X-Git-Url: http://lists.indexdata.dk/cgi-bin?a=blobdiff_plain;ds=sidebyside;f=index%2Fretrieve.c;h=0e7be09e7d62e90ee6c1e05b802e93a5b9b3c765;hb=c22b27ba78b584e1f3e6ba863b854bf7f741d0a3;hp=4ff858610229c1edd5d5f531485a26dbd9ac0e42;hpb=cd02e9c5558d2f0db2ab83fcad810a6522fd2319;p=idzebra-moved-to-github.git
diff --git a/index/retrieve.c b/index/retrieve.c
index 4ff8586..0e7be09 100644
--- a/index/retrieve.c
+++ b/index/retrieve.c
@@ -1,5 +1,5 @@
/* This file is part of the Zebra server.
- Copyright (C) 1995-2008 Index Data
+ Copyright (C) 1994-2011 Index Data
Zebra 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
@@ -148,51 +148,6 @@ struct index_spec *parse_index_spec(const char *elem, NMEM nmem,
*error = 1;
return first;
}
-
-static int parse_zebra_elem(const char *elem,
- const char **index, size_t *index_len,
- const char **type, size_t *type_len)
-{
- *index = 0;
- *index_len = 0;
-
- *type = 0;
- *type_len = 0;
-
- if (elem && *elem)
- {
- char *cp;
- /* verify that '::' is in the beginning of *elem
- and something more follows */
- if (':' != *elem
- || !(elem +1) || ':' != *(elem +1)
- || !(elem +2) || '\0' == *(elem +2))
- return 0;
-
- /* pick out info from string after '::' */
- elem = elem + 2;
- cp = strchr(elem, ':');
-
- if (!cp) /* index, no colon, no type */
- {
- *index = elem;
- *index_len = strlen(elem);
- }
- else if (cp[1] == '\0') /* colon, but no following type */
- {
- return 0;
- }
- else /* index, colon and type */
- {
- *index = elem;
- *index_len = cp - elem;
- *type = cp+1;
- *type_len = strlen(cp+1);
- }
- }
- return 1;
-}
-
static int sort_fetch(
struct special_fetch_s *fi, const char *elemsetname,
@@ -200,40 +155,22 @@ static int sort_fetch(
const Odr_oid **output_format,
WRBUF result, WRBUF addinfo)
{
- const char *retrieval_index;
- size_t retrieval_index_len;
- const char *retrieval_type;
- size_t retrieval_type_len;
- char retrieval_index_cstr[256];
- char retrieval_type_cstr[256];
int ord;
ZebraHandle zh = fi->zh;
+ int error;
+ struct index_spec *spec;
- if (!parse_zebra_elem(elemsetname,
- &retrieval_index, &retrieval_index_len,
- &retrieval_type, &retrieval_type_len))
- {
+ spec = parse_index_spec(elemsetname, fi->nmem, &error);
+ if (error)
return YAZ_BIB1_SPECIFIED_ELEMENT_SET_NAME_NOT_VALID_FOR_SPECIFIED_;
- }
-
- if (retrieval_type_len == 0)
- return -1; /* must have a register type specified */
- if (!retrieval_index_len ||
- retrieval_index_len >= sizeof(retrieval_index_cstr)-1)
- {
- return YAZ_BIB1_SPECIFIED_ELEMENT_SET_NAME_NOT_VALID_FOR_SPECIFIED_;
- }
-
- memcpy(retrieval_index_cstr, retrieval_index, retrieval_index_len);
- retrieval_index_cstr[retrieval_index_len] = '\0';
-
- memcpy(retrieval_type_cstr, retrieval_type, retrieval_type_len);
- retrieval_type_cstr[retrieval_type_len] = '\0';
+ /* for sort fetches.. We allow only one index and type must be given */
+ if (!spec || spec->next || !spec->index_type)
+ return -1;
ord = zebraExplain_lookup_attr_str(zh->reg->zei,
zinfo_index_category_sort,
- retrieval_type_cstr,
- retrieval_index_cstr);
+ spec->index_type,
+ spec->index_name);
if (ord == -1)
return -1; /* is not a sort index */
else
@@ -244,7 +181,7 @@ static int sort_fetch(
const char *string_index = 0;
WRBUF wrbuf_result = result;
int off = 0;
-
+
zebraExplain_lookup_ord(zh->reg->zei, ord, &index_type, &db,
&string_index);
if (!oid_oidcmp(input_format, yaz_oid_recsyn_xml))
@@ -252,7 +189,7 @@ static int sort_fetch(
*output_format = yaz_oid_recsyn_xml;
wrbuf_printf(wrbuf_result, ZEBRA_XML_HEADER_STR
" sysno=\"" ZINT_FORMAT "\""
- " set=\"zebra::index%s/\">\n",
+ " set=\"zebra::index%s\">\n",
fi->sysno, elemsetname);
}
else if (!oid_oidcmp(input_format, yaz_oid_recsyn_sutrs))
@@ -269,7 +206,7 @@ static int sort_fetch(
}
zebra_sort_type(zh->reg->sort_index, ord);
zebra_sort_sysno(zh->reg->sort_index, fi->sysno);
- zebra_sort_read(zh->reg->sort_index, wrbuf_str);
+ zebra_sort_read(zh->reg->sort_index, 0, wrbuf_str);
while (off != wrbuf_len(wrbuf_str))
{
@@ -309,14 +246,11 @@ static int special_index_fetch(
WRBUF result, WRBUF addinfo,
Record rec)
{
- const char *retrieval_index;
- size_t retrieval_index_len;
- const char *retrieval_type;
- size_t retrieval_type_len;
zebra_rec_keys_t keys;
int ret_code = 0;
- char retrieval_type_cstr[256];
ZebraHandle zh = fi->zh;
+ struct index_spec *spec, *spec_list;
+ int error;
/* set output variables before processing possible error states */
/* *rec_lenp = 0; */
@@ -331,33 +265,19 @@ static int special_index_fetch(
return YAZ_BIB1_NO_SYNTAXES_AVAILABLE_FOR_THIS_REQUEST;
}
- if (!parse_zebra_elem(elemsetname,
- &retrieval_index, &retrieval_index_len,
- &retrieval_type, &retrieval_type_len))
- return YAZ_BIB1_SPECIFIED_ELEMENT_SET_NAME_NOT_VALID_FOR_SPECIFIED_;
-
- if (retrieval_type_len)
+ spec_list = parse_index_spec(elemsetname, fi->nmem, &error);
+ if (error)
{
- memcpy(retrieval_type_cstr, retrieval_type, retrieval_type_len);
- retrieval_type_cstr[retrieval_type_len] = '\0';
+ return YAZ_BIB1_SPECIFIED_ELEMENT_SET_NAME_NOT_VALID_FOR_SPECIFIED_;
}
-
- if (retrieval_index_len)
- {
- char retrieval_index_cstr[256];
- if (retrieval_index_len < sizeof(retrieval_index_cstr) -1)
- {
- memcpy(retrieval_index_cstr, retrieval_index, retrieval_index_len);
- retrieval_index_cstr[retrieval_index_len] = '\0';
-
- if (zebraExplain_lookup_attr_str(zh->reg->zei,
- zinfo_index_category_index,
- (retrieval_type_len == 0 ? 0 :
- retrieval_type_cstr),
- retrieval_index_cstr) == -1)
- return YAZ_BIB1_SPECIFIED_ELEMENT_SET_NAME_NOT_VALID_FOR_SPECIFIED_;
- }
+ for (spec = spec_list; spec; spec = spec->next)
+ {
+ if (zebraExplain_lookup_attr_str(zh->reg->zei,
+ zinfo_index_category_index,
+ spec->index_type,
+ spec->index_name) == -1)
+ return YAZ_BIB1_SPECIFIED_ELEMENT_SET_NAME_NOT_VALID_FOR_SPECIFIED_;
}
keys = zebra_rec_keys_open();
@@ -374,13 +294,13 @@ static int special_index_fetch(
const char *str;
struct it_key key_in;
WRBUF wrbuf = result;
-
+
if (!oid_oidcmp(input_format, yaz_oid_recsyn_xml))
{
*output_format = input_format;
wrbuf_printf(wrbuf, ZEBRA_XML_HEADER_STR
" sysno=\"" ZINT_FORMAT "\""
- " set=\"zebra::index%s/\">\n",
+ " set=\"zebra::index%s\">\n",
fi->sysno, elemsetname);
}
else if (!oid_oidcmp(input_format, yaz_oid_recsyn_sutrs))
@@ -393,55 +313,54 @@ static int special_index_fetch(
const char *index_type;
const char *db = 0;
const char *string_index = 0;
- size_t string_index_len;
char dst_buf[IT_MAX_WORD];
+ int match = 0;
zebraExplain_lookup_ord(zh->reg->zei, ord, &index_type, &db,
&string_index);
- string_index_len = strlen(string_index);
-
- /* process only if index is not defined,
- or if defined and matching */
- if (retrieval_index == 0
- || (string_index_len == retrieval_index_len
- && !memcmp(string_index, retrieval_index,
- string_index_len)))
+ if (!spec_list)
+ match = 1; /* match all if no specs were given */
+ else
{
- /* process only if type is not defined, or is matching */
- if (retrieval_type == 0
- || !strcmp(retrieval_type_cstr, index_type))
+ for (spec = spec_list; spec; spec = spec->next)
{
- if (zebra_term_untrans(zh, index_type, dst_buf, str))
- *dst_buf = '\0'; /* untrans failed */
-
- if (!oid_oidcmp(input_format, yaz_oid_recsyn_xml))
- {
- wrbuf_printf(wrbuf, " ",
- key_in.mem[key_in.len -1]);
- wrbuf_xmlputs(wrbuf, dst_buf);
- wrbuf_printf(wrbuf, "\n");
- }
- else
- {
- wrbuf_printf(wrbuf, "%s ", string_index);
-
- wrbuf_printf(wrbuf, "%s", index_type);
-
- for (i = 1; i < key_in.len; i++)
- wrbuf_printf(wrbuf, " " ZINT_FORMAT,
- key_in.mem[i]);
-
- wrbuf_printf(wrbuf, " %s", dst_buf);
-
- wrbuf_printf(wrbuf, "\n");
-
- }
+ if ((!spec->index_type ||
+ !yaz_matchstr(spec->index_type, index_type))
+ &&
+ !yaz_matchstr(spec->index_name, string_index))
+ match = 1;
+ }
+ }
+ if (match)
+ {
+ if (zebra_term_untrans(zh, index_type, dst_buf, str))
+ *dst_buf = '\0'; /* untrans failed */
+
+ if (!oid_oidcmp(input_format, yaz_oid_recsyn_xml))
+ {
+ wrbuf_printf(wrbuf, " ",
+ key_in.mem[key_in.len -1]);
+ wrbuf_xmlputs(wrbuf, dst_buf);
+ wrbuf_printf(wrbuf, "\n");
+ }
+ else
+ {
+ wrbuf_printf(wrbuf, "%s ", string_index);
+
+ wrbuf_printf(wrbuf, "%s", index_type);
+
+ for (i = 1; i < key_in.len; i++)
+ wrbuf_printf(wrbuf, " " ZINT_FORMAT,
+ key_in.mem[i]);
+
+ wrbuf_printf(wrbuf, " %s", dst_buf);
+
+ wrbuf_printf(wrbuf, "\n");
}
}
}
@@ -484,6 +403,41 @@ static void retrieve_puts_int(WRBUF wrbuf, const char *name,
}
+static void snippet_check_fields(ZebraHandle zh, WRBUF wrbuf,
+ zebra_snippets *doc,
+ const zebra_snippet_word *doc_w,
+ const char *w_index_type)
+{
+ /* beginning of snippet. See which fields the snippet also
+ occur */
+ const zebra_snippet_word *w;
+ int no = 0;
+ for (w = zebra_snippets_constlist(doc); w; w = w->next)
+ {
+ /* same sequence but other field? */
+ if (w->seqno == doc_w->seqno && w->ord != doc_w->ord)
+ {
+ const char *index_type;
+ const char *db = 0;
+ const char *string_index = 0;
+
+ zebraExplain_lookup_ord(zh->reg->zei, w->ord,
+ &index_type, &db, &string_index);
+ /* only report for same index type */
+ if (!strcmp(w_index_type, index_type))
+ {
+ if (no == 0)
+ wrbuf_printf(wrbuf, " fields=\"%s", string_index);
+ else
+ wrbuf_printf(wrbuf, " %s", string_index);
+ no++;
+ }
+ }
+ }
+ if (no)
+ wrbuf_printf(wrbuf, "\"");
+}
+
static void snippet_xml_record(ZebraHandle zh, WRBUF wrbuf, zebra_snippets *doc)
{
const zebra_snippet_word *doc_w;
@@ -503,8 +457,11 @@ static void snippet_xml_record(ZebraHandle zh, WRBUF wrbuf, zebra_snippets *doc)
if (mark_state == 0)
{
+
wrbuf_printf(wrbuf, " ", index_type);
+ wrbuf_printf(wrbuf, " type=\"%s\"", index_type);
+ snippet_check_fields(zh, wrbuf, doc, doc_w, index_type);
+ wrbuf_printf(wrbuf, ">");
}
if (doc_w->match)
wrbuf_puts(wrbuf, "");
@@ -664,13 +621,20 @@ static int term_qsort_handle(const void *a, const void *b)
static void term_collect_freq(ZebraHandle zh,
struct term_collect *col, int no_terms_collect,
- int ord, RSET rset)
+ int ord, RSET rset, double scale_factor)
{
int i;
for (i = 0; i < no_terms_collect; i++)
{
if (col[i].term)
- col[i].set_occur = freq_term(zh, ord, col[i].term, rset);
+ {
+ if (scale_factor < 0.0)
+ {
+ col[i].set_occur = freq_term(zh, ord, col[i].term, rset);
+ }
+ else
+ col[i].set_occur = scale_factor * col[i].oc;
+ }
}
qsort(col, no_terms_collect, sizeof(*col), term_qsort_handle);
}
@@ -736,7 +700,7 @@ static int perform_facet_sort(ZebraHandle zh, int no_ord, int *ord_array,
zebra_sort_type(zh->reg->sort_index, ord_array[ord_i]);
wrbuf_rewind(w);
- if (zebra_sort_read(zh->reg->sort_index, w))
+ if (zebra_sort_read(zh->reg->sort_index, 0, w))
{
zebra_strmap_t sm = map_array[ord_i];
int off = 0;
@@ -876,7 +840,8 @@ static int perform_facet(ZebraHandle zh,
no_collect_terms = 1;
col = term_collect_create(map_array[i], no_collect_terms, nmem);
term_collect_freq(zh, col, no_collect_terms, ord_array[i],
- resultSetRef(zh, fi->setname));
+ resultSetRef(zh, fi->setname),
+ cat == zinfo_index_category_sort ? 1.0 : -1.0);
if (use_xml)
wrbuf_printf(wr, " \n",
@@ -943,8 +908,6 @@ static int facet_fetch(
/* whether sort or index based */
zinfo_index_category_t cat = zinfo_index_category_sort;
- res_get_int(zh->res, "facetNumRecs", &num_recs);
-
/* see if XML is required for response */
if (oid_oidcmp(input_format, yaz_oid_recsyn_xml) == 0)
use_xml = 1;
@@ -976,6 +939,7 @@ static int facet_fetch(
if (ord == -1)
break;
ord_array[i] = ord;
+ num_recs = 10000;
}
if (spec)
{
@@ -995,6 +959,8 @@ static int facet_fetch(
if (spec)
return YAZ_BIB1_SPECIFIED_ELEMENT_SET_NAME_NOT_VALID_FOR_SPECIFIED_;
+ res_get_int(zh->res, "facetNumRecs", &num_recs);
+
pos_array = (zint *) nmem_malloc(fi->nmem, num_recs * sizeof(*pos_array));
for (i = 0; i < num_recs; i++)
pos_array[i] = i+1;
@@ -1302,6 +1268,7 @@ int zebra_record_fetch(ZebraHandle zh, const char *setname,
/*
* Local variables:
* c-basic-offset: 4
+ * c-file-style: "Stroustrup"
* indent-tabs-mode: nil
* End:
* vim: shiftwidth=4 tabstop=8 expandtab