X-Git-Url: http://git.indexdata.com/?a=blobdiff_plain;f=index%2Frecgrs.c;h=c62d1a964ccca36840523a0cf9d4a5451155c7c9;hb=53f50a1b1dd002ef484a41f50f3598386335cae1;hp=c85b635817b96b799a515db5de23ff61e871a191;hpb=89d3a004b7c651fd5673abfc192e1472dc4d4197;p=idzebra-moved-to-github.git diff --git a/index/recgrs.c b/index/recgrs.c index c85b635..c62d1a9 100644 --- a/index/recgrs.c +++ b/index/recgrs.c @@ -1,4 +1,4 @@ -/* $Id: recgrs.c,v 1.13 2007-01-15 15:10:17 adam Exp $ +/* $Id: recgrs.c,v 1.19 2007-05-08 12:50:04 adam Exp $ Copyright (C) 1995-2007 Index Data ApS @@ -26,7 +26,8 @@ Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA #include #include -#include +#include +#include #include #include @@ -525,25 +526,6 @@ static void mk_tag_path_full(char *tag_path_full, size_t max, data1_node *n) } -static void index_staticrank(struct recExtractCtrl *p, - RecWord *wrd, - data1_absyn *absyn) -{ - const char *staticrank_index = data1_absyn_get_staticrank(absyn); - - if (staticrank_index && !strcmp(wrd->index_name, staticrank_index)) - { - char valz[20]; - size_t len = wrd->term_len; - - if (len > sizeof(valz)-1) - len = sizeof(valz)-1; - memcpy(valz, wrd->term_buf, len); - valz[len] = '\0'; - p->staticrank = atozint(valz); - } -} - static void index_xpath(struct source_parser *sp, data1_node *n, struct recExtractCtrl *p, int level, RecWord *wrd, @@ -555,7 +537,6 @@ static void index_xpath(struct source_parser *sp, data1_node *n, char tag_path_full[1024]; int termlist_only = 1; data1_termlist *tl; - int xpdone = 0; if (!n->root->u.root.absyn || @@ -570,7 +551,6 @@ static void index_xpath(struct source_parser *sp, data1_node *n, 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); @@ -607,7 +587,6 @@ static void index_xpath(struct source_parser *sp, data1_node *n, else { (*p->tokenAdd)(&wrd_tl); - index_staticrank(p, &wrd_tl, n->root->u.root.absyn); } if (wrd_tl.seqno > max_seqno) max_seqno = wrd_tl.seqno; @@ -618,7 +597,7 @@ static void index_xpath(struct source_parser *sp, data1_node *n, } /* 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'; @@ -626,8 +605,6 @@ static void index_xpath(struct source_parser *sp, data1_node *n, } break; case DATA1N_tag: - if (termlist_only) - return; mk_tag_path_full(tag_path_full, sizeof(tag_path_full), n); wrd->index_type = '0'; @@ -647,7 +624,8 @@ static void index_xpath(struct source_parser *sp, data1_node *n, { data1_xattr *xp; - (*p->tokenAdd)(wrd); /* index element pag (AKA tag path) */ + if (!termlist_only) + (*p->tokenAdd)(wrd); /* index element pag (AKA tag path) */ if (xpath_is_start == 1) /* only for the starting tag... */ { @@ -662,35 +640,38 @@ static void index_xpath(struct source_parser *sp, data1_node *n, /* 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); - /* attribute (no value) */ - wrd->index_type = '0'; - 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 (!termlist_only) { - /* attribute value exact */ - strcpy (comb, xp->name); - strcat (comb, "="); - strcat (comb, xp->value); - - wrd->index_name = ZEBRA_XPATH_ATTR_NAME; + /* attribute (no value) */ wrd->index_type = '0'; - wrd->term_buf = comb; - wrd->term_len = strlen(comb); - wrd->seqno--; + 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) + { + /* 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; + wrd->term_len = strlen(comb); + wrd->seqno--; + + (*p->tokenAdd)(wrd); + } + } i++; } @@ -702,7 +683,6 @@ static void index_xpath(struct source_parser *sp, data1_node *n, sprintf (attr_tag_path_full, "@%s/%s", xp->name, tag_path_full); - if ((tl = tll[i])) { /* If there is a termlist given (=xelm directive) */ @@ -724,8 +704,6 @@ static void index_xpath(struct source_parser *sp, data1_node *n, wrd->term_buf = xp->value; wrd->term_len = strlen(xp->value); (*p->tokenAdd)(wrd); - index_staticrank(p, wrd, - n->root->u.root.absyn); } } } @@ -733,7 +711,7 @@ static void index_xpath(struct source_parser *sp, data1_node *n, /* 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); @@ -793,7 +771,6 @@ static void index_termlist (struct source_parser *sp, data1_node *par, { wrd->index_type = *tlist->structure; wrd->index_name = tlist->index_name; - index_staticrank(p, wrd, n->root->u.root.absyn); (*p->tokenAdd)(wrd); } } @@ -906,19 +883,10 @@ static int dumpkeys(data1_node *n, struct recExtractCtrl *p, RecWord *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); */ @@ -932,8 +900,6 @@ static int grs_extract_sub(void *clientData, struct recExtractCtrl *p, { data1_node *n; struct grs_read_info gri; - oident oe; - int oidtmp[OID_SIZE]; RecWord wrd; gri.stream = p->stream; @@ -944,18 +910,8 @@ static int grs_extract_sub(void *clientData, struct recExtractCtrl *p, 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 */ @@ -1123,7 +1079,7 @@ int zebra_grs_retrieve(void *clientData, struct recRetrieveCtrl *p, struct grs_read_info gri; const char *tagname; - int requested_schema = VAL_NONE; + const Odr_oid *requested_schema = 0; data1_marctab *marctab; int dummy; @@ -1137,7 +1093,7 @@ int zebra_grs_retrieve(void *clientData, struct recRetrieveCtrl *p, node = (*grs_read)(&gri); if (!node) { - p->diagnostic = 14; + p->diagnostic = YAZ_BIB1_SYSTEM_ERROR_IN_PRESENTING_RECORDS; nmem_destroy (mem); return 0; } @@ -1185,8 +1141,14 @@ int zebra_grs_retrieve(void *clientData, struct recRetrieveCtrl *p, 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); @@ -1196,34 +1158,33 @@ int zebra_grs_retrieve(void *clientData, struct recRetrieveCtrl *p, 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; } @@ -1237,12 +1198,12 @@ int zebra_grs_retrieve(void *clientData, struct recRetrieveCtrl *p, 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; } @@ -1250,40 +1211,18 @@ int zebra_grs_retrieve(void *clientData, struct recRetrieveCtrl *p, } } 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); } } @@ -1302,10 +1241,12 @@ int zebra_grs_retrieve(void *clientData, struct recRetrieveCtrl *p, 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 @@ -1316,97 +1257,103 @@ int zebra_grs_retrieve(void *clientData, struct recRetrieveCtrl *p, 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;