ucode 0.2.1 → 0.2.3

data/lib/ucode/repo/writers/indexes_writer.rb

@@ -34,11 +34,11 @@ module Ucode
         def write
           count = 0
           count += 1 if write_atomic(Paths.names_index_path(@output_root),
-                                      to_pretty_json(@names))
+                                     to_pretty_json(@names))
           count += 1 if write_atomic(Paths.labels_index_path(@output_root),
-                                      to_pretty_json(@labels))
+                                     to_pretty_json(@labels))
           count += 1 if write_atomic(codepoint_to_block_path,
-                                      to_pretty_json(@cp_to_block))
+                                     to_pretty_json(@cp_to_block))
           count
         end
 
@@ -50,4 +50,4 @@ module Ucode
       end
     end
   end
-end
+end

data/lib/ucode/repo/writers/manifest_writer.rb

@@ -45,10 +45,10 @@ module Ucode
         def write
           path = Paths.manifest_path(@output_root)
           content = {
-            "ucd_version"     => @ucd_version,
+            "ucd_version" => @ucd_version,
             "codepoint_count" => @codepoint_count,
-            "glyph_count"     => @glyph_count,
-            "schema_version"  => SCHEMA_VERSION,
+            "glyph_count" => @glyph_count,
+            "schema_version" => SCHEMA_VERSION,
           }
           ts = preserved_or_new_timestamp(path, content)
           payload = content.merge("generated_at" => ts)
@@ -75,4 +75,4 @@ module Ucode
       end
     end
   end
-end
+end

data/lib/ucode/repo/writers/named_sequences_writer.rb

@@ -44,4 +44,4 @@ module Ucode
       end
     end
   end
-end
+end

data/lib/ucode/repo/writers/planes_writer.rb

@@ -16,16 +16,16 @@ module Ucode
         # Static metadata for the 17 Unicode planes. Planes 4–13 are
         # unassigned in Unicode 17; their entries use placeholder names.
         PLANE_TABLE = {
-          0  => ["Basic Multilingual Plane",            "BMP"],
-          1  => ["Supplementary Multilingual Plane",    "SMP"],
-          2  => ["Supplementary Ideographic Plane",     "SIP"],
-          3  => ["Tertiary Ideographic Plane",          "TIP"],
-          4  => ["Unassigned Plane 4",                  "—"],
-          5  => ["Unassigned Plane 5",                  "—"],
-          6  => ["Unassigned Plane 6",                  "—"],
-          7  => ["Unassigned Plane 7",                  "—"],
-          8  => ["Unassigned Plane 8",                  "—"],
-          9  => ["Unassigned Plane 9",                  "—"],
+          0 => ["Basic Multilingual Plane",            "BMP"],
+          1 => ["Supplementary Multilingual Plane",    "SMP"],
+          2 => ["Supplementary Ideographic Plane",     "SIP"],
+          3 => ["Tertiary Ideographic Plane",          "TIP"],
+          4 => ["Unassigned Plane 4",                  "—"],
+          5 => ["Unassigned Plane 5",                  "—"],
+          6 => ["Unassigned Plane 6",                  "—"],
+          7 => ["Unassigned Plane 7",                  "—"],
+          8 => ["Unassigned Plane 8",                  "—"],
+          9 => ["Unassigned Plane 9",                  "—"],
           10 => ["Unassigned Plane 10",                 "—"],
           11 => ["Unassigned Plane 11",                 "—"],
           12 => ["Unassigned Plane 12",                 "—"],
@@ -68,15 +68,15 @@ module Ucode
           range_first = plane_number * 0x10000
           range_last  = range_first + 0xFFFF
           to_pretty_json(
-            "number"       => plane_number,
-            "name"         => name,
-            "abbrev"       => abbrev,
-            "range_first"  => range_first,
-            "range_last"   => range_last,
-            "block_ids"    => block_ids,
+            "number" => plane_number,
+            "name" => name,
+            "abbrev" => abbrev,
+            "range_first" => range_first,
+            "range_last" => range_last,
+            "block_ids" => block_ids,
           )
         end
       end
     end
   end
-end
+end

data/lib/ucode/repo/writers/relationships_writer.rb

@@ -68,4 +68,4 @@ module Ucode
       end
     end
   end
-end
+end

data/lib/ucode/repo/writers/scripts_writer.rb

@@ -41,14 +41,14 @@ module Ucode
 
         def script_payload(code, ranges)
           to_pretty_json(
-            "code"           => code,
-            "name"           => ranges.first&.name,
-            "range_first"    => ranges.map(&:range_first).min,
-            "range_last"     => ranges.map(&:range_last).max,
-            "codepoint_ids"  => (@script_codepoint_ids[code] || []),
+            "code" => code,
+            "name" => ranges.first&.name,
+            "range_first" => ranges.map(&:range_first).min,
+            "range_last" => ranges.map(&:range_last).max,
+            "codepoint_ids" => @script_codepoint_ids[code] || [],
           )
         end
       end
     end
   end
-end
+end

data/lib/ucode/repo/writers.rb

@@ -17,4 +17,4 @@ module Ucode
       autoload :ManifestWriter,        "ucode/repo/writers/manifest_writer"
     end
   end
-end
+end

data/lib/ucode/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Ucode
-  VERSION = "0.2.1"
+  VERSION = "0.2.3"
 end

data/lib/ucode.rb

@@ -32,8 +32,6 @@ module Ucode
   autoload :UnknownVersionError, "ucode/error"
   autoload :UnknownBlockError, "ucode/error"
   autoload :GlyphError, "ucode/error"
-  autoload :PdfRenderError, "ucode/error"
-  autoload :GridDetectionError, "ucode/error"
   autoload :LastResortMissingError, "ucode/error"
   autoload :EmbeddedFontsMissingError, "ucode/error"
   autoload :CodeChartNotFoundError, "ucode/error"

data/ucode.gemspec

@@ -45,7 +45,12 @@ Gem::Specification.new do |spec|
   spec.require_paths = ["lib"]
 
   spec.add_dependency "base64"
-  spec.add_dependency "fontisan", "~> 0.2"
+  # Pin fontisan to 0.2.22 — 0.2.23+ removed
+  # `Fontisan::Commands::AuditCommand` (used by RealFonts::CoverageAuditor)
+  # and 0.4.x removed the entire Audit subsystem. Until
+  # CoverageAuditor is rewritten against the new fontisan API, this
+  # pin keeps the existing audit path working.
+  spec.add_dependency "fontisan", "= 0.2.22"
   spec.add_dependency "fontist", "~> 3.0"
   spec.add_dependency "logger"
   spec.add_dependency "lutaml-model", "~> 0.8"

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: ucode
 version: !ruby/object:Gem::Version
-  version: 0.2.1
+  version: 0.2.3
 platform: ruby
 authors:
 - Ribose Inc.
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-06-30 00:00:00.000000000 Z
+date: 2026-07-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: base64
@@ -28,16 +28,16 @@ dependencies:
   name: fontisan
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - '='
       - !ruby/object:Gem::Version
-        version: '0.2'
+        version: 0.2.22
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - '='
       - !ruby/object:Gem::Version
-        version: '0.2'
+        version: 0.2.22
 - !ruby/object:Gem::Dependency
   name: fontist
   requirement: !ruby/object:Gem::Requirement
@@ -321,7 +321,6 @@ files:
 - lib/ucode/commands/canonical_build.rb
 - lib/ucode/commands/fetch.rb
 - lib/ucode/commands/font_coverage.rb
-- lib/ucode/commands/glyphs.rb
 - lib/ucode/commands/lookup.rb
 - lib/ucode/commands/parse.rb
 - lib/ucode/commands/release.rb
@@ -342,8 +341,6 @@ files:
 - lib/ucode/fetch/ucd_zip.rb
 - lib/ucode/fetch/unihan_zip.rb
 - lib/ucode/glyphs.rb
-- lib/ucode/glyphs/cell_extractor.rb
-- lib/ucode/glyphs/dvisvgm_renderer.rb
 - lib/ucode/glyphs/embedded_fonts.rb
 - lib/ucode/glyphs/embedded_fonts/catalog.rb
 - lib/ucode/glyphs/embedded_fonts/content_stream_correlator.rb
@@ -352,9 +349,11 @@ files:
 - lib/ucode/glyphs/embedded_fonts/source.rb
 - lib/ucode/glyphs/embedded_fonts/svg.rb
 - lib/ucode/glyphs/embedded_fonts/tounicode.rb
+- lib/ucode/glyphs/embedded_fonts/trace_correlator.rb
+- lib/ucode/glyphs/embedded_fonts/trace_glyph.rb
+- lib/ucode/glyphs/embedded_fonts/trace_parser.rb
+- lib/ucode/glyphs/embedded_fonts/trace_runner.rb
 - lib/ucode/glyphs/embedded_fonts/writer.rb
-- lib/ucode/glyphs/grid.rb
-- lib/ucode/glyphs/grid_detector.rb
 - lib/ucode/glyphs/last_resort.rb
 - lib/ucode/glyphs/last_resort/cmap_index.rb
 - lib/ucode/glyphs/last_resort/contents.rb
@@ -363,14 +362,7 @@ files:
 - lib/ucode/glyphs/last_resort/source.rb
 - lib/ucode/glyphs/last_resort/svg.rb
 - lib/ucode/glyphs/last_resort/writer.rb
-- lib/ucode/glyphs/monolith_page_map.rb
-- lib/ucode/glyphs/mutool_renderer.rb
-- lib/ucode/glyphs/page_renderer.rb
-- lib/ucode/glyphs/path_bbox.rb
-- lib/ucode/glyphs/pdf2svg_renderer.rb
 - lib/ucode/glyphs/pdf_fetcher.rb
-- lib/ucode/glyphs/pdftocairo_renderer.rb
-- lib/ucode/glyphs/pipeline.rb
 - lib/ucode/glyphs/real_fonts.rb
 - lib/ucode/glyphs/real_fonts/block_coverage.rb
 - lib/ucode/glyphs/real_fonts/cmap_cache.rb
@@ -397,7 +389,6 @@ files:
 - lib/ucode/glyphs/universal_set/manifest_writer.rb
 - lib/ucode/glyphs/universal_set/pre_build_check.rb
 - lib/ucode/glyphs/universal_set/validator.rb
-- lib/ucode/glyphs/writer.rb
 - lib/ucode/index.rb
 - lib/ucode/index_builder.rb
 - lib/ucode/models.rb

data/lib/ucode/commands/glyphs.rb

@@ -1,94 +0,0 @@
-# frozen_string_literal: true
-
-require "pathname"
-
-require "ucode/glyphs"
-
-module Ucode
-  module Commands
-    # `ucode glyphs` — extract per-codepoint SVGs from Code Charts PDFs.
-    # Thin Thor-facing wrapper around {Ucode::Glyphs::Pipeline}:
-    # opt-in gate + experimental warning live here; the pipeline
-    # assembly (block loading, fetcher, per-block specs) lives in
-    # {Ucode::Glyphs::Pipeline}.
-    #
-    # **Status (v0.1): EXPERIMENTAL.** The cell-extraction pipeline
-    # currently includes cell-border decorations alongside the actual
-    # character outline because the Code Charts PDFs composite the two
-    # into a single glyph definition. The output is therefore not yet
-    # suitable for end-user display. The command is retained so the
-    # pipeline can be iterated on without churning the CLI surface, but
-    # callers MUST opt in via `include_glyphs: true` (CLI: `--include-glyphs`)
-    # and will receive a printed warning. Tracked for v0.2.
-    #
-    # Takes a resolved version string; CLI callers resolve via
-    # {VersionResolver.resolve} once and thread it through. See
-    # Candidate 4 of the 2026-06-29 architecture review.
-    class GlyphsCommand
-      ExperimentalWarning = "ucode glyphs is experimental in v0.1: " \
-                            "extracted SVGs include cell-border decorations " \
-                            "alongside the character outline."
-      private_constant :ExperimentalWarning
-
-      class << self
-        # @return [String] the experimental-status banner. Exposed so the
-        #   CLI and BuildCommand surface the same message verbatim.
-        def experimental_warning
-          ExperimentalWarning
-        end
-      end
-
-      # @param version [String] resolved UCD version
-      # @param output_root [String, Pathname]
-      # @param block_filter [Array<String>, nil] block ids to limit to;
-      #   nil = every block
-      # @param force [Boolean] re-fetch PDFs even when cached
-      # @param monolith_path [String, Pathname, nil] path to CodeCharts.pdf
-      #   for fallback slicing; defaults to ./CodeCharts.pdf
-      # @param include_glyphs [Boolean] opt-in for the experimental v0.1
-      #   pipeline. When false (default), the command returns a `skipped`
-      #   payload without touching disk.
-      # @param warn [IO, nil] when provided, the experimental warning is
-      #   written here exactly once before work begins.
-      # @return [Hash] aggregated Writer tally + version, or a `skipped`
-      #   payload when opt-in is false.
-      def call(version, output_root:,
-               block_filter: nil, force: false,
-               monolith_path: Glyphs::Pipeline::DEFAULT_MONOLITH_PATH,
-               include_glyphs: false, warn: nil)
-        return skipped(version) unless include_glyphs
-
-        warn&.puts(ExperimentalWarning)
-
-        pipeline = Glyphs::Pipeline.new(
-          version: version,
-          block_filter: block_filter,
-          monolith_path: monolith_path,
-        )
-        specs = pipeline.build_specs(force: force)
-
-        writer = Glyphs::Writer.new(
-          output_root: Pathname.new(output_root),
-          parallel_workers: workers,
-        )
-        tally = writer.write_all(specs)
-        tally.merge(version: version, block_count: specs.size)
-      end
-
-      private
-
-      def workers
-        Ucode.configuration.parallel_workers
-      end
-
-      def skipped(version)
-        {
-          version: version,
-          skipped: true,
-          reason: :experimental_v0_1,
-          warning: ExperimentalWarning,
-        }
-      end
-    end
-  end
-end

data/lib/ucode/glyphs/cell_extractor.rb

@@ -1,130 +0,0 @@
-# frozen_string_literal: true
-
-require "nokogiri"
-
-require "ucode/glyphs/path_bbox"
-
-module Ucode
-  module Glyphs
-    # Extracts a single character cell from a Code Charts SVG page and
-    # returns a normalized standalone SVG containing only that cell's
-    # vector paths.
-    #
-    # The cell is identified by codepoint. The extractor asks the Grid
-    # for the cell's anchor position, finds the `<use>` element placed
-    # at that position, resolves its glyph definition from `<defs>`,
-    # and emits a fresh `<svg>` whose viewBox is `0 0 1000 1000` and
-    # whose body is the glyph's `<path>` data translated and scaled to
-    # fit that viewBox with a small margin.
-    #
-    # Vector-only. Never rasterizes, never OCRs. If the cell is empty
-    # (no character glyph placed there, e.g. unassigned codepoint or
-    # control character), the extractor returns nil.
-    class CellExtractor
-      ViewBoxSize = 1000.0
-      MarginRatio = 0.1
-      private_constant :ViewBoxSize, :MarginRatio
-
-      # @param doc [Nokogiri::XML::Document] the rendered Code Charts page
-      def initialize(doc)
-        @doc = doc
-        @glyph_cache = {}
-      end
-
-      # @param grid [Ucode::Glyphs::Grid]
-      # @param codepoint [Integer]
-      # @return [Nokogiri::XML::Document, nil] a standalone `<svg>` doc
-      #   with viewBox `0 0 1000 1000`, or nil if the cell is empty.
-      def extract(grid, codepoint)
-        anchor = grid.cell_position(codepoint)
-        return nil unless anchor
-
-        use_node = find_use_at(anchor, grid)
-        return nil unless use_node
-
-        path_data = collect_paths(use_node["xlink:href"] || use_node["href"])
-        return nil if path_data.empty?
-
-        bbox = PathBbox.estimate(path_data.join(" "))
-        return nil if bbox.empty?
-
-        build_svg(path_data, bbox, use_node["x"].to_f, use_node["y"].to_f)
-      end
-
-      private
-
-      def find_use_at(anchor, grid)
-        tolerance_x = grid.column_pitch / 2
-        tolerance_y = grid.row_pitch / 2
-
-        candidates = @doc.css("use").select do |node|
-          href = node["xlink:href"] || node["href"] || ""
-          href.start_with?("#glyph-") &&
-            (node["x"].to_f - anchor[0]).abs <= tolerance_x &&
-            (node["y"].to_f - anchor[1]).abs <= tolerance_y
-        end
-
-        candidates.min_by do |node|
-          dx = node["x"].to_f - anchor[0]
-          dy = node["y"].to_f - anchor[1]
-          (dx * dx) + (dy * dy)
-        end
-      end
-
-      def collect_paths(href)
-        return [] unless href
-
-        glyph_id = href.sub(/\A#/, "")
-        node = glyph_definition(glyph_id)
-        return [] unless node
-
-        node.css("path").map { |p| p["d"] }.compact
-      end
-
-      def glyph_definition(glyph_id)
-        return @glyph_cache[glyph_id] if @glyph_cache.key?(glyph_id)
-
-        @glyph_cache[glyph_id] = @doc.at_css("defs ##{glyph_id}")
-      end
-
-      def build_svg(path_data, glyph_bbox, place_x, place_y)
-        placed = PathBbox::Result.new(
-          min_x: place_x + glyph_bbox.min_x,
-          min_y: place_y + glyph_bbox.min_y,
-          max_x: place_x + glyph_bbox.max_x,
-          max_y: place_y + glyph_bbox.max_y,
-        )
-
-        width = placed.width
-        height = placed.height
-        return nil if width <= 0 || height <= 0
-
-        content_size = ViewBoxSize * (1.0 - (2.0 * MarginRatio))
-        scale = [content_size / width, content_size / height].min
-        offset_x = (ViewBoxSize - (width * scale)) / 2.0
-        offset_y = (ViewBoxSize - (height * scale)) / 2.0
-        translate_x = offset_x - (placed.min_x * scale)
-        translate_y = offset_y - (placed.min_y * scale)
-
-        builder = Nokogiri::XML::Document.new
-        root = builder.create_element(
-          "svg",
-          xmlns: "http://www.w3.org/2000/svg",
-          viewBox: "0 0 #{ViewBoxSize.to_i} #{ViewBoxSize.to_i}",
-          width: ViewBoxSize.to_i,
-          height: ViewBoxSize.to_i,
-        )
-        group = builder.create_element(
-          "g",
-          transform: "scale(#{format('%.6f', scale)}) translate(#{format('%.6f', translate_x)}, #{format('%.6f', translate_y)})",
-        )
-        path_data.each do |d|
-          group.add_child(builder.create_element("path", d: d, fill: "black"))
-        end
-        root.add_child(group)
-        builder.add_child(root)
-        builder
-      end
-    end
-  end
-end

data/lib/ucode/glyphs/dvisvgm_renderer.rb

@@ -1,29 +0,0 @@
-# frozen_string_literal: true
-
-require "ucode/glyphs/page_renderer"
-
-module Ucode
-  module Glyphs
-    # `dvisvgm` — originally a DVI-to-SVG converter, also handles PDF.
-    # The `--no-fonts` flag forces outline-only output (no font subsetting
-    # artifacts), which is what we want for vector glyph extraction.
-    #
-    # Command: `dvisvgm --pdf --no-fonts --page=<n> <in.pdf> -o <out.svg>`
-    class DvisvgmRenderer < PageRenderer
-      class << self
-        def renderer_name
-          :dvisvgm
-        end
-
-        def binary_name
-          :dvisvgm
-        end
-
-        def build_command(pdf_path, page_num, out_path)
-          ["dvisvgm", "--pdf", "--no-fonts", "--page=#{page_num}",
-           pdf_path.to_s, "-o", out_path.to_s]
-        end
-      end
-    end
-  end
-end

data/lib/ucode/glyphs/grid.rb

@@ -1,30 +0,0 @@
-# frozen_string_literal: true
-
-module Ucode
-  module Glyphs
-    Grid = Struct.new(
-      :origin_x, :origin_y,
-      :column_pitch, :row_pitch,
-      :columns, :rows,
-      :block_first_cp,
-      keyword_init: true,
-    ) do
-      def cell_position(codepoint)
-        offset = codepoint - block_first_cp
-        return nil if offset.negative?
-
-        row, col = offset.divmod(columns)
-        return nil if row >= rows
-
-        [origin_x + (col * column_pitch), origin_y + (row * row_pitch)]
-      end
-
-      def codepoint_at(row, col)
-        return nil if row.negative? || row >= rows
-        return nil if col.negative? || col >= columns
-
-        block_first_cp + (row * columns) + col
-      end
-    end
-  end
-end