isodoc 1.0.26 → 1.1.1

data/lib/isodoc/{function → xref}/xref_sect_gen.rb

@@ -1,18 +1,16 @@
-module IsoDoc::Function
-  module XrefSectGen
+module IsoDoc::XrefGen
+  module Sections
     def back_anchor_names(docxml)
       docxml.xpath(ns("//annex")).each_with_index do |c, i|
         annex_names(c, (65 + i).chr.to_s)
       end
       docxml.xpath(
-           ns("//bibliography/clause[.//references[@normative = 'false']] | "\
-           "//bibliography/references[@normative = 'false']"
-           )).each do |b|
-        preface_names(b)
-      end
-      docxml.xpath(ns("//bibitem[not(ancestor::bibitem)]")).each do |ref|
-        reference_names(ref)
-      end
+           ns(@klass.bibliography_xpath)).each do |b|
+            preface_names(b)
+          end
+          docxml.xpath(ns("//bibitem[not(ancestor::bibitem)]")).each do |ref|
+            reference_names(ref)
+          end
     end
 
     def initial_anchor_names(d)
@@ -20,9 +18,7 @@ module IsoDoc::Function
       # potentially overridden in middle_section_asset_names()
       sequential_asset_names(d.xpath(ns("//preface/*")))
       n = section_names(d.at(ns("//clause[title = 'Scope']")), 0, 1)
-      n = section_names(d.at(ns(
-        "//bibliography/clause[.//references[@normative = 'true']] | "\
-        "//bibliography/references[@normative = 'true']")), n, 1)
+      n = section_names(d.at(ns(@klass.norm_ref_xpath)), n, 1)
       n = section_names(d.at(ns("//sections/terms | "\
                                 "//sections/clause[descendant::terms]")), n, 1)
       n = section_names(d.at(ns("//sections/definitions")), n, 1)
@@ -65,15 +61,14 @@ module IsoDoc::Function
 
     def middle_section_asset_names(d)
       middle_sections = "//clause[title = 'Scope'] | "\
-        "//references[@normative = 'true'] | "\
+        "#{@klass.norm_ref_xpath} | "\
         "//sections/terms | //preface/* | "\
         "//sections/definitions | //clause[parent::sections]"
       sequential_asset_names(d.xpath(ns(middle_sections)))
     end
 
     def clause_names(docxml, sect_num)
-      q = self.class::MIDDLE_CLAUSE
-      docxml.xpath(ns(q)).each_with_index do |c, i|
+      docxml.xpath(ns(@klass.middle_clause)).each_with_index do |c, i|
         section_names(c, (i + sect_num), 1)
       end
     end
@@ -82,7 +77,7 @@ module IsoDoc::Function
       return num if clause.nil?
       num = num + 1
       @anchors[clause["id"]] =
-        { label: num.to_s, xref: l10n("#{@clause_lbl} #{num}"), level: lvl,
+        { label: num.to_s, xref: l10n("#{@labels["clause"]} #{num}"), level: lvl,
           type: "clause" }
       clause.xpath(ns(SUBCLAUSES)).
         each_with_index do |c, i|
@@ -93,7 +88,7 @@ module IsoDoc::Function
 
     def section_names1(clause, num, level)
       @anchors[clause["id"]] =
-        { label: num, level: level, xref: l10n("#{@clause_lbl} #{num}"),
+        { label: num, level: level, xref: l10n("#{@labels["clause"]} #{num}"),
           type: "clause" }
       clause.xpath(ns(SUBCLAUSES)).
         each_with_index do |c, i|
@@ -102,9 +97,9 @@ module IsoDoc::Function
     end
 
     def annex_name_lbl(clause, num)
-      obl = l10n("(#{@inform_annex_lbl})")
-      obl = l10n("(#{@norm_annex_lbl})") if clause["obligation"] == "normative"
-      l10n("<b>#{@annex_lbl} #{num}</b><br/>#{obl}")
+      obl = l10n("(#{@labels["inform_annex"]})")
+      obl = l10n("(#{@labels["norm_annex"]})") if clause["obligation"] == "normative"
+      l10n("<b>#{@labels["annex"]} #{num}</b><br/>#{obl}")
     end
 
     def single_annex_special_section(clause)
@@ -119,7 +114,7 @@ module IsoDoc::Function
     def annex_names(clause, num)
       @anchors[clause["id"]] = { label: annex_name_lbl(clause, num),
                                  type: "clause",
-                                 xref: "#{@annex_lbl} #{num}", level: 1 }
+                                 xref: "#{@labels["annex"]} #{num}", level: 1 }
       if a = single_annex_special_section(clause)
         annex_names1(a, "#{num}", 1)
       else
@@ -131,11 +126,26 @@ module IsoDoc::Function
     end
 
     def annex_names1(clause, num, level)
-      @anchors[clause["id"]] = { label: num, xref: "#{@annex_lbl} #{num}",
+      @anchors[clause["id"]] = { label: num, xref: "#{@labels["annex"]} #{num}",
                                  level: level, type: "clause" }
       clause.xpath(ns(SUBCLAUSES)).each_with_index do |c, i|
         annex_names1(c, "#{num}.#{i + 1}", level + 1)
       end
     end
+
+    ISO_PUBLISHER_XPATH =
+      "./contributor[xmlns:role/@type = 'publisher']/"\
+      "organization[abbreviation = 'ISO' or xmlns:abbreviation = 'IEC' or "\
+      "xmlns:name = 'International Organization for Standardization' or "\
+      "xmlns:name = 'International Electrotechnical Commission']".freeze
+
+    def reference_names(ref)
+      isopub = ref.at(ns(ISO_PUBLISHER_XPATH))
+      ids = @klass.bibitem_ref_code(ref)
+      identifiers = @klass.render_identifier(ids)
+      date = ref.at(ns("./date[@type = 'published']"))
+      reference = @klass.docid_l10n(identifiers[0] || identifiers[1])
+      @anchors[ref["id"]] = { xref: reference }
+    end
   end
 end

data/lib/isodoc/xslfo_convert.rb

@@ -15,14 +15,26 @@ module IsoDoc
 
     def initialize(options)
       @format = :pdf
+      @suffix = "pdf"
       super
     end
 
-    def convert(filename, file = nil, debug = false)
-      file = File.read(filename, encoding: "utf-8") if file.nil?
-      docxml, outname_html, dir = convert_init(file, filename, debug)
+    def pdf_stylesheet(docxml)
+      nil
+    end
+
+    def convert(input_filename, file = nil, debug = false, output_filename = nil)
+      file = File.read(input_filename, encoding: "utf-8") if file.nil?
+      docxml, filename, dir = convert_init(file, input_filename, debug)
+      /\.xml$/.match(input_filename) or
+          input_filename = Tempfile.open([filename, ".xml"], encoding: "utf-8") do |f|
+          f.write file
+          f.path
+        end
       FileUtils.rm_rf dir
-      ::Metanorma::Output::XslfoPdf.new.convert(filename, outname_html + ".pdf", nil)
+      ::Metanorma::Output::XslfoPdf.new.convert(input_filename,
+                                                output_filename || "#{filename}.#{@suffix}", 
+                                               File.join(@libdir, pdf_stylesheet(docxml)))
     end
 
     def xref_parse(node, out)

data/spec/assets/i18n.yaml

@@ -1,2 +1,5 @@
 foreword: Enkonduko
-
+clause: klaŭzo
+table: Tabelo
+source: SOURCE
+modified: modified

data/spec/assets/odf.svg

@@ -0,0 +1,4 @@
+<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 100 100">
+  <circle fill="#009" r="45" cx="50" cy="50"/>
+  <path d="M33,26H78A37,37,0,0,1,33,83V57H59V43H33Z" fill="#FFF"/>
+</svg>

data/spec/assets/odf1.svg

@@ -0,0 +1,4 @@
+<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 100 100">
+  <circle fill="#009" r="45" cx="50" cy="50"/>
+  <path d="M33,26H78A37,37,0,0,1,33,83V57H59V43H33Z" fill="#FFF"/>
+</svg>

data/spec/isodoc/blocks_spec.rb

@@ -5,7 +5,7 @@ RSpec.describe IsoDoc do
     expect(xmlpp(IsoDoc::HtmlConvert.new({}).convert("test", <<~"INPUT", true))).to be_equivalent_to xmlpp(<<~"OUTPUT")
     <iso-standard xmlns="http://riboseinc.com/isoxml">
     <preface><foreword>
-    <note>
+    <note id="A" keep-with-next="true" keep-lines-together="true">
   <p id="_f06fd0d1-a203-4f3d-a515-0bdba0f8d83f">These results are based on a study carried out on three different types of kernel.</p>
 </note>
     </foreword></preface>
@@ -15,7 +15,7 @@ RSpec.describe IsoDoc do
                <br/>
                <div>
                  <h1 class="ForewordTitle">Foreword</h1>
-                 <div id="" class="Note">
+                 <div id="A" class="Note" style="page-break-after: avoid;page-break-inside: avoid;">
                    <p><span class="note_label">NOTE</span>&#160; These results are based on a study carried out on three different types of kernel.</p>
                  </div>
                </div>
@@ -30,7 +30,7 @@ RSpec.describe IsoDoc do
     expect(xmlpp(IsoDoc::WordConvert.new({}).convert("test", <<~"INPUT", true))).to be_equivalent_to xmlpp(<<~"OUTPUT")
     <iso-standard xmlns="http://riboseinc.com/isoxml">
     <preface><foreword>
-    <note>
+    <note id="A">
   <p id="_f06fd0d1-a203-4f3d-a515-0bdba0f8d83f">These results are based on a study carried out on three different types of kernel.</p>
 </note>
     </foreword></preface>
@@ -47,7 +47,7 @@ RSpec.describe IsoDoc do
              <p><br clear="all" style="mso-special-character:line-break;page-break-before:always"/></p>
              <div>
                <h1 class="ForewordTitle">Foreword</h1>
-               <div id="" class="Note">
+               <div id="A" class="Note">
                  <p class="Note"><span class="note_label">NOTE</span><span style="mso-tab-count:1">&#160; </span>These results are based on a study carried out on three different types of kernel.</p>
                </div>
              </div>
@@ -134,7 +134,7 @@ INPUT
                <br/>
                <div>
                  <h1 class="ForewordTitle">Foreword</h1>
-                 <div id="" class="Note">
+                 <div class="Note">
                    <p><span class="note_label">NOTE</span>&#160; These results are based on a study carried out on three different types of kernel.</p>
                    <p id="_f06fd0d1-a203-4f3d-a515-0bdba0f8d83a">These results are based on a study carried out on three different types of kernel.</p>
                  </div>
@@ -150,7 +150,7 @@ INPUT
     expect(xmlpp(IsoDoc::HtmlConvert.new({}).convert("test", <<~"INPUT", true))).to be_equivalent_to xmlpp(<<~"OUTPUT")
     <iso-standard xmlns="http://riboseinc.com/isoxml">
     <preface><foreword>
-    <note>
+    <note id="A">
     <dl>
     <dt>A</dt>
     <dd><p>B</p></dd>
@@ -165,7 +165,7 @@ INPUT
                <br/>
                <div>
                  <h1 class="ForewordTitle">Foreword</h1>
-                 <div id="" class="Note"><p><span class="note_label">NOTE</span>&#160; </p>
+                 <div id="A" class="Note"><p><span class="note_label">NOTE</span>&#160; </p>
            <dl><dt><p>A</p></dt><dd><p>B</p></dd></dl>
            <ul>
            <li>C</li></ul>
@@ -183,7 +183,7 @@ INPUT
     expect(xmlpp(IsoDoc::WordConvert.new({}).convert("test", <<~"INPUT", true))).to be_equivalent_to xmlpp(<<~"OUTPUT")
     <iso-standard xmlns="http://riboseinc.com/isoxml">
     <preface><foreword>
-    <note>
+    <note id="A">
     <dl>
     <dt>A</dt>
     <dd><p>B</p></dd>
@@ -205,7 +205,7 @@ INPUT
              <p><br clear="all" style="mso-special-character:line-break;page-break-before:always"/></p>
              <div>
                <h1 class="ForewordTitle">Foreword</h1>
-               <div id="" class="Note"><p class="Note"><span class="note_label">NOTE</span><span style="mso-tab-count:1">&#160; </span></p>
+               <div id="A" class="Note"><p class="Note"><span class="note_label">NOTE</span><span style="mso-tab-count:1">&#160; </span></p>
            <table class="dl"><tr><td valign="top" align="left"><p align="left" style="margin-left:0pt;text-align:left;">A</p></td><td valign="top"><p class="Note">B</p></td></tr></table>
            <ul>
            <li>C</li></ul>
@@ -317,7 +317,7 @@ OUTPUT
     expect(xmlpp(strip_guid(IsoDoc::HtmlConvert.new({}).convert("test", <<~"INPUT", true)))).to be_equivalent_to xmlpp(<<~"OUTPUT")
     <iso-standard xmlns="http://riboseinc.com/isoxml">
     <preface><foreword>
-    <figure id="figureA-1">
+    <figure id="figureA-1" keep-with-next="true" keep-lines-together="true">
   <name>Split-it-right <em>sample</em> divider<fn reference="1"><p>X</p></fn></name>
   <image src="rice_images/rice_image1.png" height="20" width="30" id="_8357ede4-6d44-4672-bac4-9a85e82ab7f0" mimetype="image/png" alt="alttext" title="titletxt"/>
   <image src="rice_images/rice_image1.png" height="20" width="auto" id="_8357ede4-6d44-4672-bac4-9a85e82ab7f1" mimetype="image/png"/>
@@ -345,7 +345,7 @@ B
                <br/>
                <div>
                  <h1 class="ForewordTitle">Foreword</h1>
-                 <div id="figureA-1" class="figure">
+                 <div id="figureA-1" class="figure" style='page-break-after: avoid;page-break-inside: avoid;'>
 
          <img src="rice_images/rice_image1.png" height="20" width="30" alt="alttext" title="titletxt"/>
          <img src="rice_images/rice_image1.png" height="20" width="auto"/>
@@ -355,7 +355,7 @@ B
        </div></aside>
          <p  style='page-break-after:avoid;'><b>Key</b></p><dl><dt><p>A</p></dt><dd><p>B</p></dd></dl>
        <p class="FigureTitle" style="text-align:center;">Figure 1&#160;&#8212; Split-it-right <i>sample</i> divider
-       <a rel='footnote' href='#fn:1' epub:type='footnote'>
+       <a class='FootnoteRef' href='#fn:1'>
   <sup>1</sup>
 </a>
         </p></div>
@@ -380,6 +380,7 @@ B
   end
 
   it "processes figures (Word)" do
+    FileUtils.rm_rf "spec/assets/odf1.emf"
     expect(xmlpp(strip_guid(IsoDoc::WordConvert.new({}).convert("test", <<~"INPUT", true).sub(/['"][^'".]+\.gif['"]/, "'_.gif'").gsub(/mso-bookmark:_Ref\d+/, "mso-bookmark:_Ref")))).to be_equivalent_to xmlpp(<<~"OUTPUT")
     <iso-standard xmlns="http://riboseinc.com/isoxml">
     <preface><foreword>
@@ -426,7 +427,7 @@ B
           <p class='FigureTitle' style='text-align:center;'>
    Figure 1&#160;&#8212; Split-it-right sample divider
    <span style='mso-bookmark:_Ref'>
-     <a href='#ftn1' epub:type='footnote'>
+     <a href='#ftn1' epub:type='footnote' class='FootnoteRef'>
        <sup>1</sup>
      </a>
    </span>
@@ -452,11 +453,122 @@ B
     OUTPUT
   end
 
+   it "converts SVG (Word)" do
+    FileUtils.rm_rf "spec/assets/odf1.emf"
+    expect(xmlpp(strip_guid(IsoDoc::WordConvert.new({}).convert("test", <<~"INPUT", true).gsub(/['"][^'".]+(?<!odf1)(?<!odf)\.emf['"]/, "'_.emf'").gsub(/mso-bookmark:_Ref\d+/, "mso-bookmark:_Ref")))).to be_equivalent_to xmlpp(<<~"OUTPUT")
+    <iso-standard xmlns="http://riboseinc.com/isoxml">
+    <preface><foreword>
+    <figure id="figureA-1">
+  <image src="spec/assets/odf.svg" mimetype="image/svg+xml"/>
+  <image src="spec/assets/odf1.svg" mimetype="image/svg+xml"/>
+  <image src="data:image/svg+xml;base64,PHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHZpZXdCb3g9IjAgMCAxMDAgMTAwIj4KICA8Y2lyY2xlIGZpbGw9IiMwMDkiIHI9IjQ1IiBjeD0iNTAiIGN5PSI1MCIvPgogIDxwYXRoIGQ9Ik0zMywyNkg3OEEzNywzNywwLDAsMSwzMyw4M1Y1N0g1OVY0M0gzM1oiIGZpbGw9IiNGRkYiLz4KPC9zdmc+Cg==" id="_d3731866-1a07-435a-a6c2-1acd41023a4e" mimetype="image/svg+xml" height="auto" width="auto"/>
+</figure>
+    </foreword></preface>
+    </iso-standard>
+    INPUT
+    <html xmlns:epub='http://www.idpf.org/2007/ops' lang='en'>
+<head>
+<style>
+        </style>
+  </head>
+  <body lang='EN-US' link='blue' vlink='#954F72'>
+    <div class='WordSection1'>
+      <p>&#160;</p>
+    </div>
+    <p>
+      <br clear='all' class='section'/>
+    </p>
+    <div class='WordSection2'>
+      <p>
+        <br clear='all' style='mso-special-character:line-break;page-break-before:always'/>
+      </p>
+      <div>
+        <h1 class='ForewordTitle'>Foreword</h1>
+        <div id='figureA-1' class='figure'>
+          <img src='spec/assets/odf.emf'/>
+          <img src='spec/assets/odf1.emf'/>
+          <img src='_.emf' height='auto' width='auto'/>
+          <p class='FigureTitle' style='text-align:center;'>Figure 1</p>
+        </div>
+      </div>
+      <p>&#160;</p>
+    </div>
+    <p>
+      <br clear='all' class='section'/>
+    </p>
+    <div class='WordSection3'>
+      <p class='zzSTDTitle1'/>
+    </div>
+  </body>
+</html>
+
+OUTPUT
+    end
+
+context "disable inkscape" do
+
+   it "converts SVG (Word) with inkscape disabled" do
+    FileUtils.rm_rf "spec/assets/odf1.emf"
+    allow(IsoDoc::WordFunction::Body).to receive(:inkscape_installed?).and_return(nil)
+    allow_any_instance_of(IsoDoc::WordFunction::Body).to receive(:inkscape_installed?)
+
+    expect(xmlpp(strip_guid(IsoDoc::WordConvert.new({}).convert("test", <<~"INPUT", true).gsub(/['"][^'".]+(?<!odf1)(?<!odf)\.svg['"]/, "'_.svg'").gsub(/mso-bookmark:_Ref\d+/, "mso-bookmark:_Ref")))).to be_equivalent_to xmlpp(<<~"OUTPUT")
+    <iso-standard xmlns="http://riboseinc.com/isoxml">
+    <preface><foreword>
+    <figure id="figureA-1">
+  <image src="spec/assets/odf.svg" mimetype="image/svg+xml"/>
+  <image src="spec/assets/odf1.svg" mimetype="image/svg+xml"/>
+  <image src="data:image/svg+xml;base64,PHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHZpZXdCb3g9IjAgMCAxMDAgMTAwIj4KICA8Y2lyY2xlIGZpbGw9IiMwMDkiIHI9IjQ1IiBjeD0iNTAiIGN5PSI1MCIvPgogIDxwYXRoIGQ9Ik0zMywyNkg3OEEzNywzNywwLDAsMSwzMyw4M1Y1N0g1OVY0M0gzM1oiIGZpbGw9IiNGRkYiLz4KPC9zdmc+Cg==" id="_d3731866-1a07-435a-a6c2-1acd41023a4e" mimetype="image/svg+xml" height="auto" width="auto"/>
+</figure>
+    </foreword></preface>
+    </iso-standard>
+    INPUT
+ <html xmlns:epub='http://www.idpf.org/2007/ops' lang='en'>
+<head>
+<style>
+        </style>
+  </head>
+  <body lang='EN-US' link='blue' vlink='#954F72'>
+    <div class='WordSection1'>
+      <p>&#160;</p>
+    </div>
+    <p>
+      <br clear='all' class='section'/>
+    </p>
+    <div class='WordSection2'>
+      <p>
+        <br clear='all' style='mso-special-character:line-break;page-break-before:always'/>
+      </p>
+      <div>
+        <h1 class='ForewordTitle'>Foreword</h1>
+        <div id='figureA-1' class='figure'>
+          <img src='spec/assets/odf.emf'/>
+          <img src='spec/assets/odf1.svg'/>
+          <img src='_.svg' height='auto' width='auto'/>
+          <p class='FigureTitle' style='text-align:center;'>Figure 1</p>
+        </div>
+      </div>
+      <p>&#160;</p>
+    </div>
+    <p>
+      <br clear='all' class='section'/>
+    </p>
+    <div class='WordSection3'>
+      <p class='zzSTDTitle1'/>
+    </div>
+  </body>
+</html>
+
+OUTPUT
+    end
+   end
+
+
   it "processes examples" do
     expect(xmlpp(IsoDoc::HtmlConvert.new({}).convert("test", <<~"INPUT", true))).to be_equivalent_to xmlpp(<<~"OUTPUT")
     <iso-standard xmlns="http://riboseinc.com/isoxml">
     <preface><foreword>
-    <example id="samplecode">
+    <example id="samplecode" keep-with-next="true" keep-lines-together="true">
     <name>Title</name>
   <p>Hello</p>
   <sourcecode id="X">
@@ -470,7 +582,7 @@ B
                <br/>
                <div>
                  <h1 class="ForewordTitle">Foreword</h1>
-                 <div id="samplecode" class="example">
+                 <div id="samplecode" class="example" style="page-break-after: avoid;page-break-inside: avoid;">
                  <p class="example-title">EXAMPLE&#160;&#8212; Title</p>
          <p>Hello</p>
          <pre id='X' class='prettyprint '>
@@ -489,6 +601,60 @@ B
     OUTPUT
   end
 
+   it "processes examples (Word)" do
+    expect(xmlpp(IsoDoc::WordConvert.new({}).convert("test", <<~"INPUT", true))).to be_equivalent_to xmlpp(<<~"OUTPUT")
+    <iso-standard xmlns="http://riboseinc.com/isoxml">
+    <preface><foreword>
+    <example id="samplecode" keep-with-next="true" keep-lines-together="true">
+    <name>Title</name>
+  <p>Hello</p>
+  <sourcecode id="X">
+  <name>Sample</name>
+  </sourcecode>
+</example>
+    </foreword></preface>
+    </iso-standard>
+    INPUT
+    <html  xmlns:epub='http://www.idpf.org/2007/ops' lang='en'><head><style>
+        </style>
+  </head>
+  <body lang='EN-US' link='blue' vlink='#954F72'>
+    <div class='WordSection1'>
+      <p>&#160;</p>
+    </div>
+    <p>
+      <br clear='all' class='section'/>
+    </p>
+    <div class='WordSection2'>
+      <p>
+        <br clear='all' style='mso-special-character:line-break;page-break-before:always'/>
+      </p>
+      <div>
+        <h1 class='ForewordTitle'>Foreword</h1>
+        <div id='samplecode' class='example' style='page-break-after: avoid;page-break-inside: avoid;'>
+          <p class='example-title'>EXAMPLE&#160;&#8212; Title</p>
+          <p>Hello</p>
+          <p id='X' class='Sourcecode'>
+            <br/>
+            &#160;
+            <br/>
+            &#160;
+          </p>
+          <p class='SourceTitle' style='text-align:center;'>Sample</p>
+        </div>
+      </div>
+      <p>&#160;</p>
+    </div>
+    <p>
+      <br clear='all' class='section'/>
+    </p>
+    <div class='WordSection3'>
+      <p class='zzSTDTitle1'/>
+    </div>
+  </body>
+</html>
+OUTPUT
+   end
 
   it "processes sequences of examples" do
     expect(xmlpp(IsoDoc::HtmlConvert.new({}).convert("test", <<~"INPUT", true))).to be_equivalent_to xmlpp(<<~"OUTPUT")
@@ -671,7 +837,7 @@ Que?
     expect(xmlpp(IsoDoc::HtmlConvert.new({}).convert("test", <<~"INPUT", true))).to be_equivalent_to xmlpp(<<~"OUTPUT")
     <iso-standard xmlns="http://riboseinc.com/isoxml">
     <preface><foreword>
-    <admonition id="_70234f78-64e5-4dfc-8b6f-f3f037348b6a" type="caution">
+    <admonition id="_70234f78-64e5-4dfc-8b6f-f3f037348b6a" type="caution" keep-with-next="true" keep-lines-together="true">
   <p id="_e94663cc-2473-4ccc-9a72-983a74d989f2">Only use paddy or parboiled rice for the determination of husked rice yield.</p>
 </admonition>
     </foreword></preface>
@@ -681,7 +847,7 @@ Que?
                <br/>
                <div>
                  <h1 class="ForewordTitle">Foreword</h1>
-                 <div class="Admonition"><p class="AdmonitionTitle" style="text-align:center;">CAUTION</p>
+                 <div class="Admonition" id='_70234f78-64e5-4dfc-8b6f-f3f037348b6a' style='page-break-after: avoid;page-break-inside: avoid;'><p class="AdmonitionTitle" style="text-align:center;">CAUTION</p>
          <p id="_e94663cc-2473-4ccc-9a72-983a74d989f2">Only use paddy or parboiled rice for the determination of husked rice yield.</p>
        </div>
                </div>
@@ -707,7 +873,7 @@ Que?
                <br/>
                <div>
                  <h1 class="ForewordTitle">Foreword</h1>
-                 <div class="Admonition"><p class="AdmonitionTitle" style="text-align:center;">Title</p>
+                 <div class="Admonition" id="_70234f78-64e5-4dfc-8b6f-f3f037348b6a"><p class="AdmonitionTitle" style="text-align:center;">Title</p>
          <p id="_e94663cc-2473-4ccc-9a72-983a74d989f2">Only use paddy or parboiled rice for the determination of husked rice yield.</p>
        </div>
                </div>
@@ -723,7 +889,7 @@ Que?
     expect(xmlpp(IsoDoc::HtmlConvert.new({}).convert("test", <<~"INPUT", true))).to be_equivalent_to xmlpp(<<~"OUTPUT")
     <iso-standard xmlns="http://riboseinc.com/isoxml">
     <preface><foreword>
-    <formula id="_be9158af-7e93-4ee2-90c5-26d31c181934" unnumbered="true">
+    <formula id="_be9158af-7e93-4ee2-90c5-26d31c181934" unnumbered="true"  keep-with-next="true" keep-lines-together="true">
   <stem type="AsciiMath">r = 1 %</stem>
 <dl id="_e4fe94fe-1cde-49d9-b1ad-743293b7e21d">
   <dt>
@@ -747,16 +913,16 @@ Que?
                <br/>
                <div>
                  <h1 class="ForewordTitle">Foreword</h1>
-                 <div id="_be9158af-7e93-4ee2-90c5-26d31c181934" class="formula"><p><span class="stem">(#(r = 1 %)#)</span></p></div><p style='page-break-after:avoid;'>where</p><dl id="_e4fe94fe-1cde-49d9-b1ad-743293b7e21d" class="formula_dl"><dt>
+                 <div id="_be9158af-7e93-4ee2-90c5-26d31c181934" style='page-break-after: avoid;page-break-inside: avoid;'><div class="formula"><p><span class="stem">(#(r = 1 %)#)</span></p></div><p style='page-break-after:avoid;'>where</p><dl id="_e4fe94fe-1cde-49d9-b1ad-743293b7e21d" class="formula_dl"><dt>
            <span class="stem">(#(r)#)</span>
          </dt><dd>
            <p id="_1b99995d-ff03-40f5-8f2e-ab9665a69b77">is the repeatability limit.</p>
          </dd></dl>
 
 
-           <div id="_83083c7a-6c85-43db-a9fa-4d8edd0c9fc0" class="Note"><p><span class="note_label">NOTE</span>&#160; [durationUnits] is essentially a duration statement without the "P" prefix. "P" is unnecessary because between "G" and "U" duration is always expressed.</p></div>
+           <div id="_83083c7a-6c85-43db-a9fa-4d8edd0c9fc0" class="Note"><p><span class="note_label">NOTE</span>&#160; [durationUnits] is essentially a duration statement without the "P" prefix. "P" is unnecessary because between "G" and "U" duration is always expressed.</p></div></div>
 
-                 <div id="_be9158af-7e93-4ee2-90c5-26d31c181935" class="formula"><p><span class="stem">(#(r = 1 %)#)</span>&#160; (1)</p></div>
+                 <div id="_be9158af-7e93-4ee2-90c5-26d31c181935"><div class="formula"><p><span class="stem">(#(r = 1 %)#)</span>&#160; (1)</p></div></div>
                  </div>
                <p class="zzSTDTitle1"/>
              </div>
@@ -769,7 +935,7 @@ Que?
     expect(xmlpp(IsoDoc::WordConvert.new({}).convert("test", <<~"INPUT", true))).to be_equivalent_to xmlpp(<<~"OUTPUT")
     <iso-standard xmlns="http://riboseinc.com/isoxml">
     <preface><foreword>
-    <formula id="_be9158af-7e93-4ee2-90c5-26d31c181934" unnumbered="true">
+    <formula id="_be9158af-7e93-4ee2-90c5-26d31c181934" unnumbered="true"   keep-with-next="true" keep-lines-together="true">
   <stem type="AsciiMath">r = 1 %</stem>
 <dl id="_e4fe94fe-1cde-49d9-b1ad-743293b7e21d">
   <dt>
@@ -807,7 +973,7 @@ Que?
              </p>
              <div>
                <h1 class='ForewordTitle'>Foreword</h1>
-               <div id='_be9158af-7e93-4ee2-90c5-26d31c181934' class='formula'>
+               <div id='_be9158af-7e93-4ee2-90c5-26d31c181934' style='page-break-after: avoid;page-break-inside: avoid;'><div class='formula'>
                  <p>
                    <span class='stem'>(#(r = 1 %)#)</span>
                  </p>
@@ -834,12 +1000,14 @@ Que?
                    always expressed.
                  </p>
                </div>
-               <div id='_be9158af-7e93-4ee2-90c5-26d31c181935' class='formula'>
+               </div>
+               <div id='_be9158af-7e93-4ee2-90c5-26d31c181935'><div class='formula'>
                  <p>
                    <span class='stem'>(#(r = 1 %)#)</span>
                    <span style='mso-tab-count:1'>&#160; </span>
                    (1)
                  </p>
+                 </div>
                </div>
              </div>
              <p>&#160;</p>
@@ -855,14 +1023,14 @@ Que?
     OUTPUT
   end
 
-  it "processes paragraph alignments" do
+  it "processes paragraph attributes" do
     expect(xmlpp(IsoDoc::HtmlConvert.new({}).convert("test", <<~"INPUT", true))).to be_equivalent_to xmlpp(<<~"OUTPUT")
     <iso-standard xmlns="http://riboseinc.com/isoxml">
     <preface><foreword>
     <p align="left" id="_08bfe952-d57f-4150-9c95-5d52098cc2a8">Vache Equipment<br/>
 Fictitious<br/>
 World</p>
-    <p align="justify">Justify</p>
+    <p align="justify" keep-with-next="true" keep-lines-together="true">Justify</p>
     </foreword></preface>
     </iso-standard>
     INPUT
@@ -874,7 +1042,7 @@ World
        Fictitious<br/>
        World
            </p>
-           <p style="text-align:justify;">Justify</p>
+           <p style="text-align:justify;page-break-after: avoid;page-break-inside: avoid;">Justify</p>
                </div>
                <p class="zzSTDTitle1"/>
              </div>
@@ -883,14 +1051,14 @@ World
     OUTPUT
   end
 
-  it "processes paragraph alignments (Word)" do
+  it "processes paragraph attributes (Word)" do
     expect(xmlpp(IsoDoc::WordConvert.new({}).convert("test", <<~"INPUT", true))).to be_equivalent_to xmlpp(<<~"OUTPUT")
     <iso-standard xmlns="http://riboseinc.com/isoxml">
     <preface><foreword>
     <p align="left" id="_08bfe952-d57f-4150-9c95-5d52098cc2a8">Vache Equipment<br/>
 Fictitious<br/>
 World</p>
-    <p align="justify">Justify</p>
+    <p align="justify" keep-with-next="true" keep-lines-together="true">Justify</p>
     </foreword></preface>
     </iso-standard>
     INPUT
@@ -905,11 +1073,11 @@ World
                <p><br  clear="all" style="mso-special-character:line-break;page-break-before:always"/></p>
                <div>
                  <h1 class="ForewordTitle">Foreword</h1>
-                 <p id="_08bfe952-d57f-4150-9c95-5d52098cc2a8" align="left" style="text-align:left">Vache Equipment<br/>
+                 <p id="_08bfe952-d57f-4150-9c95-5d52098cc2a8" align="left" style="text-align:left;">Vache Equipment<br/>
        Fictitious<br/>
        World
            </p>
-           <p style="text-align:justify">Justify</p>
+           <p style="text-align:justify;page-break-after: avoid;page-break-inside: avoid;">Justify</p>
                </div>
                <p>&#160;</p>
              </div>
@@ -984,7 +1152,7 @@ World
     expect(xmlpp(IsoDoc::HtmlConvert.new({}).convert("test", <<~"INPUT", true))).to be_equivalent_to xmlpp(<<~"OUTPUT")
     <iso-standard xmlns="http://riboseinc.com/isoxml">
     <preface><foreword>
-    <permission id="_">
+    <permission id="_"   keep-with-next="true" keep-lines-together="true">
   <label>/ogc/recommendation/wfs/2</label>
   <inherit>/ss/584/2015/level/1</inherit>
   <inherit><eref type="inline" bibitemid="rfc2616" citeas="RFC 2616">RFC 2616 (HTTP/1.1)</eref></inherit>
@@ -1042,7 +1210,8 @@ World
        <br/>
       <div>
         <h1 class="ForewordTitle">Foreword</h1>
-        <div class="permission"><p class="RecommendationTitle">Permission 1:<br/>/ogc/recommendation/wfs/2</p>
+        <div class="permission" id='_' style='page-break-after: avoid;page-break-inside: avoid;'>
+        <p class="RecommendationTitle">Permission 1:<br/>/ogc/recommendation/wfs/2</p>
         <p><i>Subject: user<br/>
 Inherits: /ss/584/2015/level/1
 <br/>
@@ -1058,7 +1227,7 @@ Inherits: RFC 2616 (HTTP/1.1)
   </div>
   <div class="requirement-measurement-target">
     <p id="_">The measurement target shall be measured as:</p>
-    <div id="_" class="formula"><p><span class="stem">(#(r/1 = 0)#)</span>&#160; (1)</p></div>
+    <div id="_"><div class="formula"><p><span class="stem">(#(r/1 = 0)#)</span>&#160; (1)</p></div></div>
 
 
   </div>
@@ -1088,7 +1257,7 @@ Inherits: RFC 2616 (HTTP/1.1)
     expect(xmlpp(IsoDoc::HtmlConvert.new({}).convert("test", <<~"INPUT", true))).to be_equivalent_to xmlpp(<<~"OUTPUT")
     <iso-standard xmlns="http://riboseinc.com/isoxml">
     <preface><foreword>
-    <requirement id="A" unnumbered="true">
+    <requirement id="A" unnumbered="true"  keep-with-next="true" keep-lines-together="true">
   <title>A New Requirement</title>
   <label>/ogc/recommendation/wfs/2</label>
   <inherit>/ss/584/2015/level/1</inherit>
@@ -1114,7 +1283,7 @@ Inherits: RFC 2616 (HTTP/1.1)
   <description>
     <p id="_">As for the measurement targets,</p>
   </description>
-  <measurement-target exclude="false">
+  <measurement-target exclude="false"  keep-with-next="true" keep-lines-together="true">
     <p id="_">The measurement target shall be measured as:</p>
     <formula id="B">
       <stem type="AsciiMath">r/1 = 0</stem>
@@ -1139,7 +1308,8 @@ Inherits: RFC 2616 (HTTP/1.1)
         <br/>
       <div>
         <h1 class="ForewordTitle">Foreword</h1>
-        <div class="require"><p class="RecommendationTitle">Requirement:<br/>/ogc/recommendation/wfs/2. A New Requirement</p><p><i>Subject: user<br/>Inherits: /ss/584/2015/level/1</i></p>
+        <div class="require" id='A' style='page-break-after: avoid;page-break-inside: avoid;'>
+<p class="RecommendationTitle">Requirement:<br/>/ogc/recommendation/wfs/2. A New Requirement</p><p><i>Subject: user<br/>Inherits: /ss/584/2015/level/1</i></p>
 
   <div class="requirement-description">
     <p id="_">I recommend <i>this</i>.</p>
@@ -1148,9 +1318,9 @@ Inherits: RFC 2616 (HTTP/1.1)
   <div class="requirement-description">
     <p id="_">As for the measurement targets,</p>
   </div>
-  <div class="requirement-measurement-target">
+  <div class="requirement-measurement-target"  style='page-break-after: avoid;page-break-inside: avoid;'>
     <p id="_">The measurement target shall be measured as:</p>
-    <div id="B" class="formula"><p><span class="stem">(#(r/1 = 0)#)</span>&#160; (1)</p></div>
+    <div id="B"><div class="formula"><p><span class="stem">(#(r/1 = 0)#)</span>&#160; (1)</p></div></div>
 
 
   </div>
@@ -1227,7 +1397,7 @@ Inherits: RFC 2616 (HTTP/1.1)
              <br/>
              <div>
                <h1 class='ForewordTitle'>Avant-propos</h1>
-               <div class='require'>
+               <div class='require' id="A">
                  <p class='RecommendationTitle'>
                    Exigence:
                    <br/>
@@ -1252,11 +1422,12 @@ Inherits: RFC 2616 (HTTP/1.1)
                  </div>
                  <div class='requirement-measurement-target'>
                    <p id='_'>The measurement target shall be measured as:</p>
-                   <div id='B' class='formula'>
+                   <div id='B'><div class='formula'>
                      <p>
                        <span class='stem'>(#(r/1 = 0)#)</span>
                        &#160; (1)
                      </p>
+                     </div>
                    </div>
                  </div>
                  <div class='requirement-verification'>
@@ -1287,7 +1458,7 @@ end
     expect(xmlpp(IsoDoc::HtmlConvert.new({}).convert("test", <<~"INPUT", true))).to be_equivalent_to xmlpp(<<~"OUTPUT")
     <iso-standard xmlns="http://riboseinc.com/isoxml">
     <preface><foreword>
-    <recommendation id="_" obligation="shall,could">
+    <recommendation id="_" obligation="shall,could"   keep-with-next="true" keep-lines-together="true">
   <label>/ogc/recommendation/wfs/2</label>
   <inherit>/ss/584/2015/level/1</inherit>
   <classification><tag>type</tag><value>text</value></classification>
@@ -1339,7 +1510,8 @@ end
        <br/>
       <div>
         <h1 class="ForewordTitle">Foreword</h1>
-<div class="recommend"><p class="RecommendationTitle">Recommendation 1:<br/>/ogc/recommendation/wfs/2</p><p><i>Obligation: shall,could<br/>Subject: user<br/>Inherits: /ss/584/2015/level/1<br/>Type: text<br/>Language: BASIC</i></p>
+<div class="recommend"  id='_' style='page-break-after: avoid;page-break-inside: avoid;'>
+<p class="RecommendationTitle">Recommendation 1:<br/>/ogc/recommendation/wfs/2</p><p><i>Obligation: shall,could<br/>Subject: user<br/>Inherits: /ss/584/2015/level/1<br/>Type: text<br/>Language: BASIC</i></p>
   <div class="requirement-description">
     <p id="_">I recommend <i>this</i>.</p>
   </div>
@@ -1349,7 +1521,7 @@ end
   </div>
   <div class="requirement-measurement-target">
     <p id="_">The measurement target shall be measured as:</p>
-    <div id="_" class="formula"><p><span class="stem">(#(r/1 = 0)#)</span>&#160; (1)</p></div>
+    <div id="_"><div class="formula"><p><span class="stem">(#(r/1 = 0)#)</span>&#160; (1)</p></div></div>
 
 
   </div>
@@ -1374,7 +1546,7 @@ end
     <language>en</language>
     </bibdata>
         <preface><foreword>
-  <figure id="_" class="pseudocode"><name>Label</name><p id="_">  <strong>A</strong><br/>
+  <figure id="_" class="pseudocode" keep-with-next="true" keep-lines-together="true"><name>Label</name><p id="_">  <strong>A</strong><br/>
         <smallcap>B</smallcap></p>
 <p id="_">  <em>C</em></p></figure>
 </preface></itu-standard>
@@ -1383,7 +1555,7 @@ INPUT
              <br/>
              <div>
                <h1 class="ForewordTitle">Foreword</h1>
-               <div id="_" class="pseudocode"><p id="_">&#160;&#160;<b>A</b><br/>
+               <div id="_" class="pseudocode" style='page-break-after: avoid;page-break-inside: avoid;'><p id="_">&#160;&#160;<b>A</b><br/>
        &#160;&#160;&#160;&#160;&#160;&#160;&#160;&#160;<span style="font-variant:small-caps;">B</span></p>
        <p id="_">&#160;&#160;<i>C</i></p><p class="SourceTitle" style="text-align:center;">Figure 1&#xA0;&#x2014; Label</p></div>
              </div>
@@ -1463,26 +1635,35 @@ INPUT
        end
 
          it "processes passthrough with compatible format" do
-    expect(xmlpp(IsoDoc::HtmlConvert.new({}).convert("test", <<~"INPUT", true))).to be_equivalent_to xmlpp(<<~"OUTPUT")
+               FileUtils.rm_f "test.html"
+    IsoDoc::HtmlConvert.new({}).convert("test", <<~"INPUT", false)
     <iso-standard xmlns="http://riboseinc.com/isoxml">
     <preface><foreword>
-    <passthrough format="html,rfc">&lt;A&gt;</passthrough>
+    <passthrough format="html,rfc">&lt;A&gt;</passthrough><em>Hello</em><passthrough format="html,rfc">&lt;/A&gt;</passthrough>
     </foreword></preface>
     </iso-standard>
     INPUT
-    #{HTML_HDR}
-        <br/>
-      <div>
-        <h1 class='ForewordTitle'>Foreword</h1>
-        <a/>
-      </div>
-      <p class='zzSTDTitle1'/>
-    </div>
-  </body>
-</html>
+expect(( File.read("test.html").gsub(%r{^.*<h1 class="ForewordTitle">Foreword</h1>}m, "").gsub(%r{</div>.*}m, ""))).to be_equivalent_to xmlpp(<<~"OUTPUT")
+        <A><i>Hello</i></A>
     OUTPUT
   end
 
+          it "aborts if passthrough results in malformed XML" do
+               FileUtils.rm_f "test.html"
+               FileUtils.rm_f "test.html.err"
+               begin
+    expect { IsoDoc::HtmlConvert.new({}).convert("test", <<~"INPUT", false) }.to raise_error(SystemExit)
+    <iso-standard xmlns="http://riboseinc.com/isoxml">
+    <preface><foreword>
+    <passthrough format="html,rfc">&lt;A&gt;</passthrough><em>Hello</em>
+    </foreword></preface>
+    </iso-standard>
+    INPUT
+    rescue SystemExit
+    end
+    expect(File.exist?("test.html.err")).to be true
+  end
+
      it "ignore passthrough with incompatible format" do
     expect(xmlpp(IsoDoc::HtmlConvert.new({}).convert("test", <<~"INPUT", true))).to be_equivalent_to xmlpp(<<~"OUTPUT")
     <iso-standard xmlns="http://riboseinc.com/isoxml">