metanorma-standoc 2.4.0 → 2.4.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 11179b51c3de0cefbd6cd55d2cd3e6945e0d72cb379ec9a32ed792b1239444fd
-  data.tar.gz: e3c6231ecd99e4f631da9173b1c89f10213428c331147b54efa1eccf8833a017
+  metadata.gz: 29a338b7e2acb906be1a2f9fb958fec1281991d934cae8f33279605d4c7296e7
+  data.tar.gz: 104dfd478b186a14b6a7acd276c28785ada074ce3f747839ee30a4620269b74e
 SHA512:
-  metadata.gz: d000c768270a37dacc417c3d29cab591a708bf7e7b47c51bbded56120c711c6bc82257335a44f34401cfb06d96dd8c64cb410d1664064534147fee3bf7525d85
-  data.tar.gz: 11799198a90c8b67a358365115cebf4594bfd1be0d58cbc193db94fa95096f98cbbd5974fb65f8b6dbc07349d6a76080591ed6703979d448ee768cd8e5f1ceae
+  metadata.gz: 94ef91339b3743becae21df403eeb7b9af5910257ba97756c72111ddababda0523d9e6b961aa6eab9737af6c460f4c3c89aecf0af3ecbacbb391aa9e10493af3
+  data.tar.gz: 0ef86c13ca5bea907b55aa9e9ce98b1447e9c887e388db737855f9d8f55f3db39d20cf34aa526e2c2f0ccf140792cfd6f396b4ba117a1d50f67f28a0dd027442

data/lib/metanorma/standoc/cleanup_maths.rb

@@ -11,10 +11,23 @@ module Metanorma
           .asciimath_to_mathml(text, ["<amathstem>", "</amathstem>"],
                                retain_asciimath: true)
         asciimath2mathml_wrap(text)
+      rescue StandardError => e
+        asciimath2mathml_err(text, e)
+        text
+      end
+
+      def asciimath2mathml_err(text, expr)
+        err = "Malformed MathML: #{expr}\n#{text}"
+        @log.add("Maths", nil, err)
+        @fatalerror << err
+        warn err
       end
 
       def asciimath2mathml_wrap(text)
         x = Nokogiri::XML(text)
+        x.xpath("//*[local-name() = 'math'][@display]").each do |y|
+          y.delete("display")
+        end
         x.xpath("//*[local-name() = 'math'][not(parent::stem)]").each do |y|
           y.wrap("<stem type='MathML'></stem>")
         end
@@ -43,7 +56,9 @@ module Metanorma
       end
 
       def mathml_namespace(stem)
-        stem.xpath("./math").each { |x| x.default_namespace = MATHML_NS }
+        stem.xpath("./*[local-name() = 'math']").each do |x|
+          x.default_namespace = MATHML_NS
+        end
       end
 
       def mathml_mi_italics

data/lib/metanorma/standoc/cleanup_section.rb

@@ -1,7 +1,6 @@
 require "date"
 require "htmlentities"
 require "json"
-require "mathml2asciimath"
 require_relative "cleanup_section_names"
 
 module Metanorma

data/lib/metanorma/standoc/inline.rb

@@ -3,7 +3,7 @@ require "unicode2latex"
 require "mime/types"
 require "base64"
 require "English"
-require "latexmath"
+require "plurimath"
 
 module Metanorma
   module Standoc
@@ -140,18 +140,22 @@ module Metanorma
 
       def latex_parse1(text)
         lxm_input = Unicode2LaTeX.unicode2latex(@c.decode(text))
-        results = Latexmath.parse(lxm_input).to_mathml
-        results.nil? and
+        results = Plurimath::Math.parse(lxm_input, "latex").to_mathml
+        if results.nil?
           @log.add("Math", nil,
                    "latexmlmath failed to process equation:\n#{lxm_input}")
-        results&.sub(%r{<math ([^>]+ )?display="block"}, "<math \\1")
+          return
+        end
+        results.sub(%r{<math ([^>]+ )?display="block"}, "<math \\1")
       end
 
       def stem_parse(text, xml, style)
         if /&lt;([^:>&]+:)?math(\s+[^>&]+)?&gt; |
           <([^:>&]+:)?math(\s+[^>&]+)?>/x.match? text
           math = xml_encode(text)
-          xml.stem math, type: "MathML"
+          xml.stem type: "MathML" do |s|
+            s << math
+          end
         elsif style == :latexmath then latex_parse(text, xml)
         else
           xml.stem text&.gsub(/&amp;#/, "&#"), type: "AsciiMath"

data/lib/metanorma/standoc/utils.rb

@@ -18,7 +18,7 @@ module Metanorma
       end
 
       def noko(&block)
-        Metanorma::Utils::noko(&block)
+        Metanorma::Utils::noko(@script, &block)
       end
 
       def attr_code(attributes)
@@ -68,11 +68,12 @@ module Metanorma
         para
       end
 
-       def xml_encode(text)
+      def xml_encode(text)
         @c.encode(text, :basic, :hexadecimal)
           .gsub(/&amp;gt;/, ">").gsub(/&amp;lt;/, "<").gsub(/&amp;amp;/, "&")
           .gsub(/&gt;/, ">").gsub(/&lt;/, "<").gsub(/&amp;/, "&")
           .gsub(/&quot;/, '"').gsub(/&#xa;/, "\n").gsub(/&amp;#/, "&#")
+          .gsub(/&apos;/, "'")
       end
 
       class EmptyAttr

data/lib/metanorma/standoc/validate.rb

@@ -53,10 +53,39 @@ module Metanorma
         table_validate(doc)
         @fatalerror += requirement_validate(doc)
         image_validate(doc)
+        math_validate(doc)
         @fatalerror.empty? or
           clean_abort(@fatalerror.join("\n"), doc)
       end
 
+      MATHML_NS = "http://www.w3.org/1998/Math/MathML".freeze
+
+      def math_validate(doc)
+        doc.xpath("//m:math", "m" => MATHML_NS).each do |m|
+          math = mathml_sanitise(m.dup)
+          Plurimath::Math.parse(math, "mathml").to_mathml
+        rescue StandardError => e
+          math_validate_error(math, m, e)
+        end
+      end
+
+      def mathml_sanitise(math)
+        math.to_xml(encoding: "US-ASCII").gsub(/ xmlns=["'][^"']+["']/, "")
+          .gsub(%r{<[^:/>]+:}, "<").gsub(%r{</[^:/>]+:}, "</")
+          #.gsub(/&#([^;]+);/) { |x| "&#x#{$1.to_i.to_s(16)};" }
+      end
+
+      def math_validate_error(math, elem, error)
+        a = elem.parent.at("./asciimath")
+        l = elem.parent.at("./latexmath")
+        orig = ""
+        a and orig += "\n\tAsciimath original: #{@c.decode(a.children.to_xml)}"
+        l and orig += "\n\tLatexmath original: #{@c.decode(l.children.to_xml)}"
+        @log.add("Mathematics", elem,
+                 "Invalid MathML: #{math}\n #{error}#{orig}")
+        @fatalerror << "Invalid MathML: #{math}"
+      end
+
       def nested_asset_validate(doc)
         nested_asset_validate_basic(doc)
         nested_note_validate(doc)

data/lib/metanorma/standoc/version.rb

@@ -19,6 +19,6 @@ module Metanorma
   end
 
   module Standoc
-    VERSION = "2.4.0".freeze
+    VERSION = "2.4.1".freeze
   end
 end

data/metanorma-standoc.gemspec

@@ -38,8 +38,7 @@ Gem::Specification.new do |spec|
   # relaton-cli not just relaton, to avoid circular reference in metanorma
   spec.add_dependency "asciimath2unitsml", "~> 0.4.0"
   spec.add_dependency "concurrent-ruby"
-  spec.add_dependency "latexmath"
-  spec.add_dependency "mathml2asciimath"
+  spec.add_dependency "plurimath"
   spec.add_dependency "pngcheck"
   spec.add_dependency "relaton-cli", "~> 1.15.0"
   spec.add_dependency "relaton-iev", "~> 1.1.0"

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: metanorma-standoc
 version: !ruby/object:Gem::Version
-  version: 2.4.0
+  version: 2.4.1
 platform: ruby
 authors:
 - Ribose Inc.
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2023-03-13 00:00:00.000000000 Z
+date: 2023-03-27 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: asciidoctor
@@ -137,21 +137,7 @@ dependencies:
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
-  name: latexmath
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :runtime
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-- !ruby/object:Gem::Dependency
-  name: mathml2asciimath
+  name: plurimath
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ">="