hexapdf 1.6.0 → 1.8.0

data/test/hexapdf/font/true_type/test_subsetter.rb

@@ -29,12 +29,13 @@ describe HexaPDF::Font::TrueType::Subsetter do
 
   it "doesn't use certain subset glyph IDs for performance reasons" do
     1.upto(93) {|i| @subsetter.use_glyph(i) }
-    # glyph 0, 93 used glyph, 4 special glyphs
-    assert_equal(1 + 93 + 4, @subsetter.instance_variable_get(:@glyph_map).size)
-    1.upto(12) {|i| assert_equal(i, @subsetter.subset_glyph_id(i), "id=#{i}") }
-    13.upto(38) {|i| assert_equal(i + 1, @subsetter.subset_glyph_id(i), "id=#{i}") }
-    39.upto(88) {|i| assert_equal(i + 3, @subsetter.subset_glyph_id(i), "id=#{i}") }
-    89.upto(93) {|i| assert_equal(i + 4, @subsetter.subset_glyph_id(i), "id=#{i}") }
+    # glyph 0 and 93 are used glyph, 5 special glyphs
+    assert_equal(1 + 93 + 5, @subsetter.instance_variable_get(:@glyph_map).size)
+    1.upto(9) {|i| assert_equal(i, @subsetter.subset_glyph_id(i), "id=#{i}") }
+    10.upto(11) {|i| assert_equal(i + 1, @subsetter.subset_glyph_id(i), "id=#{i}") }
+    12.upto(37) {|i| assert_equal(i + 2, @subsetter.subset_glyph_id(i), "id=#{i}") }
+    38.upto(87) {|i| assert_equal(i + 4, @subsetter.subset_glyph_id(i), "id=#{i}") }
+    88.upto(93) {|i| assert_equal(i + 5, @subsetter.subset_glyph_id(i), "id=#{i}") }
   end
 
   it "creates the subset font file" do
@@ -54,18 +55,19 @@ describe HexaPDF::Font::TrueType::Subsetter do
   end
 
   it "correctly subsets compound glyphs" do
-    font_file = "/usr/share/fonts/truetype/dejavu/DejaVuSans.ttf"
+    font_file = "/usr/share/fonts/truetype/noto/NotoSansMono-Regular.ttf"
     skip unless File.exist?(font_file)
 
     begin
       @font = HexaPDF::Font::TrueType::Font.new(File.open(font_file))
       @subsetter = HexaPDF::Font::TrueType::Subsetter.new(@font)
 
-      @subsetter.use_glyph(@font[:cmap].preferred_table['À'.ord])
+      @subsetter.use_glyph(@font[:cmap].preferred_table['ë'.ord])
       subset = HexaPDF::Font::TrueType::Font.new(StringIO.new(@subsetter.build_font))
 
-      assert_equal(4, subset[:maxp].num_glyphs)
+      assert_equal(5, subset[:maxp].num_glyphs)
       assert_equal([2, 3], subset[:glyf][1].components)
+      assert_equal([4], subset[:glyf][3].components)
     ensure
       @font.io.close
     end

data/test/hexapdf/font_loader/test_from_file.rb

@@ -30,6 +30,13 @@ describe HexaPDF::FontLoader::FromFile do
     refute(wrapper.subset?)
   end
 
+  it "raises an error if the provided font does not contain TrueType outlines" do
+    font = HexaPDF::Font::TrueType::Font.new(File.open(@font_file, 'rb'))
+    font.directory.instance_variable_get(:@tables).delete('glyf')
+    exception = assert_raises(HexaPDF::Error) { @klass.call(@doc, font) }
+    assert_match(/does not contain TrueType but CFF/, exception.message)
+  end
+
   it "returns nil if the given name doesn't represent a file" do
     assert_nil(@klass.call(@doc, "Unknown"))
   end

data/test/hexapdf/task/test_import_pages.rb

@@ -0,0 +1,126 @@
+# -*- encoding: utf-8 -*-
+
+require 'test_helper'
+require 'hexapdf/document'
+require 'hexapdf/task/import_pages'
+
+describe HexaPDF::Task::ImportPages do
+  before do
+    @doc = HexaPDF::Document.new
+    @pages = [@doc.pages.add, @doc.pages.add]
+    @pages[0][:Page1] = true
+    @pages[1][:Page2] = true
+
+    @target = HexaPDF::Document.new
+  end
+
+  describe "pages argument" do
+    it "imports all pages by default" do
+      @target.task(:import_pages, source: @doc)
+      assert_equal(2, @target.pages.count)
+      assert(@target.pages[0][:Page1])
+      assert(@target.pages[1][:Page2])
+    end
+
+    it "imports the provided page objects" do
+      @target.task(:import_pages, source: @doc, pages: @pages.reverse)
+      assert_equal(2, @target.pages.count)
+      assert(@target.pages[0][:Page2])
+      assert(@target.pages[1][:Page1])
+    end
+
+    it "imports a single page" do
+      @target.task(:import_pages, source: @doc, pages: 1)
+      assert_equal(1, @target.pages.count)
+      assert(@target.pages[0][:Page2])
+    end
+
+    it "imports a page range" do
+      @target.task(:import_pages, source: @doc, pages: 0..-1)
+      assert_equal(2, @target.pages.count)
+      assert(@target.pages[0][:Page1])
+      assert(@target.pages[1][:Page2])
+    end
+
+    it "imports multiple pages" do
+      @target.task(:import_pages, source: @doc, pages: [1, 0..-1],
+                   ocgs: :ignore, acro_form: :ignore)
+      assert_equal(2, @target.pages.count)
+      assert(@target.pages[0][:Page2])
+      assert(@target.pages[1][:Page1])
+    end
+  end
+
+  it "doesn't append the pages if specified so" do
+    result = @target.task(:import_pages, source: @doc, append: false)
+    assert_equal(0, @target.pages.count)
+    assert_equal(2, result.size)
+    assert(result[0][:Page1])
+    assert(result[1][:Page2])
+  end
+
+  it "merges the AcroForm fields" do
+    form = @doc.acro_form(create: true)
+    field = form.create_text_field("Text")
+    field.create_widget(@doc.pages[0], Rect: [0, 0, 0, 0])
+    @doc.dispatch_message(:complete_objects)
+    @doc.validate
+
+    @target.task(:import_pages, source: @doc)
+    assert_equal(1, @target.acro_form.root_fields.size)
+  end
+
+  describe "ocgs argument" do
+    before do
+      @ocg1 = @doc.optional_content.ocg('OCG')
+      @ocg1.add_to_ui(path: @ocg1)
+      @ocg2 = @doc.optional_content.ocg('OCMD')
+      @ocg2.add_to_ui(path: @ocg2)
+      @ocg2.off!
+      @ocmd = @doc.optional_content.create_ocmd(@ocg2)
+    end
+
+    it "doesn't preserve unused ocgs" do
+      @target.task(:import_pages, source: @doc)
+      assert(@target.optional_content.ocgs.empty?)
+    end
+
+    it "preserves OCGs and OCMDs in content streams" do
+      canvas = @doc.pages[0].canvas
+      canvas.optional_content(@ocg1)
+      canvas.optional_content(@ocmd)
+      @target.task(:import_pages, source: @doc)
+      assert_equal(['OCG', 'OCMD'], @target.optional_content.ocgs.map(&:name))
+      assert(@target.optional_content.ocg('OCG').on?)
+      refute(@target.optional_content.ocg('OCMD').on?)
+    end
+
+    it "preserves OCGs/OCMDs associated with XObjects" do
+      canvas = @doc.pages[0].canvas
+      form = canvas.form
+      form[:OC] = @ocg1
+      canvas.xobject(form, at: [0, 0])
+      @target.task(:import_pages, source: @doc)
+      assert_equal(['OCG'], @target.optional_content.ocgs.map(&:name))
+    end
+
+    it "preserves OCGs/OCMDs associated with annotations" do
+      annot = @doc.annotations.create_line(@doc.pages[0], start_point: [0, 0], end_point: [50, 50])
+      annot[:OC] = @ocmd
+      annot.regenerate_appearance
+      @target.task(:import_pages, source: @doc)
+      assert_equal(['OCMD'], @target.optional_content.ocgs.map(&:name))
+      refute(@target.optional_content.ocg('OCMD').on?)
+    end
+
+    it "preserves the radio button group state of imported OCGs" do
+      @doc.pages[0].canvas.optional_content(@ocg1)
+      @doc.optional_content.default_configuration[:RBGroups] = [[@ocg1, @ocg2]]
+      @target.task(:import_pages, source: @doc)
+      assert_equal(['OCG'], @target.optional_content.ocgs.map {|ocg| ocg.name })
+      rb_groups = @target.optional_content.default_configuration[:RBGroups]
+      assert_equal(1, rb_groups.size)
+      assert_equal(['OCG'], rb_groups[0].map(&:name))
+    end
+  end
+end

data/test/hexapdf/task/test_pdfa.rb

@@ -38,4 +38,76 @@ describe HexaPDF::Task::PDFA do
     assert_equal('sRGB2014.icc', oi[:Info])
     assert_kind_of(HexaPDF::Stream, oi[:DestOutputProfile])
   end
+
+  it "applies fixes based on the optional fixes argument" do
+    file = File.join(TEST_DATA_DIR, 'pdfa', 'mismatching_glyph_widths_cidfont_type2.pdf')
+
+    # Document loaded -> all fixes applied by default
+    doc = HexaPDF::Document.open(file)
+    doc.task(:pdfa, level: '3b')
+    doc.dispatch_message(:complete_objects)
+    font = HexaPDF::Font::TrueType::Font.new(StringIO.new(doc.object(10).stream))
+    assert_equal(348, font[:hmtx][1].advance_width)
+
+    # Not loaded -> fixes for loaded documents excluded
+    doc = HexaPDF::Document.open(file)
+    created = HexaPDF::Document.new
+    created.pages << created.import(doc.pages[0])
+    created.task(:pdfa, level: '3b')
+    created.dispatch_message(:complete_objects)
+    font_file = created.pages[0].resources.font(:F1).descendant_font[:FontDescriptor][:FontFile2]
+    font = HexaPDF::Font::TrueType::Font.new(StringIO.new(font_file.stream))
+    assert_equal(346, font[:hmtx][1].advance_width)
+
+    # Explicitly specify to apply all fixes
+    created.task(:pdfa, level: '3b', fixes: :all)
+    created.dispatch_message(:complete_objects)
+    font = HexaPDF::Font::TrueType::Font.new(StringIO.new(font_file.stream))
+    assert_equal(348, font[:hmtx][1].advance_width)
+  end
+
+  describe "fix_glyph_widths" do
+    before do
+      @file = File.join(TEST_DATA_DIR, 'pdfa', 'mismatching_glyph_widths_cidfont_type2.pdf')
+    end
+
+    it "fixes glyph width inconsistencies between the font and the font dictionary" do
+      doc = HexaPDF::Document.open(@file)
+      doc.task(:pdfa, level: '3b', fixes: [:fix_glyph_widths])
+
+      font = HexaPDF::Font::TrueType::Font.new(StringIO.new(doc.object(10).stream))
+      assert_equal(346, font[:hmtx][1].advance_width)
+      doc.dispatch_message(:complete_objects)
+      font = HexaPDF::Font::TrueType::Font.new(StringIO.new(doc.object(10).stream))
+      assert_equal(348, font[:hmtx][1].advance_width)
+    end
+
+    it "works if there is an explicit CIDToGIDMap stream" do
+      doc = HexaPDF::Document.open(@file)
+      doc.object(5)[:CIDToGIDMap] = doc.wrap({}, stream: [0, 1, 2, 3, 4].pack('n*'))
+      doc.task(:pdfa, level: '3b', fixes: [:fix_glyph_widths])
+      doc.dispatch_message(:complete_objects)
+      font = HexaPDF::Font::TrueType::Font.new(StringIO.new(doc.object(10).stream))
+      assert_equal(348, font[:hmtx][1].advance_width)
+    end
+
+    it "processes annotation appearances" do
+      doc = HexaPDF::Document.new
+      doc.pages.add
+      doc.annotations.create_rectangle(doc.pages[0], 20, 20, 20, 60).
+        regenerate_appearance
+      form = doc.pages[0][:Annots][0].create_appearance
+      form.canvas.
+        font(File.join(TEST_DATA_DIR, 'fonts', 'Ubuntu-Title.ttf'), size: 10).
+        text('Hola', at: [0, 0])
+
+      doc = HexaPDF::Document.new(io: StringIO.new(doc.write_to_string))
+      font = doc.pages[0][:Annots][0].appearance.resources.font(:F1).descendant_font
+      font[:W][1][0] = 10
+      doc.task(:pdfa, level: '3b', fixes: [:fix_glyph_widths])
+      doc.dispatch_message(:complete_objects)
+      font = HexaPDF::Font::TrueType::Font.new(StringIO.new(font[:FontDescriptor][:FontFile2].stream))
+      assert_equal(10, font[:hmtx][1].advance_width)
+    end
+  end
 end

data/test/hexapdf/test_document.rb

@@ -286,6 +286,13 @@ describe HexaPDF::Document do
       assert_equal({a: {b: 10}}, @doc.unwrap(value))
     end
 
+    it "doesn't unwrap PDF stream objects" do
+      stream = @io_doc.wrap({a: HexaPDF::Reference.new(1, 0)}, stream: 'data')
+      result = @io_doc.unwrap(stream)
+      assert_same(stream, result)
+      assert_equal(HexaPDF::Reference.new(1, 0), result.value[:a])
+    end
+
     it "fails to unwrap recursive structures" do
       obj1 = @doc.add({})
       obj2 = @doc.add({})
@@ -413,6 +420,12 @@ describe HexaPDF::Document do
       assert(@doc.trailer.info.key?(:Author))
     end
 
+    it "works even in case of invalid PDFs with a non-dictionary value for trailer.info" do
+      @doc.trailer[:Info] = :something_else
+      @doc.write(StringIO.new)
+      assert(@doc.trailer.info.key?(:ModDate))
+    end
+
     it "it doesn't optimize the file by default" do
       io = StringIO.new(''.b)
       @io_doc.write(io)

data/test/hexapdf/test_serializer.rb

@@ -104,7 +104,7 @@ describe HexaPDF::Serializer do
 
   it "serializes strings" do
     assert_serialized("(Hallo)", "Hallo")
-    assert_serialized("(Hallo\\r\n\t\\(\\)\\\\)", "Hallo\r\n\t()\\")
+    assert_serialized("(Hallo\\r\\n\t\\(\\)\\\\)", "Hallo\r\n\t()\\")
     assert_serialized("(\xFE\xFF\x00H\x00a\x00l\x00\f\x00\b\x00\\()".b, "Hal\f\b(")
   end
 

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

@@ -322,6 +322,12 @@ describe HexaPDF::Type::AcroForm::Form do
       assert_equal("value", field.field_value)
     end
 
+    it "ignores values for password fields" do
+      field = @acro_form.create_password_field('test')
+      @acro_form.fill("test" => "value")
+      assert_nil(field.field_value)
+    end
+
     it "works for radio buttons" do
       field = @acro_form.create_radio_button("test")
       field.create_widget(@doc.pages.add, value: :name)

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

@@ -272,10 +272,16 @@ describe HexaPDF::Type::AcroForm::TextField do
 
     it "checks that the field value has a valid type" do
       assert(@field.validate) # no field value
-      @field[:V] = :sym
+      @field[:V] = [5]
       refute(@field.validate)
     end
 
+    it "converts an invalid Symbol value to string" do
+      @field[:V] = :sym
+      assert(@field.validate)
+      assert_equal('sym', @field[:V])
+    end
+
     it "checks the field value against /MaxLen" do
       @field[:V] = 'Test'
       assert(@field.validate)

data/test/hexapdf/type/annotations/test_widget.rb

@@ -188,4 +188,15 @@ describe HexaPDF::Type::Annotations::Widget do
       end
     end
   end
+
+  describe "perform_validation" do
+    it "validates the widget as form field if they are the same" do
+      @widget[:Rect] = [0, 0, 0, 0]
+      @widget[:FT] = :Tx
+      @widget[:T] = 'field'
+      @widget[:V] = :Sym
+      assert(@widget.validate)
+      assert_equal('Sym', @widget[:V]) # this auto-correct is part of TextField
+    end
+  end
 end

data/test/hexapdf/type/test_font_type3.rb

@@ -28,6 +28,10 @@ describe HexaPDF::Type::FontType3 do
     assert_equal(0.002, @font.glyph_scaling_factor)
   end
 
+  it "always returns true for embedded?" do
+    assert(@font.embedded?)
+  end
+
   describe "validation" do
     it "works for valid objects" do
       assert(@font.validate)

data/test/hexapdf/type/test_page.rb

@@ -416,6 +416,14 @@ describe HexaPDF::Type::Page do
     end
   end
 
+  describe "extract_text" do
+    it "extracts the layouted text from the page" do
+      page = @doc.pages.add
+      page.canvas.font('Helvetica', size: 10).text('Hello', at: [10, 10])
+      assert_equal('Hello', page.extract_text(line_tolerance_factor: 5))
+    end
+  end
+
   describe "index" do
     it "returns the index of the page in the page tree" do
       kid1 = @doc.add({Type: :Pages, Parent: @doc.pages.root, Count: 4})

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: hexapdf
 version: !ruby/object:Gem::Version
-  version: 1.6.0
+  version: 1.8.0
 platform: ruby
 authors:
 - Thomas Leitner
@@ -77,6 +77,20 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: 3.1.2
+- !ruby/object:Gem::Dependency
+  name: brotli
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.7'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.7'
 - !ruby/object:Gem::Dependency
   name: kramdown
   requirement: !ruby/object:Gem::Requirement
@@ -334,6 +348,8 @@ files:
 - examples/029-composer_fallback_fonts.rb
 - examples/030-pdfa.rb
 - examples/031-acro_form_java_script.rb
+- examples/032-acro_form_list_and_fill.rb
+- examples/033-text_extraction.rb
 - examples/emoji-smile.png
 - examples/emoji-wink.png
 - examples/machupicchu.jpg
@@ -370,6 +386,7 @@ files:
 - lib/hexapdf/content/operator.rb
 - lib/hexapdf/content/parser.rb
 - lib/hexapdf/content/processor.rb
+- lib/hexapdf/content/smart_text_extractor.rb
 - lib/hexapdf/content/transformation_matrix.rb
 - lib/hexapdf/data_dir.rb
 - lib/hexapdf/dictionary.rb
@@ -408,6 +425,7 @@ files:
 - lib/hexapdf/filter.rb
 - lib/hexapdf/filter/ascii85_decode.rb
 - lib/hexapdf/filter/ascii_hex_decode.rb
+- lib/hexapdf/filter/brotli_decode.rb
 - lib/hexapdf/filter/crypt.rb
 - lib/hexapdf/filter/encryption.rb
 - lib/hexapdf/filter/flate_decode.rb
@@ -498,6 +516,7 @@ files:
 - lib/hexapdf/stream.rb
 - lib/hexapdf/task.rb
 - lib/hexapdf/task/dereference.rb
+- lib/hexapdf/task/import_pages.rb
 - lib/hexapdf/task/merge_acro_form.rb
 - lib/hexapdf/task/optimize.rb
 - lib/hexapdf/task/pdfa.rb
@@ -542,6 +561,7 @@ files:
 - lib/hexapdf/type/catalog.rb
 - lib/hexapdf/type/cid_font.rb
 - lib/hexapdf/type/cmap.rb
+- lib/hexapdf/type/document_security_store.rb
 - lib/hexapdf/type/embedded_file.rb
 - lib/hexapdf/type/file_specification.rb
 - lib/hexapdf/type/font.rb
@@ -641,6 +661,7 @@ files:
 - test/data/images/truecolour-srgb-8bit.png
 - test/data/images/ycck.jpg
 - test/data/minimal.pdf
+- test/data/pdfa/mismatching_glyph_widths_cidfont_type2.pdf
 - test/data/standard-security-handler/README
 - test/data/standard-security-handler/bothpwd-aes-128bit-V4.pdf
 - test/data/standard-security-handler/bothpwd-aes-256bit-V5-R5.pdf
@@ -678,6 +699,7 @@ files:
 - test/hexapdf/content/test_operator.rb
 - test/hexapdf/content/test_parser.rb
 - test/hexapdf/content/test_processor.rb
+- test/hexapdf/content/test_smart_text_extractor.rb
 - test/hexapdf/content/test_transformation_matrix.rb
 - test/hexapdf/digital_signature/common.rb
 - test/hexapdf/digital_signature/signing/test_default_handler.rb
@@ -711,6 +733,7 @@ files:
 - test/hexapdf/filter/common.rb
 - test/hexapdf/filter/test_ascii85_decode.rb
 - test/hexapdf/filter/test_ascii_hex_decode.rb
+- test/hexapdf/filter/test_brotli_decode.rb
 - test/hexapdf/filter/test_crypt.rb
 - test/hexapdf/filter/test_encryption.rb
 - test/hexapdf/filter/test_flate_decode.rb
@@ -778,6 +801,7 @@ files:
 - test/hexapdf/layout/test_text_shaper.rb
 - test/hexapdf/layout/test_width_from_polygon.rb
 - test/hexapdf/task/test_dereference.rb
+- test/hexapdf/task/test_import_pages.rb
 - test/hexapdf/task/test_merge_acro_form.rb
 - test/hexapdf/task/test_optimize.rb
 - test/hexapdf/task/test_pdfa.rb