added step debug

extract_amendments.rb

@@ -22,17 +22,23 @@ class AmendmentExtractor
   end
 
   def extract(opendocument_path, options = {})
+    debug "extracting content from document"
     xml = nil
     Zippy.open(opendocument_path) do |zip|
       xml = zip['content.xml']
     end
 
+    
+    debug "parsing document xml"
     doc = Nokogiri::XML::Document.parse(xml)
 
     if options[:xml_dump_path]
+      debug "dumping xml"
       File.open(options[:xml_dump_path], "w") { |f| f.write doc.to_xml(indent: 2) }
     end
 
+    
+    debug "parsing styles"
     styles = {}
     doc.css("style|style").each do |node|
       name = node["style:name"]
@@ -46,9 +52,13 @@ class AmendmentExtractor
       styles[name] = style
     end
 
+    
+    debug "extracting document text"
     text = doc.xpath('//office:text').first
     raise "no office:text found" unless text
 
+    
+    debug "extracting amendment nodes"
     amend_start = nil
     amend_nodes = []
 
@@ -67,6 +77,8 @@ class AmendmentExtractor
 
     debug amendments_found: amend_nodes.length
 
+    
+    debug "extracting info from amendments"
     amendments = []
 
     amend_nodes.each do |nodes|
@@ -78,6 +90,9 @@ class AmendmentExtractor
       num_am = amend_doc.xpath("//NumAm").first.text
       
       next if options[:parse_only_num] and num_am != options[:parse_only_num].to_s
+
+      
+      debug "parsing amendment #{num_am}"
       
       doc_amend = amend_doc.xpath("//DocAmend").first.text
       article = amend_doc.xpath("//Article").first.text
@@ -88,18 +103,23 @@ class AmendmentExtractor
         article: article,
       }
       debug amendment
+
       
+      debug "parsing amendment table"
       tables = nodes.css('table|table')
       raise "amendment table not found" if tables.size == 0
       raise "too many tables" if tables.size > 1
       table = tables.first
+
       
+      debug "converting table nodes to text"
       text_table = table.css("table|table-row").map do |row|
         row.css("table|table-cell").map do |cell|
           cell.css("text|p").map do |paragraph|
             paragraph.children.map do |element|
               text = element.text
               
+              # add mediawiki triple quote if the text is bold in the document
               if element.is_a? Nokogiri::XML::Element and element.name == 'span'
                 style_name = element["text:style-name"]
                 if style_name and styles[style_name][:bold]
@@ -114,9 +134,13 @@ class AmendmentExtractor
       end
       debug text_table: text_table
       
+
+      debug "locating header"
       header_index = text_table.index(["Text proposed by the Commission", "Amendment"])
-      raise "first row not found in table of amendment #{num_am}" unless header_index
+      raise "header not found in table of amendment #{num_am}" unless header_index
       
+
+      debug "extracting changes from table"
       changes = text_table[(header_index + 1)..-1]
       raise "amendment changes not found" if changes.size == 0
       
@@ -128,6 +152,7 @@ class AmendmentExtractor
       break if options[:parse_only_one]
     end
 
+    debug "rendering amendments"
     template = ERB.new File.read('template.erb'), nil, '-'
 
     result = []