Error: Spaces make a new line

lib/tabula_java_wrapper.rb

@@ -67,12 +67,13 @@ module Extraction
 
     def Extraction.openPDF(pdf_filename, password='')
       raise Errno::ENOENT unless File.exists?(pdf_filename)
-      document = PDDocument.load(pdf_filename)
+      document = PDDocument.load(java.io.File.new(pdf_filename))
       #document = PDDocument.loadNonSeq(java.io.File.new(pdf_filename), nil, password)
       document
     end
 
     class ObjectExtractor < Java::TechnologyTabula.ObjectExtractor
+      field_accessor :pdfDocument
 
       alias_method :close!, :close
 
@@ -85,6 +86,10 @@ def initialize(pdf_filename, pages=[1], password='', options={})
 
         super(document)
       end
+
+      def page_count
+        self.pdfDocument.get_number_of_pages
+      end
     end
 
     class PagesInfoExtractor < ObjectExtractor

lib/tabula_job_executor/jobs/generate_thumbnails.rb

@@ -12,7 +12,7 @@ def perform
     output_dir = options[:output_dir]
     thumbnail_sizes = options[:thumbnail_sizes]
 
-    generator = JPedalThumbnailGenerator.new(filepath, output_dir, thumbnail_sizes)
+    generator = PDFBox2ThumbnailGenerator.new(filepath, output_dir, thumbnail_sizes)
     generator.add_observer(self, :at)
     generator.generate_thumbnails!
 

lib/thumbnail_generator.rb

@@ -1,18 +1,18 @@
 require 'java'
 require 'observer'
 
-java.lang.System.setProperty('org.jpedal.jai', 'true')
-require_relative './jars/jpedal_lgpl.jar'
-
 java_import javax.imageio.ImageIO
 java_import java.awt.image.BufferedImage
 java_import java.awt.Image
 
-java_import org.jpedal.PdfDecoder
-java_import org.jpedal.fonts.FontMappings
+java_import org.apache.pdfbox.rendering.PDFRenderer
+java_import org.apache.pdfbox.pdmodel.PDDocument
+java_import java.io.ByteArrayOutputStream
+
 
 class AbstractThumbnailGenerator
   include Observable
+  SIZE = 800
 
   def initialize(pdf_filename, output_directory, sizes=[2048, 560])
     raise Errno::ENOENT unless File.directory?(output_directory)
@@ -48,41 +48,37 @@ def generate_thumbnails!
   end
 end
 
-class JPedalThumbnailGenerator < AbstractThumbnailGenerator
+class PDFBox2ThumbnailGenerator  < AbstractThumbnailGenerator
   def initialize(pdf_filename, output_directory, sizes=[2048, 560])
     super(pdf_filename, output_directory, sizes)
-    @decoder = PdfDecoder.new(true)
-    FontMappings.setFontReplacements
-    @decoder.openPdfFile(pdf_filename)
-    @decoder.setExtractionMode(0, 1.0)
-    @decoder.useHiResScreenDisplay(true)
+    @pdf_document = PDDocument.load(java.io.File.new(pdf_filename))
   end
-
   def generate_thumbnails!
-    total_pages = @decoder.getPageCount
-
-    total_pages.times do |i|
-
-      begin
-        image = @decoder.getPageAsImage(i+1);
-        image_w, image_h = image.getWidth, image.getHeight
-
-        @sizes.each do |s|
-          scale = s.to_f / image_w.to_f
-          bi = BufferedImage.new(s, image_h * scale, image.getType)
-          bi.getGraphics.drawImage(image.getScaledInstance(s, image_h * scale, Image::SCALE_SMOOTH), 0, 0, nil)
-          ImageIO.write(bi,
-                        'png',
-                        java.io.File.new(File.join(@output_directory,
-                                                   "document_#{s}_#{i+1}.png")))
-          changed
-          notify_observers(i+1, total_pages, "generating page thumbnails...")
-        end
-      rescue java.lang.RuntimeException
-        # TODO What?
-      end
+    renderer = PDFRenderer.new(@pdf_document);
+    total_pages = @pdf_document.get_number_of_pages
+
+    total_pages.times do |pi|
+      image = renderer.render_image_with_dpi(pi, 75);
+      imageWidth = image.width # was get_width
+      imageHeight = image.height # was get_height
+      scale = SIZE / imageWidth.to_f
+
+      bi = BufferedImage.new(SIZE, (imageHeight * scale).round, image.type);
+      bi.get_graphics.draw_image(image.get_scaled_instance(SIZE, (imageHeight * scale).round, Image::SCALE_SMOOTH), 0, 0, nil);
+
+      out = ByteArrayOutputStream.new
+      ImageIO.write(bi, "png", out);
+
+      filename = "document_#{SIZE}_#{pi + 1}.png"
+      ImageIO.write(bi,
+                    'png',
+                    java.io.File.new(File.join(@output_directory,
+                                               filename)))
+      notify_observers(pi+1, total_pages, "generating page thumbnails...")
     end
-    @decoder.closePdfFile
+
+    @pdf_document.close();
+
   end
 end
 
@@ -93,9 +89,10 @@ def update(page, total_pages)
       STDERR.puts "#{page}///#{total_pages}"
     end
   end
+  require_relative '../lib/jars/tabula-1.0.0-SNAPSHOT-jar-with-dependencies.jar'
 
-  #pdftg = JPedalThumbnailGenerator.new(ARGV[0], '/tmp', [560])
-  pdftg = MUDrawThumbnailGenerator.new(ARGV[0], '/tmp', [560])
+  # pdftg = MUDrawThumbnailGenerator.new(ARGV[0], '/tmp', [560])
+  pdftg = PDFBox2ThumbnailGenerator.new(ARGV[0], '/tmp', [560])
   pdftg.add_observer(STDERRProgressReporter.new)
   pdftg.generate_thumbnails!
 end

webapp/tabula_web.rb

@@ -9,7 +9,7 @@
 require 'fileutils'
 require 'securerandom'
 
-require_relative '../lib/jars/tabula-0.9.2-jar-with-dependencies.jar'
+require_relative '../lib/jars/tabula-1.0.0-SNAPSHOT-jar-with-dependencies.jar'
 
 require_relative '../lib/tabula_java_wrapper.rb'
 java_import 'java.io.ByteArrayOutputStream'