X-Git-Url: http://jsfdemo.indexdata.com/?a=blobdiff_plain;ds=sidebyside;f=data1%2Fd1_absyn.c;h=f19edbf5da4f19666ef47fe7676206fb3a8f2469;hb=9615a2bcb063c5abdbec9a1b8be38f150443c4e7;hp=a913e4b55ee001ed01dd3265160c0bb4c5569fa4;hpb=2e4e9c6def27f1e1463dcb6f205fab6a98054f38;p=idzebra-moved-to-github.git

diff --git a/data1/d1_absyn.c b/data1/d1_absyn.c
index a913e4b..f19edbf 100644
--- a/data1/d1_absyn.c
+++ b/data1/d1_absyn.c
@@ -1,4 +1,4 @@
-/* $Id: d1_absyn.c,v 1.10 2004-08-04 08:35:22 adam Exp $
+/* $Id: d1_absyn.c,v 1.9.2.8 2006-08-14 10:38:51 adam Exp $
    Copyright (C) 1995,1996,1997,1998,1999,2000,2001,2002
    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>
@@ -58,7 +58,7 @@ data1_absyn *data1_absyn_search (data1_handle dh, const char *name)
 
     while (p)
     {
-	if (!yaz_matchstr (name, p->name))
+	if (!strcmp (name, p->name))
 	    return p->absyn;
 	p = p->next;
     }
@@ -133,7 +133,7 @@ data1_attset *data1_attset_search_name (data1_handle dh, const char *name)
 
     while (p)
     {
-	if (!yaz_matchstr (name, p->name))
+	if (!strcmp (name, p->name))
 	    return p->attset;
 	p = p->next;
     }
@@ -293,65 +293,75 @@ void fix_element_ref (data1_handle dh, data1_absyn *absyn, data1_element *e)
 
  */
 
-const char * mk_xpath_regexp (data1_handle dh, char *expr) 
+static const char * mk_xpath_regexp (data1_handle dh, const char *expr) 
 {
-    char *p = expr;
-    char *pp;
-    char *s;
+    const char *p = expr;
     int abs = 1;
-    int i;
-    int j;
-    int e=0;
-    int is_predicate = 0;
-    
-    static char *stack[32];
-    static char res[1024];
-    char *r = "";
+    int e = 0;
+    char *stack[32];
+    char *res_p, *res = 0;
+    size_t res_size = 1;
     
-    if (*p != '/') { return (""); }
+    if (*p != '/')
+	return ("");
     p++;
-    if (*p == '/') { abs=0; p++; }
-    
-    while (*p) {
-        i=0;
-        while (*p && !strchr("/",*p)) { 
-	  i++; p++; 
-	}
-        stack[e] = (char *) nmem_malloc (data1_nmem_get (dh), i+1);
+    if (*p == '/') 
+    { 
+	abs =0;
+	p++;
+    }
+    while (*p)
+    {
+	int is_predicate = 0;
+	char *s;
+	int i, j;
+        for (i = 0; *p && !strchr("/",*p); i++, p++)
+	    ;
+	res_size += (i+3); /* we'll add / between later .. */
+        stack[e] = (char *) nmem_malloc(data1_nmem_get(dh), i+1);
 	s = stack[e];
-	for (j=0; j< i; j++) {
-	  pp = p-i+j;
-	  if (*pp == '[') {
-	    is_predicate=1;
-	  }
-	  else if (*pp == ']') {
-	    is_predicate=0;
-	  }
-	  else {
-	    if (!is_predicate) {
-	      if (*pp == '*') 
-		 *s++ = '.';
-	      *s++ = *pp;
-	    }
-	  }
+	for (j = 0; j < i; j++)
+	{
+	    const char *pp = p-i+j;
+	    if (*pp == '[')
+		is_predicate=1;
+	    else if (*pp == ']')
+		is_predicate=0;
+	    else 
+	    {
+		if (!is_predicate) {
+		    if (*pp == '*') 
+			*s++ = '.';
+		    *s++ = *pp;
+		}
+	    }
 	}
 	*s = 0;
         e++;
-        if (*p) {p++;}
+        if (*p)
+	    p++;
+    }
+    res_p = res = nmem_malloc(data1_nmem_get(dh), res_size + 10);
+
+    *res_p = '\0';
+    if (stack[e-1][0] == '@')  /* path/@attr spec (leaf is attribute) */
+	strcpy(res_p, "/");
+    else
+	strcpy(res_p, "[^@]*/");  /* path .. (index all cdata below it) */
+    res_p = res_p + strlen(res_p);
+    while (--e >= 0) {
+	sprintf(res_p, "%s/", stack[e]);
+	res_p += strlen(stack[e]) + 1;
     }
-    e--;  p = &res[0]; i=0;
-    sprintf (p, "^"); p++;
-    while (e >= 0) {
-        /* !!! res size is not checked !!! */
-        sprintf (p, "%s/",stack[e]);
-        p += strlen(stack[e]) + 1;
-        e--;
+    if (!abs)
+    {
+	sprintf(res_p, ".*"); 
+	res_p += 2;
     }
-    if (!abs) { sprintf (p, ".*"); p+=2; }
-    sprintf (p, "$"); p++;
-    r = nmem_strdup (data1_nmem_get (dh), res);
-    yaz_log(LOG_DEBUG,"Got regexp: %s",r);
-    return (r);
+    sprintf (res_p, "$");
+    res_p++;
+    yaz_log(LOG_DEBUG, "Got regexp: %s", res);
+    return res;
 }
 
 /* *ostrich*
@@ -362,28 +372,57 @@ const char * mk_xpath_regexp (data1_handle dh, char *expr)
    pop, 2002-12-13
  */
 static int parse_termlists (data1_handle dh, data1_termlist ***tpp,
-			    char *p, const char *file, int lineno,
+			    char *cp, const char *file, int lineno,
 			    const char *element_name, data1_absyn *res,
 			    int xpelement)
 {
     data1_termlist **tp = *tpp;
-    do
+    while(1)
     {
 	char attname[512], structure[512];
 	char *source;
-	int r;
-	
-	if (!(r = sscanf(p, "%511[^:,]:%511[^,]", attname,
-			 structure)))
+	int r, i;
+	int level = 0;
+	structure[0] = '\0';
+	for (i = 0; cp[i] && i<sizeof(attname)-1; i++)
+	    if (strchr(":,", cp[i]))
+		break;
+	    else
+		attname[i] = cp[i];
+	if (i == 0)
 	{
-	    yaz_log(LOG_WARN,
-		    "%s:%d: Syntax error in termlistspec '%s'",
-		    file, lineno, p);
-	    return -1;
+	    if (*cp)
+		yaz_log(LOG_WARN,
+			"%s:%d: Syntax error in termlistspec '%s'",
+			file, lineno, cp);
+	    break;
 	}
+	attname[i] = '\0';
+	r = 1;
+	cp += i;
+	if (*cp == ':')
+	    cp++;
+
+	for (i = 0; cp[i] && i<sizeof(structure)-1; i++)
+	    if (level == 0 && strchr(",", cp[i]))
+		break;
+	    else
+	    {
+		structure[i] = cp[i];
+		if (cp[i] == '(')
+		    level++;
+		else if (cp[i] == ')')
+		    level--;
+	    }
+	structure[i] = '\0';
+	if (i)
+	    r = 2;
+	cp += i;
+	if (*cp)
+	    cp++;  /* skip , */
 
 	*tp = (data1_termlist *)
-	  nmem_malloc(data1_nmem_get(dh), sizeof(**tp));
+	    nmem_malloc(data1_nmem_get(dh), sizeof(**tp));
 	(*tp)->next = 0;
         
 	if (!xpelement) {
@@ -416,11 +455,38 @@ static int parse_termlists (data1_handle dh, data1_termlist ***tpp,
 		nmem_strdup (data1_nmem_get (dh), structure);
 	tp = &(*tp)->next;
     }
-    while ((p = strchr(p, ',')) && *(++p));
+
     *tpp = tp;
     return 0;
 }
 
+/* quinn
+ * Converts a 'melm' field[$subfield] pattern to a simple xpath
+ */
+static int melm2xpath(char *melm, char *buf)
+{
+    char *dollar;
+    char *field = melm;
+    char *subfield;
+    char *fieldtype;
+    if ((dollar = strchr(melm, '$'))) {
+	*dollar = '\0';
+	subfield = ++dollar;
+    } else
+	subfield = "";
+    if (field[0] == '0' && field[1] == '0')
+	fieldtype = "controlfield";
+    else
+	fieldtype = "datafield";
+    sprintf(buf, "/*/%s[@tag=\"%s\"]", fieldtype, field);
+    if (*subfield) 
+	sprintf(buf + strlen(buf), "/subfield[@code=\"%s\"]", subfield);
+    else if (field[0] != '0' || field[1] != '0')
+	strcat(buf, "/subfield");
+    yaz_log(LOG_DEBUG, "Created xpath: '%s'", buf);
+    return 0;
+}
+
 const char *data1_systag_lookup(data1_absyn *absyn, const char *tag,
                                 const char *default_value)
 {
@@ -663,20 +729,28 @@ data1_absyn *data1_read_absyn (data1_handle dh, const char *file,
 	   pop, 2003-01-17
 	*/
 
-	else if (!strcmp(cmd, "xelm")) {
+	else if (!strcmp(cmd, "xelm") || !strcmp(cmd, "melm")) {
 
 	    int i;
 	    char *p, *xpath_expr, *termlists;
 	    const char *regexp;
 	    struct DFA *dfa = dfa = dfa_init();
 	    data1_termlist **tp;
+	    char melm_xpath[128];
             
 	    if (argc < 3)
 	    {
 		yaz_log(LOG_WARN, "%s:%d: Bad # of args to xelm", file, lineno);
 		continue;
 	    }
-	    xpath_expr = argv[1];
+	    
+	    if (!strcmp(cmd, "melm")) {
+		if (melm2xpath(argv[1], melm_xpath) < 0)
+		    continue;
+		xpath_expr = melm_xpath;
+	    } else {
+		xpath_expr = argv[1];
+	    }
 	    termlists = argv[2];
 	    regexp = mk_xpath_regexp(dh, xpath_expr);
 	    i = dfa_parse (dfa, &regexp);