/* This file is part of Pazpar2.
- Copyright (C) 2006-2011 Index Data
+ Copyright (C) 2006-2012 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
NMEM nmem;
};
-
struct word_entry {
const char *norm_str;
int termno;
+ char *ccl_field;
struct word_entry *next;
};
-static void add_word_entry(NMEM nmem,
- struct word_entry **entries,
- const char *norm_str,
- int term_no)
-{
- struct word_entry *ne = nmem_malloc(nmem, sizeof(*ne));
- ne->norm_str = nmem_strdup(nmem, norm_str);
- ne->termno = term_no;
-
- ne->next = *entries;
- *entries = ne;
-}
-
-
-int word_entry_match(struct word_entry *entries, const char *norm_str)
+static int word_entry_match(struct word_entry *entries, const char *norm_str,
+ const char *frank, int *local_mult)
{
for (; entries; entries = entries->next)
{
if (!strcmp(norm_str, entries->norm_str))
+ {
+ const char *cp = 0;
+ if (frank && (cp = strchr(frank, ' ')))
+ {
+ if ((cp - frank) == strlen(entries->ccl_field) &&
+ memcmp(entries->ccl_field, frank, cp - frank) == 0)
+ *local_mult = atoi(cp + 1);
+ }
return entries->termno;
+ }
}
return 0;
}
-static struct word_entry *build_word_entries(pp2_charset_token_t prt,
- NMEM nmem,
- const char **terms)
-{
- int termno = 1; /* >0 signals THERE is an entry */
- struct word_entry *entries = 0;
- const char **p = terms;
-
- for (; *p; p++)
- {
- const char *norm_str;
-
- pp2_charset_token_first(prt, *p, 0);
- while ((norm_str = pp2_charset_token_next(prt)))
- add_word_entry(nmem, &entries, norm_str, termno);
- termno++;
- }
- return entries;
-}
-
void relevance_countwords(struct relevance *r, struct record_cluster *cluster,
- const char *words, int multiplier, const char *name)
+ const char *words, int multiplier, const char *name,
+ const char *frank)
{
int *mult = cluster->term_frequency_vec_tmp;
const char *norm_str;
int i, length = 0;
-
pp2_charset_token_first(r->prt, words, 0);
for (i = 1; i < r->vec_len; i++)
mult[i] = 0;
while ((norm_str = pp2_charset_token_next(r->prt)))
{
- int res = word_entry_match(r->entries, norm_str);
+ int local_mult = multiplier;
+ int res = word_entry_match(r->entries, norm_str, frank, &local_mult);
if (res)
{
assert(res < r->vec_len);
- mult[res] += multiplier;
+ mult[res] += local_mult;
}
length++;
}
cluster->term_frequency_vec[0] += length;
}
-struct relevance *relevance_create(pp2_charset_fact_t pft,
- NMEM nmem, const char **terms)
+static void pull_terms(struct relevance *res, struct ccl_rpn_node *n)
{
- struct relevance *res = nmem_malloc(nmem, sizeof(struct relevance));
- const char **p;
+ char **words;
+ int numwords;
+ char *ccl_field;
+ int i;
+
+ switch (n->kind)
+ {
+ case CCL_RPN_AND:
+ case CCL_RPN_OR:
+ case CCL_RPN_NOT:
+ case CCL_RPN_PROX:
+ pull_terms(res, n->u.p[0]);
+ pull_terms(res, n->u.p[1]);
+ break;
+ case CCL_RPN_TERM:
+ nmem_strsplit(res->nmem, " ", n->u.t.term, &words, &numwords);
+ for (i = 0; i < numwords; i++)
+ {
+ const char *norm_str;
+
+ ccl_field = nmem_strdup_null(res->nmem, n->u.t.qual);
+
+ pp2_charset_token_first(res->prt, words[i], 0);
+ while ((norm_str = pp2_charset_token_next(res->prt)))
+ {
+ struct word_entry **e = &res->entries;
+ while (*e)
+ e = &(*e)->next;
+ *e = nmem_malloc(res->nmem, sizeof(**e));
+ (*e)->norm_str = nmem_strdup(res->nmem, norm_str);
+ (*e)->ccl_field = ccl_field;
+ (*e)->termno = res->vec_len++;
+ (*e)->next = 0;
+ }
+ }
+ break;
+ default:
+ break;
+ }
+}
+
+struct relevance *relevance_create_ccl(pp2_charset_fact_t pft,
+ struct ccl_rpn_node *query)
+{
+ NMEM nmem = nmem_create();
+ struct relevance *res = nmem_malloc(nmem, sizeof(*res));
int i;
- for (p = terms, i = 0; *p; p++, i++)
- ;
- res->vec_len = ++i;
- res->doc_frequency_vec = nmem_malloc(nmem, res->vec_len * sizeof(int));
- memset(res->doc_frequency_vec, 0, res->vec_len * sizeof(int));
res->nmem = nmem;
+ res->entries = 0;
+ res->vec_len = 1;
res->prt = pp2_charset_token_create(pft, "relevance");
- res->entries = build_word_entries(res->prt, nmem, terms);
+
+ pull_terms(res, query);
+
+ res->doc_frequency_vec = nmem_malloc(nmem, res->vec_len * sizeof(int));
+ for (i = 0; i < res->vec_len; i++)
+ res->doc_frequency_vec[i] = 0;
return res;
}
if (*rp)
{
pp2_charset_token_destroy((*rp)->prt);
+ nmem_destroy((*rp)->nmem);
*rp = 0;
}
}
}
}
-
void relevance_donerecord(struct relevance *r, struct record_cluster *cluster)
{
int i;
}
}
// Calculate relevance for each document
-
while (1)
{
int t;
break;
for (t = 1; t < rel->vec_len; t++)
{
- float termfreq;
-#if 1
- termfreq = (float) rec->term_frequency_vecf[t];
-#else
- if (rec->term_frequency_vec[0])
- {
- termfreq = (float)
- rec->term_frequency_vec[t] / rec->term_frequency_vec[0] ;
- }
- else
- termfreq = 0.0;
-#endif
+ float termfreq = (float) rec->term_frequency_vecf[t];
relevance += 100000 * (termfreq * idfvec[t] + 0.0000005);
}
rec->relevance_score = relevance;