X-Git-Url: http://git.indexdata.com/?p=idzebra-moved-to-github.git;a=blobdiff_plain;f=recctrl%2Frecgrs.c;h=1c34c2e72718193962ef3f0de2a09a18e62bf532;hp=cc7e2f60e181000ff6fb653c0a25e7fb48b91bfc;hb=4eb3b54bb2ca9af74f39f000d3d40dba99ded887;hpb=cc4289d73c2b872c09d221d5e1188d3cdd9da438 diff --git a/recctrl/recgrs.c b/recctrl/recgrs.c index cc7e2f6..1c34c2e 100644 --- a/recctrl/recgrs.c +++ b/recctrl/recgrs.c @@ -1,4 +1,4 @@ -/* $Id: recgrs.c,v 1.89 2004-08-24 14:29:09 adam Exp $ +/* $Id: recgrs.c,v 1.95 2004-12-13 20:51:32 adam Exp $ Copyright (C) 1995,1996,1997,1998,1999,2000,2001,2002,2003,2004 Index Data Aps @@ -25,107 +25,17 @@ Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA #include #ifndef WIN32 #include +#include #endif #include #include -#include -#include "grsread.h" +#include +#include #define GRS_MAX_WORD 512 -struct grs_handler { - RecTypeGrs type; - void *clientData; - int initFlag; - struct grs_handler *next; -}; - -struct grs_handlers { - struct grs_handler *handlers; -}; - -static int read_grs_type (struct grs_handlers *h, - struct grs_read_info *p, const char *type, - data1_node **root) -{ - struct grs_handler *gh = h->handlers; - const char *cp = strchr (type, '.'); - - if (cp == NULL || cp == type) - { - cp = strlen(type) + type; - *p->type = 0; - } - else - strcpy (p->type, cp+1); - for (gh = h->handlers; gh; gh = gh->next) - { - if (!memcmp (type, gh->type->type, cp-type) && - gh->type->type[cp-type] == '\0') - { - if (!gh->initFlag) - { - gh->initFlag = 1; - gh->clientData = (*gh->type->init)(); - } - p->clientData = gh->clientData; - *root = (gh->type->read)(p); - gh->clientData = p->clientData; - return 0; - } - } - return 1; -} - -static void grs_add_handler (struct grs_handlers *h, RecTypeGrs t) -{ - struct grs_handler *gh = (struct grs_handler *) xmalloc (sizeof(*gh)); - gh->next = h->handlers; - h->handlers = gh; - gh->initFlag = 0; - gh->clientData = 0; - gh->type = t; -} - -static void *grs_init(RecType recType) -{ - struct grs_handlers *h = (struct grs_handlers *) xmalloc (sizeof(*h)); - h->handlers = 0; - - grs_add_handler (h, recTypeGrs_sgml); - grs_add_handler (h, recTypeGrs_regx); -#if HAVE_TCL_H - grs_add_handler (h, recTypeGrs_tcl); -#endif - grs_add_handler (h, recTypeGrs_marc); - grs_add_handler (h, recTypeGrs_marcxml); -#if HAVE_EXPAT_H - grs_add_handler (h, recTypeGrs_xml); -#endif -#if HAVE_PERL - grs_add_handler (h, recTypeGrs_perl); -#endif - grs_add_handler (h, recTypeGrs_danbib); - return h; -} - -static void grs_destroy(void *clientData) -{ - struct grs_handlers *h = (struct grs_handlers *) clientData; - struct grs_handler *gh = h->handlers, *gh_next; - while (gh) - { - gh_next = gh->next; - if (gh->initFlag) - (*gh->type->destroy)(gh->clientData); - xfree (gh); - gh = gh_next; - } - xfree (h); -} - struct source_parser { int len; const char *tok; @@ -270,26 +180,26 @@ int d1_check_xpath_predicate(data1_node *n, struct xpath_predicate *p) if (p->which == XPATH_PREDICATE_RELATION) { if (p->u.relation.name[0]) { if (*p->u.relation.name != '@') { - yaz_log(LOG_WARN, + yaz_log(YLOG_WARN, " Only attributes (@) are supported in xelm xpath predicates"); - yaz_log(LOG_WARN, "predicate %s ignored", p->u.relation.name); + yaz_log(YLOG_WARN, "predicate %s ignored", p->u.relation.name); return (1); } attname = p->u.relation.name + 1; res = 0; /* looking for the attribute with a specified name */ for (attr = n->u.tag.attributes; attr; attr = attr->next) { - yaz_log(LOG_DEBUG," - attribute %s <-> %s", attname, attr->name ); + 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(LOG_WARN, + yaz_log(YLOG_WARN, "Only '=' relation is supported (%s)",p->u.relation.op); - yaz_log(LOG_WARN, "predicate %s ignored", p->u.relation.name); + yaz_log(YLOG_WARN, "predicate %s ignored", p->u.relation.name); res = 1; break; } else { - yaz_log(LOG_DEBUG," - value %s <-> %s", + yaz_log(YLOG_DEBUG," - value %s <-> %s", p->u.relation.value, attr->value ); if (!strcmp(attr->value, p->u.relation.value)) { res = 1; break; @@ -301,7 +211,7 @@ int d1_check_xpath_predicate(data1_node *n, struct xpath_predicate *p) } } } - yaz_log(LOG_DEBUG, "return %d", res); + yaz_log(YLOG_DEBUG, "return %d", res); return res; } else { return 1; @@ -316,7 +226,7 @@ int d1_check_xpath_predicate(data1_node *n, struct xpath_predicate *p) return (d1_check_xpath_predicate(n, p->u.boolean.left) || d1_check_xpath_predicate(n, p->u.boolean.right)); } else { - yaz_log(LOG_WARN, "Unknown boolean relation %s, ignored",p->u.boolean.op); + yaz_log(YLOG_WARN, "Unknown boolean relation %s, ignored",p->u.boolean.op); return 1; } } @@ -355,7 +265,7 @@ data1_termlist *xpath_termlist_by_tagpath(char *tagpath, data1_node *n) int ok = 0; sprintf (pexpr, "%s\n", tagpath); - yaz_log(LOG_DEBUG,"Checking tagpath %s",tagpath); + yaz_log(YLOG_DEBUG,"Checking tagpath %s",tagpath); while (xpe) { struct DFA_state **dfaar = xpe->dfa->states; @@ -383,9 +293,9 @@ data1_termlist *xpath_termlist_by_tagpath(char *tagpath, data1_node *n) } while (i >= 0); } if (ok) - yaz_log(LOG_DEBUG," xpath match %s",xpe->xpath_expr); + yaz_log(YLOG_DEBUG," xpath match %s",xpe->xpath_expr); else - yaz_log(LOG_DEBUG," xpath no match %s",xpe->xpath_expr); + yaz_log(YLOG_DEBUG," xpath no match %s",xpe->xpath_expr); pexpr--; if (ok) { @@ -402,11 +312,11 @@ data1_termlist *xpath_termlist_by_tagpath(char *tagpath, data1_node *n) backwards trough xpath location steps ... */ for (i=xpe->xpath_len - 1; i>0; i--) { - yaz_log(LOG_DEBUG,"Checking step %d: %s on tag %s", + 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(LOG_DEBUG," Predicates didn't match"); + yaz_log(YLOG_DEBUG," Predicates didn't match"); ok = 0; break; } @@ -426,7 +336,7 @@ data1_termlist *xpath_termlist_by_tagpath(char *tagpath, data1_node *n) xfree(pexpr); if (ok) { - yaz_log(LOG_DEBUG,"Got it"); + yaz_log(YLOG_DEBUG,"Got it"); return xpe->termlists; } else { return NULL; @@ -490,7 +400,7 @@ static void index_xpath (data1_node *n, struct recExtractCtrl *p, data1_termlist *tl; int xpdone = 0; - yaz_log(LOG_DEBUG, "index_xpath level=%d use=%d", level, use); + yaz_log(YLOG_DEBUG, "index_xpath level=%d use=%d", level, use); if ((!n->root->u.root.absyn) || (n->root->u.root.absyn->enable_xpath_indexing)) { termlist_only = 0; @@ -923,8 +833,9 @@ int grs_extract_tree(struct recExtractCtrl *p, data1_node *n) return dumpkeys(n, p, 0, &wrd); } -static int grs_extract_sub(struct grs_handlers *h, struct recExtractCtrl *p, - NMEM mem) +static int grs_extract_sub(void *clientData, struct recExtractCtrl *p, + NMEM mem, + data1_node *(*grs_read)(struct grs_read_info *)) { data1_node *n; struct grs_read_info gri; @@ -940,9 +851,9 @@ static int grs_extract_sub(struct grs_handlers *h, struct recExtractCtrl *p, gri.offset = p->offset; gri.mem = mem; gri.dh = p->dh; + gri.clientData = clientData; - if (read_grs_type (h, &gri, p->subType, &n)) - return RECCTRL_EXTRACT_ERROR_NO_SUCH_FILTER; + n = (*grs_read)(&gri); if (!n) return RECCTRL_EXTRACT_EOF; oe.proto = PROTO_Z3950; @@ -976,13 +887,12 @@ static int grs_extract_sub(struct grs_handlers *h, struct recExtractCtrl *p, return RECCTRL_EXTRACT_OK; } -static int grs_extract(void *clientData, struct recExtractCtrl *p) +int zebra_grs_extract(void *clientData, struct recExtractCtrl *p, + data1_node *(*grs_read)(struct grs_read_info *)) { int ret; NMEM mem = nmem_create (); - struct grs_handlers *h = (struct grs_handlers *) clientData; - - ret = grs_extract_sub(h, p, mem); + ret = grs_extract_sub(clientData, p, mem, grs_read); nmem_destroy(mem); return ret; } @@ -990,7 +900,8 @@ static int grs_extract(void *clientData, struct recExtractCtrl *p) /* * Return: -1: Nothing done. 0: Ok. >0: Bib-1 diagnostic. */ -static int process_comp(data1_handle dh, data1_node *n, Z_RecordComposition *c) +static int process_comp(data1_handle dh, data1_node *n, Z_RecordComposition *c, + char **addinfo, ODR o) { data1_esetname *eset; Z_Espec1 *espec = 0; @@ -1004,10 +915,11 @@ static int process_comp(data1_handle dh, data1_node *n, Z_RecordComposition *c) if (!(eset = data1_getesetbyname(dh, n->u.root.absyn, c->u.simple->u.generic))) { - yaz_log(LOG_LOG, "Unknown esetname '%s'", c->u.simple->u.generic); + yaz_log(YLOG_LOG, "Unknown esetname '%s'", c->u.simple->u.generic); + *addinfo = odr_strdup(o, c->u.simple->u.generic); return 25; /* invalid esetname */ } - yaz_log(LOG_DEBUG, "Esetname '%s' in simple compspec", + yaz_log(YLOG_DEBUG, "Esetname '%s' in simple compspec", c->u.simple->u.generic); espec = eset->spec; break; @@ -1024,23 +936,24 @@ static int process_comp(data1_handle dh, data1_node *n, Z_RecordComposition *c) data1_getesetbyname(dh, n->u.root.absyn, p->u.elementSetName))) { - yaz_log(LOG_LOG, "Unknown esetname '%s'", + yaz_log(YLOG_DEBUG, "Unknown esetname '%s'", p->u.elementSetName); + *addinfo = odr_strdup(o, p->u.elementSetName); return 25; /* invalid esetname */ } - yaz_log(LOG_DEBUG, "Esetname '%s' in complex compspec", + yaz_log(YLOG_DEBUG, "Esetname '%s' in complex compspec", p->u.elementSetName); espec = eset->spec; break; case Z_ElementSpec_externalSpec: if (p->u.externalSpec->which == Z_External_espec1) { - yaz_log(LOG_DEBUG, "Got Espec-1"); + yaz_log(YLOG_DEBUG, "Got Espec-1"); espec = p->u.externalSpec-> u.espec1; } else { - yaz_log(LOG_LOG, "Unknown external espec."); + yaz_log(YLOG_LOG, "Unknown external espec."); return 25; /* bad. what is proper diagnostic? */ } break; @@ -1052,12 +965,12 @@ static int process_comp(data1_handle dh, data1_node *n, Z_RecordComposition *c) } if (espec) { - yaz_log(LOG_DEBUG, "Element: Espec-1 match"); + yaz_log(YLOG_DEBUG, "Element: Espec-1 match"); return data1_doespec1(dh, n, espec); } else { - yaz_log(LOG_DEBUG, "Element: all match"); + yaz_log(YLOG_DEBUG, "Element: all match"); return -1; } } @@ -1110,7 +1023,8 @@ static void zebra_xml_metadata (struct recRetrieveCtrl *p, data1_node *top, data1_mk_text (p->dh, mem, i2, n); } -static int grs_retrieve(void *clientData, struct recRetrieveCtrl *p) +int zebra_grs_retrieve(void *clientData, struct recRetrieveCtrl *p, + data1_node *(*grs_read)(struct grs_read_info *)) { data1_node *node = 0, *onode = 0, *top; data1_node *dnew; @@ -1119,7 +1033,7 @@ static int grs_retrieve(void *clientData, struct recRetrieveCtrl *p) NMEM mem; struct grs_read_info gri; const char *tagname; - struct grs_handlers *h = (struct grs_handlers *) clientData; + int requested_schema = VAL_NONE; data1_marctab *marctab; int dummy; @@ -1133,14 +1047,10 @@ static int grs_retrieve(void *clientData, struct recRetrieveCtrl *p) gri.offset = 0; gri.mem = mem; gri.dh = p->dh; + gri.clientData = clientData; - yaz_log(LOG_DEBUG, "grs_retrieve"); - if (read_grs_type (h, &gri, p->subType, &node)) - { - p->diagnostic = 14; - nmem_destroy (mem); - return 0; - } + yaz_log(YLOG_DEBUG, "grs_retrieve"); + node = (*grs_read)(&gri); if (!node) { p->diagnostic = 14; @@ -1157,7 +1067,7 @@ static int grs_retrieve(void *clientData, struct recRetrieveCtrl *p) #endif top = data1_get_root_tag (p->dh, node); - yaz_log(LOG_DEBUG, "grs_retrieve: size"); + yaz_log(YLOG_DEBUG, "grs_retrieve: size"); tagname = data1_systag_lookup(node->u.root.absyn, "size", "size"); if (tagname && (dnew = data1_mk_tag_data_wd(p->dh, top, tagname, mem))) @@ -1172,7 +1082,7 @@ static int grs_retrieve(void *clientData, struct recRetrieveCtrl *p) if (tagname && p->score >= 0 && (dnew = data1_mk_tag_data_wd(p->dh, top, tagname, mem))) { - yaz_log(LOG_DEBUG, "grs_retrieve: %s", tagname); + yaz_log(YLOG_DEBUG, "grs_retrieve: %s", tagname); dnew->u.data.what = DATA1I_num; dnew->u.data.data = dnew->lbuf; sprintf(dnew->u.data.data, "%d", p->score); @@ -1184,13 +1094,17 @@ static int grs_retrieve(void *clientData, struct recRetrieveCtrl *p) if (tagname && p->localno > 0 && (dnew = data1_mk_tag_data_wd(p->dh, top, tagname, mem))) { - yaz_log(LOG_DEBUG, "grs_retrieve: %s", tagname); + yaz_log(YLOG_DEBUG, "grs_retrieve: %s", tagname); dnew->u.data.what = DATA1I_text; dnew->u.data.data = dnew->lbuf; sprintf(dnew->u.data.data, ZINT_FORMAT, p->localno); dnew->u.data.len = strlen(dnew->u.data.data); } + + if (p->input_format == VAL_TEXT_XML) + zebra_xml_metadata (p, top, mem); + #if 0 data1_pr_tree (p->dh, node, stdout); #endif @@ -1219,7 +1133,7 @@ static int grs_retrieve(void *clientData, struct recRetrieveCtrl *p) */ if (requested_schema != VAL_NONE) { - yaz_log(LOG_DEBUG, "grs_retrieve: schema mapping"); + 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) @@ -1247,7 +1161,7 @@ static int grs_retrieve(void *clientData, struct recRetrieveCtrl *p) * the overlap of schema and formatting which is inherent in the MARC * family) */ - yaz_log(LOG_DEBUG, "grs_retrieve: syntax mapping"); + yaz_log(YLOG_DEBUG, "grs_retrieve: syntax mapping"); if (node->u.root.absyn) for (map = node->u.root.absyn->maptabs; map; map = map->next) { @@ -1263,7 +1177,7 @@ static int grs_retrieve(void *clientData, struct recRetrieveCtrl *p) break; } } - yaz_log(LOG_DEBUG, "grs_retrieve: schemaIdentifier"); + yaz_log(YLOG_DEBUG, "grs_retrieve: schemaIdentifier"); if (node->u.root.absyn && node->u.root.absyn->reference != VAL_NONE && p->input_format == VAL_GRS1) @@ -1301,8 +1215,9 @@ static int grs_retrieve(void *clientData, struct recRetrieveCtrl *p) } } - yaz_log(LOG_DEBUG, "grs_retrieve: element spec"); - if (p->comp && (res = process_comp(p->dh, node, p->comp)) > 0) + yaz_log(YLOG_DEBUG, "grs_retrieve: element spec"); + if (p->comp && (res = process_comp(p->dh, node, p->comp, &p->addinfo, + p->odr)) > 0) { p->diagnostic = res; if (onode) @@ -1317,12 +1232,11 @@ static int grs_retrieve(void *clientData, struct recRetrieveCtrl *p) #if 0 data1_pr_tree (p->dh, node, stdout); #endif - yaz_log(LOG_DEBUG, "grs_retrieve: transfer syntax mapping"); + yaz_log(YLOG_DEBUG, "grs_retrieve: transfer syntax mapping"); switch (p->output_format = (p->input_format != VAL_NONE ? p->input_format : VAL_SUTRS)) { case VAL_TEXT_XML: - zebra_xml_metadata (p, top, mem); #if 0 data1_pr_tree (p->dh, node, stdout); @@ -1422,13 +1336,3 @@ static int grs_retrieve(void *clientData, struct recRetrieveCtrl *p) return 0; } -static struct recType grs_type = -{ - "grs", - grs_init, - grs_destroy, - grs_extract, - grs_retrieve -}; - -RecType recTypeGrs = &grs_type;