Add outline entries for pages in existing PDF files

README.md

@@ -104,6 +104,70 @@ The << operator defaults to secure injection by renaming references to avoid con
 pdf.pages(nil, false).each {|page| page << stamp_page}
 ```
 
+## Add outline entries pointing to existing pages
+
+You can add outline entries pointing to pages loaded or parsed from existing PDF files. The outlines dictionary allows you to create tree hierarchy.
+
+To create outline entries for PDF files:
+
+```ruby
+pdf = CombinePDF.new
+i = 1
+CombinePDF.load("file.pdf").pages.each do |page|
+  pdf.add_outline_item(page, "Page #{i}")
+  pdf << page
+  i += 1
+end
+pdf.save "outlines.pdf"
+```
+
+You also can create a tree hierarchy:
+
+```ruby
+pdf = CombinePDF.new
+i = 1
+CombinePDF.load("file.pdf").pages.each do |page|
+  if i.eql? 1
+    pdf.add_outline_grouper(page, 'Section 1')
+    pdf.add_outline_grouper(page, 'Subsection 1.1')
+  elsif i.eql? 3
+    pdf.go_out_outline_grouping_level
+    pdf.add_outline_grouper(page, 'Subsection 1.2')
+  elsif i.eql? 4
+    pdf.go_outline_root
+    pdf.add_outline_grouper(page, 'Section 2')
+  elsif i.eql? 5
+    pdf.go_out_outline_grouping_level
+  end
+
+  pdf.add_outline_item(page, "Page #{i}")
+
+  pdf << page
+  i += 1
+end
+pdf.save "outlines.pdf"
+```
+
+This will generate something like:
+
+```
+Section 1
+  Subsection 1.1
+    Page 1
+    Page 2
+  Subsection 1.2
+    Page 3
+Section 2
+  Page 4
+Page 5
+Page 6
+```
+
+Notice that if you add an outline_grouper (`add_outline_grouper`) all the future outline entries (`add_outline_item`) will be added under the last outline_grouper until you manually exit the grouping level (`go_out_grouping_level` and `go_outline_root`).
+
+`go_out_grouping_level`: exit one level in the tree hierarchy.
+`go_outline_root`: exit to the root in the tree hierarchy.
+
 ## Page Numbering
 
 adding page numbers to a PDF object or file is as simple as can be:

lib/combine_pdf/pdf_protected.rb

@@ -405,5 +405,136 @@ def rename_object(object, _dictionary)
       when Hash
       end
     end
+
+    # @private
+    # This method runs the process to add a new outline entry to the current
+    # (referenced) outline grouper (the grouper is the parent in the tree
+    # hierarchy). This method take 2 parameters:
+    #
+    # page:: the page object to which the outline will point.
+    # title:: the title for the outline.
+    def add_outline_node(page, title)
+      new_outline = new_outline_node(page, title)
+      insert_outline_node(new_outline)
+      update_children_count(actual_object(new_outline)[:Parent])
+      new_outline
+    end
+
+    # @private
+    # This method generates and returns a new outline object. This method takes
+    # 2 parameters:
+    #
+    # page:: the page to which the outline will point.
+    # title:: the title for the outline.
+    def new_outline_node(page, title)
+      {
+        is_reference_only: true,
+        referenced_object: {
+          Count: 0,
+          Title: title,
+          Dest: [
+            { is_reference_only: true, referenced_object: page },
+            :XYZ, nil, nil, nil
+          ],
+          Parent: {
+            is_reference_only: true,
+            referenced_object: @current_outline_grouper
+          }
+        }
+      }
+    end
+
+    # @private
+    # This method inserts a new outline node to the current (referenced)
+    # outline grouper (the grouper is the parent in the tree hierarchy). This
+    # method takes 1 parameter:
+    #
+    # outline_node:: the outline node to be inserted in the current outline grouper
+    def insert_outline_node(outline_node)
+      if outline_grouper_without_children?
+        insert_first_outline_child(outline_node)
+      elsif outline_grouper_with_only_one_child?
+        insert_second_outline_child(outline_node)
+      else
+        insert_last_outline_child(outline_node)
+      end
+    end
+
+    # @private
+    # This method inserts the first outline node in the current (referenced)
+    # outline grouper (the grouper is the parent in the tree hierarchy). This
+    # method takes 1 parameter:
+    #
+    # outline_node:: the outline node to be inserted in the current outline grouper
+    def insert_first_outline_child(outline_node)
+      @current_outline_grouper[:First] = outline_node
+      @current_outline_grouper[:Last] = outline_node
+    end
+
+    # @private
+    # This method inserts the second outline node in the current (referenced)
+    # outline grouper (the grouper is the parent in the tree hierarchy). This
+    # method takes 1 parameter:
+    #
+    # outline_node:: the outline node to be inserted in the current outline grouper
+    def insert_second_outline_child(outline_node)
+      actual_object(@current_outline_grouper[:First])[:Next] = outline_node
+      actual_object(outline_node)[:Prev] = @current_outline_grouper[:First]
+      @current_outline_grouper[:Last] = outline_node
+    end
+
+    # @private
+    # This method inserts one more outline node in the current (referenced)
+    # outline grouper (the grouper is the parent in the tree hierarchy), this
+    # means that the current grouper has more than 1 outline-child node. This
+    # method takes 1 parameter:
+    #
+    # outline_node:: the outline node to be inserted in the current outline grouper
+    def insert_last_outline_child(outline_node)
+      actual_object(@current_outline_grouper[:Last])[:Next] = outline_node
+      actual_object(outline_node)[:Prev] = @current_outline_grouper[:Last]
+      @current_outline_grouper[:Last] = outline_node
+    end
+
+    # @private
+    # This method is executed recursively to update the children count for the
+    # ascendant parents in the tree hierarchy of the outlines. This method takes
+    # 1 parameter:
+    #
+    # outline_grouper:: the outlien grouper to be updated in its children count.
+    def update_children_count(outline_grouper)
+      if actual_object(outline_grouper)[:Count].nil?
+        actual_object(outline_grouper)[:Count] = 0
+      end
+      actual_object(outline_grouper)[:Count] += 1
+      return if outline_root?(outline_grouper)
+
+      update_children_count(actual_object(outline_grouper)[:Parent])
+    end
+
+    # @private
+    # This method checks if the received outline node is the outline root of
+    # PDF document. This method takes 1 parameter:
+    #
+    # outline_node:: the outline object to be evaluated.
+    def outline_root?(outline_node)
+      actual_object(outline_node)[:Parent].nil?
+    end
+
+    # @private
+    # This method returns true if the current (referenced) outline grouper (the
+    # grouper is the parent in the tree hierarchy) has no outline-children
+    # nodes.
+    def outline_grouper_without_children?
+      @current_outline_grouper.exclude?(:First)
+    end
+
+    # @private
+    # This method returns true if the current (referenced) outline grouper (the
+    # grouper is the parent in the tree hierarchy) has only one outline-child
+    # node.
+    def outline_grouper_with_only_one_child?
+      @current_outline_grouper[:First].eql?(@current_outline_grouper[:Last])
+    end
   end
 end

lib/combine_pdf/pdf_public.rb

@@ -104,6 +104,7 @@ def initialize(parser = nil)
       @names = parser.names_object || {}
       @forms_data = parser.forms_object || {}
       @outlines = parser.outlines_object || {}
+      @current_outline_grouper = @outlines
       # rebuild the catalog, to fix wkhtmltopdf's use of static page numbers
       rebuild_catalog
 
@@ -503,5 +504,68 @@ def stamp_pages(stamp, options = {})
     # 	end
     # 	nil
     # end
+
+    # This method adds a new node to the Outlines dictionary and references it
+    # as the current outline grouper, this means that new feature outline nodes
+    # are going to be added as children of this one. This method takes
+    # 2 parameters:
+    #
+    # page:: the page to which the outline will point.
+    # title:: the title for the outline.
+    def add_outline_grouper(page, title)
+      # The page param must be a Hash "Page" object
+      unless page.is_a?(Hash) && actual_object(page)[:Type] == :Page
+        warn "Shouldn't point object from outline unless it is a PDF page."
+        return false
+      end
+
+      # The title param must be a string object
+      unless title.is_a?(String)
+        warn 'Title for outline should be a String object'
+        return false
+      end
+
+      # Reference the new outline node as the current outline grouper in
+      # the tree hierarchy
+      @current_outline_grouper = actual_object(add_outline_node(page, title))
+    end
+
+    # This method adds a new node to the Outlines dictionary in the current
+    # outline grouper. This method takes 2 parameters:
+    #
+    # page:: the page to which the outline will point.
+    # title:: the title for the outline.
+    def add_outline_item(page, title)
+      # The page param must be a Hash "Page" object
+      unless page.is_a?(Hash) && actual_object(page)[:Type] == :Page
+        warn "Shouldn't point object from outline unless it is a PDF page."
+        return false
+      end
+
+      # The title param must be a string object
+      unless title.is_a?(String)
+        warn 'Title for outline should be a String object'
+        return false
+      end
+
+      add_outline_node(page, title)
+    end
+
+    # This method takes the current outline grouper out one level in the tree
+    # hierarchy of the Outlines.
+    def go_out_outline_grouping_level
+      return if @current_outline_grouper.nil? || outline_root?(@current_outline_grouper)
+
+      @current_outline_grouper = actual_object(@current_outline_grouper[:Parent])
+    end
+
+    # This method takes the current outline grouper to the root level in the
+    # tree hierarchy of the Outlines.
+    def go_outline_root
+      return if @current_outline_grouper.nil? || outline_root?(@current_outline_grouper)
+
+      go_out_outline_grouping_level
+      go_outline_root
+    end
   end
 end