hexapdf 0.21.0 → 0.23.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: b666a69330b87a3ad7a5c937a1113a66a3b5ed3d0cb4f4478ee24f97bed0e411
-  data.tar.gz: 460c3cd90b8f76d3e4d32fdd6f1bc8fd71ddfef5d18799c6a207f562773371f8
+  metadata.gz: d6f11a38da2472389966c1d103b86e46ba8cd858d2927b1a65f441f56e2e7dd5
+  data.tar.gz: 74ff92dcb6ede9f137303afc7c8b9d08e0baa42b0ab0b7d4436777e6db617ea4
 SHA512:
-  metadata.gz: a9135019912d8c3d1e282797054dfbc99c028f26aa28f8412a138ab3d8d67124c4e9c52ad104eeb4289487e47dd850e3ff4211b5e928cdb0a8aab2d8ea773930
-  data.tar.gz: 070d0facdb6576fb6a0377a7139609ee648a580be879dabfd10e943bd2f62d66d9d5077058edcecf9ca3eb4e7969a2427c7a4af26bbb4d8f734e238f7022c7cd
+  metadata.gz: 468e0a5a24e06574b4a802408daafbdb59b3a1dc460ad2b6bf9f543088d1210d8295b843847819a872436ea4c380df85240822e28b548dd88b55f74b30ff8ab3
+  data.tar.gz: 29e95def1e7ccf0d55bca3cf5ad4d95a070d5910a0e1b8645d2884e0e5022367c63e14e2df3ecc0bac7cdfc329d88698c01060dc8005e1c1025a0c8337367afc

data/CHANGELOG.md

@@ -1,3 +1,81 @@
+## 0.23.0 - 2022-05-26
+
+### Added
+
+- [HexaPDF::Composer#create_stamp] for creating a form Xobject
+- [HexaPDF::Revision#reset_objects] for deleting all live loaded and added
+  objects
+- Support for removing or flattening annotations to the `hexapdf modify` command
+- Option to CLI command `hexapdf form` to allow generation of a template file
+- Support for centering a floating box in [HexaPDF::Layout::Frame]
+- [HexaPDF::Type::Catalog#names] for easier access to the name dictionary
+- [HexaPDF::Type::Names#destinations] for easier access to the destinations name
+  tree
+- [HexaPDF::Document::Destinations], accessible via
+  [HexaPDF::Document#destinations], as convenience interface for working with
+  destination arrays
+
+### Changed
+
+- **Breaking change**: Refactored the [HexaPDF::Document] interface for working
+  with objects and move parts into [HexaPDF::Revisions]
+- **Breaking change**: [HexaPDF::Layout::TextBox] to use whole available width
+  when aligning to the center or right
+- **Breaking change**: [HexaPDF::Layout::TextBox] to use whole available height
+  when vertically aligning to the center or bottom
+- CLI command `hexapdf inspect` to show the type of revisions, as well as the
+  number of objects per revision
+- [HexaPDF::Task::Optimize] to allow skipping invalid content stream operations
+- [HexaPDF::Composer#image] to allow using a form xobject in place of the image
+
+### Fixed
+
+- [HexaPDF::Writer#write] to write modified objects into the correct revision
+- [HexaPDF::Revisions::from_io] to correctly handle hybrid-reference files
+- [HexaPDF::Writer] to assign a valid object number to a created cross-reference
+  stream in all cases
+* [HexaPDF::Type::AcroForm::TextField] to validate the existence of a /MaxLen
+  value for comb text fields
+* [HexaPDF::Type::AcroForm::TextField#field_value=] to check for the existence
+  of /MaxLen when setting a value for a comb text field
+* [HexaPDF::Type::AcroForm::TextField#field_value=] to check the value against
+  /MaxLen
+* [HexaPDF::Layout::TextLayouter#fit] to not use style valign when doing
+  variable width layouting
+* [HexaPDF::Utils::SortedTreeNode#find_entry] to work in case of a node without
+  a container name or kids key
+* CLI command `hexapdf form` to allow setting array values when using a template
+* CLI command `hexapdf form` to allow setting file select fields
+
+
+## 0.22.0 - 2022-03-26
+
+### Added
+
+- Support for writing images with an ICCBased color space
+- Support for writing images with soft masks
+
+### Changed
+
+- CLI command `hexapdf form` to show a warning when working with a file
+  containing an XFA form
+
+### Fixed
+
+- [HexaPDF::Type::AcroForm::Form#field_by_name] to work correctly when field
+  name parts are UTF-16BE encoded
+- `hexapdf inspect` command 'revision' to correctly detect the end of revisions
+- [HexaPDF::DictionaryFields::StringConverter] to use correct method name
+  `HexaPDF::Document#config`
+
+
+## 0.21.1 - 2022-03-12
+
+### Fixed
+
+- Handling of invalid AES encrypted files where the padding is missing
+
+
 ## 0.21.0 - 2022-03-04
 
 ### Added
@@ -246,7 +324,7 @@
 
 ## 0.16.0 - 2021-09-28
 
-## Added
+### Added
 
 * Support for RGB color values of the form "RGB" in addition to "RRGGBB" and for
   CSS color module level 3 color names

data/Rakefile

@@ -49,7 +49,7 @@ namespace :dev do
   task :test_all do
     versions = `rbenv versions --bare | grep -i 2.[567]\\\\\\|3.`.split("\n")
     versions.each do |version|
-      sh "rbenv shell #{version} &>/dev/null && rake test"
+      sh "eval \"$(rbenv init -)\"; rbenv shell #{version} && ruby -v && rake test"
     end
     puts "Looks okay? (enter to continue, Ctrl-c to abort)"
     $stdin.gets

data/lib/hexapdf/cli/form.rb

@@ -70,6 +70,9 @@ module HexaPDF
           @template = template
           @fill = true
         end
+        options.on('--generate-template', 'Print a template for use with --template') do
+          @generate_template = true
+        end
         options.on('--flatten', 'Flatten the form fields') do
           @flatten = true
         end
@@ -85,6 +88,7 @@ module HexaPDF
         @password = nil
         @fill = false
         @flatten = false
+        @generate_template = false
         @template = nil
         @need_appearances = nil
         @incremental = true
@@ -97,6 +101,10 @@ module HexaPDF
         end
         with_document(in_file, password: @password, out_file: out_file,
                       incremental: @incremental) do |doc|
+          if doc.acro_form[:XFA]
+            $stderr.puts "Warning: Unsupported XFA form detected, some things may not work correctly"
+          end
+
           if !doc.acro_form
             raise "This PDF doesn't contain an interactive form"
           elsif out_file
@@ -113,6 +121,15 @@ module HexaPDF
               doc.catalog.delete(:AcroForm)
               doc.delete(doc.acro_form)
             end
+          elsif @generate_template
+            unsupported_fields = [:signature_field, :password_field]
+            each_field(doc) do |_, _, field, _|
+              next if unsupported_fields.include?(field.concrete_field_type)
+              name = field.full_field_name.gsub(':', "\\:")
+              Array(field.field_value).each do |val|
+                puts "#{name}: #{val.to_s.gsub(/(\r|\r\n|\n)/, '\1  ')}"
+              end
+            end
           else
             list_form_fields(doc)
           end
@@ -216,8 +233,16 @@ module HexaPDF
           field_name = scanner.scan(/(\\:|[^:])*?:/)
           break unless field_name
           field_name.gsub!(/\\:/, ':')
+          field_name.chop!
           field_value = scanner.scan(/.*?(?=^\S|\z)/m)
-          data[field_name.chop] = field_value.strip.gsub(/^\s*/, '') if field_value
+          next unless field_value
+          field_value = field_value.strip.gsub(/^\s*/, '')
+          if data.key?(field_name)
+            data[field_name] = [data[field_name]] unless data[field_name].kind_of?(Array)
+            data[field_name] << field_value
+          else
+            data[field_name] = field_value
+          end
         end
         if !scanner.eos? && command_parser.verbosity_warning?
           $stderr.puts "Warning: Some template could not be parsed"
@@ -228,8 +253,8 @@ module HexaPDF
       # Applies the given value to the field.
       def apply_field_value(field, value)
         case field.concrete_field_type
-        when :single_line_text_field, :multiline_text_field, :comb_text_field, :combo_box,
-            :list_box, :editable_combo_box
+        when :single_line_text_field, :multiline_text_field, :comb_text_field, :file_select_field,
+            :combo_box, :list_box, :editable_combo_box
           field.field_value = value
         when :check_box
           field.field_value = case value
@@ -245,6 +270,8 @@ module HexaPDF
         else
           raise "Field type #{field.concrete_field_type} not yet supported"
         end
+      rescue
+        raise "Error while setting '#{field.full_field_name}': #{$!.message}"
       end
 
       # Iterates over all non-push button fields in page order. If a field appears on multiple

data/lib/hexapdf/cli/inspect.rb

@@ -229,10 +229,19 @@ module HexaPDF
               end
               IO.copy_stream(@doc.revisions.parser.io, $stdout, length, 0)
             else
-              puts "Document has #{@doc.revisions.size} revision#{@doc.revisions.size == 1 ? '' : 's'}"
-              revision_information do |_, index, count, signature, end_offset|
+              puts "Document has #{@doc.revisions.count} revision#{@doc.revisions.count == 1 ? '' : 's'}"
+              revision_information do |rev, index, count, signature, end_offset|
+                type = if rev.trailer[:XRefStm]
+                         "xref table + stream"
+                       elsif rev.trailer[:Type] == :XRef
+                         "xref stream"
+                       else
+                         "xref table"
+                       end
                 puts "Revision #{index + 1}"
+                puts "  Type      : #{type}"
                 puts "  Objects   : #{count}"
+                puts "  Size      : #{rev.trailer[:Size]}"
                 puts "  Signed    : yes" if signature
                 puts "  Byte range: 0-#{end_offset}"
               end
@@ -342,13 +351,17 @@ module HexaPDF
             end_index = sig[:ByteRange][-2] + sig[:ByteRange][-1]
           else
             io.seek(startxrefs[index], IO::SEEK_SET)
+            buffer = ''.b
             while io.pos < startxrefs[index + 1]
-              if io.gets =~ /^\s*%%EOF\s*$/
-                end_index = io.pos
+              buffer << io.read(1_000)
+              if (buffer_index = buffer.index(/(?:\n|\r\n?)\s*%%EOF\s*(?:\n|\r\n?)/))
+                end_index = io.pos - buffer.size + buffer_index + $~[0].size
+                break
               end
+              buffer = buffer[-20..-1]
             end
           end
-          yield(rev, index, rev.next_free_oid - 1, sig, end_index)
+          yield(rev, index, rev.each.count, sig, end_index)
         end
       end
 

data/lib/hexapdf/cli/modify.rb

@@ -53,14 +53,15 @@ module HexaPDF
         super('modify', takes_commands: false)
         short_desc("Modify a PDF file")
         long_desc(<<~EOF)
-          This command modifies a PDF file. It can be used to select pages that should appear in
-          the output file and/or rotate them. The output file can also be encrypted/decrypted and
-          optimized in various ways.
+          This command modifies a PDF file. It can be used, for example, to select pages that should
+          appear in the output file and/or rotate them. The output file can also be
+          encrypted/decrypted and optimized in various ways.
         EOF
 
         @password = nil
         @pages = '1-e'
         @embed_files = []
+        @annotation_mode = nil
 
         options.on("--password PASSWORD", "-p", String,
                    "The password for decryption. Use - for reading from standard input.") do |pwd|
@@ -74,6 +75,10 @@ module HexaPDF
                    "used multiple times)") do |file|
           @embed_files << file
         end
+        options.on("--annotations MODE", [:remove, :flatten], "Handling of annotations (either " \
+                   "remove or flatten)") do |mode|
+          @annotation_mode = mode
+        end
         define_optimization_options
         define_encryption_options
       end
@@ -82,6 +87,7 @@ module HexaPDF
         maybe_raise_on_existing_file(out_file)
         with_document(in_file, password: @password, out_file: out_file) do |doc|
           arrange_pages(doc) unless @pages == '1-e'
+          handle_annotations(doc)
           @embed_files.each {|file| doc.files.add(file, embed: true) }
           apply_encryption_options(doc)
           apply_optimization_options(doc)
@@ -109,6 +115,20 @@ module HexaPDF
         doc.pages.add unless doc.pages.count > 0
       end
 
+      # Handles the annotations of all selected pages by doing nothing, removing them or flattening
+      # them.
+      def handle_annotations(doc)
+        return unless @annotation_mode
+
+        doc.pages.each do |page|
+          if @annotation_mode == :remove
+            page.delete(:Annots)
+          else
+            page.flatten_annotations
+          end
+        end
+      end
+
     end
 
   end

data/lib/hexapdf/composer.rb

@@ -313,7 +313,8 @@ module HexaPDF
 
     # Draws the given image at the current position.
     #
-    # The +file+ argument can be anything that is accepted by HexaPDF::Document::Images#add.
+    # The +file+ argument can be anything that is accepted by HexaPDF::Document::Images#add or a
+    # HexaPDF::Type::Form object.
     #
     # See #text for details on +width+, +height+, +style+ and +style_properties+.
     #
@@ -324,7 +325,7 @@ module HexaPDF
     #   composer.image(machu_picchu, height: 30)
     def image(file, width: 0, height: 0, style: nil, **style_properties)
       style = retrieve_style(style, style_properties)
-      image = document.images.add(file)
+      image = file.kind_of?(HexaPDF::Stream) ? file : document.images.add(file)
       draw_box(Layout::ImageBox.new(image, width: width, height: height, style: style))
     end
 
@@ -361,6 +362,27 @@ module HexaPDF
       end
     end
 
+    # Creates a stamp (Form XObject) which can be used like an image multiple times on a single page
+    # or on multiple pages.
+    #
+    # The width and the height of the stamp need to be set (frame.width/height or
+    # page.box.width/height might be good choices).
+    #
+    # Examples:
+    #
+    #   #>pdf-composer
+    #   stamp = composer.create_stamp(50, 50) do |canvas|
+    #     canvas.fill_color("red").line_width(5).
+    #       rectangle(10, 10, 30, 30).fill_stroke
+    #   end
+    #   composer.image(stamp, width: 20, height: 20)
+    #   composer.image(stamp, width: 50)
+    def create_stamp(width, height) # :yield: canvas
+      stamp = @document.add({Type: :XObject, Subtype: :Form, BBox: [0, 0, width, height]})
+      yield(stamp.canvas) if block_given?
+      stamp
+    end
+
     private
 
     # Creates the frame into which boxes are layed out when a new page is created.

data/lib/hexapdf/dictionary_fields.rb

@@ -241,7 +241,7 @@ module HexaPDF
           if str.valid_encoding?
             str.encode!(Encoding::UTF_8)
           else
-            document.configuration['document.on_invalid_string'].call(str)
+            document.config['document.on_invalid_string'].call(str)
           end
         else
           Utils::PDFDocEncoding.convert_to_utf8(str)