hexapdf 0.11.7 → 0.12.2

data/lib/hexapdf/cli/files.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2019 Thomas Leitner
+# Copyright (C) 2014-2020 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

data/lib/hexapdf/cli/form.rb

@@ -0,0 +1,240 @@
+# -*- encoding: utf-8; frozen_string_literal: true -*-
+#
+#--
+# This file is part of HexaPDF.
+#
+# HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
+# Copyright (C) 2014-2020 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.
+#
+# If the GNU Affero General Public License doesn't fit your need,
+# commercial licenses are available at <https://gettalong.at/hexapdf/>.
+#++
+
+require 'hexapdf/cli/command'
+require 'strscan'
+
+module HexaPDF
+  module CLI
+
+    # Processes a PDF that contains an interactive form (AcroForm).
+    class Form < Command
+
+      def initialize #:nodoc:
+        super('form', takes_commands: false)
+        short_desc("Show form fields and fill out a form")
+        long_desc(<<~EOF)
+          Use this command to process interactive PDF forms.
+
+          If the the output file name is not given, all form fields are listed in page order. Use
+          the global --verbose option to show additional information like field type and location.
+
+          If the output file name is given, the fields can be interactively filled out. By
+          additionally using the --template option, the data for the fields is read from the given
+          template file instead of the standard input.
+        EOF
+
+        options.on("--password PASSWORD", "-p", String,
+                   "The password for decryption. Use - for reading from standard input.") do |pwd|
+          @password = (pwd == '-' ? read_password : pwd)
+        end
+        options.on("--template TEMPLATE_FILE", "-t TEMPLATE_FILE",
+                   "Use the template file for the field values") do |template|
+          @template = template
+        end
+        options.on("--[no-]viewer-override", "Let the PDF viewer override the visual " \
+                   "appearance. Default: use setting from input PDF") do |need_appearances|
+          @need_appearances = need_appearances
+        end
+        options.on("--[no-]incremental-save", "Append the changes instead of rewriting the " \
+                   "whole file. Default: true") do |incremental|
+          @incremental = incremental
+        end
+
+        @password = nil
+        @template = nil
+        @need_appearances = nil
+        @incremental = true
+      end
+
+      def execute(in_file, out_file = nil) #:nodoc:
+        maybe_raise_on_existing_file(out_file) if out_file
+        with_document(in_file, password: @password, out_file: out_file,
+                      incremental: @incremental) do |doc|
+          if !doc.acro_form
+            raise "This PDF doesn't contain an interactive form"
+          elsif out_file
+            doc.acro_form[:NeedAppearances] = @need_appearances unless @need_appearances.nil?
+            if @template
+              fill_form_with_template(doc)
+            else
+              fill_form(doc)
+            end
+          else
+            list_form_fields(doc)
+          end
+        end
+      end
+
+      private
+
+      # Lists all terminal form fields.
+      def list_form_fields(doc)
+        current_page_index = -1
+        each_field(doc) do |_page, page_index, field, widget|
+          if current_page_index != page_index
+            puts "Page #{page_index + 1}"
+            current_page_index = page_index
+          end
+
+          field_name = field.full_field_name +
+            (field.alternate_field_name ? " (#{field.alternate_field_name})" : '')
+          concrete_field_type = field.concrete_field_type
+          nice_field_type = concrete_field_type.to_s.split('_').map(&:capitalize).join(' ')
+          position = "(#{widget[:Rect].left}, #{widget[:Rect].bottom})"
+
+          puts "  #{field_name}"
+          if command_parser.verbosity_info?
+            printf("    └─ %-22s | %-20s\n", nice_field_type, position)
+          end
+          puts "    └─ #{field.field_value.inspect}"
+          if command_parser.verbosity_info?
+            if field.field_type == :Ch
+              puts "    └─ Options: #{field.option_items.map(&:inspect).join(', ')}"
+            elsif concrete_field_type == :radio_button
+              puts "    └─ Options: #{field.radio_button_values.map(&:inspect).join(', ')}"
+            end
+          end
+        end
+      end
+
+      # Fills out the form by interactively asking the user for field values.
+      def fill_form(doc)
+        current_page_index = -1
+        each_field(doc) do |_page, page_index, field, _widget|
+          if current_page_index != page_index
+            puts "Page #{page_index + 1}"
+            current_page_index = page_index
+          end
+
+          field_name = field.full_field_name +
+            (field.alternate_field_name ? " (#{field.alternate_field_name})" : '')
+          concrete_field_type = field.concrete_field_type
+
+          puts "  #{field_name}"
+          puts "    └─ Current value: #{field.field_value.inspect}"
+
+          if field.field_type == :Ch
+            puts "    └─ Possible values: #{field.option_items.map(&:inspect).join(', ')}"
+          elsif concrete_field_type == :radio_button
+            puts "    └─ Possible values: #{field.radio_button_values.map(&:inspect).join(', ')}"
+          elsif concrete_field_type == :check_box
+            puts "    └─ Possible values: y(es), t(rue); n(o), f(alse)"
+          end
+
+          begin
+            print "    └─ New value: "
+            value = $stdin.readline.chomp
+            next if value.empty?
+            apply_field_value(field, value)
+          rescue HexaPDF::Error => e
+            puts "       ⚠  #{e.message}"
+            retry
+          end
+        end
+      end
+
+      # Fills out the form using the data from the provided template file.
+      def fill_form_with_template(doc)
+        data = parse_template
+        form = doc.acro_form
+        data.each do |name, value|
+          field = form.field_by_name(name)
+          raise "Field '#{name}' not found in input PDF" unless field
+          apply_field_value(field, value)
+        end
+      end
+
+      # Parses the data from the given template file.
+      def parse_template
+        data = {}
+        scanner = StringScanner.new(File.read(@template))
+        until scanner.eos?
+          field_name = scanner.scan(/(\\:|[^:])*?:/)
+          break unless field_name
+          field_name.gsub!(/\\:/, ':')
+          field_value = scanner.scan(/.*?(?=^\S|\z)/m)
+          data[field_name.chop] = field_value.strip.gsub(/^\s*/, '') if field_value
+        end
+        if !scanner.eos? && command_parser.verbosity_warning?
+          $stderr.puts "Warning: Some template could not be parsed"
+        end
+        data
+      end
+
+      # Applies the given value to the field.
+      def apply_field_value(field, value)
+        case field.concrete_field_type
+        when :single_line_text_field
+          field.field_value = value
+        when :combo_box, :list_box
+          field.field_value = value
+        when :editable_combo_box
+          field.field_value = value
+        when :check_box
+          unless value.match?(/y(es)?|t(rue)?|f(alse)?|n(o)/)
+            raise HexaPDF::Error, "Invalid input, use one of the possible values"
+          end
+          field.field_value = value.match?(/y(es)?|t(rue)?/)
+        when :radio_button
+          field.field_value = value.intern
+        else
+          raise "Field type not yet supported"
+        end
+      end
+
+      # Iterates over all non-push button fields in page order. If a field appears on multiple
+      # pages, it is only yielded on the first page.
+      def each_field(doc) # :yields: page, page_index, field
+        seen = {}
+
+        doc.pages.each_with_index do |page, page_index|
+          page[:Annots]&.each do |annotation|
+            next unless annotation[:Subtype] == :Widget
+            field = annotation.form_field
+            next if field.concrete_field_type == :push_button
+            unless seen[field]
+              yield(page, page_index, field, annotation)
+              seen[field] = true
+            end
+          end
+        end
+      end
+
+    end
+
+  end
+end

data/lib/hexapdf/cli/image2pdf.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2019 Thomas Leitner
+# Copyright (C) 2014-2020 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

data/lib/hexapdf/cli/images.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2019 Thomas Leitner
+# Copyright (C) 2014-2020 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

data/lib/hexapdf/cli/info.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2019 Thomas Leitner
+# Copyright (C) 2014-2020 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

data/lib/hexapdf/cli/inspect.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2019 Thomas Leitner
+# Copyright (C) 2014-2020 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

data/lib/hexapdf/cli/merge.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2019 Thomas Leitner
+# Copyright (C) 2014-2020 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

data/lib/hexapdf/cli/modify.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2019 Thomas Leitner
+# Copyright (C) 2014-2020 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

data/lib/hexapdf/cli/optimize.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2019 Thomas Leitner
+# Copyright (C) 2014-2020 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

data/lib/hexapdf/cli/split.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2019 Thomas Leitner
+# Copyright (C) 2014-2020 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

data/lib/hexapdf/cli/watermark.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2019 Thomas Leitner
+# Copyright (C) 2014-2020 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

data/lib/hexapdf/composer.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2019 Thomas Leitner
+# Copyright (C) 2014-2020 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
@@ -297,7 +297,7 @@ module HexaPDF
       style ||= base_style
       style = style.dup.update(**options) unless options.empty?
       style.font(base_style.font) unless style.font?
-      style.font(@document.fonts.add(style.font)) unless style.font.respond_to?(:dict)
+      style.font(@document.fonts.add(style.font)) unless style.font.respond_to?(:pdf_object)
       style
     end
 

data/lib/hexapdf/configuration.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2019 Thomas Leitner
+# Copyright (C) 2014-2020 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
@@ -48,8 +48,9 @@ module HexaPDF
   # easily be changed.
   #
   # Some options are defined as global options because they are needed on the class level - see
-  # HexaPDF::GlobalConfiguration. Other options can be configured for individual documents as they
-  # allow to fine-tune some behavior - see HexaPDF::DefaultDocumentConfiguration.
+  # HexaPDF::GlobalConfiguration[index.html#GlobalConfiguration]. Other options can be configured for
+  # individual documents as they allow to fine-tune some behavior - see
+  # HexaPDF::DefaultDocumentConfiguration[index.html#DefaultDocumentConfiguration].
   #
   # A configuration option name is dot-separted to provide a hierarchy of option names. For
   # example, io.chunk_size.
@@ -144,6 +145,43 @@ module HexaPDF
   #
   # The following options are provided:
   #
+  # acro_form.appearance_generator::
+  #    The class that should be used for generating appearances for AcroForm fields. If the value is
+  #    a String, it should contain the name of a constant to such a class.
+  #
+  #    See HexaPDF::Type::AcroForm::AppearanceGenerator
+  #
+  # acro_form.create_appearances::
+  #    A boolean specifying whether an AcroForm field's appearances should automatically be
+  #    generated if they are missing.
+  #
+  # acro_form.text_field.default_width::
+  #    A number specifying the default width of AcroForm text fields which should be auto-sized.
+  #
+  # acro_form.default_font_size::
+  #    A number specifying the default font size of AcroForm text fields which should be auto-sized.
+  #
+  # acro_form.fallback_font::
+  #    The font that should be used when a variable text field references a font that cannot be used.
+  #
+  #    Can either be the name of a font, like 'Helvetica', or an array consisting of the font name
+  #    and a hash of font options, like ['Helvetica', variant: :italic]. If set to +nil+, the use of
+  #    the fallback font is disabled.
+  #
+  #    Default is 'Helvetica'.
+  #
+  # acro_form.on_invalid_value::
+  #    Callback hook when an invalid value is set for certain types of AcroForm fields.
+  #
+  #    The value needs to be an object that responds to \#call(field, value) where +field+ is the
+  #    AcroForm field on which the value is set and +value+ is the invalid value. The returned value
+  #    is used instead of the invalid value.
+  #
+  #    The default implementation raises an error.
+  #
+  # acro_form.text_field.default_width::
+  #    A number specifying the default width of AcroForm text fields which should be auto-sized.
+  #
   # document.auto_decrypt::
   #    A boolean determining whether the document should be decrypted automatically when parsed.
   #
@@ -307,7 +345,16 @@ module HexaPDF
   # task.map::
   #    A mapping from task names to callable task objects. See HexaPDF::Task for more information.
   DefaultDocumentConfiguration =
-    Configuration.new('document.auto_decrypt' => true,
+    Configuration.new('acro_form.appearance_generator' => 'HexaPDF::Type::AcroForm::AppearanceGenerator',
+                      'acro_form.create_appearances' => true,
+                      'acro_form.default_font_size' => 10,
+                      'acro_form.fallback_font' => 'Helvetica',
+                      'acro_form.on_invalid_value' => proc do |field, value|
+                        raise HexaPDF::Error, "Invalid value #{value.inspect} for " \
+                          "#{field.concrete_field_type} field #{field.full_field_name}"
+                      end,
+                      'acro_form.text_field.default_width' => 100,
+                      'document.auto_decrypt' => true,
                       'encryption.aes' => 'HexaPDF::Encryption::FastAES',
                       'encryption.arc4' => 'HexaPDF::Encryption::FastARC4',
                       'encryption.filter_map' => {
@@ -326,12 +373,12 @@ module HexaPDF
                         Fl: 'HexaPDF::Filter::FlateDecode',
                         RunLengthDecode: 'HexaPDF::Filter::RunLengthDecode',
                         RL: 'HexaPDF::Filter::RunLengthDecode',
-                        CCITTFaxDecode: nil,
-                        CCF: nil,
-                        JBIG2Decode: nil,
-                        DCTDecode: 'HexaPDF::Filter::DCTDecode',
-                        DCT: 'HexaPDF::Filter::DCTDecode',
-                        JPXDecode: 'HexaPDF::Filter::JPXDecode',
+                        CCITTFaxDecode: 'HexaPDF::Filter::PassThrough',
+                        CCF: 'HexaPDF::Filter::PassThrough',
+                        JBIG2Decode: 'HexaPDF::Filter::PassThrough',
+                        DCTDecode: 'HexaPDF::Filter::PassThrough',
+                        DCT: 'HexaPDF::Filter::PassThrough',
+                        JPXDecode: 'HexaPDF::Filter::PassThrough',
                         Crypt: nil,
                         Encryption: 'HexaPDF::Filter::Encryption',
                       },
@@ -447,10 +494,13 @@ module HexaPDF
                         Action: 'HexaPDF::Type::Action',
                         XXLaunchActionWinParameters: 'HexaPDF::Type::Actions::Launch::WinParameters',
                         Annot: 'HexaPDF::Type::Annotation',
-                        XXAppearanceCharacteristics: 'HexaPDF::Type::Annotations::Widget::AppearanceCharacteristics',
+                        XXAppearanceCharacteristics: \
+                          'HexaPDF::Type::Annotations::Widget::AppearanceCharacteristics',
                         XXIconFit: 'HexaPDF::Type::IconFit',
                         XXAcroForm: 'HexaPDF::Type::AcroForm::Form',
                         XXAcroFormField: 'HexaPDF::Type::AcroForm::Field',
+                        XXAppearanceDictionary: 'HexaPDF::Type::Annotation::AppearanceDictionary',
+                        Border: 'HexaPDF::Type::Annotation::Border',
                       },
                       'object.subtype_map' => {
                         nil => {
@@ -492,6 +542,11 @@ module HexaPDF
                           Link: 'HexaPDF::Type::Annotations::Link',
                           Widget: 'HexaPDF::Type::Annotations::Widget',
                         },
+                        XXAcroFormField: {
+                          Tx: 'HexaPDF::Type::AcroForm::TextField',
+                          Btn: 'HexaPDF::Type::AcroForm::ButtonField',
+                          Ch: 'HexaPDF::Type::AcroForm::ChoiceField',
+                        },
                       })
 
 end