hexapdf 0.3.0 → 0.4.0

data/data/hexapdf/cmap/UniKS-UTF16-V

@@ -0,0 +1,99 @@
+%!PS-Adobe-3.0 Resource-CMap
+%%DocumentNeededResources: ProcSet (CIDInit)
+%%DocumentNeededResources: CMap (UniKS-UTF16-H)
+%%IncludeResource: ProcSet (CIDInit)
+%%IncludeResource: CMap (UniKS-UTF16-H)
+%%BeginResource: CMap (UniKS-UTF16-V)
+%%Title: (UniKS-UTF16-V Adobe Korea1 1)
+%%Version: 1.004
+%%Copyright: -----------------------------------------------------------
+%%Copyright: Copyright 1990-2015 Adobe Systems Incorporated.
+%%Copyright: All rights reserved.
+%%Copyright:
+%%Copyright: Redistribution and use in source and binary forms, with or
+%%Copyright: without modification, are permitted provided that the
+%%Copyright: following conditions are met:
+%%Copyright:
+%%Copyright: Redistributions of source code must retain the above
+%%Copyright: copyright notice, this list of conditions and the following
+%%Copyright: disclaimer.
+%%Copyright:
+%%Copyright: Redistributions in binary form must reproduce the above
+%%Copyright: copyright notice, this list of conditions and the following
+%%Copyright: disclaimer in the documentation and/or other materials
+%%Copyright: provided with the distribution. 
+%%Copyright:
+%%Copyright: Neither the name of Adobe Systems Incorporated nor the names
+%%Copyright: of its contributors may be used to endorse or promote
+%%Copyright: products derived from this software without specific prior
+%%Copyright: written permission. 
+%%Copyright:
+%%Copyright: THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND
+%%Copyright: CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,
+%%Copyright: INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+%%Copyright: MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+%%Copyright: DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR
+%%Copyright: CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+%%Copyright: SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
+%%Copyright: NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+%%Copyright: LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+%%Copyright: HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+%%Copyright: CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
+%%Copyright: OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+%%Copyright: SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+%%Copyright: -----------------------------------------------------------
+%%EndComments
+
+/CIDInit /ProcSet findresource begin
+
+12 dict begin
+
+begincmap
+
+/UniKS-UTF16-H usecmap
+
+/CIDSystemInfo 3 dict dup begin
+  /Registry (Adobe) def
+  /Ordering (Korea1) def
+  /Supplement 1 def
+end def
+
+/CMapName /UniKS-UTF16-V def
+/CMapVersion 1.004 def
+/CMapType 1 def
+
+/XUID [1 10 25545] def
+
+/WMode 1 def
+
+11 begincidchar
+<2016> 8061
+<2025> 8058
+<3013> 8075
+<ff01> 8076
+<ff0c> 8079
+<ff0e> 8080
+<ff3b> 8087
+<ff3d> 8088
+<ff3f> 8089
+<ff5e> 8062
+<ffe3> 8093
+endcidchar
+
+7 begincidrange
+<2013> <2014> 8059
+<3001> <3002> 8056
+<3008> <3011> 8065
+<3014> <3015> 8063
+<ff08> <ff09> 8077
+<ff1a> <ff1f> 8081
+<ff5b> <ff5d> 8090
+endcidrange
+
+endcmap
+CMapName currentdict /CMap defineresource pop
+end
+end
+
+%%EndResource
+%%EOF

data/data/hexapdf/cmap/V

@@ -0,0 +1,105 @@
+%!PS-Adobe-3.0 Resource-CMap
+%%DocumentNeededResources: ProcSet (CIDInit)
+%%DocumentNeededResources: CMap (H)
+%%IncludeResource: ProcSet (CIDInit)
+%%IncludeResource: CMap (H)
+%%BeginResource: CMap (V)
+%%Title: (V Adobe Japan1 1)
+%%Version: 12.004
+%%Copyright: -----------------------------------------------------------
+%%Copyright: Copyright 1990-2015 Adobe Systems Incorporated.
+%%Copyright: All rights reserved.
+%%Copyright:
+%%Copyright: Redistribution and use in source and binary forms, with or
+%%Copyright: without modification, are permitted provided that the
+%%Copyright: following conditions are met:
+%%Copyright:
+%%Copyright: Redistributions of source code must retain the above
+%%Copyright: copyright notice, this list of conditions and the following
+%%Copyright: disclaimer.
+%%Copyright:
+%%Copyright: Redistributions in binary form must reproduce the above
+%%Copyright: copyright notice, this list of conditions and the following
+%%Copyright: disclaimer in the documentation and/or other materials
+%%Copyright: provided with the distribution. 
+%%Copyright:
+%%Copyright: Neither the name of Adobe Systems Incorporated nor the names
+%%Copyright: of its contributors may be used to endorse or promote
+%%Copyright: products derived from this software without specific prior
+%%Copyright: written permission. 
+%%Copyright:
+%%Copyright: THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND
+%%Copyright: CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,
+%%Copyright: INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+%%Copyright: MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+%%Copyright: DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR
+%%Copyright: CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+%%Copyright: SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
+%%Copyright: NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+%%Copyright: LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+%%Copyright: HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+%%Copyright: CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
+%%Copyright: OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+%%Copyright: SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+%%Copyright: -----------------------------------------------------------
+%%EndComments
+
+/CIDInit /ProcSet findresource begin
+
+12 dict begin
+
+begincmap
+
+/H usecmap
+
+/CIDSystemInfo 3 dict dup begin
+  /Registry (Adobe) def
+  /Ordering (Japan1) def
+  /Supplement 1 def
+end def
+
+/CMapName /V def
+/CMapVersion 12.004 def
+/CMapType 1 def
+
+/UIDOffset 850 def
+/XUID [1 10 25340] def
+
+/WMode 1 def
+
+27 begincidrange
+<2122> <2123> 7887
+<2131> <2132> 7889
+<213c> <213e> 7891
+<2141> <2145> 7894
+<214a> <215b> 7899
+<2161> <2161> 7917
+<2421> <2421> 7918
+<2423> <2423> 7919
+<2425> <2425> 7920
+<2427> <2427> 7921
+<2429> <2429> 7922
+<2443> <2443> 7923
+<2463> <2463> 7924
+<2465> <2465> 7925
+<2467> <2467> 7926
+<246e> <246e> 7927
+<2521> <2521> 7928
+<2523> <2523> 7929
+<2525> <2525> 7930
+<2527> <2527> 7931
+<2529> <2529> 7932
+<2543> <2543> 7933
+<2563> <2563> 7934
+<2565> <2565> 7935
+<2567> <2567> 7936
+<256e> <256e> 7937
+<2575> <2576> 7938
+endcidrange
+endcmap
+CMapName currentdict /CMap defineresource pop
+end
+end
+
+%%EndResource
+%%EOF

data/examples/arc.rb

@@ -17,7 +17,7 @@ radius = 75
 
 # Left pie chart
 center = [page.box.width * 0.25, page.box.height * 0.85]
-pie = canvas.graphic_object(:solid_arc, cx: center[0], cy: center[1], 
+pie = canvas.graphic_object(:solid_arc, cx: center[0], cy: center[1],
                             outer_a: radius, outer_b: radius)
 canvas.fill_color('ddddff')
 canvas.draw(pie, start_angle: 30, end_angle: 110).fill
@@ -26,7 +26,7 @@ canvas.draw(pie, start_angle: 110, end_angle: 130).fill
 canvas.fill_color('ddffdd')
 canvas.draw(pie, start_angle: 130, end_angle: 30).fill
 
-arc = canvas.graphic_object(:arc, cx: center[0], cy: center[1], 
+arc = canvas.graphic_object(:arc, cx: center[0], cy: center[1],
                             a: radius, b: radius)
 canvas.stroke_color('0000ff')
 canvas.draw(arc, start_angle: 30, end_angle: 110).stroke
@@ -38,7 +38,7 @@ canvas.draw(arc, start_angle: 130, end_angle: 30).stroke
 # Right pie chart
 center = [page.box.width * 0.75, page.box.height * 0.85]
 canvas.stroke_color('777777')
-pie = canvas.graphic_object(:solid_arc, cx: center[0], cy: center[1], 
+pie = canvas.graphic_object(:solid_arc, cx: center[0], cy: center[1],
                             outer_a: radius, outer_b: radius)
 canvas.fill_color('ddddff')
 canvas.draw(pie, start_angle: 30, end_angle: 110).fill_stroke

data/examples/merging.rb

@@ -10,6 +10,9 @@
 # Sometimes other things like attached files or a document outline
 # should also be preserved.
 #
+# The hexapdf binary provides a command for merging files which does
+# the merging in a more sophisticated way.
+#
 # Usage:
 # : `ruby merging.rb INPUT1.PDF INPUT2.PDF ...`
 #
@@ -21,4 +24,4 @@ ARGV.each do |file|
   pdf = HexaPDF::Document.open(file)
   pdf.pages.each {|page| target.pages << target.import(page)}
 end
-target.write("merging.pdf", optimize: true)
+target.write("2.merging.pdf", optimize: true)

data/examples/optimizing.rb

@@ -7,6 +7,9 @@
 # over which aspects should be optimized. See [HexaPDF::Task::Optimize]
 # for detailed information.
 #
+# The hexapdf binary provides an optimization command which does some
+# additional operations like optimizing the page tree.
+#
 # Usage:
 # : `ruby optimizing.rb INPUT.PDF`
 #

data/examples/show_char_bboxes.rb

@@ -2,12 +2,12 @@
 #
 # This examples shows how to process the contents of a page. It finds all
 # characters on a page and surrounds them with their bounding box. Additionally,
-# all consecutive text runs are also surrounded by box.
+# all consecutive text runs are also surrounded by a box.
 #
 # The code provides two ways of generating the boxes. The commented part of
 # `ShowTextProcessor#show_text` uses a polyline since some characters may be
 # transforemd (rotated or skewed). The un-commented part uses rectangles which
-# is faster and correct in most cases.
+# is faster and correct for most but not all cases.
 #
 # Usage:
 # : `ruby show_char_boxes.rb INPUT.PDF`

data/examples/truetype.rb

@@ -17,9 +17,9 @@
 require 'hexapdf'
 
 doc = HexaPDF::Document.new
-doc.config['font.on_missing_glyph'] = ->(n,f) { f.missing_glyph_id }
+doc.config['font.on_missing_glyph'] = ->(_, f) { f.missing_glyph_id }
 doc.config['font.map'] = {
-  'myfont' => {none: ARGV.shift || File.join(__dir__, '../test/data/fonts/Ubuntu-Title.ttf')}
+  'myfont' => {none: ARGV.shift || File.join(__dir__, '../test/data/fonts/Ubuntu-Title.ttf')},
 }
 
 wrapper = doc.fonts.load('myfont')

data/lib/hexapdf/cli.rb

@@ -39,6 +39,7 @@ require 'hexapdf/cli/modify'
 require 'hexapdf/cli/merge'
 require 'hexapdf/cli/optimize'
 require 'hexapdf/cli/images'
+require 'hexapdf/cli/batch'
 require 'hexapdf/version'
 require 'hexapdf/document'
 
@@ -59,14 +60,28 @@ module HexaPDF
     # The CmdParse::CommandParser class that is used for running the CLI application.
     class Application < CmdParse::CommandParser
 
+      # Verbosity level for no output
+      VERBOSITY_QUIET = 0
+
+      # Verbosity level for warning output
+      VERBOSITY_WARNING = 1
+
+      # Verbosity level for informational output
+      VERBOSITY_INFO = 2
+
       # Specifies whether an operation should be forced. For example, if an existing file should be
       # overwritten.
       attr_reader :force
 
+      # Specifies whether strict parsing and validation should be used.
+      attr_reader :strict
+
       def initialize #:nodoc:
         super(handle_exceptions: :no_help)
         main_command.options.program_name = "hexapdf"
         main_command.options.version = HexaPDF::VERSION
+        main_command.extend(Command::Extensions)
+        main_command.define_singleton_method(:usage_commands) { "COMMAND" }
         add_command(HexaPDF::CLI::Info.new)
         add_command(HexaPDF::CLI::Files.new)
         add_command(HexaPDF::CLI::Images.new)
@@ -74,13 +89,37 @@ module HexaPDF
         add_command(HexaPDF::CLI::Modify.new)
         add_command(HexaPDF::CLI::Optimize.new)
         add_command(HexaPDF::CLI::Merge.new)
+        add_command(HexaPDF::CLI::Batch.new)
         add_command(CmdParse::HelpCommand.new)
-        add_command(CmdParse::VersionCommand.new)
+        version_command = CmdParse::VersionCommand.new(add_switches: false)
+        add_command(version_command)
+        main_options.on_tail("--version", "Show hexapdf version") { version_command.execute }
 
         @force = false
+        @verbosity = VERBOSITY_WARNING
+        @strict = false
         global_options.on("--[no-]force", "Force overwriting existing files. Default: false") do |f|
           @force = f
         end
+        global_options.on("--strict", "Enable strict parsing and validation") do |s|
+          @strict = s
+        end
+        global_options.on("--verbose", "-v", "Verbose output") do
+          @verbosity += 1
+        end
+        global_options.on("--quiet", "-q", "Suppress any output") do
+          @verbosity = VERBOSITY_QUIET
+        end
+      end
+
+      # Returns +true+ if the verbosity level warning is enabled.
+      def verbosity_warning?
+        @verbosity >= VERBOSITY_WARNING
+      end
+
+      # Returns +true+ if the verbosity level info is enabled.
+      def verbosity_info?
+        @verbosity >= VERBOSITY_INFO
       end
 
       def parse(argv = ARGV) #:nodoc:

data/lib/hexapdf/cli/batch.rb

@@ -0,0 +1,72 @@
+# -*- encoding: utf-8 -*-
+#
+#--
+# This file is part of HexaPDF.
+#
+# HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
+# Copyright (C) 2014-2017 Thomas Leitner
+#
+# HexaPDF is free software: you can redistribute it and/or modify it
+# under the terms of the GNU Affero General Public License version 3 as
+# published by the Free Software Foundation with the addition of the
+# following permission added to Section 15 as permitted in Section 7(a):
+# FOR ANY PART OF THE COVERED WORK IN WHICH THE COPYRIGHT IS OWNED BY
+# THOMAS LEITNER, THOMAS LEITNER DISCLAIMS THE WARRANTY OF NON
+# INFRINGEMENT OF THIRD PARTY RIGHTS.
+#
+# HexaPDF is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+# FITNESS FOR A PARTICULAR PURPOSE. See the GNU Affero General Public
+# License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with HexaPDF. If not, see <http://www.gnu.org/licenses/>.
+#
+# The interactive user interfaces in modified source and object code
+# versions of HexaPDF must display Appropriate Legal Notices, as required
+# under Section 5 of the GNU Affero General Public License version 3.
+#
+# In accordance with Section 7(b) of the GNU Affero General Public
+# License, a covered work must retain the producer line in every PDF that
+# is created or manipulated using HexaPDF.
+#++
+
+require 'hexapdf/cli/command'
+require 'shellwords'
+
+module HexaPDF
+  module CLI
+
+    # Execute the same command for multiple input files.
+    class Batch < Command
+
+      def initialize #:nodoc:
+        super('batch', takes_commands: false)
+        short_desc("Execute a single command on multiple files")
+        long_desc(<<-EOF.gsub!(/^ */, ''))
+          This command allows executing a single command for multiple input files, thereby reducing
+          the overall execution time.
+
+          The first argument is used as a hexapdf command line (but without the binary name) and
+          the rest as input files. The specified command will be executed for each input file, with
+          {} being replaced by the file name.
+        EOF
+      end
+
+      def execute(command, *files) #:nodoc:
+        args = Shellwords.split(command)
+        files.each do |file|
+          begin
+            HexaPDF::CLI::Application.new.parse(args.map {|a| a.gsub(/{}/, file)})
+          rescue
+            if command_parser.verbosity_warning?
+              $stderr.puts "Error processing '#{file}': #{$!.message}"
+            end
+          end
+        end
+      end
+
+    end
+
+  end
+end

data/lib/hexapdf/cli/command.rb

@@ -35,6 +35,7 @@ require 'io/console'
 require 'ostruct'
 require 'cmdparse'
 require 'hexapdf/document'
+require 'hexapdf/font/true_type'
 
 module HexaPDF
   module CLI
@@ -43,6 +44,16 @@ module HexaPDF
     # commands.
     class Command < CmdParse::Command
 
+      module Extensions #:nodoc:
+        def help_banner #:nodoc:
+          "hexapdf #{HexaPDF::VERSION} - Versatile PDF Manipulation Tool\n" \
+            "Copyright (c) 2014-2017 Thomas Leitner; licensed under the AGPLv3\n\n" <<
+            format(usage, indent: 7) << "\n\n"
+        end
+      end
+
+      include Extensions
+
       def initialize(*args, &block) #:nodoc:
         super
         @out_options = OpenStruct.new
@@ -51,6 +62,7 @@ module HexaPDF
         @out_options.object_streams = :preserve
         @out_options.xref_streams = :preserve
         @out_options.streams = :preserve
+        @out_options.optimize_fonts = false
 
         @out_options.encryption = :preserve
         @out_options.enc_user_pwd = @out_options.enc_owner_pwd = nil
@@ -62,6 +74,57 @@ module HexaPDF
 
       protected
 
+      # Creates a HexaPDF::Document instance for the PDF file and yields it.
+      #
+      # If +out_file+ is given, the document is written to it after yielding.
+      def with_document(file, password: nil, out_file: nil) #:yield: document
+        if file == out_file
+          doc = HexaPDF::Document.open(file, pdf_options(password))
+        else
+          file_io = File.open(file, 'rb')
+          doc = HexaPDF::Document.new(io: file_io, **pdf_options(password))
+        end
+
+        yield(doc)
+
+        write_document(doc, out_file)
+      ensure
+        file_io&.close
+      end
+
+      # Returns a hash with HexaPDF::Document options based on the given password and the option
+      # switches.
+      def pdf_options(password)
+        hash = {decryption_opts: {password: password}, config: {}}
+        hash[:config]['parser.on_correctable_error'] =
+          if command_parser.strict
+            proc { true }
+          else
+            proc do |_, msg, pos|
+              if command_parser.verbosity_info?
+                msg = MalformedPDFError.new(msg, pos: pos).message
+                $stderr.puts "Corrected parsing problem: #{msg}"
+              end
+              false
+            end
+          end
+        hash
+      end
+
+      # Writes the document to the given file or does nothing if +out_file+ is +nil+.
+      def write_document(doc, out_file)
+        if out_file
+          doc.validate(auto_correct: true) do |msg, correctable|
+            if command_parser.strict && !correctable
+              raise "Validation error: #{msg}"
+            elsif command_parser.verbosity_info?
+              $stderr.puts "#{correctable ? 'Corrected' : 'Ignored'} validation problem: #{msg}"
+            end
+          end
+          doc.write(out_file, validate: false)
+        end
+      end
+
       # Checks whether the given output file exists and raises an error if it does and
       # HexaPDF::CLI#force is not set.
       def maybe_raise_on_existing_file(filename)
@@ -98,6 +161,10 @@ module HexaPDF
                    "time; default: #{@out_options.compress_pages})") do |c|
           @out_options.compress_pages = c
         end
+        options.on("--[no-]optimize-fonts", "Optimize embedded font files; " \
+                   "default: #{@out_options.optimize_fonts})") do |o|
+          @out_options.optimize_fonts = o
+        end
       end
 
       # Defines the encryption options.
@@ -160,23 +227,37 @@ module HexaPDF
                  object_streams: @out_options.object_streams,
                  xref_streams: @out_options.xref_streams,
                  compress_pages: @out_options.compress_pages)
-        handle_streams(doc) unless @out_options.streams == :preserve
+        if @out_options.streams != :preserve || @out_options.optimize_fonts
+          doc.each(current: false) do |obj|
+            optimize_stream(obj)
+            optimize_font(obj)
+          end
+        end
       end
 
       IGNORED_FILTERS = { #:nodoc:
         CCITTFaxDecode: true, JBIG2Decode: true, DCTDecode: true, JPXDecode: true, Crypt: true
       }.freeze
 
-      # Applies the chosen stream mode to all streams.
-      def handle_streams(doc)
-        doc.each(current: false) do |obj|
-          next if !obj.respond_to?(:set_filter) || Array(obj[:Filter]).any? {|f| IGNORED_FILTERS[f]}
-          if @out_options.streams == :compress
-            obj.set_filter(:FlateDecode)
-          else
-            obj.set_filter(nil)
-          end
-        end
+      # Applies the chosen stream mode to the given object.
+      def optimize_stream(obj)
+        return if @out_options.streams == :preserve || !obj.respond_to?(:set_filter) ||
+          Array(obj[:Filter]).any? {|f| IGNORED_FILTERS[f]}
+
+        obj.set_filter(@out_options.streams == :compress ? :FlateDecode : nil)
+      end
+
+      # Optimize the object if it is a font object.
+      def optimize_font(obj)
+        return unless @out_options.optimize_fonts && obj.kind_of?(HexaPDF::Type::Font) &&
+          (obj[:Subtype] == :TrueType ||
+           (obj[:Subtype] == :Type0 && obj.descendant_font[:Subtype] == :CIDFontType2)) &&
+          obj.embedded?
+
+        font = HexaPDF::Font::TrueType::Font.new(StringIO.new(obj.font_file.stream))
+        data = HexaPDF::Font::TrueType::Optimizer.build_for_pdf(font)
+        obj.font_file.stream = data
+        obj.font_file[:Length1] = data.size
       end
 
       # Applies the encryption related options to the given HexaPDF::Document instance.
@@ -199,7 +280,9 @@ module HexaPDF
       ROTATE_MAP = {'l' => -90, 'r' => 90, 'd' => 180, 'n' => :none}.freeze #:nodoc:
 
       # Parses the pages specification string and returns an array of tuples containing a page
-      # number and a rotation value (either -90, 90, 180 or :none).
+      # number and a rotation value (either -90, 90, 180, :none or +nil+ where an integer means
+      # adding a rotation by that number of degrees, :none means removing any set rotation value and
+      # +nil+ means preserving the set rotation value).
       #
       # The parameter +count+ needs to be the total number of pages in the document.
       #
@@ -208,10 +291,12 @@ module HexaPDF
         range.split(',').each_with_object([]) do |str, arr|
           case str
           when /\A#{PAGE_NUMBER_SPEC}(l|r|d|n)?\z/o
-            arr << [($1 == 'e' ? count : str.to_i) - 1, ROTATE_MAP[$2]]
+            page_num = ($1 == 'e' ? count : str.to_i)
+            next if page_num > count
+            arr << [page_num - 1, ROTATE_MAP[$2]]
           when /\A#{PAGE_NUMBER_SPEC}-#{PAGE_NUMBER_SPEC}(?:\/([1-9]\d*))?(l|r|d|n)?\z/
-            start_nr = ($1 == 'e' ? count : $1.to_i) - 1
-            end_nr = ($2 == 'e' ? count : $2.to_i) - 1
+            start_nr = ($1 == 'e' ? count : [$1.to_i, count].min) - 1
+            end_nr = ($2 == 'e' ? count : [$2.to_i, count].min) - 1
             step = ($3 ? $3.to_i : 1) * (start_nr > end_nr ? -1 : 1)
             rotation = ROTATE_MAP[$4]
             start_nr.step(to: end_nr, by: step) {|n| arr << [n, rotation]}
@@ -235,6 +320,18 @@ module HexaPDF
         end
       end
 
+      # Removes unused pages and page tree nodes from the document.
+      def remove_unused_pages(doc)
+        retained = doc.pages.each_with_object({}) {|page, h| h[page.data] = true}
+        retained[doc.pages.root.data] = true
+        doc.each(current: false) do |obj|
+          next unless obj.kind_of?(HexaPDF::Dictionary)
+          if (obj.type == :Pages || obj.type == :Page) && !retained.key?(obj.data)
+            doc.delete(obj)
+          end
+        end
+      end
+
       private
 
       # Displays the given prompt, reads from the console without echo and returns the read string.