1 /* This file is part of Pazpar2.
2 Copyright (C) 2006-2011 Index Data
4 Pazpar2 is free software; you can redistribute it and/or modify it under
5 the terms of the GNU General Public License as published by the Free
6 Software Foundation; either version 2, or (at your option) any later
9 Pazpar2 is distributed in the hope that it will be useful, but WITHOUT ANY
10 WARRANTY; without even the implied warranty of MERCHANTABILITY or
11 FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
14 You should have received a copy of the GNU General Public License
15 along with this program; if not, write to the Free Software
16 Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
21 \brief Pazpar2 Character set facilities
28 #include <yaz/xmalloc.h>
29 #include <yaz/wrbuf.h>
31 #include <yaz/yaz-version.h>
37 #include "normalize7bit.h"
44 struct pp2_charset_s {
45 const char *(*token_next_handler)(pp2_relevance_token_t prt);
46 const char *(*get_sort_handler)(pp2_relevance_token_t prt);
47 const char *(*get_display_handler)(pp2_relevance_token_t prt);
50 struct icu_chain * icu_chn;
55 static const char *pp2_relevance_token_null(pp2_relevance_token_t prt);
56 static const char *pp2_relevance_token_a_to_z(pp2_relevance_token_t prt);
57 static const char *pp2_get_sort_ascii(pp2_relevance_token_t prt);
58 static const char *pp2_get_display_ascii(pp2_relevance_token_t prt);
61 static const char *pp2_relevance_token_icu(pp2_relevance_token_t prt);
62 static const char *pp2_get_sort_icu(pp2_relevance_token_t prt);
63 static const char *pp2_get_display_icu(pp2_relevance_token_t prt);
66 /* tokenzier handle */
67 struct pp2_relevance_token_s {
68 const char *cp; /* unnormalized buffer we're tokenizing */
69 const char *last_cp; /* pointer to last token we're dealing with */
70 pp2_charset_t pct; /* our main charset handle (type+config) */
71 WRBUF norm_str; /* normized string we return (temporarily) */
72 WRBUF sort_str; /* sort string we return (temporarily) */
79 pp2_charset_t pp2_charset_create_xml(xmlNode *xml_node)
82 UErrorCode status = U_ZERO_ERROR;
83 struct icu_chain *chain = 0;
85 xml_node = xml_node->children;
86 while (xml_node && xml_node->type != XML_ELEMENT_NODE)
87 xml_node = xml_node->next;
88 chain = icu_chain_xml_config(xml_node, 1, &status);
89 if (!chain || U_FAILURE(status)){
90 //xmlDocPtr icu_doc = 0;
91 //xmlChar *xmlstr = 0;
93 //xmlDocDumpMemory(icu_doc, size);
95 yaz_log(YLOG_FATAL, "Could not parse ICU chain config:\n"
97 xml_node->name, xml_node->name);
100 return pp2_charset_create(chain);
101 #else // YAZ_HAVE_ICU
102 yaz_log(YLOG_FATAL, "Error: ICU support requested with element:\n"
103 "<%s>\n ... \n</%s>",
104 xml_node->name, xml_node->name);
106 "But no ICU support is compiled into the YAZ library.");
108 #endif // YAZ_HAVE_ICU
111 void pp2_charset_incref(pp2_charset_t pct)
116 pp2_charset_t pp2_charset_create_a_to_z(void)
118 pp2_charset_t pct = pp2_charset_create(0);
119 pct->token_next_handler = pp2_relevance_token_a_to_z;
123 pp2_charset_t pp2_charset_create(struct icu_chain *icu_chn)
125 pp2_charset_t pct = xmalloc(sizeof(*pct));
127 pct->token_next_handler = pp2_relevance_token_null;
128 pct->get_sort_handler = pp2_get_sort_ascii;
129 pct->get_display_handler = pp2_get_display_ascii;
135 pct->icu_chn = icu_chn;
136 pct->icu_sts = U_ZERO_ERROR;
137 pct->token_next_handler = pp2_relevance_token_icu;
138 pct->get_sort_handler = pp2_get_sort_icu;
139 pct->get_display_handler = pp2_get_display_icu;
141 #endif // YAZ_HAVE_ICU
145 void pp2_charset_destroy(pp2_charset_t pct)
149 assert(pct->ref_count >= 1);
151 if (pct->ref_count == 0)
154 icu_chain_destroy(pct->icu_chn);
161 pp2_relevance_token_t pp2_relevance_tokenize(pp2_charset_t pct)
163 pp2_relevance_token_t prt = xmalloc(sizeof(*prt));
167 prt->norm_str = wrbuf_alloc();
168 prt->sort_str = wrbuf_alloc();
176 prt->iter = icu_iter_create(pct->icu_chn);
181 void pp2_relevance_first(pp2_relevance_token_t prt,
189 char *pout = firstword;
190 char articles[] = "the den der die des an a "; // must end in space
192 for (; *p && *p != ' ' && pout - firstword < (sizeof(firstword)-2); p++)
193 *pout++ = tolower(*(unsigned char *)p);
196 if (strstr(articles, firstword))
200 wrbuf_rewind(prt->norm_str);
201 wrbuf_rewind(prt->sort_str);
208 icu_iter_first(prt->iter, buf);
210 #endif // YAZ_HAVE_ICU
213 void pp2_relevance_token_destroy(pp2_relevance_token_t prt)
218 icu_iter_destroy(prt->iter);
221 wrbuf_destroy(prt->norm_str);
223 wrbuf_destroy(prt->sort_str);
227 const char *pp2_relevance_token_next(pp2_relevance_token_t prt)
230 return (prt->pct->token_next_handler)(prt);
233 const char *pp2_get_sort(pp2_relevance_token_t prt)
235 return prt->pct->get_sort_handler(prt);
238 const char *pp2_get_display(pp2_relevance_token_t prt)
240 return prt->pct->get_display_handler(prt);
243 #define raw_char(c) (((c) >= 'a' && (c) <= 'z') ? (c) : -1)
244 /* original tokenizer with our tokenize interface, but we
245 add +1 to ensure no '\0' are in our string (except for EOF)
247 static const char *pp2_relevance_token_a_to_z(pp2_relevance_token_t prt)
249 const char *cp = prt->cp;
252 /* skip white space */
253 while (*cp && (c = raw_char(tolower(*(const unsigned char *)cp))) < 0)
261 /* now read the term itself */
264 wrbuf_rewind(prt->norm_str);
265 while (*cp && (c = raw_char(tolower(*cp))) >= 0)
267 wrbuf_putc(prt->norm_str, c);
271 return wrbuf_cstr(prt->norm_str);
274 static const char *pp2_get_sort_ascii(pp2_relevance_token_t prt)
276 if (prt->last_cp == 0)
280 char *tmp = xstrdup(prt->last_cp);
282 result = normalize7bit_mergekey(tmp);
284 wrbuf_rewind(prt->sort_str);
285 wrbuf_puts(prt->sort_str, result);
287 return wrbuf_cstr(prt->sort_str);
291 static const char *pp2_get_display_ascii(pp2_relevance_token_t prt)
293 if (prt->last_cp == 0)
297 return wrbuf_cstr(prt->norm_str);
301 static const char *pp2_relevance_token_null(pp2_relevance_token_t prt)
303 const char *cp = prt->cp;
305 prt->last_cp = *cp ? cp : 0;
313 static const char *pp2_relevance_token_icu(pp2_relevance_token_t prt)
315 if (icu_iter_next(prt->iter))
317 return icu_iter_get_norm(prt->iter);
322 static const char *pp2_get_sort_icu(pp2_relevance_token_t prt)
324 return icu_iter_get_sortkey(prt->iter);
327 static const char *pp2_get_display_icu(pp2_relevance_token_t prt)
329 return icu_iter_get_display(prt->iter);
332 #endif // YAZ_HAVE_ICU
338 * c-file-style: "Stroustrup"
339 * indent-tabs-mode: nil
341 * vim: shiftwidth=4 tabstop=8 expandtab