document.rb in docparser-0.1.0

- old
+ new

@@ -3,30 +3,33 @@
   # The Document class loads and parses the files.
   # @see Parser
   # @see Output
   class Document
     attr_reader :filename, :doc, :encoding, :results
-    def initialize(filename, encoding: 'utf-8', parser: nil)
+    def initialize(filename: nil, encoding: 'utf-8', parser: nil)
       if encoding == 'utf-8'
         encodingstring = 'r:utf-8'
       else
         encodingstring = "r:#{encoding}:utf-8"
       end
-
+      @logger = Log4r::Logger.new('docparser::document')
+      @logger.debug { "Parsing #{filename}" }
       open(filename, encodingstring) do |f|
-        @doc = Nokogiri::HTML(f)
+        @html = f.read
+        @logger.warn "#{filename} is empty" if @html.empty?
+        @doc = Nokogiri(@html)
       end
-
       @encoding = encoding
       @parser = parser
       @filename = filename
-      @results = Array.new(@parser.outputs.length) { [] }
+      @results = Array.new(@parser.outputs ? @parser.outputs.length : 0) { [] }
     end
 
     # Adds a row to an output
     def add_row(*row, output: 0)
       output = @parser.outputs.index(output) if output.is_a? Output
+      @logger.debug { "#{filename}: Adding row #{row.flatten.to_s}" }
       results[output] << row.flatten
     end
 
     # Extracts the document title
     # @return [String] the title of the document
@@ -34,17 +37,21 @@
       @title ||= xpath_content('//head/title')
     end
 
     # @return [String] the source of the document
     def html
-      @html ||= @doc.inner_html #TODO: ??
+      @html
     end
 
     # Executes a xpath query
     def xpath(query)
       res = @doc.search(query)
-      res.each { |el| yield el } if block_given?
+      if block_given?
+        res.each { |el| yield el }
+      else
+        res
+      end
     end
 
     # Executes a xpath query and returns the content
     # @return [String] the content of the HTML node
     def xpath_content(query)
@@ -56,11 +63,11 @@
       end
     end
 
     # Matches the HTML source using a regular expression
     def regexp(regexp)
-      html.match(regexp) rescue nil
+      html.match(regexp)
     end
 
     # Parses the document
     # @return [Array] containing the parse results
     def parse!(&block)
@@ -68,12 +75,12 @@
       results
     end
 
     # @!visibility private
     def inspect
-      "<Document file:'#{@filename}'>"
+      "<Document file:'#{@filename}', encoding:'#{@encoding}'>"
     end
 
-    alias :css :xpath
-    alias :css_content :xpath_content
+    alias_method :css, :xpath
+    alias_method :css_content, :xpath_content
   end
 end
\ No newline at end of file