X-Git-Url: http://git.indexdata.com/?a=blobdiff_plain;f=index%2Frecgrs.c;h=6ebf807be559c0590fe3b4fc389e987aeb3bc675;hb=c401b3a1421e31b3d6b77bc7bac6ffb934207df1;hp=cd11edaeb4ad3c6ec2a2e769fe865d830e762a27;hpb=d1e1a6a2a574e8019e700abb9d8a9d03ba41aab5;p=idzebra-moved-to-github.git

diff --git a/index/recgrs.c b/index/recgrs.c
index cd11eda..6ebf807 100644
--- a/index/recgrs.c
+++ b/index/recgrs.c
@@ -1,4 +1,4 @@
-/* $Id: recgrs.c,v 1.2 2006-07-06 11:40:25 adam Exp $
+/* $Id: recgrs.c,v 1.9 2006-10-29 17:20:01 adam Exp $
    Copyright (C) 1995-2006
    Index Data ApS
 
@@ -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 <stdio.h>
@@ -245,7 +245,7 @@ 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));
 
@@ -392,20 +392,37 @@ 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 */
@@ -440,7 +457,7 @@ 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 {
@@ -547,6 +564,7 @@ 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 */
@@ -594,7 +612,12 @@ static void index_xpath(struct source_parser *sp, data1_node *n,
 		    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... */
@@ -937,6 +960,8 @@ int grs_extract_tree(struct recExtractCtrl *p, data1_node *n)
     }
     (*p->init)(p, &wrd);
 
+    /* data1_pr_tree(p->dh, n, stdout); */ 
+
     return dumpkeys(n, p, &wrd);
 }
 
@@ -950,12 +975,7 @@ 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;
@@ -980,6 +1000,9 @@ static int grs_extract_sub(void *clientData, struct recExtractCtrl *p,
     /* ensure our data1 tree is UTF-8 */
     data1_iconv (p->dh, mem, n, "UTF-8", data1_get_encoding(p->dh, n));
 
+
+    data1_remove_idzebra_subtree (p->dh, n);
+
 #if 0
     data1_pr_tree (p->dh, n, stdout);
 #endif
@@ -1144,12 +1167,7 @@ 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;
@@ -1164,6 +1182,8 @@ int zebra_grs_retrieve(void *clientData, struct recRetrieveCtrl *p,
     }
     data1_concat_text(p->dh, mem, node);
 
+    data1_remove_idzebra_subtree (p->dh, node);
+
 #if 0
     data1_pr_tree (p->dh, node, stdout);
 #endif