isodoc 0.4.5 → 0.5.5

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 92ad102a0e0bc53916aaec5d7a4f23d71828f3d4
-  data.tar.gz: 7dbbd1969e0e1714c9074ac455a335939f634335
+  metadata.gz: 32b644d32aaf9d2170de736e389cec4a7b0b1a4d
+  data.tar.gz: afc5f22b859a533e4244758b80a3f4acc9658924
 SHA512:
-  metadata.gz: 1b570dec93a170716366cf0d433d20dc614026ebe2224b02e659216fccf006b1680a1126e763a953f8d61c8154a0c9b133dfd989b05e25e675b5b3aa1f10d980
-  data.tar.gz: b9499b701f2eaca4a27d71ff85ff27926bd58e43a2b255dd429180eaf8b8c72704f8ba3768b2eb894f41128136f887308e6f2a29e89c000b7f02cf2dcb391e51
+  metadata.gz: 22698380e197c7f94fa0e690dfa181bc2312aa130aeab631fa6cd8ca515c1cd6c312b3f37b56823eef43d079c24ee1c2b539173c4dcd5f13c156dec975fdfce8
+  data.tar.gz: 6b95287839fe12f19e777cfdf62bc06b317247e79b1f6cf084b4ab6b938b561d763aebf9f6ec7bcd0ce280a3cea19ded061cd2fbb8e086c5591221ea9c768a4e

data/bin/rspec

@@ -0,0 +1,18 @@
+#!/usr/bin/env ruby
+  
+# This file was generated by Bundler.
+#
+# The application 'rspec' is installed as part of a gem, and
+# this file is here to facilitate running it.
+#
+
+require "pathname"
+ENV["BUNDLE_GEMFILE"] ||= File.expand_path(
+  "../../Gemfile", Pathname.new(__FILE__).realpath
+)
+
+require "rubygems"
+require "bundler/setup"
+
+load Gem.bin_path("rspec-core", "rspec")
+

data/isodoc.gemspec

@@ -32,7 +32,7 @@ Gem::Specification.new do |spec|
   spec.add_dependency "htmlentities", "~> 4.3.4"
   spec.add_dependency "image_size"
   spec.add_dependency "mime-types"
-  spec.add_dependency "nokogiri", "= 1.8.1"
+  spec.add_dependency "nokogiri"
   spec.add_dependency "ruby-xslt"
   spec.add_dependency "thread_safe"
   spec.add_dependency "uuidtools"

data/lib/isodoc.rb

@@ -23,6 +23,7 @@ require_relative "isodoc/inline"
 require_relative "isodoc/notes"
 require_relative "isodoc/xref_gen"
 require_relative "isodoc/html"
+require_relative "isodoc/i18n"
 require "pp"
 
 module IsoDoc
@@ -36,6 +37,9 @@ module IsoDoc
     # wordcoverpage: Cover page for Word
     # htmlintropage: Introductory page for HTML
     # wordintropage: Introductory page for Word
+    # i18nyaml: YAML file for internationalisation of text
+    # ulstyle: list style in Word CSS for unordered lists
+    # olstyle: list style in Word CSS for ordered lists
     def initialize(options)
       @htmlstylesheet = options[:htmlstylesheet]
       @wordstylesheet = options[:wordstylesheet]
@@ -45,6 +49,9 @@ module IsoDoc
       @wordcoverpage = options[:wordcoverpage]
       @htmlintropage = options[:htmlintropage]
       @wordintropage = options[:wordintropage]
+      @i18nyaml = options[:i18nyaml]
+      @ulstyle = options[:ulstyle]
+      @olstyle = options[:olstyle]
       @termdomain = ""
       @termexample = false
       @note = false
@@ -59,19 +66,41 @@ module IsoDoc
       @in_table = false
       @in_figure = false
       @seen_footnote = Set.new
+      @c = HTMLEntities.new
+      @openmathdelim = "`"
+      @closemathdelim = "`"
+      @lang = "en"
+      @script = "Latn"
     end
 
-    def convert(filename)
-      docxml = Nokogiri::XML(File.read(filename))
-      filename, dir = init_file(filename)
-      docxml.root.default_namespace = ""
-      result = noko do |xml|
+    def convert1(docxml, filename, dir)
+      noko do |xml|
         xml.html do |html|
           html.parent.add_namespace("epub", "http://www.idpf.org/2007/ops")
           html_header(html, docxml, filename, dir)
           make_body(html, docxml)
         end
       end.join("\n")
+    end
+
+    def convert_init(file, filename, debug)
+      docxml = Nokogiri::XML(file)
+      filename, dir = init_file(filename, debug)
+      docxml.root.default_namespace = ""
+      i18n_init(docxml&.at(ns("//bibdata/language"))&.text || "en",
+                docxml&.at(ns("//bibdata/script"))&.text || "Latn")
+      [docxml, filename, dir]
+    end
+
+    def convert(filename, debug = false)
+      convert_file(File.read(filename), filename, debug)
+    end
+
+    def convert_file(file, filename, debug)
+      @openmathdelim, @closemathdelim = extract_delims(file)
+      docxml, filename, dir = convert_init(file, filename, debug)
+      result = convert1(docxml, filename, dir)
+      return result if debug
       postprocess(result, filename, dir)
     end
   end

data/lib/isodoc/blocks.rb

@@ -1,30 +1,11 @@
 module IsoDoc
   class Convert
-    #attr_accessor :termdomain, :termexample, :sourcecode, :note
-    def set_termdomain(termdomain)
-      @termdomain = termdomain
-    end
-
-    def get_termexample
-      @termexample
-    end
-
-    def set_termexample(value)
-      @termexample = value
-    end
-
-    def in_sourcecode
-      @sourcecode
-    end
-
-    def is_note
-      @note
-    end
+    @annotation = false
 
     def note_label(node)
-      n = get_anchors()[node["id"]]
-      return "NOTE" if n.nil?
-      n[:label]
+      n = get_anchors[node["id"]]
+      return @note_lbl if n.nil? || n[:label].empty?
+      l10n("#{@note_lbl} #{n[:label]}")
     end
 
     def note_p_parse(node, div)
@@ -36,17 +17,21 @@ module IsoDoc
       node.element_children[1..-1].each { |n| parse(n, div) }
     end
 
+    def note_parse1(node, div)
+      div.p **{ class: "Note" } do |p|
+        p << note_label(node)
+        insert_tab(p, 1)
+      end
+      node.children.each { |n| parse(n, div) }
+    end
+
     def note_parse(node, out)
       @note = true
       out.div **{ id: node["id"], class: "Note" } do |div|
         if node.first_element_child.name == "p"
           note_p_parse(node, div)
         else
-          div.p **{ class: "Note" } do |p|
-            p << note_label(node)
-            insert_tab(p, 1)
-          end
-          node.children.each { |n| parse(n, div) }
+          note_parse1(node, div)
         end
       end
       @note = false
@@ -55,15 +40,15 @@ module IsoDoc
     def figure_name_parse(node, div, name)
       div.p **{ class: "FigureTitle", align: "center" } do |p|
         p.b do |b|
-          b << "#{get_anchors()[node['id']][:label]}&nbsp;&mdash; "
-          b << name.text
+          b << l10n("#{@figure_lbl} #{get_anchors[node['id']][:label]}")
+          b << "&nbsp;&mdash; #{name.text}" if name
         end
       end
     end
 
     def figure_key(out)
-      out.p do |p| 
-        p.b { |b| b << "Key" }
+      out.p do |p|
+        p.b { |b| b << @key_lbl }
       end
     end
 
@@ -80,17 +65,40 @@ module IsoDoc
       @in_figure = false
     end
 
-        def example_parse(node, out)
-      name = node.at(ns("./name"))
-      out.div **attr_code(id: node["id"], class: "figure") do |div|
+    def example_label(node)
+      n = get_anchors[node["id"]]
+      return @example_lbl if n.nil? || n[:label].empty?
+      l10n("#{@example_lbl} #{n[:label]}")
+    end
+
+    EXAMPLE_TBL_ATTR =
+      { width: "110pt", valign: "top",
+        style: "width:82.8pt;padding:.75pt .75pt .75pt .75pt" }.freeze
+
+    # used if we are boxing examples
+    def example_div_parse(node, out)
+      out.div **attr_code(id: node["id"], class: "example") do |div|
+        out.p { |p| p << example_label(node) }
         node.children.each do |n|
-          parse(n, div) unless n.name == "name"
+          parse(n, div)
         end
-        figure_name_parse(node, div, name) if name
       end
     end
 
-    def sourcecode_name_parse(node, div, name)
+    def example_parse(node, out)
+      out.table **attr_code(id: node["id"], class: "example") do |t|
+        t.tr do |tr|
+          tr.td **EXAMPLE_TBL_ATTR do |td|
+            td << example_label(node)
+          end
+          tr.td **{ valign: "top" } do |td|
+            node.children.each { |n| parse(n, td) }
+          end
+        end
+      end
+    end
+
+    def sourcecode_name_parse(_node, div, name)
       div.p **{ class: "FigureTitle", align: "center" } do |p|
         p.b do |b|
           b << name.text
@@ -111,9 +119,13 @@ module IsoDoc
     end
 
     def annotation_parse(node, out)
-      out.p **{ class: "Sourcecode" } do |li|
-        node.children.each { |n| parse(n, li) }
+      @sourcecode = false
+      @annotation = true
+      out.span **{ class: "zzMoveToFollowing" } do |s|
+        s  << "&lt;#{node.at(ns("//callout[@target='#{node['id']}']")).text}&gt; "
       end
+        node.children.each { |n| parse(n, out) }
+      @annotation = false
     end
 
     def admonition_parse(node, out)
@@ -127,29 +139,29 @@ module IsoDoc
     end
 
     def formula_where(dl, out)
-             out.p { |p| p << "where" }
-        parse(dl, out)
+      return unless dl
+      out.p { |p| p << @where_lbl }
+      parse(dl, out)
     end
 
     def formula_parse(node, out)
-      dl = node.at(ns("./dl"))
       out.div **attr_code(id: node["id"], class: "formula") do |div|
         parse(node.at(ns("./stem")), out)
         insert_tab(div, 1)
-        div << "(#{get_anchors()[node['id']][:label]})"
+        div << "(#{get_anchors[node['id']][:label]})"
       end
-        formula_where(dl, out) if dl
+      formula_where(node.at(ns("./dl")), out)
     end
 
     def para_attrs(node)
       classtype = nil
       classtype = "Note" if @note
-      # classtype = "MsoFootnoteText" if in_footnote
       classtype = "MsoCommentText" if in_comment
+      classtype = "Sourcecode" if @annotation
       attrs = { class: classtype, id: node["id"] }
       unless node["align"].nil?
         attrs[:align] = node["align"] unless node["align"] == "justify"
-        attrs[:style] = "text-align:#{node["align"]}"
+        attrs[:style] = "text-align:#{node['align']}"
       end
       attrs
     end
@@ -169,7 +181,7 @@ module IsoDoc
       source = node.at(ns("./source"))
       out.p **{ class: "QuoteAttribution" } do |p|
         p << "&mdash; #{author.text}, " if author
-        eref_parse(source, p)
+        eref_parse(source, p) if source
       end
     end
 
@@ -177,8 +189,8 @@ module IsoDoc
       attrs = para_attrs(node)
       attrs[:class] = "Quote"
       out.div **attr_code(attrs) do |p|
-        node.children.each do 
-          |n| parse(n, p) unless ["author", "source"].include? n.name
+        node.children.each do |n|
+          parse(n, p) unless ["author", "source"].include? n.name
         end
         quote_attribution(node, out)
       end

data/lib/isodoc/cleanup.rb

@@ -11,8 +11,8 @@ module IsoDoc
       inline_header_cleanup(docxml)
       figure_cleanup(docxml)
       table_cleanup(docxml)
+      symbols_cleanup(docxml)
       admonition_cleanup(docxml)
-      docxml
     end
 
     def admonition_cleanup(docxml)
@@ -21,12 +21,13 @@ module IsoDoc
         n = title.next_element
         n&.children&.first&.add_previous_sibling(title.text + "—")
       end
+      docxml
     end
 
     def figure_get_or_make_dl(t)
       dl = t.at(".//dl")
       if dl.nil?
-        t.add_child("<p><b>Key</b></p><dl></dl>")
+        t.add_child("<p><b>#{@key_lbl}</b></p><dl></dl>")
         dl = t.at(".//dl")
       end
       dl
@@ -62,13 +63,12 @@ module IsoDoc
 
     def inline_header_cleanup(docxml)
       docxml.xpath('//span[@class="zzMoveToFollowing"]').each do |x|
+        x.delete("class")
         n = x.next_element
         if n.nil?
-          html = Nokogiri::XML.fragment("<p></p>")
-          html.parent = x.parent
-          x.parent = html
+          x.name = "p"
         else
-          n.children.first.add_previous_sibling(x.remove)
+          n.children.first.previous = x.remove
         end
       end
     end
@@ -86,10 +86,8 @@ module IsoDoc
       n&.children&.first&.add_previous_sibling(fn.remove)
     end
 
-    TABLE_WITH_FOOTNOTES = "//table[descendant::aside]".freeze
-
     def table_footnote_cleanup(docxml)
-      docxml.xpath(TABLE_WITH_FOOTNOTES).each do |t|
+      docxml.xpath("//table[descendant::aside]").each do |t|
         t.xpath(".//aside").each do |a|
           merge_fnref_into_fn_text(a)
           a.name = "div"
@@ -139,12 +137,38 @@ module IsoDoc
                    "[ancestor::*[@class = 'Note']]").each do |p|
         p["class"] = "Note"
       end
-
     end
 
     def table_cleanup(docxml)
       table_footnote_cleanup(docxml)
       table_note_cleanup(docxml)
     end
+
+    # We assume AsciiMath. Indices sort after letter but before any following
+    # letter (x, x_m, x_1, xa); we use colon to force that sort order.
+    # Numbers sort *after* letters; we use thorn to force that sort order.
+    def symbol_key(x)
+      HTMLEntities.new.decode(x.text).gsub(/_/, ":").gsub(/`/, "").
+        gsub(/[0-9]+/, "þ\\1")
+    end
+
+    def extract_symbols_list(dl)
+      dl_out = []
+      dl.xpath("./dt | ./dd").each do |dtd|
+        if dtd.name == "dt"
+          dl_out << { dt: dtd.remove, key: symbol_key(dtd) }
+        else
+          dl_out.last[:dd] = dtd.remove
+        end
+      end
+      dl_out
+    end
+
+    def symbols_cleanup(docxml)
+      dl = docxml.at("//div[@class = 'Symbols']/dl") || return
+      dl_out = extract_symbols_list(dl)
+      dl_out.sort! { |a, b| a[:key] <=> b[:key] }
+      dl.replace(dl_out.map { |d| d[:dt].to_s + d[:dd].to_s }.join("\n"))
+    end
   end
 end

data/lib/isodoc/html.rb

@@ -1,8 +1,6 @@
 module IsoDoc
   class Convert
-
     def toHTML(result, filename)
-      # result = html_cleanup(Nokogiri::HTML(result)).to_xml
       result = from_xhtml(html_cleanup(to_xhtml(result)))
       result = populate_template(result, :html)
       File.open("#{filename}.html", "w") do |f|
@@ -11,19 +9,37 @@ module IsoDoc
     end
 
     def html_cleanup(x)
-      footnote_backlinks(move_images(html_footnote_filter(htmlPreface(htmlstyle(x)))))
+      footnote_backlinks(
+        move_images(html_footnote_filter(html_preface(htmlstyle(x))))
+      )
     end
 
-    def htmlPreface(docxml)
+    MATHJAX_ADDR =
+      "https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.1/MathJax.js".freeze
+    MATHJAX = <<~"MATHJAX".freeze
+      <script type="text/x-mathjax-config">
+        MathJax.Hub.Config({
+          asciimath2jax: {
+            delimiters: [['OPEN', 'CLOSE']]
+          }
+       });
+      </script>
+      <script src="#{MATHJAX_ADDR}?config=AM_HTMLorMML"></script>
+    MATHJAX
+
+    def mathjax(open, close)
+      MATHJAX.gsub("OPEN", open).gsub("CLOSE", close)
+    end
+
+    def html_preface(docxml)
       cover = Nokogiri::HTML(File.read(@htmlcoverpage, encoding: "UTF-8"))
       d = docxml.at('//div[@class="WordSection1"]')
-      d.children.first.add_previous_sibling cover.to_xml(encoding: 'US-ASCII')
+      d.children.first.add_previous_sibling cover.to_xml(encoding: "US-ASCII")
       cover = Nokogiri::HTML(File.read(@htmlintropage, encoding: "UTF-8"))
       d = docxml.at('//div[@class="WordSection2"]')
-      d.children.first.add_previous_sibling cover.to_xml(encoding: 'US-ASCII')
-      body = docxml.at("//*[local-name() = 'body']")
-      body << '<script src="https://cdn.mathjax.org/mathjax/latest/'\
-        'MathJax.js?config=AM_HTMLorMML"></script>'
+      d.children.first.add_previous_sibling cover.to_xml(encoding: "US-ASCII")
+      docxml.at("//*[local-name() = 'body']") << mathjax(@openmathdelim,
+                                                         @closemathdelim)
       docxml
     end
 
@@ -45,8 +61,7 @@ module IsoDoc
       docxml
     end
 
-    def update_footnote_filter(docxml, x, i, seen)
-      fn = docxml.at(%<//*[@id = '#{x['href'].sub(/^#/, '')}']>) || return
+    def update_footnote_filter(fn, x, i, seen)
       if seen[fn.text]
         x.at("./sup").content = seen[fn.text][:num].to_s
         fn.remove unless x["href"] == seen[fn.text][:href]
@@ -63,7 +78,8 @@ module IsoDoc
       seen = {}
       i = 1
       docxml.xpath('//a[@epub:type = "footnote"]').each do |x|
-        i, seen = update_footnote_filter(docxml, x, i, seen)
+        fn = docxml.at(%<//*[@id = '#{x['href'].sub(/^#/, '')}']>) || next
+        i, seen = update_footnote_filter(fn, x, i, seen)
       end
       docxml
     end
@@ -73,22 +89,21 @@ module IsoDoc
       docxml.xpath('//a[@epub:type = "footnote"]').each_with_index do |x, i|
         next if seen[x["href"]]
         seen[x["href"]] = true
-        sup = x.at("./sup").text
         fn = docxml.at(%<//*[@id = '#{x['href'].sub(/^#/, '')}']>) || next
         x["id"] || x["id"] = "_footnote#{i + 1}"
-        fn.elements.first.children.first.
-          add_previous_sibling("<a href='##{x['id']}'>#{sup}) </a>")
+        fn.elements.first.children.first.previous =
+          "<a href='##{x['id']}'>#{x.at('./sup').text}) </a>"
       end
       docxml
     end
 
+    # presupposes that the image source is local
     def move_images(docxml)
       system "rm -r _images; mkdir _images"
       docxml.xpath("//*[local-name() = 'img']").each do |i|
         matched = /\.(?<suffix>\S+)$/.match i["src"]
         uuid = UUIDTools::UUID.random_create.to_s
         new_full_filename = File.join("_images", "#{uuid}.#{matched[:suffix]}")
-        # presupposes that the image source is local
         system "cp #{i['src']} #{new_full_filename}"
         i["src"] = new_full_filename
         i["width"], i["height"] = Html2Doc.image_resize(i, 800, 1200)