stationery 0.10.0 → 0.10.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 41ce5a83b1405e6756b6b2238b4796d1c7331e82f1abb51df2747befa61dbc26
-  data.tar.gz: a1c9fab02c1ce65eb761c96084c81b72117aee45ac2c0db8c1326ce94651f7a4
+  metadata.gz: 94c8fd60adf62c2f1ac3dd92b4a45b5b4254d11284c255c37c86f4d7887d0f14
+  data.tar.gz: 704273654b825478068079526b21ec8b4f32c9b1b93d4f5c4854a47b22e6f0b7
 SHA512:
-  metadata.gz: 9c1a85c89e74b002d14d6dd2aaa7153e34e746636ad6059b8db6a83f302b7058d4efb7430346463fc614ba0b61b7991226c44d724b62479d22ab5b1a007b419a
-  data.tar.gz: dfc30baf5b1803d38d7087eed1b24960fe79c3021f119b859d695458e128ed23baf74d961779b2bdb39f1071a0ed7cd09f51d1aa3881eb3076190f1a8d5a63b0
+  metadata.gz: 07175d385d770227a5d5af7f542f608d86215ecf3172852a7f75738c426273dfe75fc272db81b929136be496e9b0b7394e1febb5b2a326ec2e06d53b012242e7
+  data.tar.gz: 864623a98fc783adecd1f8e51a7ae85bbeb1a2bd7c393a6cf72fcf8a548d48221333e29d6ca1336a100bda62576869ccae1fce084ec57b3393758226fd85ac4f

data/CHANGELOG.md

@@ -1,5 +1,9 @@
 # Changelog
 
+## 0.10.1 (2026-09-28)
+
+- Extracted text keeps characters no font has: each run of `.notdef` glyphs is wrapped in a `/Span` with `/ActualText` holding the written characters, and ToUnicode maps glyph 0 to U+FFFD instead of the first missing character drawn, so `日本語` set in a font without those glyphs copies out as `日本語`, not `日日日`. The `MissingGlyph` warning and the bytes of text the fonts cover are unchanged.
+
 ## 0.10.0 (2026-09-28)
 
 CSS for `html`, better line and page breaking, timestamped signatures, and streaming to a block.

data/README.md

@@ -794,7 +794,9 @@ a figure space, a narrow no-break space, …) is drawn as a blank of the
 character's conventional width, never as `.notdef`. Any other glyph no font
 has is drawn as the family's `.notdef` and reported as a
 `Warnings::MissingGlyph` counting each drawn occurrence (so `strict` raises
-on it). Fallback covers every text element,
+on it). The characters themselves travel as the `ActualText` of a `Span`
+around the glyphs, so the text still extracts, copies and reads aloud as
+written. Fallback covers every text element,
 table cell, list marker, table of contents entry and page template text;
 direct `canvas.text` calls draw with the font they are given.
 

data/lib/stationery/fonts/font.rb

@@ -24,6 +24,7 @@ module Stationery
                           0x3000 => 1.0 }.freeze
       SPACE_LIKE = { 0x2007 => "0", 0x2008 => "." }.freeze
       NO_FEATURES = [].freeze
+      REPLACEMENT = "\uFFFD"
 
       attr_reader :ttf
 
@@ -107,9 +108,11 @@ module Stationery
       end
 
       # { code => text } for every glyph drawn so far; a ligature's text is
-      # every character it stands for.
+      # every character it stands for. Glyph 0, .notdef, stands for every
+      # character the font lacks, so it maps to the replacement character
+      # and the run's ActualText carries the real ones (see GlyphRun).
       def used_codes
-        @used.to_h { |gid, char| [code(gid), char] }
+        @used.to_h { |gid, char| [code(gid), gid.zero? ? REPLACEMENT : char] }
       end
 
       def build(writer)

data/lib/stationery/fonts/glyph_run.rb

@@ -7,6 +7,11 @@ module Stationery
     # source text of each glyph (several characters for a ligature). All-zero
     # adjustments draw as a plain Tj string; otherwise as a TJ array. Glyphs
     # are written as the font's character codes (see Font#code).
+    #
+    # A character no font has is glyph 0, .notdef, which stands for no
+    # character in particular: each stretch of them is shown inside a Span
+    # whose ActualText is the characters, so the text still extracts, copies
+    # and reads aloud as written.
     GlyphRun = Data.define(:font, :gids, :adjust, :chars) do
       def width(size, letter_spacing: 0)
         units = gids.sum { |gid| font.ttf.advance(gid) }
@@ -18,20 +23,45 @@ module Stationery
         with(adjust: adjust.each_with_index.map { |a, i| chars[i] == " " ? a + extra : a })
       end
 
+      def missing? = gids.include?(0)
+
       def to_operator
-        return "<#{hex(gids)}> Tj" unless adjusted?
+        return show unless missing?
+
+        last = gids.size - 1
+        gids.each_index.chunk_while { |a, b| gids[a].zero? == gids[b].zero? }.map do |indices|
+          piece = with(gids: gids.values_at(*indices), adjust: adjust.values_at(*indices),
+                       chars: chars.values_at(*indices))
+          piece.marked(trailing: indices.last < last)
+        end.join("\n")
+      end
+
+      protected
+
+      # One stretch of a run: .notdef glyphs inside their Span. `trailing`
+      # keeps the adjustment after the last glyph, which a stretch followed
+      # by another needs and the end of a run does not.
+      def marked(trailing:)
+        return show(trailing:) unless missing?
+
+        "/Span <</ActualText <FEFF#{chars.join.encode(Encoding::UTF_16BE).unpack1("H*").upcase}>>> BDC\n" \
+          "#{show(trailing:)}\nEMC"
+      end
+
+      private
+
+      def show(trailing: false)
+        return "<#{hex(gids)}> Tj" unless adjusted?(trailing)
 
         parts = gids.each_index.slice_after { |i| !adjust[i].zero? }.flat_map do |indices|
           last = indices.last
           chunk = ["<#{hex(indices.map { |i| gids[i] })}>"]
-          last == gids.size - 1 || adjust[last].zero? ? chunk : chunk << PDF::Serializer.number(-adjust[last])
+          (last == gids.size - 1 && !trailing) || adjust[last].zero? ? chunk : chunk << PDF::Serializer.number(-adjust[last])
         end
         "[#{parts.join(" ")}] TJ"
       end
 
-      private
-
-      def adjusted? = adjust[0...-1].any? { |a| !a.zero? }
+      def adjusted?(trailing) = (trailing ? adjust : adjust[0...-1]).any? { |a| !a.zero? }
 
       def hex(ids) = ids.map { |gid| font.code(gid) }.pack("n*").unpack1("H*").upcase
     end

data/lib/stationery/testing/marked_text.rb

@@ -16,6 +16,8 @@ module Stationery
                  set_character_spacing set_word_spacing set_text_rise set_horizontal_text_scaling
                  set_text_rendering_mode].freeze
 
+      UTF16_BOM = "\xFE\xFF".b
+
       def_delegators :@state, *STATE
 
       # { MCID => text } and the strings shown outside marked content.
@@ -31,20 +33,30 @@ module Stationery
         @unmarked = []
       end
 
+      # A sequence with ActualText reads as that text, whatever it shows.
       def begin_marked_content_with_pl(_tag, properties)
-        @open << (properties[:MCID] if properties.is_a?(Hash))
+        properties = {} unless properties.is_a?(Hash)
+        @open << properties[:MCID]
+        return unless properties[:ActualText]
+
+        @actual = decode(properties[:ActualText])
+        @actual_depth = @open.size
       end
 
       def begin_marked_content(_tag) = @open << nil
-      def end_marked_content = @open.pop
+
+      def end_marked_content
+        @actual = nil if @actual_depth && @open.size <= @actual_depth
+        @open.pop
+      end
+
       def show_text(string) = append(string)
       def show_text_with_positioning(params) = params.grep(String).each { |string| append(string) }
 
       private
 
       def append(string)
-        font = @state.current_font
-        text = font.unpack(string).map { |code| font.to_utf8(code) }.join
+        text = shown(string)
         return @unmarked << text if @open.empty?
 
         mcid = @open.compact.last
@@ -55,6 +67,24 @@ module Stationery
         @texts[mcid] = "#{@texts[mcid]}#{separator}#{text}"
         @baselines[mcid] = baseline
       end
+
+      # The characters a string shows: its ActualText once, when it has one.
+      def shown(string)
+        if @actual
+          text = @actual
+          @actual = ""
+          return text
+        end
+
+        font = @state.current_font
+        font.unpack(string).map { |code| font.to_utf8(code) }.join
+      end
+
+      def decode(text)
+        return text.dup.force_encoding(Encoding::UTF_8) unless text.b.start_with?(UTF16_BOM)
+
+        text.b.byteslice(2..).force_encoding(Encoding::UTF_16BE).encode(Encoding::UTF_8)
+      end
     end
   end
 end

data/lib/stationery/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Stationery
-  VERSION = "0.10.0"
+  VERSION = "0.10.1"
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: stationery
 version: !ruby/object:Gem::Version
-  version: 0.10.0
+  version: 0.10.1
 platform: ruby
 authors:
 - Mikael Henriksson