hexapdf 0.32.0 → 0.32.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 3c74f0e152b612b628321f312413a0eb88e52a41a809f7b37e993acb9e0436b2
-  data.tar.gz: 9dfe8cfaaa10527695b70d9019636ed9cc6059367125a45a3dacde7bc1d8c3bb
+  metadata.gz: f0907362b897b39d61fa8f859510ccfb31ebd5ad88b94a9e36ff62c62fae813a
+  data.tar.gz: 469465941253bad5e769fc0dd5a57e437c578fdecda52a0fea908370ef8b7169
 SHA512:
-  metadata.gz: f9ee0dbe507a6d31cbf024d2f34e69c15e491bc05a64a0f4039f38bcf6d52192245d6c7d2ac4cb65ae1eb110f8af735afcf13366e15975d96bcc17dda8e2a618
-  data.tar.gz: 846ce0611e5f125011aa0535d2de69c15c356f6164f9b09f0a314f9968e34e4addda5396c03d82cd70dffc96eaa0408e1a6dd7a6fc55ce59646be5c3f8b4f9a6
+  metadata.gz: 2fc77d33ddc2c6fb377a778b82f92f01fed93d9a4e494cd57c97dc0a00fce4baa7cab185776f8a3303507d68b544e4404ac2f33b72c3b1a7cc7c65ee838c7dd7
+  data.tar.gz: 21a08984f424b91594a2d8e6fc705a7f9458a990918ee77adef469e758ae2fee976ba905a7a4a696c5c2728831d02f8845dc9b5316c28de4ff5c7f062521a7e3

data/CHANGELOG.md

@@ -1,3 +1,38 @@
+## 0.32.2 - 2023-05-06
+
+### Changed
+
+* Cross-reference table reconstruction to be more relaxed concerning the
+  `endobj` keyword
+
+### Fixed
+
+* [HexaPDF::Type::ObjectStream] to not compress any encryption dictionary
+  instead of only the current one
+
+
+## 0.32.1 - 2023-04-20
+
+### Added
+
+* [HexaPDF::Type::FontType0#font_descriptor] and
+  [HexaPDF::Type::FontSimple#font_descriptor] for easy access to the font
+  descriptor
+
+### Changed
+
+* [HexaPDF::Content::Canvas#color_from_specification] to allow strings and color
+  objects without a wrapping array
+
+### Fixed
+
+* AES 128bit encryption to include unnecessary field in encryption dictionary to
+  work around buggy PDF libraries
+* [HexaPDF::Layout::Style::LinkLayer] to correctly process the border color
+* [HexaPDF::Type::AcroForm::AppearanceGenerator] to use fallback for font cap
+  height value when necessary
+
+
 ## 0.32.0 - 2023-03-08
 
 ### Added

data/lib/hexapdf/cli.rb

@@ -62,6 +62,10 @@ module HexaPDF
       Application.new.parse(args)
     rescue StandardError => e
       $stderr.puts "Problem encountered: #{e.message}"
+      unless e.kind_of?(HexaPDF::Error)
+        $stderr.puts "--> The problem might indicate a faulty PDF or a bug in HexaPDF."
+        $stderr.puts "--> Please report this at https://github.com/gettalong/hexapdf/issues - thanks!"
+      end
       exit(1)
     end
 

data/lib/hexapdf/content/canvas.rb

@@ -2363,6 +2363,7 @@ module HexaPDF
       # This utility method is meant for use by higher-level methods that need to convert a color
       # specification into a color object for this Canvas object.
       def color_from_specification(spec)
+        spec = Array(spec)
         if spec.length == 1 && spec[0].kind_of?(String)
           ColorSpace.device_color_from_specification(spec)
         elsif spec.length == 1 && spec[0].respond_to?(:color_space)

data/lib/hexapdf/encryption/security_handler.rb

@@ -358,7 +358,7 @@ module HexaPDF
             raise(HexaPDF::UnsupportedEncryptionError,
                   "Invalid key length #{key_length} specified")
           end
-        dict[:Length] = key_length if dict[:V] == 2
+        dict[:Length] = key_length if dict[:V] == 4 || dict[:V] == 2
 
         if ![:aes, :arc4].include?(algorithm)
           raise(HexaPDF::UnsupportedEncryptionError,

data/lib/hexapdf/layout/style.rb

@@ -508,6 +508,12 @@ module HexaPDF
                          *matrix.evaluate(box.width, box.height), *matrix.evaluate(0, box.height)]
           x_minmax = quad_points.values_at(0, 2, 4, 6).minmax
           y_minmax = quad_points.values_at(1, 3, 5, 7).minmax
+          border_color = case @border_color
+                         when [], nil
+                           @border_color
+                         else
+                           canvas.color_from_specification(@border_color).components
+                         end
           annot = {
             Subtype: :Link,
             Rect: [x_minmax[0], y_minmax[0], x_minmax[1], y_minmax[1]],
@@ -515,7 +521,7 @@ module HexaPDF
             Dest: @dest,
             A: @action,
             Border: @border,
-            C: @border_color && canvas.color_from_specification(@border_color).components,
+            C: border_color,
           }
           (page[:Annots] ||= []) << page.document.add(annot)
         end

data/lib/hexapdf/parser.rb

@@ -458,7 +458,7 @@ module HexaPDF
               linearized = obj.kind_of?(Hash) && obj.key?(:Linearized)
               @tokenizer.pos = pos
             end
-            @tokenizer.scan_until(/(?:\n|\r\n?)endobj\b/)
+            @tokenizer.scan_until(/\bendobj\b/)
           end
         elsif token.kind_of?(Tokenizer::Token) && token == 'trailer'
           obj = @tokenizer.next_object rescue nil

data/lib/hexapdf/type/acro_form/appearance_generator.rb

@@ -395,7 +395,9 @@ module HexaPDF
 
           # Adobe seems to be vertically centering based on the cap height, if enough space is
           # available
-          cap_height = style.font.wrapped_font.cap_height * style.font.scaling_factor / 1000.0 *
+          tmp_cap_height = style.font.wrapped_font.cap_height ||
+            style.font.pdf_object.font_descriptor&.[](:CapHeight)
+          cap_height = tmp_cap_height * style.font.scaling_factor / 1000.0 *
             style.font_size
           y = padding + (height - 2 * padding - cap_height) / 2.0
           y = padding - style.scaled_font_descender if y < 0

data/lib/hexapdf/type/font_simple.rb

@@ -53,6 +53,13 @@ module HexaPDF
       define_field :FontDescriptor, type: :FontDescriptor, indirect: true
       define_field :Encoding, type: [Dictionary, Symbol]
 
+      # Returns the font descriptor. May be +nil+ for a standard 14 font.
+      #
+      # The font descriptor is required except for the standard 14 fonts in PDF version up to 1.7.
+      def font_descriptor
+        self[:FontDescriptor]
+      end
+
       # Returns the encoding object used for this font.
       #
       # Note that the encoding is cached internally when accessed the first time.

data/lib/hexapdf/type/font_type0.rb

@@ -63,6 +63,11 @@ module HexaPDF
         end
       end
 
+      # Returns the font descriptor of the descendant font.
+      def font_descriptor
+        descendant_font[:FontDescriptor]
+      end
+
       # Returns the writing mode which is either :horizontal or :vertical.
       def writing_mode
         cmap.wmode == 0 ? :horizontal : :vertical

data/lib/hexapdf/type/object_stream.rb

@@ -172,13 +172,16 @@ module HexaPDF
         serializer = Serializer.new
         obj_to_stm = {}
 
-        encrypt_dict = document.trailer[:Encrypt]
+        is_encrypt_dict = document.revisions.each.with_object({}) do |rev, hash|
+          hash[rev.trailer[:Encrypt]] = true
+        end
         while index < objects.size / 2
           obj = revision.object(objects[index])
 
           # Due to a bug in Adobe Acrobat, the Catalog may not be in an object stream if the
           # document is encrypted
-          if obj.nil? || obj.null? || obj.gen != 0 || obj.kind_of?(Stream) || obj == encrypt_dict ||
+          if obj.nil? || obj.null? || obj.gen != 0 || obj.kind_of?(Stream) ||
+              is_encrypt_dict[obj] ||
               obj.type == :Catalog ||
               obj.type == :Sig || obj.type == :DocTimeStamp ||
               (obj.respond_to?(:key?) && obj.key?(:ByteRange) && obj.key?(:Contents))

data/lib/hexapdf/version.rb

@@ -37,6 +37,6 @@
 module HexaPDF
 
   # The version of HexaPDF.
-  VERSION = '0.32.0'
+  VERSION = '0.32.2'
 
 end

data/test/hexapdf/content/test_canvas.rb

@@ -1280,4 +1280,30 @@ describe HexaPDF::Content::Canvas do
       end
     end
   end
+
+  describe "color_from_specification "do
+    it "accepts a color string" do
+      assert_equal([1, 0, 0], @canvas.color_from_specification("red").components)
+    end
+
+    it "accepts a color string wrapped in an array" do
+      assert_equal([1, 0, 0], @canvas.color_from_specification(["red"]).components)
+    end
+
+    it "accepts a color object" do
+      color = @canvas.color_from_specification("red")
+      assert_equal(color, @canvas.color_from_specification(color))
+    end
+
+    it "accepts a color object wrapped in an array" do
+      color = @canvas.color_from_specification("red")
+      assert_equal(color, @canvas.color_from_specification([color]))
+    end
+
+    it "accepts an array with 1, 3, or 4 color values" do
+      assert_equal([1], @canvas.color_from_specification([255]).components)
+      assert_equal([1, 0, 0], @canvas.color_from_specification([255, 0, 0]).components)
+      assert_equal([1, 0, 0, 0], @canvas.color_from_specification([100, 0, 0, 0]).components)
+    end
+  end
 end

data/test/hexapdf/encryption/test_security_handler.rb

@@ -141,6 +141,11 @@ describe HexaPDF::Encryption::SecurityHandler do
         @handler.set_up_encryption(key_length: key_length, algorithm: algorithm)
         assert(result == @handler.dict[:Length])
       end
+
+      # Work-around buggy software
+      @handler.set_up_encryption(key_length: 128, algorithm: :aes)
+      assert_equal(4, @handler.dict[:V])
+      assert_equal(128, @handler.dict[:Length])
     end
 
     it "calls the prepare_encryption method" do

data/test/hexapdf/layout/test_style.rb

@@ -573,10 +573,19 @@ describe HexaPDF::Layout::Style::LinkLayer do
       assert_equal([0, 0, 1], annot[:Border].value)
     end
 
-    it "uses the specified border and border color" do
-      annot = call_link(dest: true, border: [10, 10, 2], border_color: [255])
+    it "uses the specified border" do
+      annot = call_link(dest: true, border: [10, 10, 2])
       assert_equal([10, 10, 2], annot[:Border].value)
-      assert_equal([1.0], annot[:C].value)
+    end
+
+    it "uses the specified border color" do
+      annot = call_link(dest: true, border_color: "red")
+      assert_equal([1.0, 0, 0], annot[:C].value)
+    end
+
+    it "works when the border color is transparent" do
+      annot = call_link(dest: true, border_color: [])
+      assert_equal([], annot[:C].value)
     end
 
     it "works for simple destinations" do

data/test/hexapdf/test_parser.rb

@@ -609,7 +609,7 @@ describe HexaPDF::Parser do
     end
 
     it "serially parses the contents" do
-      create_parser("1 0 obj\n5\nendobj\n1 0 obj\n6\nendobj\ntrailer\n<</Size 1>>")
+      create_parser("1 0 obj\n5 endobj\n1 0 obj\n6\nendobj\ntrailer\n<</Size 1>>")
       assert_equal(6, @parser.load_object(@xref).value)
     end
 

data/test/hexapdf/test_writer.rb

@@ -40,7 +40,7 @@ describe HexaPDF::Writer do
       219
       %%EOF
       3 0 obj
-      <</Producer(HexaPDF version 0.32.0)>>
+      <</Producer(HexaPDF version 0.32.2)>>
       endobj
       xref
       3 1
@@ -72,7 +72,7 @@ describe HexaPDF::Writer do
       141
       %%EOF
       6 0 obj
-      <</Producer(HexaPDF version 0.32.0)>>
+      <</Producer(HexaPDF version 0.32.2)>>
       endobj
       2 0 obj
       <</Length 10>>stream
@@ -214,7 +214,7 @@ describe HexaPDF::Writer do
       <</Type/Page/MediaBox[0 0 595 842]/Parent 2 0 R/Resources<<>>>>
       endobj
       5 0 obj
-      <</Producer(HexaPDF version 0.32.0)>>
+      <</Producer(HexaPDF version 0.32.2)>>
       endobj
       4 0 obj
       <</Root 1 0 R/Info 5 0 R/Size 6/Type/XRef/W[1 1 2]/Index[0 6]/Filter/FlateDecode/DecodeParms<</Columns 4/Predictor 12>>/Length 33>>stream

data/test/hexapdf/type/acro_form/test_appearance_generator.rb

@@ -519,6 +519,19 @@ describe HexaPDF::Type::AcroForm::AppearanceGenerator do
                            range: 7)
         end
 
+        it "falls back to the cap height in the font descriptor for vertical alignment" do
+          font_metrics = @form.default_resources.font(:F1).font_wrapper.wrapped_font.metrics
+          cap_height = font_metrics.cap_height
+          font_metrics.cap_height = nil
+
+          @generator.create_appearances
+          assert_operators(@widget[:AP][:N].stream,
+                           [:set_text_matrix, [1, 0, 0, 1, 2, 6.41]],
+                           range: 7)
+        ensure
+          font_metrics.cap_height = cap_height
+        end
+
         it "vertically aligns to the font descender if the text is too high" do
           @widget[:Rect].height = 5
           @generator.create_appearances

data/test/hexapdf/type/test_font_simple.rb

@@ -13,15 +13,19 @@ describe HexaPDF::Type::FontSimple do
       <22> <0042>
       endbfchar
     EOF
-    font_descriptor = @doc.add({Type: :FontDescriptor, FontName: :Embedded, Flags: 0b100,
-                                FontBBox: [0, 1, 2, 3], ItalicAngle: 0, Ascent: 900,
-                                MissingWidth: 500, Descent: -100, CapHeight: 800, StemV: 20})
+    @font_descriptor = @doc.add({Type: :FontDescriptor, FontName: :Embedded, Flags: 0b100,
+                                 FontBBox: [0, 1, 2, 3], ItalicAngle: 0, Ascent: 900,
+                                 MissingWidth: 500, Descent: -100, CapHeight: 800, StemV: 20})
     @font = @doc.add({Type: :Font, Encoding: :WinAnsiEncoding,
-                      BaseFont: :Embedded, FontDescriptor: font_descriptor, ToUnicode: cmap,
+                      BaseFont: :Embedded, FontDescriptor: @font_descriptor, ToUnicode: cmap,
                       FirstChar: 32, LastChar: 34, Widths: [600, 0, 700]},
                      type: HexaPDF::Type::FontSimple)
   end
 
+  it "returns the font descriptor of the font" do
+    assert_same(@font_descriptor, @font.font_descriptor)
+  end
+
   describe "encoding" do
     it "fails if /Encoding is absent because encoding_from_font is not implemented" do
       @font.delete(:Encoding)

data/test/hexapdf/type/test_font_type0.rb

@@ -7,8 +7,8 @@ require 'hexapdf/type/font_type0'
 describe HexaPDF::Type::FontType0 do
   before do
     @doc = HexaPDF::Document.new
-    fd = @doc.add({Type: :FontDescriptor, FontBBox: [0, 1, 2, 3]})
-    @cid_font = @doc.wrap({Type: :Font, Subtype: :CIDFontType2, W: [633, [100]], FontDescriptor: fd,
+    @fd = @doc.add({Type: :FontDescriptor, FontBBox: [0, 1, 2, 3]})
+    @cid_font = @doc.wrap({Type: :Font, Subtype: :CIDFontType2, W: [633, [100]], FontDescriptor: @fd,
                            CIDSystemInfo: {Registry: 'Adobe', Ordering: 'Japan1', Supplement: 1}})
     @font = @doc.wrap({Type: :Font, Subtype: :Type0, Encoding: :H, DescendantFonts: [@cid_font]})
   end
@@ -26,6 +26,10 @@ describe HexaPDF::Type::FontType0 do
     assert_equal(@cid_font.value, @font.descendant_font.value)
   end
 
+  it "returns the font descriptor of the descendant font" do
+    assert_same(@fd, @font.font_descriptor)
+  end
+
   it "uses the descendant font for getting the width of a code point" do
     assert_equal(100, @font.width(0x2121))
   end

data/test/hexapdf/type/test_object_stream.rb

@@ -23,9 +23,14 @@ describe HexaPDF::Type::ObjectStream do
   before do
     @doc = Object.new
     @doc.instance_variable_set(:@version, '1.5')
-    def (@doc).trailer
-      @trailer ||= {Encrypt: HexaPDF::Object.new({}, oid: 9)}
+    @doc.define_singleton_method(:revisions) do
+      rev1 = Object.new
+      def rev1.trailer; {Encrypt: HexaPDF::Object.new({}, oid: 10)}; end
+      rev2 = Object.new
+      def rev2.trailer; {Encrypt: HexaPDF::Object.new({}, oid: 9)}; end
+      @revisions ||= [rev1, rev2]
     end
+    @doc.define_singleton_method(:trailer) { revisions.last.trailer }
     @obj = HexaPDF::Type::ObjectStream.new({N: 2, First: 8}, oid: 1, document: @doc,
                                            stream: "1 0 5 2 5 [1 2]")
   end
@@ -96,8 +101,9 @@ describe HexaPDF::Type::ObjectStream do
       assert_equal("", @obj.stream)
     end
 
-    it "doesn't allow the encryption dictionary to be compressed" do
+    it "doesn't allow an encryption dictionary to be compressed" do
       @obj.add_object(@doc.trailer[:Encrypt])
+      @obj.add_object(@doc.revisions[0].trailer[:Encrypt])
       @obj.write_objects(@revision)
       assert_equal(0, @obj.value[:N])
       assert_equal(0, @obj.value[:First])

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: hexapdf
 version: !ruby/object:Gem::Version
-  version: 0.32.0
+  version: 0.32.2
 platform: ruby
 authors:
 - Thomas Leitner
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2023-03-08 00:00:00.000000000 Z
+date: 2023-05-06 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: cmdparse