class PxTodo

# File lib/px_todo.rb, line 12
def initialize(raw_s, filepath: '.', ignore_headings: false, &blk)
  
  Dir.chdir  filepath
  s, type = RXFHelper.read(raw_s)
  
  @filepath = filepath
  @filepath = File.dirname(raw_s) if type == :file and filepath == '.'
  
  @fields = %w( title heading when duration priority status note tags)    
  
  s.gsub!(/^#.*/,'') if ignore_headings
  
  s =~ /<?xml/ ? load_px(s) : import_txt(s, &blk)  

end

# File lib/px_todo.rb, line 29
def to_px
  @px
end

# File lib/px_todo.rb, line 35
def import_txt(s, &blk)
  
  # remove the file heading
  lines = s.lines
  lines.shift 3
  
  declar = "<?ph schema='items[title]/todo[#{@fields.join(',')}]'" + 
      " format_masks[0]='[!title]'?>"

  # add a marker to identify the headings after parsing the records
  
  s2 = lines.join.gsub(/^#+\s+/,'\0:')
  
  @px = PolyrexHeadings.new(declar + "\n" + s2).to_polyrex
  
  @px.each_recursive do |x, parent, level|

    blk.call x if block_given?

    todo = x.title
    raw_status = todo.slice!(/\[.*\]\s+/)
    x.title = todo

    status  = raw_status =~ /\[\s*x\s*\]/ ? 'done' : ''      

    x.status = status
          
    # is there a note?
    
    note = todo[/^note:\s+(.*)/i,1]
    
    if note and parent.is_a? PolyrexObjects::Todo then
      
      parent.note = note
      x.delete
        
    end
    
    # is it a heading?
    
    heading = todo[/^:(.*)/,1]
    
    if heading then

      # does the heading contain tags?
      
      raw_tags = heading.slice!(/\s+#.*$/)
      x.tags = raw_tags[/#\s+(.*)/,1] if raw_tags
      x.heading = heading
      x.title = ''        
      
    end
    
  end  
  
end

# File lib/px_todo.rb, line 92
def load_px(s)
  @px = Polyrex.new s
end