X-Git-Url: http://git.indexdata.com/?a=blobdiff_plain;f=index%2Frecgrs.c;h=5b0a7837ef04b86a4c16015c2268a9f25ce1a1f8;hb=775bf7d99438335ba135b673dfb3e19fe9bcba24;hp=8a31f3ce87d600ffa6a1e052a83c30a38204a9ba;hpb=c23dcf8e79596d2ae1fc05596e1d60ed318962e7;p=idzebra-moved-to-github.git diff --git a/index/recgrs.c b/index/recgrs.c index 8a31f3c..5b0a783 100644 --- a/index/recgrs.c +++ b/index/recgrs.c @@ -1,5 +1,5 @@ -/* $Id: recgrs.c,v 1.3 2006-07-06 12:42:22 marc Exp $ - Copyright (C) 1995-2006 +/* $Id: recgrs.c,v 1.15 2007-02-02 12:16:38 adam Exp $ + Copyright (C) 1995-2007 Index Data ApS This file is part of the Zebra server. @@ -15,9 +15,9 @@ FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License -along with Zebra; see the file LICENSE.zebra. If not, write to the -Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA -02111-1307, USA. +along with this program; if not, write to the Free Software +Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ #include @@ -27,6 +27,7 @@ Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA #include #include +#include #include #include @@ -245,7 +246,7 @@ static int sp_expr(struct source_parser *sp, data1_node *n, RecWord *wrd) return 1; } -static struct source_parser *source_parser_create() +static struct source_parser *source_parser_create(void) { struct source_parser *sp = xmalloc(sizeof(*sp)); @@ -295,8 +296,6 @@ int d1_check_xpath_predicate(data1_node *n, struct xpath_predicate *p) 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 != '=') { @@ -305,8 +304,6 @@ int d1_check_xpath_predicate(data1_node *n, struct xpath_predicate *p) 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; } @@ -317,7 +314,6 @@ int d1_check_xpath_predicate(data1_node *n, struct xpath_predicate *p) } } } - yaz_log(YLOG_DEBUG, "return %d", res); return res; } else { return 1; @@ -392,20 +388,37 @@ pop, 2003-01-17 data1_termlist *xpath_termlist_by_tagpath(char *tagpath, data1_node *n) { data1_absyn *abs = n->root->u.root.absyn; - data1_xpelement *xpe = abs->xp_elements; + + data1_xpelement *xpe = 0; data1_node *nn; #ifdef ENHANCED_XELM struct xpath_location_step *xp; #endif char *pexpr = xmalloc(strlen(tagpath)+5); - int ok = 0; sprintf (pexpr, "/%s\n", tagpath); - for (; xpe; xpe = xpe->next) + + for (xpe = abs->xp_elements; xpe; xpe = xpe->next) + xpe->match_state = -1; /* don't know if it matches yet */ + + for (xpe = abs->xp_elements; xpe; xpe = xpe->next) { int i; - ok = dfa_match_first(xpe->dfa->states, pexpr); - + int ok = xpe->match_state; + if (ok == -1) + { /* don't know whether there is a match yet */ + data1_xpelement *xpe1; + + assert(xpe->dfa); + ok = dfa_match_first(xpe->dfa->states, pexpr); + +#if OPTIMIZE_MELM + /* mark this and following ones with same regexp */ + for (xpe1 = xpe; xpe1; xpe1 = xpe1->match_next) + xpe1->match_state = ok; +#endif + } + assert (ok == 0 || ok == 1); if (ok) { #ifdef ENHANCED_XELM /* we have to check the perdicates up to the root node */ @@ -419,12 +432,8 @@ 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(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; } @@ -440,7 +449,7 @@ data1_termlist *xpath_termlist_by_tagpath(char *tagpath, data1_node *n) xfree(pexpr); - if (ok) { + if (xpe) { yaz_log(YLOG_DEBUG, "Got it"); return xpe->termlists; } else { @@ -529,11 +538,15 @@ static void index_xpath(struct source_parser *sp, data1_node *n, 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: @@ -547,6 +560,7 @@ static void index_xpath(struct source_parser *sp, data1_node *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 */ @@ -555,46 +569,33 @@ static void index_xpath(struct source_parser *sp, data1_node *n, 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 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 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 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... */ @@ -606,6 +607,8 @@ 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'; @@ -624,27 +627,8 @@ static void index_xpath(struct source_parser *sp, data1_node *n, 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) { - (*p->tokenAdd)(wrd); /* index element pag (AKA tag path) */ - } + + (*p->tokenAdd)(wrd); /* index element pag (AKA tag path) */ if (xpath_is_start == 1) /* only for the starting tag... */ { @@ -652,17 +636,8 @@ static void index_xpath(struct source_parser *sp, data1_node *n, 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 */ @@ -671,44 +646,31 @@ static void index_xpath(struct source_parser *sp, data1_node *n, 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])) - { - for (; tl; tl = tl->next) - { - if (!tl->index_name) - do_xpindex = 1; - } - } + /* 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); - if (do_xpindex) { + 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); - /* attribute (no value) */ + wrd->index_name = ZEBRA_XPATH_ATTR_NAME; wrd->index_type = '0'; - wrd->index_name = ZEBRA_XPATH_ATTR_NAME; - wrd->term_buf = xp->name; - wrd->term_len = strlen(xp->name); - + wrd->term_buf = comb; + wrd->term_len = strlen(comb); 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); - } + (*p->tokenAdd)(wrd); } i++; } @@ -952,12 +914,7 @@ static int grs_extract_sub(void *clientData, struct recExtractCtrl *p, int oidtmp[OID_SIZE]; RecWord wrd; - gri.readf = p->readf; - gri.seekf = p->seekf; - gri.tellf = p->tellf; - gri.endf = p->endf; - gri.fh = p->fh; - gri.offset = p->offset; + gri.stream = p->stream; gri.mem = mem; gri.dh = p->dh; gri.clientData = clientData; @@ -1149,12 +1106,7 @@ int zebra_grs_retrieve(void *clientData, struct recRetrieveCtrl *p, int dummy; mem = nmem_create(); - gri.readf = p->readf; - gri.seekf = p->seekf; - gri.tellf = p->tellf; - gri.endf = NULL; - gri.fh = p->fh; - gri.offset = 0; + gri.stream = p->stream; gri.mem = mem; gri.dh = p->dh; gri.clientData = clientData; @@ -1163,7 +1115,7 @@ 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; } @@ -1239,7 +1191,7 @@ int zebra_grs_retrieve(void *clientData, struct recRetrieveCtrl *p, 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; } @@ -1268,7 +1220,7 @@ int zebra_grs_retrieve(void *clientData, struct recRetrieveCtrl *p, 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; }