RubyGems - html2doc - Versions diffs - 1.5.3 → 1.5.5 - Mend

html2doc 1.5.3 → 1.5.5

Files changed (7) hide show

checksums.yaml CHANGED Viewed

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 114c21f3bbc33c244fb49577d4c85334789a388be61d65c04765d33ed1913208
-  data.tar.gz: 60c00e3a300eb16db1f2332393d5ff9a45f65c60a8a1d40a1cfaba2767744249
+  metadata.gz: 46856bf56ad5dd95f8f5781dc11049bb4600060c28c49715a262837ece8028bf
+  data.tar.gz: 866ba19867f233b45aeee436df719679623d671902b30476b61952f7a6357e1f
 SHA512:
-  metadata.gz: 606e87a4dcfd0a3e270588461c6150dd38b1baad57bfc90b3965806477945deeae0b2eaeca87af9800b70d214133b11588a66f8205108e3af87f13464306b34a
-  data.tar.gz: ecc2ad3631a0cafb9ff3b7f04022b9a977efac71099f7b256f840ec6d15e418bb18f426c729d6a40e9a9cc196f3e02d4cf0c5c448a99c6097c2da9d6b2bfb4d0
+  metadata.gz: b949f47c356437ce418f65ce7fd1c497648d0d0e960fe1e05d7318d280ddf6de23ddad8e5ab94a18f447b3eaba948b2a4db69ca2d00f0dcfebd692933a64c1da
+  data.tar.gz: 953999bd39aa1c1b6a0e1a34c939dcbdba01242c898f5a587eab546b4c9a4578e8051c9c68b49900390fcb3ab5d2de4ad6f0f1bce1af5a8ee6e7bd4daf300966

data/lib/html2doc/base.rb CHANGED Viewed

@@ -30,8 +30,7 @@ class Html2Doc
   end
   def process_header(headerfile)
-    return if headerfile.nil?
+    headerfile.nil? and return
     doc = File.read(headerfile, encoding: "utf-8")
     doc = header_image_cleanup(doc, @dir1, @filename,
                                File.dirname(@filename))
@@ -66,6 +65,7 @@ class Html2Doc
   end
   def cleanup(docxml)
+    locate_landscape(docxml)
     namespace(docxml.root)
     image_cleanup(docxml, @dir1, @imagedir)
     mathml_to_ooml(docxml)
@@ -76,76 +76,11 @@ class Html2Doc
     docxml
   end
-  NOKOHEAD = <<~HERE.freeze
-    <!DOCTYPE html SYSTEM
-    "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
-    <html xmlns="http://www.w3.org/1999/xhtml">
-    <head> <title></title> <meta charset="UTF-8" /> </head>
-    <body> </body> </html>
-  HERE
-  def to_xhtml(xml)
-    xml.gsub!(/<\?xml[^>]*>/, "")
-    unless /<!DOCTYPE /.match? xml
-      xml = '<!DOCTYPE html SYSTEM
-          "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">' + xml
-    end
-    xml = xml.gsub(/<!--\s*\[([^\]]+)\]>/, "<!-- MSWORD-COMMENT \\1 -->")
-      .gsub(/<!\s*\[endif\]\s*-->/, "<!-- MSWORD-COMMENT-END -->")
-    Nokogiri::XML.parse(xml)
-  end
-  DOCTYPE = <<~"DOCTYPE".freeze
-    <!DOCTYPE html SYSTEM "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
-  DOCTYPE
-  def from_xhtml(xml)
-    xml.to_xml.sub(%{ xmlns="http://www.w3.org/1999/xhtml"}, "")
-      .sub(DOCTYPE, "").gsub(%{ />}, "/>")
-      .gsub(/<!-- MSWORD-COMMENT (.+?) -->/, "<!--[\\1]>")
-      .gsub(/<!-- MSWORD-COMMENT-END -->/, "<![endif]-->")
-      .gsub("\n--&gt;\n", "\n-->\n")
-  end
-  def msword_fix(doc)
-    # brain damage in MSWord parser
-    doc.gsub!(%r{<w:DoNotOptimizeForBrowser></w:DoNotOptimizeForBrowser>},
-              "<w:DoNotOptimizeForBrowser/>")
-    doc.gsub!(%r{<span style="mso-special-character:footnote"/>},
-              '<span style="mso-special-character:footnote"></span>')
-    doc.gsub!(%r{<div style="mso-element:footnote-list"></div>},
-              '<div style="mso-element:footnote-list"/>')
-    doc.gsub!(%r{(<a style="mso-comment-reference:[^>/]+)/>}, "\\1></a>")
-    doc.gsub!(%r{<link rel="File-List"}, "<link rel=File-List")
-    doc.gsub!(%r{<meta http-equiv="Content-Type"},
-              "<meta http-equiv=Content-Type")
-    doc.gsub!(%r{></m:jc>}, "/>")
-    doc.gsub!(%r{></v:stroke>}, "/>")
-    doc.gsub!(%r{></v:f>}, "/>")
-    doc.gsub!(%r{></v:path>}, "/>")
-    doc.gsub!(%r{></o:lock>}, "/>")
-    doc.gsub!(%r{></v:imagedata>}, "/>")
-    doc.gsub!(%r{></w:wrap>}, "/>")
-    doc.gsub!(%r{<(/)?m:(span|em)\b}, "<\\1\\2")
-    doc.gsub!(%r{&tab;|&amp;tab;},
-              '<span style="mso-tab-count:1">&#xA0; </span>')
-    doc.split(%r{(<m:oMath>|</m:oMath>)}).each_slice(4).map do |a|
-      a.size > 2 and a[2] = a[2].gsub(/>\s+</, "><")
-      a
-    end.join
-  end
-  PRINT_VIEW = <<~XML.freeze
-    <xml>
-    <w:WordDocument>
-    <w:View>Print</w:View>
-    <w:Zoom>100</w:Zoom>
-    <w:DoNotOptimizeForBrowser/>
-    </w:WordDocument>
-    </xml>
-    <meta http-equiv='Content-Type' content="text/html; charset=utf-8"/>
-  XML
+  def locate_landscape(_docxml)
+    css = read_stylesheet(@stylesheet)
+    @landscape = css.scan(/div\.\S+\s+\{\s*page:\s*[^;]+L;\s*\}/m)
+      .map { |e| e.sub(/^div\.(\S+).*$/m, "\\1") }
+  end
   def define_head1(docxml, _dir)
     docxml.xpath("//*[local-name() = 'head']").each do |h|
@@ -168,18 +103,21 @@ class Html2Doc
   end
   def stylesheet(_filename, _header_filename, cssname)
-    (cssname.nil? || cssname.empty?) and
-      cssname = File.join(File.dirname(__FILE__), "wordstyle.css")
-    stylesheet = File.read(cssname, encoding: "UTF-8")
+    stylesheet = read_stylesheet(cssname)
     xml = Nokogiri::XML("<style/>")
     # s = Nokogiri::XML::CDATA.new(xml, "\n#{stylesheet}\n")
     # xml.children.first << Nokogiri::XML::Comment.new(xml, s)
     xml.children.first << Nokogiri::XML::CDATA
       .new(xml, "\n<!--\n#{stylesheet}\n-->\n")
     xml.root.to_s
   end
+  def read_stylesheet(cssname)
+    (cssname.nil? || cssname.empty?) and
+      cssname = File.join(File.dirname(__FILE__), "wordstyle.css")
+    File.read(cssname, encoding: "UTF-8")
+  end
   def define_head(docxml)
     title = docxml.at("//*[local-name() = 'head']/*[local-name() = 'title']")
     head = docxml.at("//*[local-name() = 'head']")
@@ -195,30 +133,15 @@ class Html2Doc
       head.add_child css
     elsif title.nil?
       head.children.first.add_previous_sibling css
-    else
-      title.add_next_sibling css
+    else title.add_next_sibling css
     end
   end
-  def namespace(root)
-    {
-      o: "urn:schemas-microsoft-com:office:office",
-      w: "urn:schemas-microsoft-com:office:word",
-      v: "urn:schemas-microsoft-com:vml",
-      m: "http://schemas.microsoft.com/office/2004/12/omml",
-    }.each { |k, v| root.add_namespace_definition(k.to_s, v) }
-  end
-  def rootnamespace(root)
-    root.add_namespace(nil, "http://www.w3.org/TR/REC-html40")
-  end
   def bookmarks(docxml)
     docxml.xpath("//*[@id][not(@name)][not(@style = 'mso-element:footnote')]")
       .each do |x|
-      next if x["id"].empty? ||
-        %w(shapetype v:shapetype shape v:shape).include?(x.name)
+      (x["id"].empty? ||
+        %w(shapetype v:shapetype shape v:shape).include?(x.name)) and next
       if x.children.empty? then x.add_child("<a name='#{x['id']}'></a>")
       else x.children.first.previous = "<a name='#{x['id']}'></a>"
       end

data/lib/html2doc/mime.rb CHANGED Viewed

@@ -76,11 +76,9 @@ class Html2Doc
     end
   end
-  # max width for Word document is 400, max height is 680
   def image_resize(img, path, maxheight, maxwidth)
     s, realsize = get_image_size(img, path)
-    return s if s[0] == nil && s[1] == nil
+    s[0] == nil && s[1] == nil and return s
     if img.name == "svg" && !img["viewBox"]
       img["viewBox"] = "0 0 #{s[0]} #{s[1]}"
     end
@@ -115,21 +113,100 @@ class Html2Doc
   # only processes locally stored images
   def image_cleanup(docxml, dir, localdir)
+    maxheight, maxwidth = page_dimensions(docxml)
     docxml.traverse do |i|
-      src = i["src"]
-      next unless i.element? && %w(img v:imagedata).include?(i.name)
-      next if src.nil? || src.empty? || /^http/.match?(src)
-      next if %r{^data:(image|application)/[^;]+;base64}.match? src
-      local_filename = localname(src, localdir)
-      new_filename = "#{mkuuid}#{File.extname(src)}"
-      FileUtils.cp local_filename, File.join(dir, new_filename)
-      i["width"], i["height"] = image_resize(i, local_filename, 680, 400)
-      i["src"] = File.join(File.basename(dir), new_filename)
+      skip_image_cleanup?(i) and next
+      local_filename = rename_image(i, dir, localdir)
+      i["width"], i["height"] =
+        if landscape?(i)
+          image_resize(i, local_filename, maxwidth, maxheight)
+        else
+          image_resize(i, local_filename, maxheight, maxwidth)
+        end
     end
     docxml
   end
+  def landscape?(img)
+    img.ancestors.each do |a|
+      a.name == "div" or next
+      @landscape.include?(a["class"]) and return true
+    end
+    false
+  end
+  def rename_image(img, dir, localdir)
+    local_filename = localname(img["src"], localdir)
+    new_filename = "#{mkuuid}#{File.extname(img['src'])}"
+    FileUtils.cp local_filename, File.join(dir, new_filename)
+    img["src"] = File.join(File.basename(dir), new_filename)
+    local_filename
+  end
+  def skip_image_cleanup?(img)
+    src = img["src"]
+    (img.element? && %w(img v:imagedata).include?(img.name)) or return true
+    (src.nil? || src.empty? || /^http/.match?(src) ||
+      %r{^data:(image|application)/[^;]+;base64}.match?(src)) and return true
+    false
+  end
+  # we are going to use the 2nd instance of @page in the Word CSS,
+  # skipping the cover page. Currently doesn't deal with Landscape.
+  # Scan both @stylesheet and docxml.to_xml (where @standardstylesheet has ended up)
+  # Allow 0.9 * height to fit caption
+  def page_dimensions(docxml)
+    stylesheet = read_stylesheet(@stylesheet)
+    page_size = find_page_size_in_doc(stylesheet, docxml.to_xml) or
+      return [680, 400]
+    m_size = /size:\s*(\S+)\s+(\S+)\s*;/.match(page_size) or return [680, 400]
+    m_marg = /margin:\s*(\S+)\s+(\S+)\s*(\S+)\s*(\S+)\s*;/.match(page_size) or
+      return [680, 400]
+    [0.9 * (units_to_px(m_size[2]) - units_to_px(m_marg[1]) - units_to_px(m_marg[3])),
+     units_to_px(m_size[1]) - units_to_px(m_marg[2]) - units_to_px(m_marg[4])]
+  rescue StandardError
+    [680, 400]
+  end
+  def find_page_size_in_doc(stylesheet, doc)
+    find_page_size(stylesheet, "WordSection2", false) ||
+      find_page_size(stylesheet, "WordSection3", false) ||
+      find_page_size(doc, "WordSection2", true) ||
+      find_page_size(doc, "WordSection3", true) ||
+      find_page_size(stylesheet, "", false) || find_page_size(doc, "", true)
+  end
+  # if in_xml, CSS is embedded in XML <style> tag
+  def find_page_size(stylesheet, klass, in_xml)
+    xml_found = false
+    found = false
+    ret = ""
+    stylesheet&.lines&.each do |l|
+      in_xml && l.include?("<style") and xml_found = true and found = false
+      in_xml && l.include?("</style>") and xml_found = false
+      /^\s*@page\s+#{klass}/.match?(l) and found = true
+      found && /^\s*\{?size:/.match?(l) and ret += l
+      found && /^\s*\{?margin:/.match?(l) and ret += l
+      if found && /}/.match?(l)
+        !ret.blank? && (!in_xml || xml_found) and return ret
+        ret = ""
+        found = false
+      end
+    end
+    nil
+  end
+  def units_to_px(measure)
+    m = /^(\S+)(pt|cm)/.match(measure)
+    ret = case m[2]
+          when "px" then (m[1].to_f * 0.75)
+          when "pt" then m[1].to_f
+          when "cm" then (m[1].to_f * 28.346456693)
+          when "in" then (m[1].to_f * 72)
+          end
+    ret.to_i
+  end
   # do not parse the header through Nokogiri, since it will contain
   # non-XML like <![if !supportFootnotes]>
   def header_image_cleanup(doc, dir, filename, localdir)
@@ -155,8 +232,7 @@ class Html2Doc
       f.write %{<xml xmlns:o="urn:schemas-microsoft-com:office:office">
         <o:MainFile HRef="../#{filename}.htm"/>}
       Dir.entries(dir).sort.each do |item|
-        next if item == "." || item == ".." || /^\./.match(item)
+        (item == "." || item == ".." || /^\./.match(item)) and next
         f.write %{  <o:File HRef="#{item}"/>\n}
       end
       f.write("</xml>\n")

data/lib/html2doc/version.rb CHANGED Viewed

@@ -1,3 +1,3 @@
 class Html2Doc
-  VERSION = "1.5.3".freeze
+  VERSION = "1.5.5".freeze
 end

data/lib/html2doc/xml.rb ADDED Viewed

@@ -0,0 +1,83 @@
+class Html2Doc
+  NOKOHEAD = <<~HERE.freeze
+    <!DOCTYPE html SYSTEM
+    "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
+    <html xmlns="http://www.w3.org/1999/xhtml">
+    <head> <title></title> <meta charset="UTF-8" /> </head>
+    <body> </body> </html>
+  HERE
+  def to_xhtml(xml)
+    xml.gsub!(/<\?xml[^>]*>/, "")
+    unless /<!DOCTYPE /.match? xml
+      xml = '<!DOCTYPE html SYSTEM
+          "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">' + xml
+    end
+    xml = xml.gsub(/<!--\s*\[([^\]]+)\]>/, "<!-- MSWORD-COMMENT \\1 -->")
+      .gsub(/<!\s*\[endif\]\s*-->/, "<!-- MSWORD-COMMENT-END -->")
+    Nokogiri::XML.parse(xml)
+  end
+  DOCTYPE = <<~DOCTYPE.freeze
+    <!DOCTYPE html SYSTEM "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
+  DOCTYPE
+  def from_xhtml(xml)
+    xml.to_xml.sub(%{ xmlns="http://www.w3.org/1999/xhtml"}, "")
+      .sub(DOCTYPE, "").gsub(%{ />}, "/>")
+      .gsub(/<!-- MSWORD-COMMENT (.+?) -->/, "<!--[\\1]>")
+      .gsub(/<!-- MSWORD-COMMENT-END -->/, "<![endif]-->")
+      .gsub("\n--&gt;\n", "\n-->\n")
+  end
+  def msword_fix(doc)
+    # brain damage in MSWord parser
+    doc.gsub!(%r{<w:DoNotOptimizeForBrowser></w:DoNotOptimizeForBrowser>},
+              "<w:DoNotOptimizeForBrowser/>")
+    doc.gsub!(%r{<span style="mso-special-character:footnote"/>},
+              '<span style="mso-special-character:footnote"></span>')
+    doc.gsub!(%r{<div style="mso-element:footnote-list"></div>},
+              '<div style="mso-element:footnote-list"/>')
+    doc.gsub!(%r{(<a style="mso-comment-reference:[^>/]+)/>}, "\\1></a>")
+    doc.gsub!(%r{<link rel="File-List"}, "<link rel=File-List")
+    doc.gsub!(%r{<meta http-equiv="Content-Type"},
+              "<meta http-equiv=Content-Type")
+    doc.gsub!(%r{></m:jc>}, "/>")
+    doc.gsub!(%r{></v:stroke>}, "/>")
+    doc.gsub!(%r{></v:f>}, "/>")
+    doc.gsub!(%r{></v:path>}, "/>")
+    doc.gsub!(%r{></o:lock>}, "/>")
+    doc.gsub!(%r{></v:imagedata>}, "/>")
+    doc.gsub!(%r{></w:wrap>}, "/>")
+    doc.gsub!(%r{<(/)?m:(span|em)\b}, "<\\1\\2")
+    doc.gsub!(%r{&tab;|&amp;tab;},
+              '<span style="mso-tab-count:1">&#xA0; </span>')
+    doc.split(%r{(<m:oMath>|</m:oMath>)}).each_slice(4).map do |a|
+      a.size > 2 and a[2] = a[2].gsub(/>\s+</, "><")
+      a
+    end.join
+  end
+  PRINT_VIEW = <<~XML.freeze
+    <xml>
+    <w:WordDocument>
+    <w:View>Print</w:View>
+    <w:Zoom>100</w:Zoom>
+    <w:DoNotOptimizeForBrowser/>
+    </w:WordDocument>
+    </xml>
+    <meta http-equiv='Content-Type' content="text/html; charset=utf-8"/>
+  XML
+  def namespace(root)
+    { o: "urn:schemas-microsoft-com:office:office",
+      w: "urn:schemas-microsoft-com:office:word",
+      v: "urn:schemas-microsoft-com:vml",
+      m: "http://schemas.microsoft.com/office/2004/12/omml" }.each { |k, v| root.add_namespace_definition(k.to_s, v) }
+  end
+  def rootnamespace(root)
+    root.add_namespace(nil, "http://www.w3.org/TR/REC-html40")
+  end
+end

data/lib/html2doc.rb CHANGED Viewed

@@ -4,3 +4,4 @@ require_relative "html2doc/mime"
 require_relative "html2doc/notes"
 require_relative "html2doc/math"
 require_relative "html2doc/lists"
+require_relative "html2doc/xml"

metadata CHANGED Viewed

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: html2doc
 version: !ruby/object:Gem::Version
-  version: 1.5.3
+  version: 1.5.5
 platform: ruby
 authors:
 - Ribose Inc.
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2023-04-10 00:00:00.000000000 Z
+date: 2023-06-29 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: htmlentities
@@ -293,6 +293,7 @@ files:
 - lib/html2doc/notes.rb
 - lib/html2doc/version.rb
 - lib/html2doc/wordstyle.css
+- lib/html2doc/xml.rb
 homepage: https://github.com/metanorma/html2doc
 licenses:
 - CC-BY-SA-3.0