Patch to htdig 3.0.8b2 by Reni Seindal (seindal@webadm.kb.dk)

I made a mistake in the previous patch I described here, causing, among
other things, the url to used instead of the title as the document
headline.

A patch to fix this is found alongside the first patch, at
ftp://webadm.kb.dk/pub/htdig.patch2 
------------------------------------------------------------------------
diff -rc htdig-3.0.8b2.orig/htdig/HTML.cc htdig-3.0.8b2/htdig/HTML.cc
*** htdig-3.0.8b2.orig/htdig/HTML.cc    Mon Aug 18 12:52:18 1997
--- htdig-3.0.8b2/htdig/HTML.cc Mon Aug 18 22:23:11 1997
***************
*** 103,120 ****
        (unsigned char *) new char[strlen((char *)position)+1];
      unsigned char     *ptext = text;
  
-     start = position;
      title = 0;
      head = 0;
      doindex = 1;
      in_heading = 0;
      in_title = 0;
      in_ref = 0;
        
      while (*position)
      {
-       offset = position - start;
- 
        if (strncmp((char *)position, "<!--", 4) == 0)
        {
            //
--- 103,118 ----
        (unsigned char *) new char[strlen((char *)position)+1];
      unsigned char     *ptext = text;
  
      title = 0;
      head = 0;
      doindex = 1;
      in_heading = 0;
      in_title = 0;
      in_ref = 0;
+     in_space = 0;
        
      while (*position)
      {
        if (strncmp((char *)position, "<!--", 4) == 0)
        {
            //
***************
*** 129,145 ****
        }
        if (*position == '<')
        {
-           //
            // Start of a tag.  Since tags cannot be nested, we can simply
            // search for the closing '>'
            //
            q = (unsigned char*)strchr((char *)position, '>');
!           if (!q)
!               return; // Syntax error in the doc.  Tag never ends.
!           tag = 0;
!           tag.append((char*)position, q - position + 1);
!           do_tag(retriever, tag);
!           position = q+1;
        }
        else if (*position == '&')
        {
--- 127,147 ----
        }
        if (*position == '<')
        {
            // Start of a tag.  Since tags cannot be nested, we can simply
            // search for the closing '>'
+           // Tags are copied to text-string
            //
            q = (unsigned char*)strchr((char *)position, '>');
!           if (q) 
!           { // copy tag
!               while (position <= q)
!                   *ptext++ = *position++;
!           } 
!           else
!           { // copy rest of text, as tag does not end
!               while (*position)
!                   *ptext++ = *position++;
!           }
        }
        else if (*position == '&')
        {
***************
*** 153,162 ****
      *ptext++ = '\0';
  
      position = text;
      while (*position)
      {
        word = 0;
!       if (*position > 0 && (isalnum(*position)))
        {
            //
            // Start of a word.  Try to find the whole thing
--- 155,181 ----
      *ptext++ = '\0';
  
      position = text;
+     start = position;
+ 
      while (*position)
      {
+       offset = position - start;
+ 
        word = 0;
!       if (*position == '<')
!       {
!           // Start of a tag.  Since tags cannot be nested, we can simply
!           // search for the closing '>'
!           //
!           q = (unsigned char*)strchr((char *)position, '>');
!           if (!q)
!               return; // Syntax error in the doc.  Tag never ends.
!           tag = 0;
!           tag.append((char*)position, q - position + 1);
!           do_tag(retriever, tag);
!           position = q+1;
!       }
!       else if (*position > 0 && (isalnum(*position)))
        {
            //
            // Start of a word.  Try to find the whole thing