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 1.13 2007-05-01 12:54:44 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 **aliases;
106 int case_sensitive = cclp->ccl_case_sensitive;
109 aliases = ccl_qual_search_special(cclp->bibset, "case");
111 case_sensitive = atoi(aliases[0]);
113 for (i = 0; kw[i]; i++)
115 if (token->len == strlen(kw[i]))
119 if (!memcmp(kw[i], token->name, token->len))
124 if (!ccl_memicmp(kw[i], token->name, token->len))
133 * ccl_tokenize: tokenize CCL command string.
134 * return: CCL token list.
136 struct ccl_token *ccl_parser_tokenize(CCL_parser cclp, const char *command)
138 const char **aliases;
139 const unsigned char *cp = (const unsigned char *) command;
140 struct ccl_token *first = NULL;
141 struct ccl_token *last = NULL;
142 cclp->start_pos = command;
146 const unsigned char *cp0 = cp;
147 while (*cp && strchr(" \t\r\n", *cp))
151 first = last = (struct ccl_token *)xmalloc(sizeof(*first));
157 last->next = (struct ccl_token *)xmalloc(sizeof(*first));
158 ccl_assert(last->next);
159 last->next->prev = last;
162 last->ws_prefix_buf = (const char *) cp0;
163 last->ws_prefix_len = cp - cp0;
165 last->name = (const char *) cp;
170 last->kind = CCL_TOK_EOL;
173 last->kind = CCL_TOK_LP;
176 last->kind = CCL_TOK_RP;
179 last->kind = CCL_TOK_COMMA;
183 last->kind = CCL_TOK_PROX;
193 if (*cp == '=' || *cp == '<' || *cp == '>')
196 last->kind = CCL_TOK_REL;
199 else if (cp[-1] == '=')
200 last->kind = CCL_TOK_EQ;
202 last->kind = CCL_TOK_REL;
205 last->kind = CCL_TOK_TERM;
206 last->name = (const char *) cp;
208 while (*cp && *cp != '\"')
217 if (!strchr("(),%!><= \t\n\r", cp[-1]))
219 while (*cp && !strchr("(),%!><= \t\n\r", *cp))
225 last->kind = CCL_TOK_TERM;
227 aliases = ccl_qual_search_special(cclp->bibset, "and");
229 aliases = cclp->ccl_token_and;
230 if (token_cmp(cclp, aliases, last))
231 last->kind = CCL_TOK_AND;
233 aliases = ccl_qual_search_special(cclp->bibset, "or");
235 aliases = cclp->ccl_token_or;
236 if (token_cmp(cclp, aliases, last))
237 last->kind = CCL_TOK_OR;
239 aliases = ccl_qual_search_special(cclp->bibset, "not");
241 aliases = cclp->ccl_token_not;
242 if (token_cmp(cclp, aliases, last))
243 last->kind = CCL_TOK_NOT;
245 aliases = ccl_qual_search_special(cclp->bibset, "set");
247 aliases = cclp->ccl_token_set;
249 if (token_cmp(cclp, aliases, last))
250 last->kind = CCL_TOK_SET;
256 struct ccl_token *ccl_token_add(struct ccl_token *at)
258 struct ccl_token *n = (struct ccl_token *)xmalloc(sizeof(*n));
266 n->kind = CCL_TOK_TERM;
269 n->ws_prefix_buf = 0;
270 n->ws_prefix_len = 0;
275 * ccl_token_del: delete CCL tokens
277 void ccl_token_del(struct ccl_token *list)
279 struct ccl_token *list1;
289 static const char **create_ar(const char *v1, const char *v2)
291 const char **a = xmalloc(3 * sizeof(*a));
303 static void destroy_ar(const char **a)
308 for (i = 0; a[i]; i++)
309 xfree((char *) a[i]);
314 CCL_parser ccl_parser_create(CCL_bibset bibset)
316 CCL_parser p = (CCL_parser)xmalloc(sizeof(*p));
319 p->look_token = NULL;
324 p->ccl_token_and = create_ar("and", 0);
325 p->ccl_token_or = create_ar("or", 0);
326 p->ccl_token_not = create_ar("not", "andnot");
327 p->ccl_token_set = create_ar("set", 0);
328 p->ccl_case_sensitive = 1;
333 void ccl_parser_destroy(CCL_parser p)
337 destroy_ar(p->ccl_token_and);
338 destroy_ar(p->ccl_token_or);
339 destroy_ar(p->ccl_token_not);
340 destroy_ar(p->ccl_token_set);
344 void ccl_parser_set_case(CCL_parser p, int case_sensitivity_flag)
347 p->ccl_case_sensitive = case_sensitivity_flag;
350 int ccl_parser_get_error(CCL_parser cclp, int *pos)
352 if (pos && cclp->error_code)
353 *pos = cclp->error_pos - cclp->start_pos;
354 return cclp->error_code;
360 * indent-tabs-mode: nil
362 * vim: shiftwidth=4 tabstop=8 expandtab