X-Git-Url: http://git.nguyen.vg/gitweb/?a=blobdiff_plain;f=XMLTree.h;h=759aff83251fe7b820c22bbb6fe836a8d8891a39;hb=2220c6ca1fb8ebefd48d6f5d9a6be3afc7325da5;hp=e8cc76a42879450f1695f08e6150088a0f8c3e38;hpb=f6c4aa7602efed65bb647488104b04ba89adf4e2;p=SXSI%2FXMLTree.git

diff --git a/XMLTree.h b/XMLTree.h
index e8cc76a..759aff8 100644
--- a/XMLTree.h
+++ b/XMLTree.h
@@ -25,6 +25,12 @@
 #include <stdio.h>
 #include <stdlib.h>
 #include <cstring>
+
+//clash between TextCollection/Tools.h and libcds/includes/basics.h
+#undef W
+#undef WW
+#undef Wminusone
+
 #include "bp.h"
 #include <static_bitsequence.h>
 #include <alphabet_mapper.h>
@@ -90,6 +96,8 @@ class XMLTree {
    int parArraySize;
    int ntagnames;
    unsigned int *empty_texts_aux;
+   bool found_attributes;
+   bool disable_tc;
    
 public:
 
@@ -209,14 +217,16 @@ public:
    /** ParentNode(d): returns the parent node of document identifier d. */
    treeNode ParentNode(DocID d);
 
-   /** OpenDocument(empty_texts,sample_rate_text): initilizes the construction
+   /** OpenDocument(empty_texts,sample_rate_text,dtc): initilizes the construction
     * of the data structure for the XML document. Parameter empty_texts 
     * indicates whether we index empty texts in document or not. Parameter 
     * sample_rate_text indicates the sampling rate for the text searching data
     * structures (small values get faster searching but a bigger space 
-    * requirement). Returns a non-zero value upon success, NULLT in case of 
+    * requirement). dtc disable the use of the TextCollection
+    * (i.e. everything is considered an empty text *)
+    * Returns a non-zero value upon success, NULLT in case of 
     * error. */
-   int OpenDocument(bool empty_texts, int sample_rate_text);
+   int OpenDocument(bool empty_texts, int sample_rate_text, bool dtc);
 
    /** CloseDocument(): finishes the construction of the data structure for 
     * the XML document. Tree and tags are represented in the final form, 
@@ -254,7 +264,9 @@ public:
     * Returns NULL in case that the tag identifier is not valid.*/
    unsigned char *GetTagName(TagType tagid);
 
-   /** Prefix(s): search for texts prefixed by string s. */
+
+   TagType RegisterTag(unsigned char *tagname);
+
    bool EmptyText(DocID i) {
       return Text->EmptyText(i);
    }