# -*- 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) 2014-2021 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 'set'
require 'hexapdf/serializer'
require 'hexapdf/content/parser'
require 'hexapdf/content/operator'

module HexaPDF
  module Task

    # Task for optimizing the PDF document.
    #
    # For a list of optimization methods this task can perform have a look at the ::call method.
    module Optimize

      # Optimizes the PDF document.
      #
      # The field entries that are optional and set to their default value are always deleted.
      # Additional optimization methods are performed depending on the values of the following
      # arguments:
      #
      # compact::
      #   Compacts the object space by merging the revisions and then deleting null and unused
      #   values if set to +true+.
      #
      # object_streams::
      #   Specifies if and how object streams should be used: For :preserve, existing object
      #   streams are preserved; for :generate objects are packed into object streams as much as
      #   possible; and for :delete existing object streams are deleted.
      #
      # xref_streams::
      #  Specifies if cross-reference streams should be used. Can be :preserve (no modifications),
      #  :generate (use cross-reference streams) or :delete (remove cross-reference streams).
      #
      #  If +object_streams+ is set to :generate, this option is implicitly changed to :generate.
      #
      # compress_pages::
      #   Compresses the content streams of all pages if set to +true+. Note that this can take a
      #   *very* long time because each content stream has to be unfiltered, parsed, serialized
      #   and then filtered again.
      #
      # prune_page_resources::
      #   Removes all unused XObjects from the resources dictionaries of all pages. It is
      #   recommended to also set the +compact+ argument because otherwise the unused XObjects won't
      #   be deleted from the document.
      #
      #   This is sometimes necessary after importing pages from other PDF files that use a single
      #   resources dictionary for all pages.
      def self.call(doc, compact: false, object_streams: :preserve, xref_streams: :preserve,
                    compress_pages: false, prune_page_resources: false)
        used_refs = compress_pages(doc) if compress_pages
        prune_page_resources(doc, used_refs) if prune_page_resources

        if compact
          compact(doc, object_streams, xref_streams)
        elsif object_streams != :preserve
          process_object_streams(doc, object_streams, xref_streams)
        elsif xref_streams != :preserve
          process_xref_streams(doc, xref_streams)
        else
          doc.each(only_current: false, &method(:delete_fields_with_defaults))
        end
      end

      # Compacts the document by merging all revisions into one, deleting null and unused entries
      # and renumbering the objects.
      #
      # For the meaning of the other arguments see ::call.
      def self.compact(doc, object_streams, xref_streams)
        doc.revisions.merge
        unused = Set.new(doc.task(:dereference))
        rev = doc.revisions.add

        oid = 1
        doc.revisions[0].each do |obj|
          if obj.null? || unused.include?(obj) || (obj.type == :ObjStm) ||
              (obj.type == :XRef && xref_streams != :preserve)
            obj.data.value = nil
            next
          end

          delete_fields_with_defaults(obj)
          obj.oid = oid
          obj.gen = 0
          rev.add(obj)
          oid += 1
        end
        doc.revisions.delete(0)

        if object_streams == :generate
          process_object_streams(doc, :generate, xref_streams)
        elsif xref_streams == :generate
          doc.add({Type: :XRef})
        end
      end

      # Processes the object streams in each revision according to method: For :preserve, nothing
      # is done, for :delete all object streams are deleted and for :generate objects are packed
      # into object streams as much as possible.
      def self.process_object_streams(doc, method, xref_streams)
        case method
        when :delete
          doc.revisions.each_with_index do |rev, rev_index|
            xref_stream = false
            objects_to_delete = []
            rev.each do |obj|
              case obj.type
              when :ObjStm
                objects_to_delete << obj
              when :XRef
                xref_stream = true
                objects_to_delete << obj if xref_streams == :delete
              else
                delete_fields_with_defaults(obj)
              end
            end
            objects_to_delete.each {|obj| rev.delete(obj) }
            if xref_streams == :generate && !xref_stream
              doc.add({Type: :XRef}, revision: rev_index)
            end
          end
        when :generate
          doc.revisions.each_with_index do |rev, rev_index|
            xref_stream = false
            count = 0
            objstms = [doc.wrap({Type: :ObjStm})]
            old_objstms = []
            rev.each do |obj|
              case obj.type
              when :XRef
                xref_stream = true
              when :ObjStm
                old_objstms << obj
              end
              delete_fields_with_defaults(obj)

              next if obj.respond_to?(:stream)

              objstms[-1].add_object(obj)
              count += 1
              if count == 200
                objstms << doc.wrap({Type: :ObjStm})
                count = 0
              end
            end
            old_objstms.each {|objstm| rev.delete(objstm) }
            objstms.each {|objstm| doc.add(objstm, revision: rev_index) }
            doc.add({Type: :XRef}, revision: rev_index) unless xref_stream
          end
        end
      end

      # Processes the cross-reference streams in each revision according to method: For :preserve,
      # nothing is done, for :delete all cross-reference streams are deleted and for :generate
      # cross-reference streams are added.
      def self.process_xref_streams(doc, method)
        case method
        when :delete
          doc.each(only_current: false) do |obj, rev|
            if obj.type == :XRef
              rev.delete(obj)
            else
              delete_fields_with_defaults(obj)
            end
          end
        when :generate
          doc.revisions.each_with_index do |rev, rev_index|
            xref_stream = false
            rev.each do |obj|
              xref_stream = true if obj.type == :XRef
              delete_fields_with_defaults(obj)
            end
            doc.add({Type: :XRef}, revision: rev_index) unless xref_stream
          end
        end
      end

      # Deletes field entries of the object that are optional and currently set to their default
      # value.
      def self.delete_fields_with_defaults(obj)
        return unless obj.kind_of?(HexaPDF::Dictionary) && !obj.null?
        obj.each do |name, value|
          if (field = obj.class.field(name)) && !field.required? && field.default? &&
              value == field.default
            obj.delete(name)
          end
        end
      end

      # Compresses the contents of all pages by parsing and then serializing again. The HexaPDF
      # serializer is already optimized for small output size so nothing else needs to be done.
      #
      # Returns a hash of the form key=>true where the keys are the used XObjects (for use with
      # #prune_page_resources).
      def self.compress_pages(doc)
        used_refs = {}
        doc.pages.each do |page|
          processor = SerializationProcessor.new
          HexaPDF::Content::Parser.parse(page.contents, processor)
          page.contents = processor.result
          page[:Contents].set_filter(:FlateDecode)
          xobjects = page.resources[:XObject]
          processor.used_references.each {|ref| used_refs[xobjects[ref]] = true }
        end
        used_refs
      end

      # Deletes all XObject entries from the resources dictionaries of all pages whose names do not
      # match the keys in +used_refs+.
      def self.prune_page_resources(doc, used_refs)
        unless used_refs
          used_refs = {}
          doc.pages.each do |page|
            xobjects = page.resources[:XObject]
            HexaPDF::Content::Parser.parse(page.contents) do |op, operands|
              used_refs[xobjects[operands[0]]] = true if op == :Do
            end
          end
        end

        doc.pages.each do |page|
          xobjects = page.resources[:XObject]
          xobjects.each do |key, obj|
            next if used_refs[obj]
            xobjects.delete(key)
          end
        end
      end

      # This processor is used when compressing pages.
      class SerializationProcessor #:nodoc:

        attr_reader :result #:nodoc:

        # Contains all found references
        attr_reader :used_references

        def initialize #:nodoc:
          @result = ''.b
          @serializer = HexaPDF::Serializer.new
          @used_references = []
        end

        def process(op, operands) #:nodoc:
          @result << HexaPDF::Content::Operator::DEFAULT_OPERATORS[op].
            serialize(@serializer, *operands)
          @used_references << operands[0] if op == :Do
        end

      end

    end

  end
end