X-Git-Url: http://git.indexdata.com/?p=idzebra-moved-to-github.git;a=blobdiff_plain;f=index%2Fmod_dom.c;h=fd5986669becc8c40b9135c4816f064388cfce66;hp=0dcf999820d9bc686a8445311a5fe3e9778101e8;hb=a66b7d79383ae700f3358731eecfe2aafed0e90d;hpb=50d5984f510e0434dc6811b46c822fc2301c4450 diff --git a/index/mod_dom.c b/index/mod_dom.c index 0dcf999..fd59866 100644 --- a/index/mod_dom.c +++ b/index/mod_dom.c @@ -1,5 +1,5 @@ /* This file is part of the Zebra server. - Copyright (C) 1995-2008 Index Data + Copyright (C) 1994-2010 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 @@ -338,7 +338,7 @@ static ZEBRA_RES parse_convert(struct filter_info *tinfo, xmlNodePtr ptr, *l = p; l = &p->next; } - else if (!XML_STRCMP(ptr->name, "meta")) + else if (!XML_STRCMP(ptr->name, "process-meta")) { struct _xmlAttr *attr; struct convert_s *p = nmem_malloc(tinfo->nmem_config, sizeof(*p)); @@ -376,13 +376,12 @@ static int process_meta(struct filter_info *tinfo, xmlDocPtr doc, xmlNodePtr nod struct _xmlAttr *attr; for (attr = node->properties; attr; attr = attr->next) { - if (attr_content(attr, "element_set_name", &element_set_name)) + if (attr_content(attr, "name", &element_set_name)) ; else { dom_log(YLOG_WARN, tinfo, node, - "bad attribute @%s, expected @element_set_name", - attr->name); + "bad attribute @%s, expected @name", attr->name); } } if (element_set_name) @@ -400,11 +399,11 @@ static int process_meta(struct filter_info *tinfo, xmlDocPtr doc, xmlNodePtr nod if (ret == 0) { xmlDocPtr sub_doc = - xmlParseMemory( wrbuf_buf(result), wrbuf_len(result)); + xmlParseMemory(wrbuf_buf(result), wrbuf_len(result)); if (sub_doc) { xmlNodePtr t = xmlDocGetRootElement(sub_doc); - xmlAddChild(node, xmlCopyNode(t, 1)); + xmlReplaceNode(node, xmlCopyNode(t, 1)); xmlFreeDoc(sub_doc); } } @@ -814,10 +813,18 @@ static void index_value_of(struct filter_info *tinfo, { xmlChar *text = xmlNodeGetContent(node); size_t text_len = strlen((const char *)text); - + /* if there is no text, we do not need to proceed */ if (text_len) { + /* keep seqno base so that all text will have + identical seqno's for multiple fields , e.g + .. */ + + zint seqno_base = recword->seqno; + zint seqno_max = recword->seqno; + + const char *look = index_p; const char *bval; const char *eval; @@ -865,6 +872,7 @@ static void index_value_of(struct filter_info *tinfo, /* actually indexing the text given */ + recword->seqno = seqno_base; recword->index_name = (const char *)index; if (*type) recword->index_type = (const char *) type; @@ -879,12 +887,16 @@ static void index_value_of(struct filter_info *tinfo, (extctr->tokenAdd)(recword); + if (seqno_max < recword->seqno) + seqno_max = recword->seqno; + /* eat whitespaces */ if (*look && ' ' == *look) { look++; } } + recword->seqno = seqno_max; } xmlFree(text); } @@ -909,7 +921,13 @@ static void set_record_info(struct filter_info *tinfo, if (id_p && *id_p) - sscanf((const char *)id_p, "%255s", extctr->match_criteria); + { + size_t l = strlen(id_p); + if (l >= sizeof(extctr->match_criteria)) + l = sizeof(extctr->match_criteria)-1; + memcpy(extctr->match_criteria, id_p, l); + extctr->match_criteria[l] = '\0'; + } if (rank_p && *rank_p) extctr->staticrank = atozint((const char *)rank_p); @@ -1009,9 +1027,6 @@ static int attr_content_pi(const char **c_ptr, const char *name, const char *look = *c_ptr; int ret = 0; - *value = '\0'; - while (*look && ' ' == *look) - look++; if (strlen(look) > name_len) { if (look[name_len] == '=' && !memcmp(look, name, name_len)) @@ -1028,8 +1043,6 @@ static int attr_content_pi(const char **c_ptr, const char *name, ret = 1; } } - while (*look && ' ' == *look) - look++; *c_ptr = look; return ret; } @@ -1057,18 +1070,27 @@ static void process_xml_pi_node(struct filter_info *tinfo, *rank = '\0'; *type = '\0'; look += 6; - while (*look) + for (;;) + { + /* eat whitespace */ + while (' ' == *look) + look++; + if (*look == '\0') + break; if (attr_content_pi(&look, "id", id, sizeof(id))) ; else if (attr_content_pi(&look, "rank", rank, sizeof(rank))) ; else if (attr_content_pi(&look, "type", type, sizeof(type))) + ; + else { dom_log(YLOG_WARN, tinfo, node, "content '%s', can not parse '%s'", pi_p, look); break; } + } set_record_info(tinfo, extctr, node, id, rank, type); } /* parsing index instruction */ @@ -1151,7 +1173,6 @@ static int convert_extract_doc(struct filter_info *tinfo, int len_out; const char *params[10]; xsltStylesheetPtr last_xsp = 0; - xmlDocPtr store_doc = 0; /* per default do not ingest record */ tinfo->record_info_invoked = 0; @@ -1168,41 +1189,44 @@ static int convert_extract_doc(struct filter_info *tinfo, { xmlChar *buf_out; int len_out; + xmlDocDumpMemory(doc, &buf_out, &len_out); #if 0 FILE *outf = fopen("extract.xml", "w"); - xmlDocDumpMemory(doc, &buf_out, &len_out); fwrite(buf_out, 1, len_out, outf); -#endif - yaz_log(YLOG_LOG, "Extract Doc: %.*s", len_out, buf_out); -#if 0 fclose(outf); #endif + yaz_log(YLOG_LOG, "Extract Doc: %.*s", len_out, buf_out); } - /* input conversion */ - perform_convert(tinfo, p, 0, input->convert, params, &doc, 0); - - if (tinfo->store) + if (p->setStoreData) { - /* store conversion */ - store_doc = xmlCopyDoc(doc, 1); - perform_convert(tinfo, p, 0, tinfo->store->convert, - params, &store_doc, &last_xsp); - } - - /* saving either store doc or original doc in case no store doc exists */ - if (last_xsp) - xsltSaveResultToString(&buf_out, &len_out, - store_doc ? store_doc : doc, last_xsp); - else - xmlDocDumpMemory(store_doc ? store_doc : doc, &buf_out, &len_out); + xmlDocPtr store_doc = 0; - if (p->setStoreData) - (*p->setStoreData)(p, buf_out, len_out); - xmlFree(buf_out); + /* input conversion */ + perform_convert(tinfo, p, 0, input->convert, params, &doc, 0); + + if (tinfo->store) + { + /* store conversion */ + store_doc = xmlCopyDoc(doc, 1); + perform_convert(tinfo, p, 0, tinfo->store->convert, + params, &store_doc, &last_xsp); + } + + /* saving either store doc or original doc in case no store doc exists */ + if (last_xsp) + xsltSaveResultToString(&buf_out, &len_out, + store_doc ? store_doc : doc, last_xsp); + else + xmlDocDumpMemory(store_doc ? store_doc : doc, &buf_out, &len_out); + + if (p->setStoreData) + (*p->setStoreData)(p, buf_out, len_out); + xmlFree(buf_out); + if (store_doc) + xmlFreeDoc(store_doc); + } - if (store_doc) - xmlFreeDoc(store_doc); /* extract conversion */ perform_convert(tinfo, p, 0, tinfo->extract->convert, params, &doc, 0); @@ -1258,7 +1282,7 @@ static int extract_xml_split(struct filter_info *tinfo, ptr = xmlTextReaderExpand(input->u.xmlreader.reader); if (ptr) - { + { /* we have a new document */ xmlNodePtr ptr2 = xmlCopyNode(ptr, 1); @@ -1404,7 +1428,8 @@ static int filter_extract(void *clientData, struct recExtractCtrl *p) static int ioread_ret(void *context, char *buffer, int len) { struct recRetrieveCtrl *p = context; - return p->stream->readf(p->stream, buffer, len); + int r = p->stream->readf(p->stream, buffer, len); + return r; } static int ioclose_ret(void *context) @@ -1442,7 +1467,7 @@ static int filter_retrieve(void *clientData, struct recRetrieveCtrl *p) { p->diagnostic = YAZ_BIB1_SPECIFIED_ELEMENT_SET_NAME_NOT_VALID_FOR_SPECIFIED_; - p->addinfo = odr_strdup(p->odr, esn); + p->addinfo = odr_strdup_null(p->odr, esn); return 0; } @@ -1549,6 +1574,7 @@ idzebra_filter /* * Local variables: * c-basic-offset: 4 + * c-file-style: "Stroustrup" * indent-tabs-mode: nil * End: * vim: shiftwidth=4 tabstop=8 expandtab