X-Git-Url: http://git.indexdata.com/?a=blobdiff_plain;f=index%2Fzrpn.c;h=80d66c76b4f31362d946d2a526ac1670c625fb06;hb=587c1d5cac956ec0cf482fc14ecea40240e85fd8;hp=6f084f1316be07e9c7a2f1691a95434af9fc8007;hpb=896c0427df9d8eff5de6a1735dcd992e067df844;p=idzebra-moved-to-github.git

diff --git a/index/zrpn.c b/index/zrpn.c
index 6f084f1..80d66c7 100644
--- a/index/zrpn.c
+++ b/index/zrpn.c
@@ -1,4 +1,4 @@
-/* $Id: zrpn.c,v 1.120 2002-08-02 19:26:56 adam Exp $
+/* $Id: zrpn.c,v 1.125 2002-10-03 10:16:23 adam Exp $
    Copyright (C) 1995,1996,1997,1998,1999,2000,2001,2002
    Index Data Aps
 
@@ -46,7 +46,20 @@ struct rpn_char_map_info {
 static const char **rpn_char_map_handler (void *vp, const char **from, int len)
 {
     struct rpn_char_map_info *p = (struct rpn_char_map_info *) vp;
-    return zebra_maps_input (p->zm, p->reg_type, from, len);
+    const char **out = zebra_maps_input (p->zm, p->reg_type, from, len);
+#if 0
+    if (out && *out)
+    {
+        const char *outp = *out;
+        yaz_log (LOG_LOG, "---");
+        while (*outp)
+        {
+            yaz_log (LOG_LOG, "%02X", *outp);
+            outp++;
+        }
+    }
+#endif
+    return out;
 }
 
 static void rpn_char_map_prepare (struct zebra_register *reg, int reg_type,
@@ -167,17 +180,18 @@ struct grep_info {
 static void term_untrans  (ZebraHandle zh, int reg_type,
 			   char *dst, const char *src)
 {
+    int len = 0;
     while (*src)
     {
         const char *cp = zebra_maps_output (zh->reg->zebra_maps,
 					    reg_type, &src);
-	if (!cp)
-	    *dst++ = *src++;
+	if (!cp && len < IT_MAX_WORD-1)
+	    dst[len++] = *src++;
 	else
-	    while (*cp)
-		*dst++ = *cp++;
+	    while (*cp && len < IT_MAX_WORD-1)
+		dst[len++] = *cp++;
     }
-    *dst = '\0';
+    dst[len] = '\0';
 }
 
 static void add_isam_p (const char *name, const char *info,
@@ -220,7 +234,7 @@ static void add_isam_p (const char *name, const char *info,
     {
 	const char *db;
 	int set, use;
-	char term_tmp[512];
+	char term_tmp[IT_MAX_WORD];
 	int su_code = 0;
 	int len = key_SU_decode (&su_code, name);
 	
@@ -266,7 +280,7 @@ static int term_pre (ZebraMaps zebra_maps, int reg_type, const char **src,
     return *s0;
 }
 
-#define REGEX_CHARS "[]()|.*+!"
+#define REGEX_CHARS " []()|.*+?!"
 
 /* term_100: handle term, where trunc=none (no operators at all) */
 static int term_100 (ZebraMaps zebra_maps, int reg_type,
@@ -982,11 +996,11 @@ static int string_term (ZebraHandle zh, Z_AttributesPlusTerm *zapt,
 				  attributeSet,
 				  reg_type, space_split, term_dst))
 		return 0;
-	    logf (LOG_DEBUG, "dict_lookup_grep: %s", term_dict+prefix_len);
+	    logf (LOG_LOG, "dict_lookup_grep: %s", term_dict+prefix_len);
 	    r = dict_lookup_grep (zh->reg->dict, term_dict, 0,
 				  grep_info, &max_pos, 0, grep_handle);
 	    if (r)
-		logf (LOG_WARN, "dict_lookup_grep fail, rel=gt: %d", r);
+		logf (LOG_WARN, "dict_lookup_grep fail %d", r);
 	    break;
 	case 1:          /* right truncation */
 	    term_dict[j++] = '(';
@@ -1096,8 +1110,9 @@ static int string_term (ZebraHandle zh, Z_AttributesPlusTerm *zapt,
 }
 
 
-static int trans_term (ZebraHandle zh, Z_AttributesPlusTerm *zapt,
-                       char *termz)
+/* convert APT search term to UTF8 */
+static int zapt_term_to_utf8 (ZebraHandle zh, Z_AttributesPlusTerm *zapt,
+                              char *termz)
 {
     size_t sizez;
     Z_Term *term = zapt->term;
@@ -1105,8 +1120,7 @@ static int trans_term (ZebraHandle zh, Z_AttributesPlusTerm *zapt,
     switch (term->which)
     {
     case Z_Term_general:
-#if HAVE_ICONV_H
-        if (zh->iconv_to_utf8 != (iconv_t)(-1))
+        if (zh->iconv_to_utf8 != 0)
         {
             char *inbuf = term->u.general->buf;
             size_t inleft = term->u.general->len;
@@ -1114,24 +1128,24 @@ static int trans_term (ZebraHandle zh, Z_AttributesPlusTerm *zapt,
             size_t outleft = IT_MAX_WORD-1;
             size_t ret;
 
-            yaz_log (LOG_DEBUG, "converting general from ISO-8859-1");
-            ret = iconv(zh->iconv_to_utf8, &inbuf, &inleft,
+            ret = yaz_iconv(zh->iconv_to_utf8, &inbuf, &inleft,
                         &outbuf, &outleft);
             if (ret == (size_t)(-1))
             {
-                ret = iconv(zh->iconv_to_utf8, 0, 0, 0, 0);
+                ret = yaz_iconv(zh->iconv_to_utf8, 0, 0, 0, 0);
                 zh->errCode = 125;
                 return -1;
             }
             *outbuf = 0;
-            return 0;
         }
-#endif
-        sizez = term->u.general->len;
-        if (sizez > IT_MAX_WORD-1)
-            sizez = IT_MAX_WORD-1;
-        memcpy (termz, term->u.general->buf, sizez);
-        termz[sizez] = '\0';
+        else
+        {
+            sizez = term->u.general->len;
+            if (sizez > IT_MAX_WORD-1)
+                sizez = IT_MAX_WORD-1;
+            memcpy (termz, term->u.general->buf, sizez);
+            termz[sizez] = '\0';
+        }
         break;
     case Z_Term_characterString:
         sizez = strlen(term->u.characterString);
@@ -1142,38 +1156,47 @@ static int trans_term (ZebraHandle zh, Z_AttributesPlusTerm *zapt,
         break;
     default:
         zh->errCode = 124;
+        return -1;
     }
     return 0;
 }
 
-static void trans_scan_term (ZebraHandle zh, Z_AttributesPlusTerm *zapt,
-                             char *termz, int reg_type)
+/* convert APT SCAN term to internal cmap */
+static int trans_scan_term (ZebraHandle zh, Z_AttributesPlusTerm *zapt,
+                            char *termz, int reg_type)
 {
-    Z_Term *term = zapt->term;
-    const char **map;
-    const char *cp = (const char *) term->u.general->buf;
-    const char *cp_end = cp + term->u.general->len;
-    const char *src;
-    int i = 0;
-    const char *space_map = NULL;
-    int len;
-    
-    while ((len = (cp_end - cp)) > 0)
+    char termz0[IT_MAX_WORD];
+
+    if (zapt_term_to_utf8(zh, zapt, termz0))
+        return -1;    /* error */
+    else
     {
-        map = zebra_maps_input (zh->reg->zebra_maps, reg_type, &cp, len);
-        if (**map == *CHR_SPACE)
-            space_map = *map;
-        else
+        const char **map;
+        const char *cp = (const char *) termz0;
+        const char *cp_end = cp + strlen(cp);
+        const char *src;
+        int i = 0;
+        const char *space_map = NULL;
+        int len;
+            
+        while ((len = (cp_end - cp)) > 0)
         {
-            if (i && space_map)
-                for (src = space_map; *src; src++)
+            map = zebra_maps_input (zh->reg->zebra_maps, reg_type, &cp, len);
+            if (**map == *CHR_SPACE)
+                space_map = *map;
+            else
+            {
+                if (i && space_map)
+                    for (src = space_map; *src; src++)
+                        termz[i++] = *src;
+                space_map = NULL;
+                for (src = *map; *src; src++)
                     termz[i++] = *src;
-            space_map = NULL;
-            for (src = *map; *src; src++)
-                termz[i++] = *src;
+            }
         }
+        termz[i] = '\0';
     }
-    termz[i] = '\0';
+    return 0;
 }
 
 static RSET rpn_prox (ZebraHandle zh, RSET *rset, int rset_no,
@@ -2136,12 +2159,28 @@ static RSET xpath_trunc(ZebraHandle zh, NMEM stream,
     int prefix_len = 0;
     int ord = zebraExplain_lookupSU (zh->reg->zei, curAttributeSet, use);
     int ord_len, i, r, max_pos;
+    int term_type = Z_Term_characterString;
+    const char *flags = "void";
 
     if (grep_info_prepare (zh, 0 /* zapt */, &grep_info, '0', stream))
-	return 0;
+    {
+	rset_null_parms parms;
+	
+	parms.rset_term = rset_term_create (term, strlen(term),
+					    flags, term_type);
+	parms.rset_term->nn = 0;
+	return rset_create (rset_kind_null, &parms);
+    }
 
     if (ord < 0)
-        return 0;
+    {
+	rset_null_parms parms;
+	
+	parms.rset_term = rset_term_create (term, strlen(term),
+					    flags, term_type);
+	parms.rset_term->nn = 0;
+	return rset_create (rset_kind_null, &parms);
+    }
     if (prefix_len)
         term_dict[prefix_len++] = '|';
     else
@@ -2166,7 +2205,7 @@ static RSET xpath_trunc(ZebraHandle zh, NMEM stream,
              grep_info.isam_p_indx);
     rset = rset_trunc (zh, grep_info.isam_p_buf,
                        grep_info.isam_p_indx, term, strlen(term),
-                       "void", 1, Z_Term_characterString);
+                       flags, 1, term_type);
     grep_info_delete (&grep_info);
     return rset;
 }
@@ -2294,7 +2333,7 @@ static RSET rpn_search_xpath (ZebraHandle zh,
             
                 rset_end_tag = xpath_trunc(zh, stream,
                                        '0', xpath_rev, 2, curAttributeSet);
-            
+
                 parms.key_size = sizeof(struct it_key);
                 parms.cmp = key_compare_it;
                 parms.rset_l = rset_start_tag;
@@ -2337,7 +2376,7 @@ static RSET rpn_search_APT (ZebraHandle zh, Z_AttributesPlusTerm *zapt,
     logf (LOG_DEBUG, "search_type=%s", search_type);
     logf (LOG_DEBUG, "rank_type=%s", rank_type);
 
-    if (trans_term (zh, zapt, termz))
+    if (zapt_term_to_utf8(zh, zapt, termz))
         return 0;
 
     if (sort_flag)
@@ -2612,12 +2651,33 @@ static int scan_handle (char *name, const char *info, int pos, void *client)
 static void scan_term_untrans (ZebraHandle zh, NMEM stream, int reg_type,
 			       char **dst, const char *src)
 {
-    char term_dst[1024];
-    
-    term_untrans (zh, reg_type, term_dst, src);
+    char term_src[IT_MAX_WORD];
+    char term_dst[IT_MAX_WORD];
     
-    *dst = (char *) nmem_malloc (stream, strlen(term_dst)+1);
-    strcpy (*dst, term_dst);
+    term_untrans (zh, reg_type, term_src, src);
+
+    if (zh->iconv_from_utf8 != 0)
+    {
+        int len;
+        char *inbuf = term_src;
+        size_t inleft = strlen(term_src);
+        char *outbuf = term_dst;
+        size_t outleft = sizeof(term_dst)-1;
+        size_t ret;
+        
+        ret = yaz_iconv (zh->iconv_from_utf8, &inbuf, &inleft,
+                         &outbuf, &outleft);
+        if (ret == (size_t)(-1))
+            len = 0;
+        else
+            len = outbuf - term_dst;
+        *dst = nmem_malloc (stream, len + 1);
+        if (len > 0)
+            memcpy (*dst, term_dst, len);
+        (*dst)[len] = '\0';
+    }
+    else
+        *dst = nmem_strdup (stream, term_src);
 }
 
 static void count_set (RSET r, int *count)
@@ -2761,7 +2821,8 @@ void rpn_scan (ZebraHandle zh, ODR stream, Z_AttributesPlusTerm *zapt,
         termz[prefix_len] = 0;
         strcpy (scan_info->prefix, termz);
 
-        trans_scan_term (zh, zapt, termz+prefix_len, reg_id);
+        if (trans_scan_term (zh, zapt, termz+prefix_len, reg_id))
+            return ;
                     
         dict_scan (zh->reg->dict, termz, &before_tmp, &after_tmp,
 		   scan_info, scan_handle);