X-Git-Url: http://jsfdemo.indexdata.com/?a=blobdiff_plain;f=index%2Fextract.c;h=1b8fdb7dd971d2a76583be31b44028340260acbf;hb=d534dea1aa3550da87ef9c183c9dc6ebd78d371e;hp=9dd5f7618acad28ba93987e22553d0b2247ad9f1;hpb=6c7e7bbe11beb9394c5e65510dee80ffe69ec637;p=idzebra-moved-to-github.git

diff --git a/index/extract.c b/index/extract.c
index 9dd5f76..1b8fdb7 100644
--- a/index/extract.c
+++ b/index/extract.c
@@ -1,4 +1,4 @@
-/* $Id: extract.c,v 1.154 2004-06-03 11:38:34 adam Exp $
+/* $Id: extract.c,v 1.157.2.1 2004-09-16 14:07:50 adam Exp $
    Copyright (C) 1995,1996,1997,1998,1999,2000,2001,2002,2003,2004
    Index Data Aps
 
@@ -776,16 +776,12 @@ int fileExtract (ZebraHandle zh, SYSNO *sysno, const char *fname,
             break;
         }
     /* determine file type - depending on extension */
-    yaz_log(LOG_LOG, "recordType 1=%s", zh->m_record_type ? 
-	    zh->m_record_type : "<none>");
     original_record_type = zh->m_record_type;
     if (!zh->m_record_type)
     {
         sprintf (ext_res, "%srecordType.%s", gprefix, ext);
         zh->m_record_type = res_get (zh->res, ext_res);
     }
-    yaz_log(LOG_LOG, "recordType 2=%s", zh->m_record_type ? 
-	    zh->m_record_type : "<none>");
     if (!zh->m_record_type)
     {
 	if (zh->records_processed < zh->m_file_verbose_limit)
@@ -1036,7 +1032,7 @@ int buffer_extract_record (ZebraHandle zh,
 		extract_flushSortKeys (zh, *sysno, -1, &zh->reg->sortKeys);
 		rec_rm (&rec);
 		logRecord(zh);
-		return 0;
+		return -1;
 	    }
 	}
 
@@ -1606,7 +1602,7 @@ static void extract_add_incomplete_field (RecWord *p)
     const char **map = 0;
 
     if (remain > 0)
-	map = zebra_maps_input(p->zebra_maps, p->reg_type, &b, remain);
+	map = zebra_maps_input(p->zebra_maps, p->reg_type, &b, remain, 0);
 
     while (map)
     {
@@ -1618,7 +1614,7 @@ static void extract_add_incomplete_field (RecWord *p)
 	{
 	    remain = p->length - (b - p->string);
 	    if (remain > 0)
-		map = zebra_maps_input(p->zebra_maps, p->reg_type, &b, remain);
+		map = zebra_maps_input(p->zebra_maps, p->reg_type, &b, remain, 0);
 	    else
 		map = 0;
 	}
@@ -1633,7 +1629,7 @@ static void extract_add_incomplete_field (RecWord *p)
 		buf[i++] = *(cp++);
 	    remain = p->length - (b - p->string);
 	    if (remain > 0)
-		map = zebra_maps_input(p->zebra_maps, p->reg_type, &b, remain);
+		map = zebra_maps_input(p->zebra_maps, p->reg_type, &b, remain, 0);
 	    else
 		map = 0;
 	}
@@ -1650,17 +1646,24 @@ static void extract_add_complete_field (RecWord *p)
     char buf[IT_MAX_WORD+1];
     const char **map = 0;
     int i = 0, remain = p->length;
+    int first; /* first position */
+
+yaz_log(LOG_DEBUG, "Complete field, w='%s'", p->string);
 
     if (remain > 0)
-	map = zebra_maps_input (p->zebra_maps, p->reg_type, &b, remain);
+	map = zebra_maps_input (p->zebra_maps, p->reg_type, &b, remain, 1);
 
     while (remain > 0 && i < IT_MAX_WORD)
     {
 	while (map && *map && **map == *CHR_SPACE)
 	{
 	    remain = p->length - (b - p->string);
+
 	    if (remain > 0)
-		map = zebra_maps_input(p->zebra_maps, p->reg_type, &b, remain);
+	    {
+		first = i ? 0 : 1;
+		map = zebra_maps_input(p->zebra_maps, p->reg_type, &b, remain, first);
+	    }
 	    else
 		map = 0;
 	}
@@ -1673,14 +1676,24 @@ static void extract_add_complete_field (RecWord *p)
 	{
 	    const char *cp = *map;
 
-	    if (i >= IT_MAX_WORD)
-		break;
-	    while (i < IT_MAX_WORD && *cp)
-		buf[i++] = *(cp++);
+	    if (**map == *CHR_CUT)
+	    {
+		i = 0;
+	    }
+	    else
+	    {
+		if (i >= IT_MAX_WORD)
+		    break;
+yaz_log(LOG_DEBUG, "Adding string to index '%d'", *map);
+		while (i < IT_MAX_WORD && *cp)
+		    buf[i++] = *(cp++);
+	    }
 	    remain = p->length  - (b - p->string);
 	    if (remain > 0)
+	    {
 		map = zebra_maps_input (p->zebra_maps, p->reg_type, &b,
-					remain);
+					remain, 0);
+	    }
 	    else
 		map = 0;
 	}