libis-format 1.2.9 → 2.0.0

data/lib/libis/format/converter/image_splitter.rb

@@ -0,0 +1,80 @@
+# encoding: utf-8
+
+require_relative 'base'
+require 'libis/format/identifier'
+
+require 'mini_magick'
+
+MiniMagick.logger.level = ::Logger::UNKNOWN
+
+MiniMagick.configure do |config|
+  # config.cli = :graphicsmagick
+  config.validate_on_create = false
+  config.validate_on_write = false
+  config.whiny = false
+end
+
+module Libis
+  module Format
+    module Converter
+
+      class ImageSplitter < Libis::Format::Converter::Base
+
+        def self.input_types
+          [:PDF, :TIFF, :GIF, :PBM, :PGM, :PPM]
+        end
+
+        def self.output_types(format = nil)
+          return [] unless input_types.include?(format) if format
+          [:TIFF, :JPG, :PNG, :BMP, :GIF, :PDF, :JP2]
+        end
+
+        def self.multipage?(format)
+          [:PDF, :TIFF, :GIF, :PBM, :PGM, :PPM].include?(format)
+        end
+
+        def image_split(_)
+          #force usage of this converter
+        end
+
+        def quiet(v)
+          @quiet = !!v
+        end
+
+        def convert(source, target, format, opts = {})
+          super
+
+          FileUtils.mkpath(File.dirname(target))
+
+          if self.class.multipage?(format)
+            target = File.join(File.dirname(target), "#{File.basename(target, '.*')}-%d#{File.extname(target)}")
+          end
+
+          result = split_image(source, target, format)
+          return nil unless result
+          target
+
+        end
+
+        private
+
+        def split_image(source, target, format)
+
+          MiniMagick::Tool::Convert.new do |convert|
+            convert.quiet if @quiet
+            convert << source
+            convert.format(format)
+            convert << target
+
+            debug "ImageMagick command: '#{convert.command.join(' ')}'"
+          end
+
+          target
+
+        end
+
+      end
+
+    end
+  end
+end

data/lib/libis/format/converter/image_watermarker.rb

@@ -0,0 +1,261 @@
+# encoding: utf-8
+
+require_relative 'base'
+require 'libis/format/identifier'
+
+require 'mini_magick'
+# noinspection RubyResolve
+require 'fileutils'
+
+MiniMagick.logger.level = ::Logger::UNKNOWN
+
+MiniMagick.configure do |config|
+  # config.cli = :graphicsmagick
+  config.validate_on_create = false
+  config.validate_on_write = false
+  config.whiny = false
+end
+
+module Libis
+  module Format
+    module Converter
+
+      # noinspection RubyTooManyInstanceVariablesInspection
+      class ImageWatermarker < Libis::Format::Converter::Base
+
+        def self.input_types
+          [:TIFF, :JPG, :PNG, :BMP, :GIF, :PDF, :JP2]
+        end
+
+        def self.output_types(format = nil)
+          return [] unless input_types.include?(format) if format
+          [format]
+        end
+
+        def self.multipage?(format)
+          [:PDF, :TIFF, :GIF, :PBM, :PGM, :PPM].include?(format)
+        end
+
+        def initialize
+          super
+          @quiet = true
+          @wm_image = nil
+          @wm_file = nil
+          @wm_text = '© LIBIS'
+          @wm_tiles = 4
+          @wm_resize = nil
+          @wm_gap = 20
+          @wm_gravity = 'Center'
+          @wm_rotation = 30
+          @wm_composition ='modulate'
+          @wm_composition_args = '10'
+        end
+
+        def image_watermark(_)
+          #force usage of this converter
+          self
+        end
+
+        def quiet(v)
+          @quiet = !!v
+          self
+        end
+
+        def page(v)
+          @page = v.to_i
+          self
+        end
+
+        # watermark image to use
+        def file(v)
+          @wm_file = v.blank? ? nil : v
+          self
+        end
+
+        # text to create a watermark from
+        def text(v)
+          @wm_text = v.blank? ? nil : v
+          self
+        end
+
+        # rotation of the watermark text (counter clockwise in degrees; integer number)
+        # default 30
+        def rotation(v)
+          @wm_rotation = v.to_i
+          self
+        end
+
+        # number of tiles of the watermark
+        # default 4
+        # 0: no tiling, so only 1 watermark will be placed with the original size
+        # 1: 1 tile, so the watermark will be scaled up to fill the image
+        # n > 1: minimum n tiles in both directions
+        # n < 0: tile without scaling the watermark
+        def tiles(v)
+          @wm_tiles = v.to_i
+          self
+        end
+
+        # fraction 0.0 - 1.0
+        def resize(v)
+          @wm_resize = (v.to_f * 100).to_i
+          self
+        end
+
+        # size of the gap between watermark instances. Fractions as percentage of widht/height
+        # default 0.2
+        def gap(v)
+          @wm_gap = (v.to_f * 100).to_i
+          self
+        end
+
+        # center point for the watermark overlay
+        # default 'center'
+        def gravity(v)
+          @wm_gravity = v.blank? ? nil : v
+          self
+        end
+
+        # the image composition method for merging the watermark image
+        # default 'modulate'
+        # See https://imagemagick.org/script/compose.php for more information
+        def composition(v)
+          @wm_composition = v.blank? ? nil : v
+          self
+        end
+
+        # arguments for the composition method
+        # default '10'
+        # See https://imagemagick.org/script/compose.php for more information
+        def composition_args(v)
+          @wm_composition_args = v.blank? ? nil : v
+          self
+        end
+
+
+        def convert(source, target, format, opts = {})
+          super
+
+          FileUtils.mkpath(File.dirname(target))
+
+          if source.is_a?(Array) || File.directory?(source)
+            error 'Only a single image file is allowed for input'
+          else
+            image = MiniMagick::Image.open(source) { |b| b.quiet }
+
+            if image.pages.size > 1
+              if @page
+                convert_image(image.pages[@page].path, target, format)
+              else
+                error 'multipage input file detecte; you need to supply a page number'
+              end
+            else
+              convert_image(source, target, format)
+            end
+          end
+
+          target
+
+        end
+
+        protected
+
+        # noinspection DuplicatedCode
+        def convert_image(source, target, format)
+
+          wm_image = watermark_image
+          return nil unless wm_image
+          image_info = MiniMagick::Image::Info.new(source) {|b| b.quiet}
+
+          MiniMagick::Tool::Convert.new do |convert|
+            convert.quiet if @quiet
+
+            # adapt watermark image to tile size and apply gap and resize if necessary
+            convert << @wm_image.path
+            # noinspection RubyResolve
+            convert.bordercolor('transparent').border("#{@wm_gap}%") if @wm_gap > 0
+            convert.filter('Lagrange')
+            convert.resize("#{image_info['width'] / @wm_tiles}x#{image_info['height'] / @wm_tiles}") if @wm_tiles > 0
+            convert.resize("#{@wm_resize}%") if @wm_resize
+            convert.write('mpr:watermark').delete.+
+
+            # convert the source image
+            convert << source
+            if @wm_tiles >= 0 and @wm_tiles <= 1
+              # only 1 watermark required (tiles = 0/1 => scaled no/yes)
+              convert << 'mpr:watermark'
+            else
+              # fill the image size with a pattern of the watermark image
+              convert.stack do |stack|
+                stack.size("#{image_info['width']}x#{image_info['height']}")
+                stack << 'xc:transparent'
+                # noinspection RubyResolve
+                stack.tile('mpr:watermark')
+                # noinspection RubyResolve
+                stack.draw "rectangle 0,0,#{image_info['width']},#{image_info['height']}"
+              end
+            end
+            # perform the blending operation
+            convert.compose(@wm_composition).gravity(@wm_gravity).define("compose:args=#{@wm_composition_args}").composite
+
+            convert.format(format)
+            convert << target
+
+            debug "ImageMagick command: '#{convert.command.join(' ')}'"
+          end
+
+          target
+
+        end
+
+        private
+
+        # Create or use a watermark image.
+        #
+        # If both text and image are set, the file will be used as-is if it exists and is a valid image file. Otherwise the
+        # file will be created or overwritten with a newly created watermark image.
+        #
+        # The created watermark file will be a 2000x2000 pixels PNG image with transparent background
+        #
+        # The text will be slanted by given rotation degrees counter-clockwise
+        def watermark_image
+          rotation = 360 - @wm_rotation
+          @wm_image = MiniMagick::Image.new(@wm_file) if @wm_file
+          # only create image if file is not an image
+          unless @wm_image&.valid?
+            # noinspection RubyResolve
+            # Create image file (as given or temp file)
+            image = @wm_file || (Dir::Tmpname.create(%w(wm_image .png)) {|_|})
+            # noinspection RubyResolve
+            MiniMagick::Tool::Convert.new do |convert|
+              convert.quiet # allways quiet
+              convert.background 'transparent'
+              convert.size('2000x2000')
+              convert.gravity 'Center'
+              convert.font('Helvetica').fill('black').pointsize(72) #.stroke('black').strokewidth(1)
+              convert << "label:#{@wm_text}"
+              convert.rotate rotation
+              convert.trim.repage.+
+              convert << image
+            end
+            if @wm_file
+              @wm_image = MiniMagick::Image.new(image)
+            else # delete temp file
+              @wm_image = MiniMagick::Image.open(image)
+              File.delete(image)
+            end
+            # noinspection RubyResolve
+            unless @wm_image.valid?
+              error "Problem creating watermark image '#{image}'."
+              @wm_image = nil
+            end
+            @wm_image
+          end
+        end
+
+
+      end
+
+    end
+  end
+end

data/lib/libis/format/converter/jp2_converter.rb

@@ -15,7 +15,7 @@ module Libis
         end
 
         def self.output_types(format = nil)
-          return [] unless input_types.include?(format)
+          return [] unless input_types.include?(format) if format
           [:JP2]
         end
 
@@ -90,16 +90,16 @@ module Libis
           end
 
 
-          result = Libis::Tools::Command.run(
+          Libis::Tools::Command.run(
               Libis::Format::Config[:j2k_cmd],
               '--input-file-name', source,
               '--set-output-type', 'JP2',
               *options,
               '--output-file-name', target,
-          )
 
-          result.merge(files: [target], converter: self.class.name)
+          )
 
+          target
         end
       end
     end

data/lib/libis/format/converter/office_converter.rb

@@ -3,7 +3,7 @@
 require_relative 'base'
 
 require 'libis/format/tool/office_to_pdf'
-require 'libis/format/type_database'
+require 'libis/format/library'
 
 module Libis
   module Format
@@ -32,7 +32,7 @@ module Libis
         end
 
         def self.output_types(format = nil)
-          return [] unless input_types.include?(format)
+          return [] unless input_types.include?(format) if format
           [:PDF]
         end
 
@@ -43,7 +43,9 @@ module Libis
         def convert(source, target, format, opts = {})
           super
 
-          Format::Tool::OfficeToPdf.run(source, target)
+          return nil unless Format::Tool::OfficeToPdf.run(source, target)
+
+          target
 
         end
 

data/lib/libis/format/converter/pdf_assembler.rb

@@ -0,0 +1,66 @@
+# encoding: utf-8
+
+require_relative 'base'
+
+require 'libis/format/tool/pdf_merge'
+
+module Libis
+  module Format
+    module Converter
+
+      # noinspection DuplicatedCode
+      class PdfAssembler < Libis::Format::Converter::Base
+
+        def self.input_types
+          [:PDF]
+        end
+
+        def self.output_types(format = nil)
+          return [] unless input_types.include?(format) if format
+          [:PDF]
+        end
+
+        def self.category
+          :assembler
+        end
+
+        def pdf_assemnble(_)
+          #force usage of this converter
+        end
+
+        def convert(source, target, format, opts = {})
+          super
+
+          result = if source.is_a? Array
+                     assemble(source, target)
+                   elsif File.directory?(source)
+                     source_list = Dir[File.join(source, '**', '*')].reject {|p| File.directory? p}
+                     assemble(source_list, target)
+                   else
+                     assemble([source], target)
+                   end
+          return nil unless result
+
+          result
+        end
+
+        private
+
+        def assemble(source, target)
+
+          result = Libis::Format::Tool::PdfMerge.run(source, target)
+
+          unless result[:err].empty?
+            error("PdfMerge encountered errors:\n%s", result[:err].join(join("\n")))
+            return nil
+          end
+
+          target
+        end
+
+      end
+
+    end
+
+  end
+end

data/lib/libis/format/converter/pdf_converter.rb

@@ -5,13 +5,13 @@ require_relative 'base'
 require 'libis/tools/extend/hash'
 require 'libis/format/tool/pdf_copy'
 require 'libis/format/tool/pdf_to_pdfa'
-require 'libis/format/tool/pdfa_validator'
 require 'libis/format/tool/pdf_optimizer'
 
 module Libis
   module Format
     module Converter
 
+      # noinspection DuplicatedCode
       class PdfConverter < Libis::Format::Converter::Base
 
         def self.input_types
@@ -19,84 +19,34 @@ module Libis
         end
 
         def self.output_types(format = nil)
-          return [] unless input_types.include?(format)
+          return [] unless input_types.include?(format) if format
           [:PDF, :PDFA]
         end
 
-        def pdf_convert(_)
-          #force usage of this converter
+        def title(v)
+          @options[:title] = v.blank? ? nil : v
         end
 
-        # Set metadata for Pdf file
-        #
-        # valid metadata keys are):
-        #       - title
-        #       - author
-        #       - creator
-        #       - keywords
-        #       - subject
-        #
-        # @param [Hash] values list of metadata values to set
-        def metadata(values = {})
-          values.key_strings_to_symbols!
-          values.each do |k, v|
-            next unless [:title, :author, :creator, :keywords, :subject].include?(k)
-            @options["md_#{k}"] = v
-          end
+        def author(v)
+          @options[:author] = v.blank? ? nil : v
         end
 
-        # Select a partial list of pages
-        # @param [String] selection as described in com.itextpdf.text.pdf.SequenceList: [!][o][odd][e][even]start-end
-        def range(selection)
-          @options[:ranges] = selection
+        def creator(v)
+          @options[:creator] = v.blank? ? nil : v
         end
 
-        # Create or use a watermark image.
-        #
-        # The watermark options are (use symbols):
-        #     - text: text to create a watermark from
-        #     - file: watermark image to use
-        #     - rotation: rotation of the watermark text (in degrees; integer number)
-        #     - size: font size of the watermark text
-        #     - opacity: opacity of the watermark (fraction 0.0 - 1.0)
-        #     - gap: size of the gap between watermark instances. Integer value is absolute size in points (1/72 inch). Fractions are percentage of widht/height.
-        # If both options are given, the file will be used as-is if it exists and is a valid image file. Otherwise the
-        # file will be created or overwritten with a newly created watermark image.
-        #
-        # The created watermark file will be a PNG image with transparent background containing the supplied text
-        # slanted by 30 degrees counter-clockwise.
-        #
-        # @param [Hash] options Hash of options for watermark creation.
-        def watermark(options = {})
-          options.key_strings_to_symbols!
-          if options[:file] && File.exist?(options[:file])
-            @options['wm_image'] = options[:file]
-          else
-            @options['wm_text'] = (options[:text] || '© LIBIS').split('\n')
-            @options['wm_text_rotation'] = options[:rotation] if options[:rotation]
-            @options['wm_font_size'] = options[:size] if options[:size]
-          end
-          @options['wm_opacity'] = options[:opacity] || '0.3'
-          @options['wm_gap_ratio'] = options[:gap] if options[:gap].to_s =~ /^\s*(0+\.\d+|1\.0+)\s*$/
-          @options['wm_gap_size'] = options[:gap] if options[:gap].to_s =~ /^\s*\d+\s*$/
+        def keywords(v)
+          @options[:keywords] = v.blank? ? nil : v
         end
 
-        # Optimize the PDF
-        #
-        # This reduces the graphics quality to a level in order to limit file size. This option relies on the
-        # presence of ghostscript and takes one argument: the quality level. It should be one of:
-        #
-        # - 0 : lowest quality (Acrobat Distiller 'Screen Optimized' equivalent)
-        # - 1 : medium quality (Acrobat Distiller 'eBook' equivalent)
-        # - 2 : good quality
-        # - 3 : high quality (Acrobat Distiller 'Print Optimized' equivalent)
-        # - 4 : highest quality (Acrobat Distiller 'Prepress Optimized' equivalent)
-        #
-        # Note that the optimization is intended to be used with PDF's containing high-resolution images.
-        #
-        # @param [Integer] setting quality setting. [0-4]
-        def optimize(setting = 1)
-          @options['optimize'] = %w(screen ebook default printer prepress)[setting] if (0..4) === setting
+        def subject(v)
+          @options[:subject] = v.blank? ? nil : v
+        end
+
+        # Select a partial list of pages
+        # @param [String] selection as described in com.itextpdf.text.pdf.SequenceList: [!][o][odd][e][even]start-end
+        def range(selection)
+          @options[:ranges] = selection.blank? ? nil : selection
         end
 
         def convert(source, target, format, opts = {})
@@ -104,12 +54,6 @@ module Libis
 
           result = nil
 
-          if (quality = @options.delete('optimize'))
-            result = optimize_pdf(source, target, quality)
-            return nil unless result
-            source = result
-          end
-
           unless @options.empty?
             result = convert_pdf(source, target)
             return nil unless result
@@ -120,24 +64,17 @@ module Libis
             result = pdf_to_pdfa(source, target)
           end
 
-          { 
-            files: [result],
-            converter: self.class.name
-          }
+          result
 
         end
 
-        def optimize_pdf(source, target, quality)
-
-          using_temp(target) do |tmpname|
-            result = Libis::Format::Tool::PdfOptimizer.run(source, tmpname, quality)
-            unless result[:status] == 0
-              error("Pdf optimization encountered errors:\n%s", (result[:err] + result[:out]).join("\n"))
-              next nil
-            end
-            tmpname
-          end
-        end
+        OPTIONS_TABLE = {
+            title: 'md_title',
+            author: 'md_author',
+            creator: 'md_creator',
+            keywords: 'md_keywords',
+            subject: 'md_subject'
+        }
 
         def convert_pdf(source, target)
 
@@ -148,8 +85,9 @@ module Libis
                   if v.nil?
                     nil
                   else
+                    k = OPTIONS_TABLE[k] || k
                     ["--#{k}", (v.is_a?(Array) ? v : v.to_s)]
-                  end}.flatten
+                  end}.compact.flatten
             )
             unless result[:err].empty?
               error("Pdf conversion encountered errors:\n%s", result[:err].join(join("\n")))
@@ -164,16 +102,11 @@ module Libis
 
           using_temp(target) do |tmpname|
             result = Libis::Format::Tool::PdfToPdfa.run source, tmpname
-
-            if result[:command][:status] != 0
-              error("Pdf/A conversion encountered errors:\n%s", (result[:command][:out] + result[:command][:err]).join("\n"))
+            if result[:status] != 0
+              error("Pdf/A conversion encountered errors:\n%s", result[:err].join("\n"))
               next nil
             else
-              r = Libis::Format::Tool::PdfaValidator.run tmpname
-              if r[:status] != 0
-                error "Pdf/A file failed to validate with following errors:\n%s", (r[:err] || r[:out] || []).join("\n")
-                next nil
-              end
+              warn("Pdf/A conversion warnings:\n%s", result[:err].join("\n")) unless result[:err].empty?
             end
             tmpname
           end