hexapdf 1.11.1 → 1.11.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 8cf7d72b14ba19f840f63adbcfa3eb5b1fd7b876731c52c615788f395d23d10d
-  data.tar.gz: 48d506505dcc50d5cc5c596f7e11b8d0961dbea82cea7223c1ae8d705bb9b57e
+  metadata.gz: 3167ea077cd1778bb88e07ea026c1dfc58585624b79130f38ae995ca7829077e
+  data.tar.gz: 534b414e43ce0a459a75d1b89b05793480cfc4ceb1537855a28b0bac32a46d51
 SHA512:
-  metadata.gz: 47be93a799dd59bcab3fcb92c150a0e1b43b8b115dc4d1b38fc029907222a856dde0e717309c85c312fbd3351f0483ccd50f72194f5fce1352eeb1330d92ef39
-  data.tar.gz: 154edc9c8bf7f2e4f3e354c38c5b52a6f1099760ace8d86c4feb38e0092d0dd0ee3630f615196890e98db339204e636b43fdaac9dd2884b21b36780b59da7bde
+  metadata.gz: 5d70d58918d199a7a48090e1d57ec9a8ce88ad69f53ab5935c157bb526c034f597716e566bcf2d1b969f0edaaf0c54a4baa2883e4e9953ca79dcfa4ac68c7b59
+  data.tar.gz: 425b4a95c457e909e6bb31eaee868f23fc329472e1e068222f9dc5c73c9745a023b46c741644b93fd4433bdd740c9259f5614dd481298a8aa4f18e244799020f

data/CHANGELOG.md

@@ -1,3 +1,20 @@
+## 1.11.2 - 2026-10-20
+
+### Added
+
+* [HexaPDF::Type::OptionalContentMembership#ocgs] for retrieving the list of
+  optional content groups referenced by the membership dictionary
+
+### Fixed
+
+* Regression in [HexaPDF::Task::ImportPages] concerning files with shared
+  resources
+* [HexaPDF::Type::Catalog#acro_form] to treat a malformed /AcroForm value
+  like a missing AcroForm
+* [HexaPDF::Task::MergeAcroForm] for PDFs with no default resource dictionary
+* [HexaPDF::Layout::Frame] to work when degenerate polygons are in play
+
+
 ## 1.11.1 - 2026-10-06
 
 ### Changed

data/lib/hexapdf/layout/frame.rb

@@ -438,6 +438,11 @@ module HexaPDF
         # Just use the second top-most segment
         # TODO: not the optimal solution!
         index = segments.rindex {|s| s.start_point.y < @y }
+        unless index # degenerate case where all segments are on the same horizontal line
+          @x = @y = @available_width = @available_height = 0
+          return
+        end
+
         segment = segments[index]
         y = segment.start_point.y
         polygon = if segment.min.x == @x
@@ -477,7 +482,7 @@ module HexaPDF
       # Returns the horizontal segments of the frame's shape, sorted by maximum y-, then minimum
       # x-coordinate.
       def sorted_horizontal_segments
-        @shape.each_segment.select(&:horizontal?).sort! do |a, b|
+        @shape.each_segment.select {|s| s.horizontal? && !s.degenerate? }.sort! do |a, b|
           if a.start_point.y == b.start_point.y
             b.start_point.x <=> a.start_point.x
           else

data/lib/hexapdf/task/import_pages.rb

@@ -138,14 +138,16 @@ module HexaPDF
       def self.preserve_ocgs(doc, source, pages)
         # Find all OCGs used on all pages
         ocgs = Set.new
+        seen_resources = Set.new
         process_ocg_or_ocmd = lambda do |obj|
           if obj.type == :OCG
             ocgs << obj
           elsif obj.type == :OCMD
-            ocgs.merge(obj[:OCGs].to_ary)
+            ocgs.merge(obj.ocgs)
           end
         end
         process_page_or_form = lambda do |page_or_form|
+          next unless seen_resources.add?(page_or_form.resources.data) # handle shared resources dict
           page_or_form.resources[:Properties]&.each do |name, obj|
             process_ocg_or_ocmd.call(obj) if obj
           end
@@ -154,17 +156,11 @@ module HexaPDF
             process_page_or_form.call(obj) if obj[:Subtype] == :Form
           end
         end
-        seen_resources = {}
         pages.each do |page|
-          unless seen_resources[page.resources] # handle case when pages share the resources dict
-            process_page_or_form.call(page)
-          end
-
+          process_page_or_form.call(page)
           page.each_annotation do |annot|
             process_ocg_or_ocmd.call(annot[:OC]) if annot.key?(:OC)
           end
-
-          seen_resources[page.resources] = true
         end
 
         return if ocgs.empty?

data/lib/hexapdf/task/merge_acro_form.rb

@@ -128,7 +128,7 @@ module HexaPDF
         font_name_mapping = {}
         serializer = HexaPDF::Serializer.new
 
-        source_form.default_resources[:Font].each do |font_name, value|
+        source_form.default_resources[:Font]&.each do |font_name, value|
           new_name = target_resources.add_font(target_form.document.import(value))
           font_name_mapping[serializer.serialize(font_name)] = serializer.serialize(new_name)
         end

data/lib/hexapdf/type/catalog.rb

@@ -133,16 +133,16 @@ module HexaPDF
 
       # Returns the main AcroForm object.
       #
-      # * If an AcroForm object exists, the +create+ argument is not used.
+      # * If a valid AcroForm object exists, it is returned regardless of the +create+ argument.
       #
-      # * If no AcroForm object exists and +create+ is +true+, a new AcroForm object with default
-      #   settings will be created and returned.
+      # * If no valid AcroForm object exists and +create+ is +true+, a new AcroForm object with
+      #   default settings will be created and returned.
       #
       # * If no AcroForm object exists and +create+ is +false+, +nil+ is returned.
       #
       # See: AcroForm::Form
       def acro_form(create: false)
-        if (form = self[:AcroForm])
+        if (form = self[:AcroForm]).kind_of?(Dictionary)
           form
         elsif create
           form = self[:AcroForm] = document.add({}, type: :XXAcroForm)

data/lib/hexapdf/type/optional_content_membership.rb

@@ -57,6 +57,19 @@ module HexaPDF
                    allowed_values: [:AllOn, :AnyOn, :AnyOff, :AllOff]
       define_field :VE,   type: PDFArray
 
+      # Returns an array of OCGs whose states determine the visibility of content controlled by this
+      # membership dictionary.
+      def ocgs
+        if (ve = self[:VE])
+          ocgs = ve.to_ary.flatten.map! {|entry| document.deref(entry) }
+          ocgs.reject! {|entry| entry.kind_of?(Symbol) || entry.nil? || entry.null? }
+          ocgs.uniq!
+          ocgs
+        else
+          Array(self[:OCGs]).compact.uniq
+        end
+      end
+
     end
 
   end

data/lib/hexapdf/type/struct_tree_root.rb

@@ -54,7 +54,7 @@ module HexaPDF
       define_field :ParentTreeNextKey,     type: Integer
       define_field :RoleMap,               type: Dictionary
       define_field :ClassMap,              type: Dictionary
-      define_field :Namespaces,            type: PDFArray
+      define_field :Namespaces,            type: PDFArray, version: '2.0'
       define_field :PronounciationLexicon, type: PDFArray, version: '2.0'
       define_field :AF,                    type: PDFArray, version: '2.0'
 

data/lib/hexapdf/version.rb

@@ -37,6 +37,6 @@
 module HexaPDF
 
   # The version of HexaPDF.
-  VERSION = '1.11.1'
+  VERSION = '1.11.2'
 
 end

data/test/hexapdf/layout/test_frame.rb

@@ -502,6 +502,22 @@ describe HexaPDF::Layout::Frame do
       check_regions(frame, [[0, 100, 30, 100], [0, 60, 100, 60]])
     end
 
+    # o--------+
+    #
+    # Such a degenerate case can occur under special circumstances with e.g. a ListBox:
+    #
+    #   composer.list(height: 10, style: {padding: 5}) do |list|
+    #     list.text('test')
+    #   end
+    #
+    # Padding exactly fills the box, leaving zero height, and the algorithm in ListBox creates a
+    # degenerate Polygon.
+    it "works for a flat region with no height" do
+      shape = Geom2D::Polygon([0, 0], [100, 0], [100, 0], [0, 0])
+      frame = HexaPDF::Layout::Frame.new(0, 0, 100, 0, shape: shape)
+      check_regions(frame, [[0, 0, 100, 0]])
+    end
+
     # o---+     +--+
     # |   |  +--+  |
     # |   +--+     |

data/test/hexapdf/task/test_import_pages.rb

@@ -100,6 +100,7 @@ describe HexaPDF::Task::ImportPages do
       form = canvas.form
       form[:OC] = @ocg1
       form.canvas.optional_content(@ocmd)
+      form.resources.add_xobject(form)
       canvas.xobject(form, at: [0, 0])
       @target.task(:import_pages, source: @doc)
       assert_equal(['OCG', 'OCMD'], @target.optional_content.ocgs.map(&:name))

data/test/hexapdf/task/test_merge_acro_form.rb

@@ -47,6 +47,12 @@ describe HexaPDF::Task::MergeAcroForm do
     assert(@doc.acro_form.default_resources[:Font].key?(:F2))
   end
 
+  it "handles AcroForm dictionaries without /DR entry" do
+    @doc1.acro_form.delete(:DR)
+    @doc.task(:merge_acro_form, source: @doc1, pages: @pages)
+    refute(@doc.acro_form.default_resources[:Font].key?(:F2))
+  end
+
   it "updates the /SigFlags if necessary" do
     @doc.task(:merge_acro_form, source: @doc1, pages: [@pages[0]])
     refute(@doc.acro_form.signature_flag?(:signatures_exist))

data/test/hexapdf/test_document.rb

@@ -498,8 +498,14 @@ describe HexaPDF::Document do
   describe "acro_form" do
     it "returns the main AcroForm object" do
       assert_nil(@doc.acro_form)
+
+      @doc.acro_form(create: true)
+      refute_nil(@doc.acro_form)
+    end
+
+    it "treats a malformed /AcroForm value like a missing AcroForm" do
       @doc.catalog[:AcroForm] = 5
-      assert_equal(5, @doc.acro_form)
+      assert_nil(@doc.acro_form)
     end
 
     it "create the AcroForm object if instructed" do

data/test/hexapdf/type/test_catalog.rb

@@ -62,19 +62,32 @@ describe HexaPDF::Type::Catalog do
 
   describe "acro_form" do
     it "returns an existing form object" do
-      @catalog[:AcroForm] = :test
-      assert_equal(:test, @catalog.acro_form)
+      @catalog[:AcroForm] = {}
+      form = @catalog[:AcroForm]
+      assert_same(form, @catalog.acro_form)
+    end
+
+    it "creates a new AcroForm object with defaults if create: true" do
+      form = @catalog.acro_form(create: true)
+      assert_kind_of(HexaPDF::Type::AcroForm::Form, form)
+      assert(form[:DA])
     end
 
     it "returns an existing form object even if create: true" do
+      form = @catalog.acro_form(create: true)
+      assert_same(form, @catalog.acro_form(create: true))
+    end
+
+    it "treats an invalid /AcroForm value like a missing AcroForm" do
       @catalog[:AcroForm] = :test
-      assert_equal(:test, @catalog.acro_form(create: true))
+      assert_nil(@catalog.acro_form)
     end
 
-    it "creates a new AcroForm object with defaults if create: true" do
+    it "replaces an invalid /AcroForm value if create: true" do
+      @catalog[:AcroForm] = :test
       form = @catalog.acro_form(create: true)
       assert_kind_of(HexaPDF::Type::AcroForm::Form, form)
-      assert(form[:DA])
+      assert_same(form, @catalog.acro_form)
     end
   end
 

data/test/hexapdf/type/test_optional_content_membership.rb

@@ -0,0 +1,40 @@
+# -*- encoding: utf-8 -*-
+
+require 'test_helper'
+require 'hexapdf/type/optional_content_membership'
+require 'hexapdf/document'
+
+describe HexaPDF::Type::OptionalContentMembership do
+  before do
+    @doc = HexaPDF::Document.new
+    @ocmd = @doc.add({Type: :OCMD})
+  end
+
+  describe "ocgs" do
+    before do
+      @ocg = @doc.add({Type: :OCG, Name: "test"})
+    end
+
+    it "uses the /OCGs entry if /VE is not available" do
+      assert_equal([], @ocmd.ocgs)
+
+      @ocmd[:OCGs] = []
+      assert_equal([], @ocmd.ocgs)
+
+      @ocmd[:OCGs] = @ocg
+      assert_equal([@ocg], @ocmd.ocgs)
+
+      null = @doc.add({})
+      null.data.value = nil
+      @ocmd[:OCGS] = [@ocg, nil, null, @ocg]
+      assert_equal([@ocg], @ocmd.ocgs)
+    end
+
+    it "uses the /VE entry if available" do
+      null = @doc.add({})
+      null.data.value = nil
+      @ocmd[:VE] = [:Or, @ocg, nil, [:Not, @ocg], [:And, @ocg, [:Or, null, @ocg]]]
+      assert_equal([@ocg], @ocmd.ocgs)
+    end
+  end
+end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: hexapdf
 version: !ruby/object:Gem::Version
-  version: 1.11.1
+  version: 1.11.2
 platform: ruby
 authors:
   - Thomas Leitner
@@ -922,6 +922,7 @@ files:
   - test/hexapdf/type/test_object_stream.rb
   - test/hexapdf/type/test_optional_content_configuration.rb
   - test/hexapdf/type/test_optional_content_group.rb
+  - test/hexapdf/type/test_optional_content_membership.rb
   - test/hexapdf/type/test_optional_content_properties.rb
   - test/hexapdf/type/test_outline.rb
   - test/hexapdf/type/test_outline_item.rb