-/* $Id: charsets.c,v 1.5 2007-05-25 10:32:55 marc Exp $
- Copyright (c) 2006-2007, Index Data.
-
-This file is part of Pazpar2.
+/* This file is part of Pazpar2.
+ Copyright (C) 2006-2010 Index Data
Pazpar2 is free software; you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free
for more details.
You should have received a copy of the GNU General Public License
-along with Pazpar2; see the file LICENSE. If not, write to the
-Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA
-02111-1307, USA.
- */
+along with this program; if not, write to the Free Software
+Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+
+*/
/** \file charsets.c
\brief Pazpar2 Character set facilities
*/
#if HAVE_CONFIG_H
-#include "cconfig.h"
+#include <config.h>
#endif
#include <yaz/xmalloc.h>
#include <yaz/wrbuf.h>
#include <yaz/log.h>
+#include <yaz/yaz-version.h>
#include <ctype.h>
#include <assert.h>
+#include <string.h>
#include "charsets.h"
-//#include "config.h"
-//#include "parameters.h"
+#include "normalize7bit.h"
-#ifdef HAVE_ICU
-#include "icu_I18N.h"
-#endif // HAVE_ICU
+#if YAZ_HAVE_ICU
+#include <yaz/icu.h>
+#endif
/* charset handle */
struct pp2_charset_s {
const char *(*token_next_handler)(pp2_relevance_token_t prt);
- /* other handlers will come as we see fit */
-#ifdef HAVE_ICU
+ const char *(*get_sort_handler)(pp2_relevance_token_t prt);
+ int ref_count;
+#if YAZ_HAVE_ICU
struct icu_chain * icu_chn;
UErrorCode icu_sts;
-#endif // HAVE_ICU
+#endif
};
static const char *pp2_relevance_token_a_to_z(pp2_relevance_token_t prt);
+static const char *pp2_get_sort_ascii(pp2_relevance_token_t prt);
-#ifdef HAVE_ICU
+#if YAZ_HAVE_ICU
static const char *pp2_relevance_token_icu(pp2_relevance_token_t prt);
-#endif // HAVE_ICU
+static const char *pp2_get_sort_icu(pp2_relevance_token_t prt);
+#endif
/* tokenzier handle */
struct pp2_relevance_token_s {
const char *cp; /* unnormalized buffer we're tokenizing */
+ const char *last_cp; /* pointer to last token we're dealing with */
pp2_charset_t pct; /* our main charset handle (type+config) */
WRBUF norm_str; /* normized string we return (temporarily) */
+ WRBUF sort_str; /* sort string we return (temporarily) */
+ yaz_icu_iter_t iter;
};
+
+pp2_charset_t pp2_charset_create_xml(xmlNode *xml_node)
+{
+#if YAZ_HAVE_ICU
+ UErrorCode status = U_ZERO_ERROR;
+ struct icu_chain *chain = 0;
+ if (xml_node)
+ xml_node = xml_node->children;
+ while (xml_node && xml_node->type != XML_ELEMENT_NODE)
+ xml_node = xml_node->next;
+ chain = icu_chain_xml_config(xml_node, 1, &status);
+ if (!chain || U_FAILURE(status)){
+ //xmlDocPtr icu_doc = 0;
+ //xmlChar *xmlstr = 0;
+ //int size = 0;
+ //xmlDocDumpMemory(icu_doc, size);
+
+ yaz_log(YLOG_FATAL, "Could not parse ICU chain config:\n"
+ "<%s>\n ... \n</%s>",
+ xml_node->name, xml_node->name);
+ return 0;
+ }
+ return pp2_charset_create(chain);
+#else // YAZ_HAVE_ICU
+ yaz_log(YLOG_FATAL, "Error: ICU support requested with element:\n"
+ "<%s>\n ... \n</%s>",
+ xml_node->name, xml_node->name);
+ yaz_log(YLOG_FATAL,
+ "But no ICU support is compiled into the YAZ library.");
+ return 0;
+#endif // YAZ_HAVE_ICU
+}
+
+void pp2_charset_incref(pp2_charset_t pct)
+{
+ (pct->ref_count)++;
+}
+
pp2_charset_t pp2_charset_create(struct icu_chain * icu_chn)
{
pp2_charset_t pct = xmalloc(sizeof(*pct));
pct->token_next_handler = pp2_relevance_token_a_to_z;
-#ifdef HAVE_ICU
+ pct->get_sort_handler = pp2_get_sort_ascii;
+ pct->ref_count = 1;
+#if YAZ_HAVE_ICU
pct->icu_chn = 0;
- if (icu_chn){
+ if (icu_chn)
+ {
pct->icu_chn = icu_chn;
pct->icu_sts = U_ZERO_ERROR;
pct->token_next_handler = pp2_relevance_token_icu;
+ pct->get_sort_handler = pp2_get_sort_icu;
}
- #endif // HAVE_ICU
+#endif // YAZ_HAVE_ICU
return pct;
}
void pp2_charset_destroy(pp2_charset_t pct)
{
- xfree(pct);
+ if (pct)
+ {
+ assert(pct->ref_count >= 1);
+ --(pct->ref_count);
+ if (pct->ref_count == 0)
+ {
+#if YAZ_HAVE_ICU
+ icu_chain_destroy(pct->icu_chn);
+#endif
+ xfree(pct);
+ }
+ }
}
-pp2_relevance_token_t pp2_relevance_tokenize(pp2_charset_t pct,
- const char *buf)
+pp2_relevance_token_t pp2_relevance_tokenize(pp2_charset_t pct)
{
pp2_relevance_token_t prt = xmalloc(sizeof(*prt));
assert(pct);
prt->norm_str = wrbuf_alloc();
- prt->cp = buf;
+ prt->sort_str = wrbuf_alloc();
+ prt->cp = 0;
+ prt->last_cp = 0;
prt->pct = pct;
-#ifdef HAVE_ICU
+#if YAZ_HAVE_ICU
+ prt->iter = 0;
if (pct->icu_chn)
- {
- pct->icu_sts = U_ZERO_ERROR;
- int ok = 0;
- ok = icu_chain_assign_cstr(pct->icu_chn, buf, &pct->icu_sts);
- //printf("\nfield ok: %d '%s'\n", ok, buf);
- prt->pct = pct;
- prt->norm_str = 0;
- }
-#endif // HAVE_ICU
+ prt->iter = icu_iter_create(pct->icu_chn);
+#endif
return prt;
}
+void pp2_relevance_first(pp2_relevance_token_t prt,
+ const char *buf,
+ int skip_article)
+{
+ if (skip_article)
+ {
+ const char *p = buf;
+ char firstword[64];
+ char *pout = firstword;
+ char articles[] = "the den der die des an a "; // must end in space
+
+ while (*p && !isalnum(*(unsigned char *)p))
+ p++;
+ for (; *p && *p != ' ' && pout - firstword < (sizeof(firstword)-2); p++)
+ *pout++ = tolower(*(unsigned char *)p);
+ *pout++ = ' ';
+ *pout++ = '\0';
+ if (strstr(articles, firstword))
+ buf = p;
+ }
+
+ wrbuf_rewind(prt->norm_str);
+ wrbuf_rewind(prt->sort_str);
+ prt->cp = buf;
+ prt->last_cp = 0;
+
+#if YAZ_HAVE_ICU
+ if (prt->iter)
+ {
+ icu_iter_first(prt->iter, buf);
+ }
+#endif // YAZ_HAVE_ICU
+}
void pp2_relevance_token_destroy(pp2_relevance_token_t prt)
{
assert(prt);
+#if YAZ_HAVE_ICU
+ if (prt->iter)
+ icu_iter_destroy(prt->iter);
+#endif
if(prt->norm_str)
wrbuf_destroy(prt->norm_str);
+ if(prt->sort_str)
+ wrbuf_destroy(prt->sort_str);
xfree(prt);
}
return (prt->pct->token_next_handler)(prt);
}
-#define raw_char(c) (((c) >= 'a' && (c) <= 'z') ? (c) - 'a' + 1 : -1)
+const char *pp2_get_sort(pp2_relevance_token_t prt)
+{
+ return prt->pct->get_sort_handler(prt);
+}
+
+#define raw_char(c) (((c) >= 'a' && (c) <= 'z') ? (c) : -1)
/* original tokenizer with our tokenize interface, but we
add +1 to ensure no '\0' are in our string (except for EOF)
*/
int c;
/* skip white space */
- while (*cp && (c = raw_char(tolower(*cp))) < 0)
+ while (*cp && (c = raw_char(tolower(*(const unsigned char *)cp))) < 0)
cp++;
if (*cp == '\0')
{
prt->cp = cp;
+ prt->last_cp = 0;
return 0;
}
/* now read the term itself */
+
+ prt->last_cp = cp;
wrbuf_rewind(prt->norm_str);
while (*cp && (c = raw_char(tolower(*cp))) >= 0)
{
return wrbuf_cstr(prt->norm_str);
}
+static const char *pp2_get_sort_ascii(pp2_relevance_token_t prt)
+{
+ if (prt->last_cp == 0)
+ return 0;
+ else
+ {
+ char *tmp = xstrdup(prt->last_cp);
+ char *result = 0;
+ result = normalize7bit_mergekey(tmp);
+
+ wrbuf_rewind(prt->sort_str);
+ wrbuf_puts(prt->sort_str, result);
+ xfree(tmp);
+ return wrbuf_cstr(prt->sort_str);
+ }
+}
+
-#ifdef HAVE_ICU
+#if YAZ_HAVE_ICU
static const char *pp2_relevance_token_icu(pp2_relevance_token_t prt)
{
- //&& U_SUCCESS(pct->icu_sts))
- if (icu_chain_next_token(prt->pct->icu_chn, &prt->pct->icu_sts)){
- //printf("'%s' ", icu_chain_get_norm(prt->pct->icu_chn));
- if (U_FAILURE(prt->pct->icu_sts))
- {
- //printf("ICU status failure\n ");
- return 0;
- }
-
- return icu_chain_get_norm(prt->pct->icu_chn);
+ if (icu_iter_next(prt->iter))
+ {
+ return icu_iter_get_norm(prt->iter);
}
- //printf ("EOF\n");
return 0;
-};
-#endif // HAVE_ICU
+}
+static const char *pp2_get_sort_icu(pp2_relevance_token_t prt)
+{
+ return icu_iter_get_sortkey(prt->iter);
+}
+
+#endif // YAZ_HAVE_ICU
/*
* Local variables:
* c-basic-offset: 4
+ * c-file-style: "Stroustrup"
* indent-tabs-mode: nil
* End:
* vim: shiftwidth=4 tabstop=8 expandtab
*/
+