-/* $Id: recgrs.c,v 1.7 2006-09-29 10:02:47 adam Exp $
- Copyright (C) 1995-2006
+/* $Id: recgrs.c,v 1.19 2007-05-08 12:50:04 adam Exp $
+ Copyright (C) 1995-2007
Index Data ApS
This file is part of the Zebra server.
#include <ctype.h>
#include <yaz/log.h>
-#include <yaz/oid.h>
+#include <yaz/oid_db.h>
+#include <yaz/diagbib1.h>
#include <d1_absyn.h>
#include <idzebra/recgrs.h>
return 1;
}
-static struct source_parser *source_parser_create()
+static struct source_parser *source_parser_create(void)
{
struct source_parser *sp = xmalloc(sizeof(*sp));
res = 0;
/* looking for the attribute with a specified name */
for (attr = n->u.tag.attributes; attr; attr = attr->next) {
- yaz_log(YLOG_DEBUG," - attribute %s <-> %s", attname, attr->name );
-
if (!strcmp(attr->name, attname)) {
if (p->u.relation.op[0]) {
if (*p->u.relation.op != '=') {
yaz_log(YLOG_WARN, "predicate %s ignored", p->u.relation.name);
res = 1; break;
} else {
- yaz_log(YLOG_DEBUG," - value %s <-> %s",
- p->u.relation.value, attr->value );
if (!strcmp(attr->value, p->u.relation.value)) {
res = 1; break;
}
}
}
}
- yaz_log(YLOG_DEBUG, "return %d", res);
return res;
} else {
return 1;
backwards trough xpath location steps ... */
for (i = xpe->xpath_len - 1; i>0; i--)
{
- yaz_log(YLOG_DEBUG, "Checking step %d: %s on tag %s",
- i, xp[i].part, nn->u.tag.tag);
-
if (!d1_check_xpath_predicate(nn, xp[i].predicate))
{
- yaz_log(YLOG_DEBUG, " Predicates didn't match");
ok = 0;
break;
}
char tag_path_full[1024];
int termlist_only = 1;
data1_termlist *tl;
- int xpdone = 0;
- if ((!n->root->u.root.absyn) ||
- (n->root->u.root.absyn->xpath_indexing == DATA1_XPATH_INDEXING_ENABLE)) {
+
+ if (!n->root->u.root.absyn
+ ||
+ n->root->u.root.absyn->xpath_indexing == DATA1_XPATH_INDEXING_ENABLE)
+ {
termlist_only = 0;
}
+
switch (n->which)
{
case DATA1N_data:
wrd->term_buf = n->u.data.data;
wrd->term_len = n->u.data.len;
- xpdone = 0;
mk_tag_path_full(tag_path_full, sizeof(tag_path_full), n);
if (n->root->u.root.absyn &&
(tl = xpath_termlist_by_tagpath(tag_path_full, n)))
{
+ zint max_seqno = 0;
for (; tl; tl = tl->next)
{
/* need to copy recword because it may be changed */
memcpy (&wrd_tl, wrd, sizeof(*wrd));
if (tl->source)
sp_parse(sp, n, &wrd_tl, tl->source);
- if (!tl->index_name)
- {
- /* this is the ! case, so structure is for the xpath index */
- wrd_tl.index_name = xpath_index;
- if (p->flagShowRecords)
- {
- int i;
- printf("%*sXPath index", (level + 1) * 4, "");
- printf (" XData:\"");
- for (i = 0; i<wrd_tl.term_len && i < 40; i++)
- fputc (wrd_tl.term_buf[i], stdout);
- fputc ('"', stdout);
- if (wrd_tl.term_len > 40)
- printf (" ...");
- fputc ('\n', stdout);
- }
- else
- (*p->tokenAdd)(&wrd_tl);
- xpdone = 1;
- } else {
- /* this is just the old fashioned attribute based index */
- wrd_tl.index_name = tl->index_name;
- if (p->flagShowRecords)
- {
- int i;
- printf("%*sIdx: [%s]", (level + 1) * 4, "",
- tl->structure);
- printf("%s %s", tl->index_name, tl->source);
- printf (" XData:\"");
- for (i = 0; i<wrd_tl.term_len && i < 40; i++)
- fputc (wrd_tl.term_buf[i], stdout);
- fputc ('"', stdout);
- if (wrd_tl.term_len > 40)
- printf (" ...");
- fputc ('\n', stdout);
- }
- else
- (*p->tokenAdd)(&wrd_tl);
- }
+
+ /* this is just the old fashioned attribute based index */
+ wrd_tl.index_name = tl->index_name;
+ if (p->flagShowRecords)
+ {
+ int i;
+ printf("%*sIdx: [%s]", (level + 1) * 4, "",
+ tl->structure);
+ printf("%s %s", tl->index_name, tl->source);
+ printf (" XData:\"");
+ for (i = 0; i<wrd_tl.term_len && i < 40; i++)
+ fputc (wrd_tl.term_buf[i], stdout);
+ fputc ('"', stdout);
+ if (wrd_tl.term_len > 40)
+ printf (" ...");
+ fputc ('\n', stdout);
+ }
+ else
+ {
+ (*p->tokenAdd)(&wrd_tl);
+ }
+ if (wrd_tl.seqno > max_seqno)
+ max_seqno = wrd_tl.seqno;
}
+ if (max_seqno)
+ wrd->seqno = max_seqno;
+
}
/* xpath indexing is done, if there was no termlist given,
or no ! in the termlist, and default indexing is enabled... */
- if (!p->flagShowRecords && !xpdone && !termlist_only)
+ if (!p->flagShowRecords && !termlist_only)
{
wrd->index_name = xpath_index;
wrd->index_type = 'w';
else
{
data1_xattr *xp;
- data1_termlist *tl;
- int do_xpindex;
-
- /* Add tag start/end xpath index, only when there is a ! in
- the apropriate xelm directive, or default xpath indexing
- is enabled
- */
- if (!(do_xpindex = 1 - termlist_only))
- {
- if ((tl = xpath_termlist_by_tagpath(tag_path_full, n)))
- {
- for (; tl; tl = tl->next)
- {
- if (!tl->index_name)
- do_xpindex = 1;
- }
- }
- }
- if (do_xpindex) {
+
+ if (!termlist_only)
(*p->tokenAdd)(wrd); /* index element pag (AKA tag path) */
- }
if (xpath_is_start == 1) /* only for the starting tag... */
{
data1_termlist *tll[MAX_ATTR_COUNT];
int i = 0;
-
- /* get termlists for attributes, and find out, if we have to do xpath indexing */
- for (xp = n->u.tag.attributes; xp; xp = xp->next) {
- i++;
- }
-
- i = 0;
for (xp = n->u.tag.attributes; xp; xp = xp->next) {
char comb[512];
- int do_xpindex = 1 - termlist_only;
- data1_termlist *tl;
char attr_tag_path_full[1024];
/* this could be cached as well */
sprintf (attr_tag_path_full, "@%s/%s",
xp->name, tag_path_full);
-
+
tll[i] = xpath_termlist_by_tagpath(attr_tag_path_full,n);
- /* if there is a ! in the xelm termlist, or default indexing is on,
- proceed with xpath idx */
- if ((tl = tll[i]))
+ if (!termlist_only)
{
- for (; tl; tl = tl->next)
- {
- if (!tl->index_name)
- do_xpindex = 1;
- }
- }
-
- if (do_xpindex) {
-
/* attribute (no value) */
wrd->index_type = '0';
- wrd->index_name = ZEBRA_XPATH_ATTR_NAME;
+ wrd->index_name = ZEBRA_XPATH_ATTR_NAME;
wrd->term_buf = xp->name;
wrd->term_len = strlen(xp->name);
wrd->seqno--;
(*p->tokenAdd)(wrd);
- if (xp->value &&
- strlen(xp->name) + strlen(xp->value) < sizeof(comb)-2) {
-
+ if (xp->value
+ &&
+ strlen(xp->name) + strlen(xp->value) < sizeof(comb)-2)
+ {
/* attribute value exact */
strcpy (comb, xp->name);
strcat (comb, "=");
strcat (comb, xp->value);
-
+
wrd->index_name = ZEBRA_XPATH_ATTR_NAME;
wrd->index_type = '0';
wrd->term_buf = comb;
(*p->tokenAdd)(wrd);
}
- }
+ }
i++;
}
sprintf (attr_tag_path_full, "@%s/%s",
xp->name, tag_path_full);
-
if ((tl = tll[i]))
{
/* If there is a termlist given (=xelm directive) */
/* if there was no termlist for the given path,
or the termlist didn't have a ! element, index
the attribute as "w" */
- if ((!xpdone) && (!termlist_only))
+ if (!xpdone && !termlist_only)
{
index_xpath_attr (attr_tag_path_full, xp->name,
xp->value, "w", p, wrd);
int grs_extract_tree(struct recExtractCtrl *p, data1_node *n)
{
- oident oe;
- int oidtmp[OID_SIZE];
RecWord wrd;
- oe.proto = PROTO_Z3950;
- oe.oclass = CLASS_SCHEMA;
- if (n->u.root.absyn)
- {
- oe.value = n->u.root.absyn->reference;
-
- if ((oid_ent_to_oid (&oe, oidtmp)))
- (*p->schemaAdd)(p, oidtmp);
- }
+ if (n->u.root.absyn && n->u.root.absyn->oid)
+ (*p->schemaAdd)(p, n->u.root.absyn->oid);
(*p->init)(p, &wrd);
/* data1_pr_tree(p->dh, n, stdout); */
{
data1_node *n;
struct grs_read_info gri;
- oident oe;
- int oidtmp[OID_SIZE];
RecWord wrd;
gri.stream = p->stream;
n = (*grs_read)(&gri);
if (!n)
return RECCTRL_EXTRACT_EOF;
- oe.proto = PROTO_Z3950;
- oe.oclass = CLASS_SCHEMA;
-#if 0
- if (!n->u.root.absyn)
- return RECCTRL_EXTRACT_ERROR;
-#endif
- if (n->u.root.absyn)
- {
- oe.value = n->u.root.absyn->reference;
- if ((oid_ent_to_oid (&oe, oidtmp)))
- (*p->schemaAdd)(p, oidtmp);
- }
+ if (n->u.root.absyn && n->u.root.absyn->oid)
+ (*p->schemaAdd)(p, n->u.root.absyn->oid);
data1_concat_text(p->dh, mem, n);
/* ensure our data1 tree is UTF-8 */
struct grs_read_info gri;
const char *tagname;
- int requested_schema = VAL_NONE;
+ const Odr_oid *requested_schema = 0;
data1_marctab *marctab;
int dummy;
node = (*grs_read)(&gri);
if (!node)
{
- p->diagnostic = 14;
+ p->diagnostic = YAZ_BIB1_SYSTEM_ERROR_IN_PRESENTING_RECORDS;
nmem_destroy (mem);
return 0;
}
dnew->u.data.len = strlen(dnew->u.data.data);
}
- if (p->input_format == VAL_TEXT_XML)
- zebra_xml_metadata (p, top, mem);
+ if (!p->input_format)
+ { /* SUTRS is default input_format */
+ p->input_format = yaz_oid_recsyn_sutrs;
+ }
+ assert(p->input_format);
+
+ if (!oid_oidcmp(p->input_format, yaz_oid_recsyn_xml))
+ zebra_xml_metadata (p, top, mem);
#if 0
data1_pr_tree (p->dh, node, stdout);
p->comp->u.complex->generic->which == Z_Schema_oid &&
p->comp->u.complex->generic->schema.oid)
{
- oident *oe = oid_getentbyoid (p->comp->u.complex->generic->schema.oid);
- if (oe)
- requested_schema = oe->value;
+ requested_schema = p->comp->u.complex->generic->schema.oid;
}
/* If schema has been specified, map if possible, then check that
* we got the right one
*/
- if (requested_schema != VAL_NONE)
+ if (requested_schema)
{
yaz_log(YLOG_DEBUG, "grs_retrieve: schema mapping");
for (map = node->u.root.absyn->maptabs; map; map = map->next)
{
- if (map->target_absyn_ref == requested_schema)
+ // if (map->target_absyn_ref == requested_schema)
+ if (!oid_oidcmp(map->oid, requested_schema))
{
onode = node;
if (!(node = data1_map_record(p->dh, onode, map, mem)))
{
- p->diagnostic = 14;
+ p->diagnostic = YAZ_BIB1_SYSTEM_ERROR_IN_PRESENTING_RECORDS;
nmem_destroy (mem);
return 0;
}
break;
}
}
- if (node->u.root.absyn &&
- requested_schema != node->u.root.absyn->reference)
+ if (node->u.root.absyn
+ && oid_oidcmp(requested_schema, node->u.root.absyn->oid))
{
- p->diagnostic = 238;
+ p->diagnostic = YAZ_BIB1_RECORD_NOT_AVAILABLE_IN_REQUESTED_SYNTAX;
nmem_destroy (mem);
return 0;
}
if (node->u.root.absyn)
for (map = node->u.root.absyn->maptabs; map; map = map->next)
{
- if (map->target_absyn_ref == p->input_format)
+ if (!oid_oidcmp(map->oid, p->input_format))
{
onode = node;
if (!(node = data1_map_record(p->dh, onode, map, mem)))
{
- p->diagnostic = 14;
+ p->diagnostic = YAZ_BIB1_SYSTEM_ERROR_IN_PRESENTING_RECORDS;
nmem_destroy (mem);
return 0;
}
}
}
yaz_log(YLOG_DEBUG, "grs_retrieve: schemaIdentifier");
- if (node->u.root.absyn &&
- node->u.root.absyn->reference != VAL_NONE &&
- p->input_format == VAL_GRS1)
+ if (node->u.root.absyn && node->u.root.absyn->oid
+ && !oid_oidcmp(p->input_format, yaz_oid_recsyn_grs_1))
{
- oident oe;
- Odr_oid *oid;
- int oidtmp[OID_SIZE];
-
- oe.proto = PROTO_Z3950;
- oe.oclass = CLASS_SCHEMA;
- oe.value = node->u.root.absyn->reference;
-
- if ((oid = oid_ent_to_oid (&oe, oidtmp)))
- {
- char tmp[128];
- data1_handle dh = p->dh;
- char *p = tmp;
- int *ii;
-
- for (ii = oid; *ii >= 0; ii++)
- {
- if (p != tmp)
- *(p++) = '.';
- sprintf(p, "%d", *ii);
- p += strlen(p);
- }
- if ((dnew = data1_mk_tag_data_wd(dh, top,
- "schemaIdentifier", mem)))
- {
- dnew->u.data.what = DATA1I_oid;
- dnew->u.data.data = (char *) nmem_malloc(mem, p - tmp);
- memcpy(dnew->u.data.data, tmp, p - tmp);
- dnew->u.data.len = p - tmp;
- }
+ char oid_str[OID_STR_MAX];
+ char *dot_str = oid_oid_to_dotstring(node->u.root.absyn->oid, oid_str);
+
+ if (dot_str && (dnew = data1_mk_tag_data_wd(p->dh, top,
+ "schemaIdentifier", mem)))
+ {
+ dnew->u.data.what = DATA1I_oid;
+ dnew->u.data.data = (char *) nmem_strdup(mem, dot_str);
+ dnew->u.data.len = strlen(dot_str);
}
}
data1_pr_tree (p->dh, node, stdout);
#endif
yaz_log(YLOG_DEBUG, "grs_retrieve: transfer syntax mapping");
- switch (p->output_format = (p->input_format != VAL_NONE ?
- p->input_format : VAL_SUTRS))
+
+ p->output_format = p->input_format;
+
+ assert(p->input_format);
+ if (!oid_oidcmp(p->input_format, yaz_oid_recsyn_xml))
{
- case VAL_TEXT_XML:
#if 0
data1_pr_tree (p->dh, node, stdout);
#endif
if (!(p->rec_buf = data1_nodetoidsgml(p->dh, node, selected,
&p->rec_len)))
- p->diagnostic = 238;
+ p->diagnostic = YAZ_BIB1_RECORD_NOT_AVAILABLE_IN_REQUESTED_SYNTAX;
else
{
char *new_buf = (char*) odr_malloc (p->odr, p->rec_len);
memcpy (new_buf, p->rec_buf, p->rec_len);
p->rec_buf = new_buf;
}
- break;
- case VAL_GRS1:
+ }
+ else if (!oid_oidcmp(p->input_format, yaz_oid_recsyn_grs_1))
+ {
data1_iconv (p->dh, mem, node, "UTF-8", data1_get_encoding(p->dh, node));
dummy = 0;
if (!(p->rec_buf = data1_nodetogr(p->dh, node, selected,
p->odr, &dummy)))
- p->diagnostic = 238; /* not available in requested syntax */
+ p->diagnostic = YAZ_BIB1_RECORD_NOT_AVAILABLE_IN_REQUESTED_SYNTAX;
else
p->rec_len = -1;
- break;
- case VAL_EXPLAIN:
+ }
+ else if (!oid_oidcmp(p->input_format, yaz_oid_recsyn_explain))
+ {
/* ensure our data1 tree is UTF-8 */
data1_iconv (p->dh, mem, node, "UTF-8", data1_get_encoding(p->dh, node));
if (!(p->rec_buf = data1_nodetoexplain(p->dh, node, selected,
p->odr)))
- p->diagnostic = 238;
+ p->diagnostic = YAZ_BIB1_RECORD_NOT_AVAILABLE_IN_REQUESTED_SYNTAX;
else
p->rec_len = -1;
- break;
- case VAL_SUMMARY:
+ }
+ else if (!oid_oidcmp(p->input_format, yaz_oid_recsyn_summary))
+ {
/* ensure our data1 tree is UTF-8 */
data1_iconv (p->dh, mem, node, "UTF-8", data1_get_encoding(p->dh, node));
if (!(p->rec_buf = data1_nodetosummary(p->dh, node, selected,
p->odr)))
- p->diagnostic = 238;
+ p->diagnostic = YAZ_BIB1_RECORD_NOT_AVAILABLE_IN_REQUESTED_SYNTAX;
else
p->rec_len = -1;
- break;
- case VAL_SUTRS:
+ }
+ else if (!oid_oidcmp(p->input_format, yaz_oid_recsyn_sutrs))
+ {
if (p->encoding)
data1_iconv (p->dh, mem, node, p->encoding,
data1_get_encoding(p->dh, node));
if (!(p->rec_buf = data1_nodetobuf(p->dh, node, selected,
&p->rec_len)))
- p->diagnostic = 238;
+ p->diagnostic = YAZ_BIB1_RECORD_NOT_AVAILABLE_IN_REQUESTED_SYNTAX;
else
{
char *new_buf = (char*) odr_malloc (p->odr, p->rec_len);
memcpy (new_buf, p->rec_buf, p->rec_len);
p->rec_buf = new_buf;
}
- break;
- case VAL_SOIF:
+ }
+ else if (!oid_oidcmp(p->input_format, yaz_oid_recsyn_soif))
+ {
if (p->encoding)
data1_iconv (p->dh, mem, node, p->encoding,
data1_get_encoding(p->dh, node));
if (!(p->rec_buf = data1_nodetosoif(p->dh, node, selected,
&p->rec_len)))
- p->diagnostic = 238;
+ p->diagnostic = YAZ_BIB1_RECORD_NOT_AVAILABLE_IN_REQUESTED_SYNTAX;
else
{
char *new_buf = (char*) odr_malloc (p->odr, p->rec_len);
memcpy (new_buf, p->rec_buf, p->rec_len);
p->rec_buf = new_buf;
}
- break;
- default:
+ }
+ else
+ {
if (!node->u.root.absyn)
- {
- p->diagnostic = 238;
- break;
- }
- for (marctab = node->u.root.absyn->marc; marctab;
- marctab = marctab->next)
- if (marctab->reference == p->input_format)
- break;
- if (!marctab)
- {
- p->diagnostic = 238;
- break;
- }
- if (p->encoding)
- data1_iconv (p->dh, mem, node, p->encoding,
- data1_get_encoding(p->dh, node));
- if (!(p->rec_buf = data1_nodetomarc(p->dh, marctab, node,
- selected, &p->rec_len)))
- p->diagnostic = 238;
- else
- {
- char *new_buf = (char*) odr_malloc (p->odr, p->rec_len);
- memcpy (new_buf, p->rec_buf, p->rec_len);
- p->rec_buf = new_buf;
- }
+ p->diagnostic = YAZ_BIB1_RECORD_NOT_AVAILABLE_IN_REQUESTED_SYNTAX;
+ else
+ {
+ for (marctab = node->u.root.absyn->marc; marctab;
+ marctab = marctab->next)
+ if (marctab->oid && !oid_oidcmp(marctab->oid, p->input_format))
+ break;
+ if (!marctab)
+ p->diagnostic = YAZ_BIB1_RECORD_NOT_AVAILABLE_IN_REQUESTED_SYNTAX;
+ else
+ {
+ if (p->encoding)
+ data1_iconv (p->dh, mem, node, p->encoding,
+ data1_get_encoding(p->dh, node));
+ if (!(p->rec_buf = data1_nodetomarc(p->dh, marctab, node,
+ selected, &p->rec_len)))
+ p->diagnostic = YAZ_BIB1_RECORD_NOT_AVAILABLE_IN_REQUESTED_SYNTAX;
+ else
+ {
+ char *new_buf = (char*) odr_malloc (p->odr, p->rec_len);
+ memcpy (new_buf, p->rec_buf, p->rec_len);
+ p->rec_buf = new_buf;
+ }
+ }
+ }
}
nmem_destroy(mem);
return 0;