add bates example Rack application

examples/bates/README.md

@@ -0,0 +1,3 @@
+# Bates numbering with CombinePDF and Rack
+
+This folder holds a demo application combining Iodine's `X-Sendfile` support with the [`combine_pdf` gem](https://github.com/boazsegev/combine_pdf) to create a "bates numbering" tool that is common to the one used by some court systems (such as Israeli courts).

examples/bates/config.ru

@@ -0,0 +1,342 @@
+require 'combine_pdf'
+require 'rack'
+require 'base64'
+# require_relative 'pdf_controller'
+
+module BatesAPP
+  # This is the HTTP response object according to the Rack specification.
+  STATIC_RESPONSE = [200, { 'Content-Type' => 'text/html', 'X-Sendfile' => File.expand_path('./public/index.html') }.freeze, []].freeze
+  WS_RESPONSE = [403, { 'Content-Type' => 'text/html' }, "Forbidden"]
+
+   # this is function will be called by the Rack server (iodine) for every request.
+   def self.call env
+    req = Rack::Request.new(env)
+    # check if this is an upgrade request (WebsSocket / SSE).
+    return WS_RESPONSE if(env['rack.upgrade?'.freeze])
+    # simply return the RESPONSE object, no matter what request was received.
+    return STATIC_RESPONSE if req.params.empty?
+    return bates(req.params)
+   end
+
+
+  def self.bates(params)
+    params['file'] = params['file'].values # convert to Array
+    # catch any exception and tell the user which PDF caused the exception.
+    begin
+      # set the file_name variable
+      # this will be used in case an exception is caught
+      # to state which file caused the exception.
+      file_name = ''
+
+      # container for the complete pdf
+      # the complete pdf container will hold all
+      # the merged pdf data.
+      completed_pdf = CombinePDF.new
+
+      # get the output file name
+      # this will be used to name the download for the client's browser
+      output_name = params['bates']['output'].to_s + '.pdf'
+      output_name = "unknown.pdf" if output_name == '.pdf'
+
+      # get some paramaters that will be used while combining pages
+      params['bates'] ||= {}
+      first_page_number = params['bates']['first_page_number'].to_i
+      first_index_number = params['bates']['first_index_number'].to_i
+
+      # we will add an option for the stamping to ignore the first pdf
+      # this is useful for the court cases that use bates numbering
+      # (the "cover PDF" will be the briefs of submissions that contain exhibits to be bates stamped)
+      ignore_first_file = params['bates']['first_pdf_is_cover']
+      first_page = nil
+
+      # we will pick some data up while running combining the different pdf files.
+      # this will be used for the table of contents later on.
+      pdfs_pages_count = []
+      pdf_dates = []
+      pdf_titles = []
+
+      # we will be creating title pages before each PDF file.
+      # the title pages will be sized using the mediabox variable
+      # wich will be set with the dimentions of each pdf file's first page.
+      mediabox = nil
+
+      # register UNICODE fonts if necessary
+      # in this example I will register the Hebrew font David from an existing PDF file.
+      unless CombinePDF::Fonts.get_font :my_new_david
+        # if your running Rails, consider Rails.root instead of Root
+        fonts = CombinePDF.new(File.expand_path('./david+bold.pdf').to_s).fonts(true)
+        # I know the first font is David regular, after using the
+        # ruby console and looking at the fonts array for the file.
+        CombinePDF.register_font_from_pdf_object :my_new_david, fonts[0]
+        # the second font of the array was the latin font for a newline and space... useless
+        # the third was the david bold. I will now add that font.
+        CombinePDF.register_font_from_pdf_object :my_new_david_bold, fonts[2]
+      end
+
+      # iterate through the different files sent from the client's browser's web form
+      params['file'].each do |v|
+        # set the file_name variable in case an exception will be raised
+        file_name = v['name']
+
+        # parse the pdf data
+        # we will use the CombinePDF.parse method which allows us
+        # to parse data without saving the PDF to the file system.
+        # our javascript encoded the file data using base64, which we will need to decode.
+        # (this is specific to my form which uses the HTML5 File API in this specific manner)
+        # begin
+        puts "parsing file: #{file_name}"
+        pdf_file = CombinePDF.parse(
+          Base64.urlsafe_decode64(
+            v['data'].slice('data:application/pdf;base64,'.length,
+                           v['data'].length)
+          )
+        )
+        # rescue
+        #   puts "Failed when parsing #{file_name} with data:\n#{v['data'].inspect}"
+        #   raise
+        # end
+
+        # we will use the pages array a few times, so in order to avoid
+        # recomputing the array every time, we will save it to a local variable.
+        pdf_file_pages = pdf_file.pages
+
+        # we will add the page count to the page count array,
+        # used by the table of contents
+        pdfs_pages_count << pdf_file_pages.length
+
+        ######
+        # create and add title page to arrays.
+
+        if ignore_first_file && first_page.nil?
+          # if the first PDF file is a "cover page" PDF,
+          # we will not add a title, nor add the file.
+          # instead we will save the data in a variable to add it after we're done
+          pdf_dates << ''
+          pdf_titles << ''
+          first_page = pdf_file
+        else
+          # set title page mediabox size
+          # the mediabox data (page size) is contained in the page's
+          # :CropBox or :MediaBox keys (pages are Hash objects).
+          mediabox ||= pdf_file_pages[0][:CropBox] || pdf_file_pages[0][:MediaBox]
+
+          # create a title page unless we're only merging or there is no indexing
+          if params['bates']['should_index'] && params['bates']['numbering'] != 3
+
+            # create an empty page object
+            title_page = CombinePDF.create_page mediabox
+
+            # write the content to the title page.
+            # we will be using the I18n.t shortcut to write some of the data.
+            # the rest of the data, like the title, we got from the form.
+            title_page.textbox("#{params['bates']['title_type']} #{pdfs_pages_count.length + first_index_number - (ignore_first_file ? 2 : 1)}",
+                               max_font_size: 34,
+                               font: :my_new_david,
+                               y: (mediabox[3] - mediabox[1]) / 2) unless params['bates']['title_type'].to_s.empty?
+            title_page.textbox v['title'].to_s, max_font_size: 36, font: :my_new_david_bold
+            title_page.textbox v['date'].to_s, max_font_size: 24, font: :my_new_david, height: (mediabox[3] - mediabox[1]) / 2
+
+            # we will add the page object to the completed pdf object.
+            # notice that page objects are created as "floating" pages,
+            # not attached to any specific PDF file/object.
+            completed_pdf << title_page
+
+            # we will add some data that will be used to create the
+            # table of contents at a later stage.
+            page_count = pdfs_pages_count.pop + 1
+            pdfs_pages_count << page_count
+            pdf_dates << v['date'].to_s
+            pdf_titles << v['title'].to_s
+          end
+
+          # now we are ready to add the pdf file data to the completed pdf object.
+          # there is no need to add each page, we can add the pdf as a whole.
+          # (it's actually faster, as the PDF page catalog isn't recomputed for each page)
+          completed_pdf << pdf_file
+        end
+      end
+
+      ##########
+      # create the index pdf...
+      # ...unless we're only merging or there is no indexing
+      if params['bates']['should_index'] && params['bates']['numbering'].to_i != 3
+
+        # set the fonts and formatting for the table of contents.
+        #
+        # also, add an empty array for the table data.
+        #
+        # the table data array will contain arrays of String objects, each one
+        # corresponding to a row in the table.
+        table_options = { font: :my_new_david,
+                          header_font: :my_new_david_bold,
+                          max_font_size: 12,
+                          column_widths:  (params['bates']['date_header'].to_s.empty? ? [3, 40, 4] : [3, 10, 30, 4]),
+                          table_data: [] }
+
+        # set the table header array.
+        # this is an array of strings.
+        # to chose the localized strings
+        table_options[:headers] = [params['bates']['number_header'],
+                                   (params['bates']['date_header'].to_s.empty? ? nil : params['bates']['date_header']),
+                                   params['bates']['title_header'].to_s,
+                                   params['bates']['page_header'].to_s]
+        table_options[:headers].compact!
+
+        # by default, there are 25 rows per page for table pdf created by CombinePDF
+        # we can override this in the formatting (but we didn't).
+        #
+        # the 25 rows include 1 header row per page - so there are only 24 effective rows.
+        #
+        # we will calculate how many pages the table of contents pdf will have once completes,
+        # so we can add the count to the page numbers in the index.
+        index_page_length = pdfs_pages_count.length / 24
+        index_page_length += 1 if pdfs_pages_count.length % 24 > 0
+
+        # set the page number for the first entry in the table of contents.
+        page_number = first_page_number + index_page_length
+
+        # set the index count to 0, we will use it to change the index for each entry.
+        # we need a different variable in case the first PDF file is a "cover page".
+        index_count = 0
+
+        # iterate over the data we collected before and add it to the table data.
+        pdfs_pages_count.each_index do |i|
+          # add the data unless it is set to be ignored
+          unless ignore_first_file
+
+            # add an array of strings to the :table_data array,
+            # representing a row in our table.
+            # remember there might not be a date column.
+            if params['bates']['date_header'].to_s.empty?
+              table_options[:table_data] << [(first_index_number + index_count).to_s,
+                                             pdf_titles[i], page_number]
+            else
+              table_options[:table_data] << [(first_index_number + index_count).to_s,
+                                             pdf_dates[i],
+                                             pdf_titles[i], page_number]
+            end
+
+            # if the data was added to the index table, bump the index count
+            index_count += 1
+
+          end
+
+          # make sure future data will not be ignored
+          ignore_first_file = false
+
+          # add the page count to the page number, so that the next
+          # index's page number is up to date.
+          page_number += pdfs_pages_count[i]
+        end
+
+        # if out current locale is hebrew, which is a right to left language,
+        # set the direction for the table to Right-To-left (:rtl).
+        #
+        # notice that RTL text should be automatically recognized, but that
+        # feature isn't available (and shouldn't be available) for tables.
+        table_options[:direction] = :rtl if params['bates']['dir'] == 'rtl'
+
+        # if there is table data, we will create an index pdf.
+        unless table_options[:table_data].empty?
+
+          # create the index PDF from the table data and options we have.
+          index_pdf = CombinePDF.create_table table_options
+
+          # We will now add the words "Table of Contents" (or the I18n equivilant)
+          # to the first page of our new index_pdf PDF object.
+          #
+          # the table PDF object was created by CombinePDF using writable PDF pages,
+          # so we have properties like .mediabox and methods like .textbox
+          # at our disposal.
+
+          # get the first page of the index_pdf object, we will use this reference a lot.
+          title_page = index_pdf.pages[0]
+
+          # write the textbox, using the mediabox page data [x,y,width,height] to place
+          # the text we want to write.
+          #
+          # we will use the I18n.t shortcut to choose the text to write down.
+          title_page.textbox params['bates']['index_title'],
+                             y: ((title_page.mediabox[3] - title_page.mediabox[1]) * 0.91),
+                             height: ((title_page.mediabox[3] - title_page.mediabox[1]) * 0.03),
+                             font: :my_new_david,
+                             max_font_size: 36,
+                             text_valign: :bottom
+
+          # now we will add the index_pdf to the BEGINING of the completed pdf.
+          # for this we will use the >> operator instead of the << operator.
+          completed_pdf >> index_pdf
+
+        end
+
+      end
+
+      # add first file if it was skipped
+      completed_pdf >> first_page unless first_page.nil?
+
+      #####
+      # number pages
+      # unless no numbering
+      if params['bates']['numbering'].to_i != 3
+        # list the numbering options
+        numbering_options = [[:top, :bottom], [:top_left, :bottom_left], [:top_right, :bottom_right]]
+
+        # set the first visible page number to the page where numbering starts
+        # this assumes that the bates numbering include the numbering of the "cover page",
+        # yet at the same time the numbering isn't visible on the "cover page"
+        first_page_number += pdfs_pages_count[0] if params['bates']['first_pdf_is_cover'] && params['bates']['skip_cover']
+
+        # call the page numbering method and
+        # add the special properties we want for the textbox
+        completed_pdf.number_pages(start_at: first_page_number,
+                                   page_range: params['bates']['skip_cover'] ? (pdfs_pages_count[0].to_i..-1) : nil,
+                                   font_name: :my_new_david,
+                                   font_size: 14,
+                                   font_color: [0, 0, 0.4],
+                                   box_color: [0.8, 0.8, 0.8],
+                                   border_width: 1,
+                                   border_color: [0.3, 0.3, 0.3],
+                                   box_radius: 8,
+                                   number_location: numbering_options[params['bates']['numbering'].to_i],
+                                   opacity: 0.75)
+      end
+
+      # send the completed PDF to the client.
+      # if the completed PDF is empty, raise an error.
+      if completed_pdf.pages.empty?
+        # inform the client there was an unknown error.
+        response = STATIC_RESPONSE.dup
+        response[1] = response[1].dup
+        response[1]['set-cookie'] = "Unknown error - 0 pages."
+        return response
+      end
+
+      # make sure the PDF version is high enough for the opacity we used in the page numbering.
+      completed_pdf.version = [completed_pdf.version, 1.6].max
+
+      # we will format the PDF to a pdf file WITHOUT saving it to the file system,
+      # using the .to_pdf method (instead of the .save method).
+      # we will send the raw PDF data stream.
+      return [200, {'content-type' => "application/pdf",
+                    'content-disposition' => "attachment; filename=\"#{output_name}\""}, completed_pdf.to_pdf]
+
+    rescue Exception => e
+      puts e.message, e.backtrace
+      puts "The file causing the exception is: #{file_name}"
+      # if an exception was raised, tell the user which PDF caused the exception
+      response = STATIC_RESPONSE.dup
+      response[1] = response[1].dup
+      response[1]['set-cookie'] = "notice=Unsupported or error reading file: #{file_name}"
+      return response
+    end
+  end
+end
+
+Iodine.workers = 1
+Iodine.threads = 5
+Iodine::DEFAULT_SETTINGS[:public] ||= './public'
+
+p Iodine::DEFAULT_SETTINGS
+Iodine::DEFAULT_SETTINGS[:max_body] = 70 if Iodine::DEFAULT_SETTINGS[:max_body].to_i < 5
+
+run BatesAPP