oa_templater 0.5.5

data/lib/oa_templater/oa_templater.rb

@@ -0,0 +1,862 @@
+# encoding: UTF-8
+
+require 'support/oa_regexes'
+require 'support/formats'
+
+require 'nkf'
+require 'fileutils'
+require 'date'
+require 'sablon'
+require 'yaml'
+require 'csv'
+require 'charlock_holmes'
+require 'kakasi'
+
+module OaTemplater
+  class OA
+    attr_accessor :outputfile
+    attr_reader :props
+
+    def initialize(sourcefile, casenumber = '11110')
+      @sourcefile = sourcefile
+      @casenumber = casenumber
+      read_oa_data
+      set_templates
+      init_instance_vars
+      set_reasons_file
+      read_templatable_file #chunks of text for swapping
+    end
+
+    # require template files, not included because of NDA
+    def set_templates(options = {})
+      defaults = {  kyozetsuriyu: File.join(File.dirname(__FILE__), 'default_riyu.docx'),
+                    shinpankyozetsuriyu: File.join(File.dirname(__FILE__), 'default_shinpankyozetsuriyu.docx'),
+                    kyozetsusatei: File.join(File.dirname(__FILE__), 'default_satei.docx'),
+                    shinnen: File.join(File.dirname(__FILE__), 'default_shinnen.docx'),
+                    shireisho: File.join(File.dirname(__FILE__), 'default_shireisho.docx'),
+                    rejectamendments: File.join(File.dirname(__FILE__), 'default_rejectamendments.docx'),
+                    examiners: File.join(File.dirname(__FILE__), 'examiners.txt')
+                  }
+      @templates = defaults.merge(options)
+      pick_template
+    end
+
+    # require reason file, not included because of NDA
+    def set_reasons_file(r = File.join(File.dirname(__FILE__), 'reasons.yml'))
+      @reasons = YAML.load_file(r)
+    end
+
+    # A bunch of set phrases that can be swapped out
+    def read_templatable_file(r = File.join(File.dirname(__FILE__), 'support', 'templatable.yml'))
+      @templatables = YAML.load_file(r)
+    end
+
+    def parse_appeal_drafted
+      capture_the(:appeal_drafted, R_CAPTURE_APPEAL_DRAFTED)  # year/month/day
+      return if @scrapes[:appeal_drafted].nil?
+
+      set_prop(:appeal_drafted, format_date('%04u/%02u/%02u', @scrapes[:appeal_drafted]))
+    end
+
+    def parse_drafted
+      capture_the(:drafted, R_CAPTURE_DRAFTED)  # year/month/day
+      return if @scrapes[:drafted].nil?
+
+      set_prop(:drafted, format_date('%04u/%02u/%02u', @scrapes[:drafted]))
+    end
+
+    def parse_mailing_date(demarker = '-')
+      @outputfile = 'oa_template'
+      capture_the(:mailing_date, R_CAPTURE_MAILING_DATE)
+      return if @scrapes[:mailing_date].nil?
+
+      @outputfile = "ALP#{@casenumber}#{demarker}#{@template_name}#{demarker}#{format_date('%04u%02u%02u', @scrapes[:mailing_date])}.docx"
+
+      set_prop(:mailing_date, format_date('%04u/%02u/%02u', @scrapes[:mailing_date]))
+    end
+
+    def parse_satei_previous_oa
+      capture_the(:satei_previous_oa, R_CAPTURE_PREVIOUS_OA)
+      return if @scrapes[:satei_previous_oa].nil?
+
+      set_prop(:satei_previous_oa, format_date('%04u/%02u/%02u', @scrapes[:satei_previous_oa]))
+
+      # set "and Amendments"
+      set_prop(:and_amendments, R_AND_AMENDMENTS =~ @data ? 'Remarks and Amendments' : 'Remarks')
+      
+      # set the reason
+      # default
+      set_prop(:satei_reasons, 'the reasons')
+      r = @data.match(R_SATEI_REASONS)
+      m = r.nil? ? '理由' : r[1].gsub!(/\s+/, '')
+      set_prop(:satei_reasons, m == '理由' ? 'the reasons' : format_headers(m))
+    end
+
+    def parse_amendments_date
+      capture_the(:amendments_date, R_CAPTURE_AMENDMENTS_DATE)
+      return if @scrapes[:amendments_date].nil?
+
+      set_prop(:amendments_date, format_date('%04u/%02u/%02u', @scrapes[:amendments_date]))
+    end
+
+    def parse_retroactive
+      capture_the(:retroactive, R_CAPTURE_RETROACTIVE)
+      return if @scrapes[:retroactive].nil?
+
+      set_prop(:retroactive, format_date("\nFiling Date (Retroactive Date) \t%04u/%02u/%02u\n \n", @scrapes[:retroactive]))
+    end
+
+    def parse_appeal_no
+      capture_the(:appeal_no, R_CAPTURE_APPEAL_NO)
+      return if @scrapes[:appeal_no].nil?
+
+      set_prop(:appeal_no, NKF.nkf('-m0Z1 -w', @scrapes[:appeal_no][1]) + '-' + NKF.nkf('-m0Z1 -w', @scrapes[:appeal_no][2]))
+    end
+
+    def parse_shireisho_app
+      capture_the(:shireisho_num, R_CAPTURE_SHIREISHO_APP)
+      return if @scrapes[:shireisho_num].nil?
+
+      set_prop(:shireisho_num, NKF.nkf('-m0Z1 -w', @scrapes[:shireisho_num][1]) + '-' + NKF.nkf('-m0Z1 -w', @scrapes[:shireisho_num][2]))
+    end
+
+    def parse_shireisho_code
+      capture_the(:scode, R_CAPTURE_SHIREISHO_CODE)
+      return if @scrapes[:scode].nil?
+
+      set_prop(:scode, NKF.nkf('-m0Z1 -w', @scrapes[:scode][1]) + ' ' + NKF.nkf('-m0Z1 -w', @scrapes[:scode][2]))
+    end
+
+    def parse_app_no
+      capture_the(:app_no, R_CAPTURE_APP_NO)
+      if @scrapes[:app_no].nil?
+        #try for the appeal format if nothing came up
+        capture_the(:app_no, R_CAPTURE_APPEAL_APP_NO)
+      end
+      return if @scrapes[:app_no].nil?
+
+      set_prop(:app_no, NKF.nkf('-m0Z1 -w', @scrapes[:app_no][1]) + '-' + NKF.nkf('-m0Z1 -w', @scrapes[:app_no][2]))
+    end
+
+    # definitely need to fix this up later, haha
+    def parse_examiner(do_examiner = false)
+      capture_the(:taro, R_CAPTURE_TARO) # 1, 2 (codes are #3, 4)
+
+      #if there was no normal appeal examiner, try an appeal examiner
+      if @scrapes[:taro].nil?
+        capture_the(:taro, R_CAPTURE_APPEAL_TARO)
+      end
+
+      return if @scrapes[:taro].nil?
+
+      found = false
+
+      if do_examiner
+        last, first = @scrapes[:taro][1], @scrapes[:taro][2]
+
+        CSV.foreach(@templates[:examiners]) do |r|
+          if NKF.nkf('-m0Z1 -w', r[1]).eql? (' ' + last + ' ' + first)
+            set_prop(:taro, r[0])
+            found = true
+            break
+          end
+        end
+
+        unless found
+          found = true
+          first, last = Kakasi.kakasi('-Ja', first).capitalize, Kakasi.kakasi('-Ja', last).upcase
+
+          # use kakashi to romajify the Examiner names
+          set_prop(:taro, "#{first} #{last} #{@scrapes[:taro][1]} #{@scrapes[:taro][2]}")
+        end
+      end
+
+      set_prop(:taro, @scrapes[:taro][1] + ' ' + @scrapes[:taro][2]) unless found
+
+      # always set examiners numbers
+      set_prop(:code, NKF.nkf('-m0Z1 -w', @scrapes[:taro][3]) + ' ' + NKF.nkf('-m0Z1 -w', @scrapes[:taro][4]))
+    end
+
+    def parse_appeal_examiner
+      capture_the(:appeal_taro, R_CAPTURE_APPEAL_TARO) # 1, 2
+      return if @scrapes[:appeal_taro].nil?
+
+      set_prop(:appeal_taro, @scrapes[:appeal_taro][1] + ' ' + @scrapes[:appeal_taro][2])
+    end
+
+    def parse_final_oa
+      set_prop(:reason_for_final, '')
+      capture_the(:final_oa, /＜＜＜＜\p{Z}+最\p{Z}{0,6}後\p{Z}+＞＞＞＞/)
+      return if @scrapes[:final_oa].nil?
+      set_prop(:final_oa, "\n<<<<    FINAL    >>>>\n")
+      set_prop(:reason_for_final, Sablon.content(:word_ml, FINALWML))
+    end
+
+    def parse_see_list
+      set_prop(:see_list, /引用文献等については引用文献等一覧参照/ =~ @data ? "  \n(See the List of Citations for the cited publications)  \n" : '')
+    end
+
+    def parse_response_period
+      set_prop(:response_period, R_RESPONSE_PERIOD =~ @data ? '60 days' : 'three months')
+    end
+
+    def parse_our_lawyer
+      capture_the(:our_lawyer, /[特許出願人]*代理人[弁理士]*[弁護士]*\p{Zs}+(\S+?)\p{Zs}(\S+?)/)
+      return if @scrapes[:our_lawyer].nil?
+
+      # only check last name
+      case @scrapes[:our_lawyer][1]
+      when '村山'
+        set_prop(:our_lawyer, 'Yasuhiko MURAYAMA')
+      when '志賀'
+        set_prop(:our_lawyer, 'Masatake SHIGA')
+      when '佐伯'
+        set_prop(:our_lawyer, 'Yoshifumi SAEKI')
+      when '渡邊'
+        set_prop(:our_lawyer, 'Takashi WATANABE')
+      when '実広'
+        set_prop(:our_lawyer, 'Shinya JITSUHIRO')
+      when '棚井'
+        set_prop(:our_lawyer, 'Sumio TANAI')
+      else
+        set_prop(:our_lawyer, 'Taro TOKKYO')
+      end
+    end
+
+    def parse_note_to_applicant
+      capture_the(:note_to_applicant, /本願出願時に公開されており、/)
+      return if @scrapes[:note_to_applicant].nil?
+
+      set_prop(:note_to_applicant, "\t• Request to the Applicant\r\n\tCitation 1 was already published at the time of filing of the present application and has a common applicant or inventor with the present application.  Citation 1 alone would be a bar to the novelty or inventive step of more than one claim of the present application.\r\n\tBased on this type of citation, appropriately evaluating the invention in advance can be thought to be beneficial to the applicant while creating appropriate claims, as well as helpful to the Examiner for an efficient and accurate examination.  We request that the applicant disclose this type of citation that the applicant is already aware of when filing the application or a request for examination, as well as requesting that the applicant evaluates whether or not the invention for which a patent is sought has patentability based on this type of citation. ")
+    end
+
+    def parse_currently_known
+      case @data
+      when /拒絶の理由を発見しない請求項/
+        if m = @data.match(R_CAPTURE_NO_REJECT_CLAIMS)
+          set_prop(:currently_known, "<Claims for which no reasons for rejection have been found>\r\n \tNo reasons for rejection are currently known for #{format_headers(m[1])} which were not indicated in this Notice of Reasons for Rejection.  The applicant will be notified of new reasons for rejection if such reasons for rejection are found.")
+        else
+          set_prop(:currently_known, "<Claims for which no reasons for rejection have been found>\r\n \tNo reasons for rejection are currently known for the claims which were not indicated in this Notice of Reasons for Rejection.  The applicant will be notified of new reasons for rejection if such reasons for rejection are found.")
+        end
+      when /拒絶の理由が通知される/
+        set_prop(:currently_known, 'The applicant will be notified of new reasons for rejection if such reasons for rejection are found.')
+      else
+        set_prop(:currently_known, '')
+      end
+    end
+
+    def parse_ipc
+      ipc_text = ''
+
+      if m = @data.match(/先行技術文献(?:等{0,1})調査結果(.*?)..先行技術文献/m)
+        data = m[1]
+        ipc_list_end = m.end(0)
+        if m = data.match(/(Ｉ|I)(Ｐ|P)(Ｃ|C)/)
+          data = data[m.begin(0)..-2]
+          ipc_text = NKF.nkf('-m0Z1 -w', data).gsub('IPC', 'IPC:').gsub('DB名', "\tDB Name:").gsub('^\p{Z}{3,8}', "\t ")
+          parse_ipc_references(ipc_list_end)
+        end
+      end
+
+      set_prop(:ipc_list, ipc_text)
+    end
+
+    def parse_ipc_references(ipc_list_end)
+      ipc_reference_text = ''
+      data = @data[ipc_list_end..-1]
+
+      if m = data.match(/(^.*先行技術文献調査結果|この拒絶理由通知の内容)/)
+        @cits ||= YAML.load_file(CITATIONS_FILE)
+        data = data[0..m.begin(0)]
+        oldmatch = false
+        count = 1
+
+        data.each_line do |line|
+          match = false
+          @cits.each do |_n, a|
+            if m = line.match(a['japanese'])
+              match = true
+              ipc_reference_text += "#{count}.  #{convert_pub_no(m, a['english'])}\n"
+            end
+          end # cits.each
+
+          unless match
+            # if no match, change 全角 to 半角
+            line = NKF.nkf('-m0Z1 -w', line)
+
+            # first line of non-match
+            if oldmatch and (!match)
+              line.gsub!(/^/, "#{count}. ") if line.length > 4
+            end
+
+            # >1st line of non-match
+            if (!oldmatch) and (!match)
+              count -= 1 #decrease count to that it stays the same after being increased below
+              #remove newlines since it's probably a big english title
+              ipc_reference_text.gsub!(/\r\n$/,"\n") if line.length > 4
+            end
+
+            line = pad_spaces(line)
+
+            ipc_reference_text += line
+          end
+
+          # increase count
+          count += 1
+
+          oldmatch = match
+        end
+      end
+
+      set_prop(:ipc_reference_text, ipc_reference_text)
+    end
+
+    def parse_headers(dh)
+      oa_headers_text = ''
+
+      if dh and m = @data.match(/(?:理\p{Z}{0,2}由.*^\p{Z}*先行技術文献調査結果の記録|理\p{Z}{0,2}由.*^－－－－－－－－－－－|\p{Z}記\p{Z}.*引\p{Z}?用\p{Z}?文\p{Z}?献\p{Z}?等\p{Z}?一\p{Z}?覧|\p{Z}記\p{Z}.*^－－－－－－－－－－－|理\p{Z}{0,2}由.*引\p{Z}?用\p{Z}?文\p{Z}?献\p{Z}?等\p{Z}?一\p{Z}?覧|理\p{Z}{0,2}由.*最後の拒絶理由通知とする理由|検討しましたが.*|\p{Z}理\p{Z}{0,2}由.*この通知に関するお問い合わせがございましたら)/mi)
+        # gsub to strip HTML tags from japanese text
+        tdata = @data[m.begin(0)..m.end(0)].gsub(%r{</?[^>]+?>}, '').gsub("\r\n", "\n")
+        #
+        # matches stuff like this
+        # （理由１）
+        # ＜請求項１－１１＞
+        # ・引用文献１
+        # 引用文献１
+        # 引用文献：１
+        # 備考
+
+        line = ''
+        until tdata.nil? do
+          m = tdata.match(/^(.*?)\n/)
+          break if m.nil?
+
+          line = m[1] =~ /^\s*$/ ? "\n" : m[1] #save a newline if it's empty
+
+          tdata = tdata[m.end(0)..-1]
+
+          if res = check_for_templatable_portion(line, tdata)
+            tex, tdata = res
+            # added a match against unnecessary IPC lines
+            oa_headers_text += tex
+          elsif m = line.match(R_HEADER_TYPES)
+            tex = m[1]
+
+            oa_headers_text += tex =~ /^\s*$/ ? "\n" : format_headers(tex) + "\n" unless mistaken_header?(tex)
+          end
+        end
+
+        3.times { oa_headers_text.gsub!("\n\n\n", "\n") }
+
+        oa_headers_text.encode!(:xml => :text)
+        #replace OAOA_TEMPLATER_TAB with word_ml newline + tab
+        oa_headers_text.gsub!(/OAOA_TEMPLATER_TAB/, STOPSTARTTAB) 
+        #replace newlines with word_ml newlines
+        oa_headers_text.gsub!(/\n/, STOPSTARTP) 
+
+
+        set_prop(:oa_headers, Sablon.content(:word_ml, sprintf(HEADERS_FMT, oa_headers_text)))
+      end
+    end
+
+    def check_for_templatable_portion(line, tdata)
+      odata = [line, tdata].join.gsub("\n", '') #unshift the first line back to tdata
+      @templatables.each do |t, a|
+        if line =~ a['detect'] 
+          if m = odata.match(a['full'])
+            #this starting offset should actually be m.end(0) - line.length + (the number of newline characters up to the match)
+            tdata = [line, "\n", tdata].join
+            tdata = tdata[m.end(0) .. -1]  
+            tex = odata[m.begin(0)..m.end(0)]
+            tex.gsub!(a['full']) do 
+              res = a['text']
+              Regexp.last_match.captures.each_with_index do |match, i|
+                if match =~ /^請求項[\p{N},～、－及びおよ]+$/
+                  res = res.gsub(/\\#{i+1}/, format_headers(match))
+                elsif match =~ /^請求項[\p{N},～、－及びおよ]+に係る発明$/
+                  res = res.gsub(/\\#{i+1}/, format_invention_according_to(match))
+                else
+                  res = res.gsub(/\\#{i+1}/, NKF.nkf('-m0Z1 -w', match))
+                end
+              end
+              res
+            end
+            return  [tex, tdata]
+          end
+        end
+      end
+
+      return nil
+    end
+
+    def format_invention_according_to(str)
+      res = format_headers(str.gsub('に係る発明','').gsub('請求項', 'the invention according to 請求項'))
+      res += res =~ /inventions/ ? ' are' : ' is'
+    end
+
+    def mistaken_header?(tex)
+      val = false
+      val = true if tex =~ /調査/ || /先行技術文/ =~ tex || /注意/ =~ tex and !(/検討しましたが/ =~ tex)
+      val = true if (tex =~ /段/) || (tex =~/段/) || (tex =~ /には/) || (tex =~ /が/)
+      val
+    end
+
+    def parse_citations
+      citation_text = ''
+
+      if m = @data.match(R_CITATIONS_START)
+        @cits ||= YAML.load_file(CITATIONS_FILE)
+        count = 0
+        data = @data[m.end(0) - 2..-1].gsub(%r{</?[^>]+?>}, '') # end minus '1.', gsub to remove html
+
+        catch :done_scanning do
+          data.each_line do |line|
+            tex = line
+            throw :done_scanning if (/^\s*$/ =~ line) || (line[0..2].eql?('－－－'))
+
+            old_citation_text = citation_text
+            if /^\p{Z}*\p{N}+((?:\.|．|：)+.*?)/m =~ tex
+              count += 1
+            end
+
+            @cits.each do |_n, a|
+              if m = tex.match(a['japanese'])
+                if /United States/ =~ a['english']
+                  # citation is in English (no prime needed)
+                  citation_text += sprintf(CIT_SIMPLE, count, convert_pub_no(m, a['english']))
+                else # normal
+                  if /Published Japanese Translation No./ =~ a['english']
+                    citation_text += sprintf(CIT_WITH_PRIME_PCT, count, convert_pub_no(m, a['english']), count)
+                  else
+                    citation_text += sprintf(CIT_WITH_PRIME, count, convert_pub_no(m, a['english']), count)
+                  end
+                end
+              end
+            end # cits
+
+            if old_citation_text == citation_text
+              tex = NKF.nkf('-m0Z1 -w', tex)
+              # strip blank dos lines
+              tex.gsub!(/\p{Z}*\r\n/, '')
+              tex = pad_spaces(tex)
+
+              # if no match was found, just copy the japanese, skip first character (it's a period from the regex)
+              # should have the correct number from the actual source (not from count variable)
+              tex.encode!(:xml => :text) if tex
+              citation_text += sprintf(CIT_MISS, tex)
+            end
+          end # each line
+        end # catch
+      end # if citations found
+
+      set_prop(:citation_list, Sablon.content(:word_ml, citation_text))
+    end
+
+    def pad_spaces (tex)
+      # add space after period, add space after comma, remove year kanji, reduce multiple spaces to 1
+      tex.gsub!(/\.(\p{N})/, '. \1')
+      tex.gsub!(/\,/, ', ')
+      tex.gsub!(/年/, '')
+      tex.gsub!(/p{Z}*/, ' ')
+      tex.gsub!(/. ,/, '.,')
+      tex
+    end
+
+    def convert_pub_no(m, eng)
+      #m is MatchData object, handle different styles of citations
+      #by using the number of captures
+      case m.length
+      when 2
+        pub = (eng =~ /United States Patent No/) ? eng.gsub('CIT_NO', NKF.nkf('-m0Z1 -w', m[1]).to_i.commas) : (eng =~ /European Patent/ ? eng.gsub('CIT_NO', NKF.nkf('-m0Z1 -w', m[1]).to_i.eurostyle) : eng.gsub('CIT_NO', NKF.nkf('-m0Z1 -w', m[1])))
+      when 3
+        pub = eng.gsub('CIT_NO', (NKF.nkf('-m0Z1 -w', m[1]) + '/' + NKF.nkf('-m0Z1 -w', m[2])))
+      when 4, 5
+        pub = eng.gsub('CIT_NO', convert_possible_heisei(m[2], m[3], m[4]))
+      when 9
+        pub = eng.gsub(/CIT_NO /, convert_possible_heisei(m[2], m[3], m[4]) + ' ').gsub('CIT_NO2', convert_possible_heisei(m[6], m[7], m[8]))
+      end
+
+      pub
+    end
+
+    # matches /([昭|平]*)(\p{N}+?).(?:\p{Z}*)(\p{N}+?)号/
+    # convert 平０９－０６０２７４ into H09-060274 or ２００８-００３７４９ into 2008-003748
+    def convert_possible_heisei(hs, first, last)
+      no = ''
+      case hs
+      when'平'
+        no += 'H' + sprintf('%02u', NKF.nkf('-m0Z1 -w', first).to_i(10)) + '-' + NKF.nkf('-m0Z1 -w', last)
+      when '昭'
+        no += 'S' + sprintf('%02u', NKF.nkf('-m0Z1 -w', first).to_i(10)) + '-' + NKF.nkf('-m0Z1 -w', last)
+      else
+        no += NKF.nkf('-m0Z1 -w', first) + '-' + NKF.nkf('-m0Z1 -w', last)
+      end
+
+      no
+    end
+
+    def parse_articles
+      count = 1
+      articles_text = '<w:p><w:pPr><w:kinsoku w:val="0"/><w:spacing w:line="360" w:lineRule="atLeast"/></w:pPr><w:r w:rsidR="006A661C"><w:rPr><w:b/><w:noProof/></w:rPr><w:t>Cited Articles:</w:t><w:tab/><w:tab/><w:tab/>'
+      reasons_for_text = '<w:p><w:pPr><w:kinsoku w:val="0"/><w:spacing w:line="360" w:lineRule="atLeast"/></w:pPr><w:r w:rsidR="006A661C"><w:rPr><w:noProof/></w:rPr>'
+
+      found_articles = []
+      original_length = articles_text.length
+
+      m = @data.match(R_ARTICLE_SECTION)
+      target_data = m ? m[0] : @data
+
+      line_num = 0
+
+      while (line = target_data.lines[line_num..line_num+1]) do
+        #last line+1 will return nil
+        if line
+          line = line.join.gsub(/\r\n|\r|\n/,'') #merge 2 lines and remove newline
+        else
+          break
+        end
+
+        line_num += 1
+
+        @reasons.each do |_r, a|
+          if line =~ a['japanese']
+            found_articles << a['short']
+
+            unless /#{Regexp.quote(a["english"])}/m =~ reasons_for_text 
+              reasons_for_text += "<w:t>#{count}.</w:t><w:tab/><w:t>#{a['english']}</w:t><w:br/><w:br/>"
+              count += 1
+            end
+          end
+        end
+      end
+
+      found_articles.uniq.sort.each do |a|
+        # skip tab on first reason
+        articles_text += "<w:tab/><w:tab/><w:tab/><w:tab/><w:tab/>" unless articles_text.length == original_length
+        # only add short text once (36 shows up multiple times)
+        articles_text += "<w:t>#{a}</w:t><w:br/>" 
+      end
+
+      # remove number if only 1 article listed
+      reasons_for_text.gsub!(/<w:t>1.<\/w:t><w:tab\/><w:t>/, '<w:tab\/><w:t>') if count == 2
+
+      #remove final word_ml newline
+      reasons_for_text.gsub!(/<w:br\/><w:br\/>$/, '')
+      articles_text.gsub!(/<w:br\/>$/, '')
+
+      #close the paragraph
+      articles_text += '</w:r></w:p>'
+      reasons_for_text += '</w:r></w:p>'
+
+      set_prop(:articles, Sablon.content(:word_ml, articles_text))
+      set_prop(:reasons_for, Sablon.content(:word_ml, reasons_for_text))
+    end
+
+    def finish(options = {})
+      defaults = {  sablon: true
+                  }
+      options = defaults.merge(options)
+
+      stemplate = Sablon.template(@template)
+      stemplate.render_to_file @outputfile, @props
+    end
+
+    def scan(options = {})
+      defaults = {  do_headers: false,
+                    do_dashes: 0,
+                    do_examiner: false,
+                    sablon: true
+                  }
+      options = defaults.merge(options)
+
+      parse_mailing_date options[:do_dashes]
+      parse_examiner options[:do_examiner]
+      parse_app_no
+      parse_drafted
+      parse_our_lawyer
+      parse_response_period
+      parse_see_list
+      parse_final_oa
+      parse_amendments_date
+      parse_satei_previous_oa
+      parse_articles
+      parse_currently_known
+      parse_citations
+      parse_ipc
+      parse_appeal_examiner
+      parse_appeal_drafted
+      parse_appeal_no
+      parse_retroactive
+      parse_note_to_applicant
+      parse_shireisho_app
+      parse_shireisho_code
+
+      parse_headers options[:do_headers]
+    end
+
+    private
+
+    def squish!(t)
+      t.gsub!(/\A[[:space:]]+/, '')
+      t.gsub!(/[[:space:]]+\z/, '')
+      t.gsub!(/[[:space:]]+/, ' ')
+    end
+
+    def format_headers(tex, options = {})
+      defaults = {  replace_toh: false,
+                    ignore_toh: true,
+                    markdown: false
+                  }
+      options = defaults.merge(options)
+
+      squish! tex
+
+      # try to handle when Examiners put multiple groups separated by : or /
+      # on the same line like 引用文献１：請求項１，２/ bla
+      formatted_text = ''
+      if R_HEADER_SEPARATOR_DETECT =~ tex
+        # super fragile. If regex is changed
+        # demarker = NKF.nkf('-m0Z1 -w', '#{$&[1,1]} ')
+        demarker = NKF.nkf('-m0Z1 -w', "#{$1} ") #$~ is last matchdata
+        tex.split(R_HEADER_SEPARATOR).each do |section|
+          formatted_text += demarker unless formatted_text.length == 0
+
+          if section =~ R_JPL_DETECT
+            formatted_text += handle_jpl(section)
+          else
+            #no jpl to handle
+            formatted_text += format_headers(section, options)
+          end
+
+        end
+      else
+        if /#{R_HEADER_REASONS}/x =~ tex
+          #handle special Reason lines
+          if /及び|、/ =~ tex
+            tex.split(/及び|、/).each do |section|
+              section = format_number_listing(section)
+
+              formatted_text += ' and ' unless formatted_text.length == 0
+              formatted_text += "#{replace_common_phrases(section, options)}".gsub('(', ' (') #add space before parenthasis
+            end
+          else
+            if tex =~ R_JPL_DETECT
+              formatted_text += handle_jpl(tex)
+            else
+              #no jpl to handle
+              formatted_text = "#{replace_common_phrases(tex, options)}"
+            end
+          end
+          formatted_text.gsub!('Reason(', 'Reason (')
+          formatted_text.gsub!(/(\p{N})\(/, '\1 (')
+        else
+          if tex =~ R_JPL_DETECT
+            #note : it cant get here
+            formatted_text += handle_jpl(tex)
+          else
+            #no jpl to handle
+            formatted_text = "#{replace_common_phrases(tex, options)}"
+          end
+        end
+      end
+
+      formatted_text
+    end
+
+    def replace_common_phrases(tex, options = {})
+      defaults = {  replace_toh: false,
+                    ignore_toh: true
+      }
+      options = defaults.merge(options)
+
+      tex = NKF.nkf('-m0Z1 -w', tex)
+      tex = swap_words(tex)
+      tex.gsub!('等', '') if options[:ignore_toh]
+      tex.gsub!('等', ', etc.') if options[:replace_toh]
+
+      # strip abberant \r characters
+      tex.gsub!("\r", '')
+
+      tex = format_number_listing(tex)
+    end
+
+    def handle_jpl(tex)
+      #comes in looking something like "（Ａ）理由１（特許法２９条１項３号）"
+      jpl = ''
+      tex = NKF.nkf('-m0Z1 -w', tex)
+      jpl = tex.gsub(/(.*)特許法第?(\p{N}+)条第?(\p{N}+)項(?:第?(\p{N}+.*)号)*/){ 
+        replace_common_phrases(format_number_listing($1)) + "Japanese Patent Law, Article #{$2}, Paragraph #{$3}, Number #{$4}"
+      }
+      jpl.gsub!(/, Number\p{Z}+\)(?:について)?$/, ')') #if it doesnt have a \4
+      jpl.gsub!(')R', ') R')
+      jpl.gsub!('Reason(', 'Reason (')
+      jpl.gsub!(/(\p{N})\(/, '\1 (')
+      jpl
+    end
+
+    #do actual swapping of japanese and english words
+    def swap_words(tex)
+      tex.gsub!('、', ',')
+      tex.gsub!('，', ',')
+      tex.gsub!('拡大先願', 'Expansion of Application')
+      tex.gsub!('ないし', 'to')
+      tex.gsub!('について', '')
+      tex.gsub!('のいずれか', 'any one of')
+      tex.gsub!('及び', ',')
+      tex.gsub!('および', ',')
+      tex.gsub!('進歩性', 'Inventive Step')
+      tex.gsub!('実施可能要件', 'Enablement Requirements')
+      tex.gsub!('産業上の利用可能性', 'Industrial Applicability')
+      tex.gsub!('発明の単一性', 'Unity of Invention')
+      tex.gsub!('明確性', 'Clarity')
+      tex.gsub!('サポート要件', 'Support Requirements')
+      tex.gsub!('新規性', 'Novelty')
+      tex.gsub!(/請\p{Z}*求\p{Z}*項/, 'Claim')
+      tex.gsub!('引用文献', 'Citation')
+      tex.gsub!(/引\p{Z}*用\p{Z}*例/, 'Citation')
+      tex.gsub!(/実\p{Z}*施\p{Z}*例/, 'Embodiment')
+      tex.gsub!(/理\p{Z}*由/, 'Reason')
+      tex.gsub!(/先\p{Z}*願/, 'Prior Application')
+      tex.gsub!('－', 'to')
+      tex.gsub!('-', 'to')
+      tex.gsub!('～', 'to')
+      tex.gsub!('乃至', 'to')
+      tex.gsub!('理由', 'Reason')
+
+      # match 備考:
+      tex.gsub!('備考', 'Notes')
+
+      tex
+    end
+
+    # formats a number listing assuming only one list in the string
+    # one level up, format_headers breaks single lines into a plurality of these
+    # ex: 請求項３，１７，３１，４５
+    def format_number_listing(tex)
+      tex = NKF.nkf('-m0Z1 -w', tex)
+
+      # if no numbers (like 'Notes:') then do nothing
+      if m = tex.match(/(?:...)(.*?)\p{N}/) # skip first two charcters in case it's something like '１．理由１，２について'
+        # opening, numbers, close
+        op = tex[0..m.end(1) - 1]
+        num_start = m.end(1)
+        m = tex.match(/\p{N}(?!.*\p{N})/)
+        cl = tex[m.end(0)..-1]
+        nums = tex[num_start..m.end(0) - 1]
+
+        parsed = nums.split(/((?:～|-)*\p{N}+(?:to\p{N}+)*,*)/).reject(&:empty?)
+
+        # change ['1to2,', '3'] to ['1', '2', '3']
+        parsed.each_index do |el|
+          if /to\p{N}/ =~ parsed[el]
+            parts = parsed[el].split(/to/)
+            if parts[0].to_i(10) == (parts[1].to_i(10) - 1)
+              parsed[el] = parts[0] + ','
+              parsed.insert(el + 1, parts[1])
+            end
+          end
+        end
+
+        if parsed.length > 1
+          parsed.insert(-2, 'and')
+          parsed[0].gsub!(',', '') if parsed.length == 3
+        end
+
+        tex = "#{op} #{parsed.join(' ')}#{cl}"
+
+        if (parsed.length > 2) || (/\p{N}to\p{N}/ =~ tex)
+          tex.gsub!('Claim', 'Claims')
+          tex.gsub!('Citation', 'Citations')
+          tex.gsub!('Embodiment', 'Embodiments')
+          tex.gsub!('Reason', 'Reasons')
+          tex.gsub!('invention', 'inventions')
+          tex.gsub!('Prior Application', 'Prior Applications')
+        end
+        tex.gsub!('to', ' to ')
+
+        # remove extra spaces
+        tex.gsub!(/\p{Z}+/, ' ')
+      end
+
+      # dont feel like trackign this bug down, cludge
+      tex.gsub!('( ', ' (')
+
+      tex
+    end
+
+    # the @props hash is passed to docx_templater gem
+    def set_prop(prop, value)
+      @props[prop] = value
+    end
+
+    def init_instance_vars
+      @props = {}
+      @scrapes = {}
+      @props[:citaton_list] = ''
+      capture_the(:mailing_no, /発送番号\p{Z}+(\S+)/)
+      capture_the(:ref_no, /整理番号\p{Z}+(\S+)/)
+      capture_the(:ipc_list, /調査した分野$/)
+      set_prop(:ipc_reference_text, '')
+    end
+
+    def read_oa_data
+      # read in OA data
+      begin
+        @data = File.read(@sourcefile)
+      rescue
+        raise 'oa_templater_exception'
+      end
+
+      begin
+        # convert detected encoding (usually SHIFT_JIS Japanese) to UTF-8
+        detection = CharlockHolmes::EncodingDetector.detect(@data)
+        @data = CharlockHolmes::Converter.convert @data, detection[:encoding], 'UTF-8'
+      rescue
+        raise 'oa_templater_exception'
+      end
+    end
+
+    def capture_the(prop, reg, offset = 0)
+      matches = @data.match(reg, offset)
+      @scrapes[prop] = matches ? matches : nil
+      @props[prop] = matches ? matches[1] : ''
+    end
+
+    def format_date(format, date)
+      #date is MatchData object with three captures, the first being Heisei year
+      #convert from 全角文字 to normal ascii 
+      return '' if date.nil?
+      y = (NKF.nkf('-m0Z1 -w', date[1]).to_i + 1988).to_s
+      m = (NKF.nkf('-m0Z1 -w', date[2]).to_i).to_s
+      d = (NKF.nkf('-m0Z1 -w', date[3]).to_i).to_s
+      sprintf(format, y, m, d)
+    end
+
+    def pick_template
+      case @data
+      when /審判請求の番.*不服.*特許出願の番号.*特願.*起案日.*審判長.*代理人弁理士/m
+        @template = @templates[:shinpankyozetsuriyu]
+        @template_name = '審判拒絶理由'
+      when /<TITLE>拒絶理由通知書<\/TITLE>/i
+        @template = @templates[:kyozetsuriyu]
+        @template_name = '拒絶理由'
+      when /<TITLE>補正の却下の決定<\/TITLE>/i
+        @template = @templates[:rejectamendments]
+        @template_name = '補正の却下の決定'
+      when /<TITLE>拒絶査定<\/TITLE>/i
+        @template = @templates[:kyozetsusatei]
+        @template_name = '拒絶査定'
+      when /<TITLE>審尋（審判官）<\/TITLE>/i
+        @template = @templates[:shinnen]
+        @template_name = '審尋'
+      when /<TITLE>同一出願人による同日出願通知書<\/TITLE>/i
+        @template = @templates[:shireisho]
+        @template_name = '指令書'
+      else
+        # not satei or riyu, default to riyu
+        @template = @templates[:kyozetsuriyu]
+        @template_name = '拒絶理由'
+      end
+    end
+  end
+end