pdfrb 0.7.1 → 0.7.10

data/lib/pdfrb/task/thumbnail.rb

@@ -0,0 +1,133 @@
+# frozen_string_literal: true
+
+module Pdfrb
+  module Task
+    # Generates a low-resolution thumbnail for each page (or a single
+    # page) and attaches it via the page-level /Thumb key. PDF
+    # viewers use /Thumb to render page previews quickly without
+    # parsing the full content stream.
+    #
+    # The thumbnail is a Form XObject rendered to a small image
+    # XObject. Pure-Ruby implementation: renders only the page's text
+    # (no image / form XObject content) at the target resolution
+    # using the standard 14 AFM metrics when available, otherwise a
+    # placeholder rectangle.
+    module Thumbnail
+      DEFAULT_DPI = 72
+      DEFAULT_MAX_WIDTH = 200
+
+      module_function
+
+      # Generate thumbnails for every page in +document+. Returns the
+      # count of thumbnails added. Idempotent: existing /Thumb
+      # entries are left alone unless +force:+ is true.
+      def call(document, dpi: DEFAULT_DPI, max_width: DEFAULT_MAX_WIDTH, force: false)
+        count = 0
+        document.pages.each do |page|
+          next if page.value[:Thumb] && !force
+
+          thumb = build_thumbnail(document, page, dpi: dpi, max_width: max_width)
+          next unless thumb
+
+          page.value[:Thumb] = Pdfrb::Model::Reference.new(thumb.oid, thumb.gen)
+          count += 1
+        end
+        count
+      end
+
+      # Build a single thumbnail image XObject for +page+ at the
+      # given DPI, capped at +max_width+ pixels. Returns the image
+      # XObject or nil if the page has no media box.
+      def build_thumbnail(document, page, dpi: DEFAULT_DPI, max_width: DEFAULT_MAX_WIDTH)
+        media_box = page.value[:MediaBox]
+        return nil unless media_box.is_a?(::Array) && media_box.length == 4
+
+        width, height = page_pixel_dimensions(media_box, dpi, max_width)
+        return nil unless width.positive? && height.positive?
+
+        # Single-component grayscale image — 1 byte per pixel.
+        # White background, black "text" rectangle placeholder. Real
+        # text rendering would require running the content processor
+        # against a rasteriser; that's a much larger feature.
+        pixel_data = render_placeholder(width, height, page)
+        compressed = require_zlib.deflate(pixel_data)
+
+        image = document.add(
+          {
+            Type: :XObject, Subtype: :Image,
+            Width: width, Height: height,
+            ColorSpace: :DeviceGray,
+            BitsPerComponent: 8,
+            Filter: :FlateDecode,
+            Length: compressed.bytesize
+          },
+          type: Pdfrb::Model::Cos::Stream
+        )
+        image.stream = compressed
+        image
+      end
+
+      # Compute thumbnail pixel dimensions, capped at +max_width+.
+      def page_pixel_dimensions(media_box, dpi, max_width)
+        _x0, _y0, x1, y1 = media_box
+        pt_w = x1.to_f
+        pt_h = y1.to_f
+        return [0, 0] if pt_w.zero? || pt_h.zero?
+
+        scale = (dpi / 72.0) * (max_width.to_f / pt_w)
+        scale = 1.0 if scale > 1.0
+        [(pt_w * scale).round, (pt_h * scale).round]
+      end
+
+      # Render a 1-bpp-style grayscale placeholder: white background,
+      # a black border, and a few black "text line" rectangles
+      # representing where text content might appear. Pure visual
+      # cue, not a real render.
+      def render_placeholder(width, height, _page)
+        bytes = Array.new(width * height, 255) # white
+
+        draw_rect(bytes, width, height, 0, 0, width, height, 0) # border (black)
+        # Determine text area: roughly 80% of page width, centred
+        margin = (width * 0.1).round
+        text_top = (height * 0.85).round
+        line_height = (height * 0.02).round
+        line_height = 1 if line_height.zero?
+        line_count = [(height * 0.6 / line_height).floor, 3].max
+        line_count.times do |i|
+          y = text_top - (i * line_height * 2)
+          draw_horizontal_line(bytes, width, height, margin, y,
+                               width - margin, 0)
+        end
+        bytes.pack("C*").b
+      end
+
+      def draw_rect(bytes, width, height, x0, y0, x1, y1, value)
+        x0.upto([x1 - 1, width - 1].min) do |x|
+          set_pixel(bytes, width, height, x, y0, value)
+          set_pixel(bytes, width, height, x, y1 - 1, value)
+        end
+        y0.upto([y1 - 1, height - 1].min) do |y|
+          set_pixel(bytes, width, height, x0, y, value)
+          set_pixel(bytes, width, height, x1 - 1, y, value)
+        end
+      end
+
+      def draw_horizontal_line(bytes, width, height, x0, y, x1, value)
+        x0.upto([x1 - 1, width - 1].min) do |x|
+          set_pixel(bytes, width, height, x, y, value)
+        end
+      end
+
+      def set_pixel(bytes, width, height, x, y, value)
+        return if x.negative? || y.negative? || x >= width || y >= height
+
+        bytes[(y * width) + x] = value
+      end
+
+      def require_zlib
+        require "zlib"
+        ::Zlib
+      end
+    end
+  end
+end

data/lib/pdfrb/task.rb

@@ -10,5 +10,6 @@ module Pdfrb
     autoload :Benchmark, "pdfrb/task/benchmark"
     autoload :MemoryProfile, "pdfrb/task/memory_profile"
     autoload :RegenerateAppearances, "pdfrb/task/regenerate_appearances"
+    autoload :Thumbnail, "pdfrb/task/thumbnail"
   end
 end

data/lib/pdfrb/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Pdfrb
-  VERSION = "0.7.1"
+  VERSION = "0.7.10"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: pdfrb
 version: !ruby/object:Gem::Version
-  version: 0.7.1
+  version: 0.7.10
 platform: ruby
 authors:
 - Ribose Inc.
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-08-08 00:00:00.000000000 Z
+date: 2026-08-11 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: thor
@@ -699,6 +699,7 @@ files:
 - data/pdfrb/arlington/latest/_UniversalArray.tsv
 - data/pdfrb/arlington/latest/_UniversalDictionary.tsv
 - data/pdfrb/glyphlist.txt
+- data/pdfrb/layout/hyphenation_en.txt
 - data/pdfrb/zapfdingbats.txt
 - exe/pdfrb
 - lib/pdfrb.rb
@@ -746,18 +747,28 @@ files:
 - lib/pdfrb/cli/watermark.rb
 - lib/pdfrb/color.rb
 - lib/pdfrb/color/color_space.rb
+- lib/pdfrb/color/default_profile.rb
 - lib/pdfrb/color/icc_profile.rb
+- lib/pdfrb/color/icc_validator.rb
 - lib/pdfrb/compare.rb
 - lib/pdfrb/compare/comparator.rb
 - lib/pdfrb/compare/report.rb
 - lib/pdfrb/composer.rb
 - lib/pdfrb/configuration.rb
 - lib/pdfrb/conformance.rb
+- lib/pdfrb/conformance/ltv.rb
+- lib/pdfrb/conformance/pades.rb
+- lib/pdfrb/conformance/pdf_2_af.rb
 - lib/pdfrb/conformance/pdf_a.rb
+- lib/pdfrb/conformance/pdf_a4_deep.rb
 - lib/pdfrb/conformance/pdf_ua.rb
+- lib/pdfrb/conformance/pdf_ua2_deep.rb
+- lib/pdfrb/conformance/pdf_ua_tagging_deep.rb
+- lib/pdfrb/conformance/pdf_vt.rb
 - lib/pdfrb/conformance/pdf_x.rb
 - lib/pdfrb/conformance/rule.rb
 - lib/pdfrb/conformance/structure_elements.rb
+- lib/pdfrb/conformance/tagged_pdf.rb
 - lib/pdfrb/conformance/verapdf_bridge.rb
 - lib/pdfrb/content.rb
 - lib/pdfrb/content/canvas.rb
@@ -796,6 +807,7 @@ files:
 - lib/pdfrb/digital_signature/cms_handler.rb
 - lib/pdfrb/digital_signature/handler.rb
 - lib/pdfrb/digital_signature/signing.rb
+- lib/pdfrb/digital_signature/timestamp_client.rb
 - lib/pdfrb/digital_signature/timestamp_handler.rb
 - lib/pdfrb/digital_signature/verification.rb
 - lib/pdfrb/digital_signature/verification_result.rb
@@ -826,9 +838,11 @@ files:
 - lib/pdfrb/encryption/aes.rb
 - lib/pdfrb/encryption/identity.rb
 - lib/pdfrb/encryption/password_verification.rb
+- lib/pdfrb/encryption/public_key_security_handler.rb
 - lib/pdfrb/encryption/rc4.rb
 - lib/pdfrb/encryption/security_handler.rb
 - lib/pdfrb/encryption/standard_security_handler.rb
+- lib/pdfrb/encryption/v5_writer.rb
 - lib/pdfrb/error.rb
 - lib/pdfrb/filter.rb
 - lib/pdfrb/filter/ascii_85_decode.rb
@@ -881,15 +895,18 @@ files:
 - lib/pdfrb/font_loader/standard14.rb
 - lib/pdfrb/font_loader/true_type.rb
 - lib/pdfrb/font_loader/type1.rb
+- lib/pdfrb/font_loader/type3.rb
 - lib/pdfrb/font_loader/variant_from_name.rb
 - lib/pdfrb/font_resolver.rb
 - lib/pdfrb/image.rb
 - lib/pdfrb/image/audit.rb
 - lib/pdfrb/image/downsampler.rb
 - lib/pdfrb/image_loader.rb
+- lib/pdfrb/image_loader/gif.rb
 - lib/pdfrb/image_loader/jpeg.rb
 - lib/pdfrb/image_loader/pdf.rb
 - lib/pdfrb/image_loader/png.rb
+- lib/pdfrb/image_loader/tiff.rb
 - lib/pdfrb/importer.rb
 - lib/pdfrb/layout.rb
 - lib/pdfrb/layout/bidi.rb
@@ -897,18 +914,25 @@ files:
 - lib/pdfrb/layout/box_fitter.rb
 - lib/pdfrb/layout/column_box.rb
 - lib/pdfrb/layout/container_box.rb
+- lib/pdfrb/layout/font_fallback.rb
 - lib/pdfrb/layout/frame.rb
+- lib/pdfrb/layout/hyphenation.rb
 - lib/pdfrb/layout/image_box.rb
 - lib/pdfrb/layout/inline_box.rb
+- lib/pdfrb/layout/justification_kashidas.rb
 - lib/pdfrb/layout/line.rb
 - lib/pdfrb/layout/list_box.rb
+- lib/pdfrb/layout/multi_cell_text_layout.rb
+- lib/pdfrb/layout/multi_page_table_box.rb
 - lib/pdfrb/layout/numeric_refinements.rb
 - lib/pdfrb/layout/page_style.rb
+- lib/pdfrb/layout/polygon_frame.rb
 - lib/pdfrb/layout/style.rb
 - lib/pdfrb/layout/table_box.rb
 - lib/pdfrb/layout/text_box.rb
 - lib/pdfrb/layout/text_fragment.rb
 - lib/pdfrb/layout/text_layouter.rb
+- lib/pdfrb/layout/text_shaper.rb
 - lib/pdfrb/linearization.rb
 - lib/pdfrb/linearization/hint_stream.rb
 - lib/pdfrb/linearization/writer.rb
@@ -1145,6 +1169,7 @@ files:
 - lib/pdfrb/source/header_reader.rb
 - lib/pdfrb/source/helpers.rb
 - lib/pdfrb/source/linearization_detection.rb
+- lib/pdfrb/source/linearization_reader.rb
 - lib/pdfrb/source/object_reader.rb
 - lib/pdfrb/source/object_stream_reader.rb
 - lib/pdfrb/source/parser.rb
@@ -1164,6 +1189,7 @@ files:
 - lib/pdfrb/task/merge.rb
 - lib/pdfrb/task/optimize.rb
 - lib/pdfrb/task/regenerate_appearances.rb
+- lib/pdfrb/task/thumbnail.rb
 - lib/pdfrb/test_utils.rb
 - lib/pdfrb/validator.rb
 - lib/pdfrb/version.rb