995/05/16 09:39:28 adam
58 * Revision 1.8 1995/05/11 14:03:57 adam
59 * Changes in the reading of qualifier(s). New function: ccl_qual_fitem.
60 * New variable ccl_case_sensitive, which controls whether reserved
61 * words and field names are case sensitive or not.
63 * Revision 1.7 1995/04/19 12:11:24 adam
66 * Revision 1.6 1995/04/17 09:31:48 adam
67 * Improved handling of qualifiers. Aliases or reserved words.
69 * Revision 1.5 1995/02/23 08:32:00 adam
72 * Revision 1.3 1995/02/15 17:42:16 adam
73 * Minor changes of the api of this module. FILE* argument added
76 * Revision 1.2 1995/02/14 19:55:13 adam
77 * Header files ccl.h/cclp.h are gone! They have been merged an
78 * moved to ../include/ccl.h.
79 * Node kind(s) in ccl_rpn_node have changed names.
81 * Revision 1.1 1995/02/13 12:35:21 adam
82 * First version of CCL. Qualifiers aren't handled yet.
93 * token_cmp: Compare token with keyword(s)
94 * kw: Keyword list. Each keyword is separated by space.
96 * return: 1 if token string matches one of the keywords in list;
99 static int token_cmp (CCL_parser cclp, const char *kw, struct ccl_token *token)
101 const char *cp1 = kw;
104 int case_sensitive = cclp->ccl_case_sensitive;
106 aliases = ccl_qual_search_special(cclp->bibset, "case");
108 case_sensitive = atoi(aliases);
111 while ((cp2 = strchr (cp1, ' ')))
113 if (token->len == (size_t) (cp2-cp1))
117 if (!memcmp (cp1, token->name, token->len))
122 if (!ccl_memicmp (cp1, token->name, token->len))
129 return token->len == strlen(cp1)
130 && !memcmp (cp1, token->name, token->len);
131 return token->len == strlen(cp1) &&
132 !ccl_memicmp (cp1, token->name, token->len);
136 * ccl_token_simple: tokenize CCL raw tokens
138 struct ccl_token *ccl_token_simple (const char *command)
140 const char *cp = command;
141 struct ccl_token *first = NULL;
142 struct ccl_token *last = NULL;
146 while (*cp && strchr (" \t\r\n", *cp))
153 first = last = (struct ccl_token *)xmalloc (sizeof (*first));
159 last->next = (struct ccl_token *)xmalloc (sizeof(*first));
160 ccl_assert (last->next);
161 last->next->prev = last;
170 last->kind = CCL_TOK_EOL;
173 last->kind = CCL_TOK_TERM;
176 while (*cp && *cp != '\"')
185 while (*cp && !strchr (" \t\n\r", *cp))
190 last->kind = CCL_TOK_TERM;
198 * ccl_tokenize: tokenize CCL command string.
199 * return: CCL token list.
201 struct ccl_token *ccl_parser_tokenize (CCL_parser cclp, const char *command)
204 const char *cp = command;
205 struct ccl_token *first = NULL;
206 struct ccl_token *last = NULL;
210 while (*cp && strchr (" \t\r\n", *cp))
217 first = last = (struct ccl_token *)xmalloc (sizeof (*first));
223 last->next = (struct ccl_token *)xmalloc (sizeof(*first));
224 ccl_assert (last->next);
225 last->next->prev = last;
234 last->kind = CCL_TOK_EOL;
237 last->kind = CCL_TOK_LP;
240 last->kind = CCL_TOK_RP;
243 last->kind = CCL_TOK_COMMA;
247 last->kind = CCL_TOK_PROX;
257 if (*cp == '=' || *cp == '<' || *cp == '>')
260 last->kind = CCL_TOK_REL;
263 else if (cp[-1] == '=')
264 last->kind = CCL_TOK_EQ;
266 last->kind = CCL_TOK_REL;
269 last->kind = CCL_TOK_TERM;
272 while (*cp && *cp != '\"')
281 if (!strchr ("(),%!><= \t\n\r", cp[-1]))
283 while (*cp && !strchr ("(),%!><= \t\n\r", *cp))
289 last->kind = CCL_TOK_TERM;
291 aliases = ccl_qual_search_special(cclp->bibset, "and");
293 aliases = cclp->ccl_token_and;
294 if (token_cmp (cclp, aliases, last))
295 last->kind = CCL_TOK_AND;
297 aliases = ccl_qual_search_special(cclp->bibset, "or");
299 aliases = cclp->ccl_token_or;
300 if (token_cmp (cclp, aliases, last))
301 last->kind = CCL_TOK_OR;
303 aliases = ccl_qual_search_special(cclp->bibset, "not");
305 aliases = cclp->ccl_token_not;
306 if (token_cmp (cclp, aliases, last))
307 last->kind = CCL_TOK_NOT;
309 aliases = ccl_qual_search_special(cclp->bibset, "set");
311 aliases = cclp->ccl_token_set;
313 if (token_cmp (cclp, aliases, last))
314 last->kind = CCL_TOK_SET;
320 struct ccl_token *ccl_tokenize (const char *command)
322 CCL_parser cclp = ccl_parser_create ();
323 struct ccl_token *list;
325 list = ccl_parser_tokenize (cclp, command);
327 ccl_parser_destroy (cclp);
332 * ccl_token_del: delete CCL tokens
334 void ccl_token_del (struct ccl_token *list)
336 struct ccl_token *list1;
346 char *ccl_strdup (const char *str)
348 int len = strlen(str);
349 char *p = (char*) xmalloc (len+1);
354 CCL_parser ccl_parser_create (void)
356 CCL_parser p = (CCL_parser)xmalloc (sizeof(*p));
359 p->look_token = NULL;
364 p->ccl_token_and = ccl_strdup("and");
365 p->ccl_token_or = ccl_strdup("or");
366 p->ccl_token_not = ccl_strdup("not andnot");
367 p->ccl_token_set = ccl_strdup("set");
368 p->ccl_case_sensitive = 1;
373 void ccl_parser_destroy (CCL_parser p)
377 xfree (p->ccl_token_and);
378 xfree (p->ccl_token_or);
379 xfree (p->ccl_token_not);
380 xfree (p->ccl_token_set);
384 void ccl_parser_set_op_and (CCL_parser p, const char *op)
388 if (p->ccl_token_and)
389 xfree (p->ccl_token_and);
390 p->ccl_token_and = ccl_strdup (op);
394 void ccl_parser_set_op_or (CCL_parser p, const char *op)
399 xfree (p->ccl_token_or);
400 p->ccl_token_or = ccl_strdup (op);
403 void ccl_parser_set_op_not (CCL_parser p, const char *op)
407 if (p->ccl_token_not)
408 xfree (p->ccl_token_not);
409 p->ccl_token_not = ccl_strdup (op);
412 void ccl_parser_set_op_set (CCL_parser p, const char *op)
416 if (p->ccl_token_set)
417 xfree (p->ccl_token_set);
418 p->ccl_token_set = ccl_strdup (op);
422 void ccl_parser_set_case (CCL_parser p, int case_sensitivity_flag)
425 p->ccl_case_sensitive = case_sensitivity_flag;