isodoc-i18n 1.1.1 → 1.1.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 8f00540677aae6f53c0c517b376529ce6eb5094c6cebd1d1a9b31ce20e346266
-  data.tar.gz: 69dd44b3aedc560540c5de52f05f0ae11b1332edccb4565817845648f23a9631
+  metadata.gz: 573397fe43411b36723793fd50893ce1fd5eda84ca84f8e00ba608f6e50f3cd5
+  data.tar.gz: 3117720a218de8fe171ee8c4d3237c97408a87beac521d4bb13d0d1021d329a9
 SHA512:
-  metadata.gz: 8b3a53fe9058b2cba6f929938e61147adcb6eaa0782705ca2097cf448f63a76500abbe4b4d3513b9d0cfca04456547d8127e278ef7a521ec3b1cff6de8302f58
-  data.tar.gz: 24ff8195e4172769e91fc4f879fe6835b4f24170041e16198ab3a405625f07ed56bce3a702604b5cb25102b4d6f21ce87d5d92889122cd524a4694549ba8d0c8
+  metadata.gz: 865a347ccd8f837f4a88f950179d375a79db880ca61de6880d545163ad5e8ca19488410f7dc0a03cf8205a3d948256b58dae49909a442427ffb409635ed7d9c5
+  data.tar.gz: a0a3ba98de7b47388e81f9bbdc8dac86e95f9c91aaca0290b1952d9bf7b9f98d23b5725bc70c75f8018c65d61bbaaa629fe65ee2edb868f6cb1dab1223afa8a3

data/lib/isodoc/i18n.rb

@@ -5,10 +5,13 @@ require "twitter_cldr"
 
 module IsoDoc
   class I18n
+    Hash.include Metanorma::Utils::Hash
+
     def load_yaml(lang, script, i18nyaml = nil, i18nhash = nil)
       ret = load_yaml1(lang, script)
-      return normalise_hash(ret.merge(YAML.load_file(i18nyaml))) if i18nyaml
-      return normalise_hash(ret.merge(i18nhash)) if i18nhash
+      i18nyaml and
+        return normalise_hash(ret.deep_merge(YAML.load_file(i18nyaml)))
+      i18nhash and return normalise_hash(ret.deep_merge(i18nhash))
 
       normalise_hash(ret)
     end
@@ -98,26 +101,27 @@ module IsoDoc
     end
 
     def l10n_zh(text)
-      xml = Nokogiri::HTML::DocumentFragment.parse(text)
+      xml = Nokogiri::XML::DocumentFragment.parse(text)
       xml.traverse do |n|
         next unless n.text?
 
         n.replace(cleanup_entities(l10_zh1(n.text), is_xml: false))
       end
-      xml.to_xml.gsub(/<b>/, "").gsub("</b>", "").gsub(/<\?[^>]+>/, "")
+      xml.to_xml(encoding: "UTF-8").gsub(/<b>/, "").gsub("</b>", "")
+        .gsub(/<\?[^>]+>/, "")
     end
 
     def l10n_fr(text, locale)
-      xml = Nokogiri::HTML::DocumentFragment.parse(text)
+      xml = Nokogiri::XML::DocumentFragment.parse(text)
       xml.traverse do |n|
         next unless n.text?
 
         n.replace(cleanup_entities(l10n_fr1(n.text, locale), is_xml: false))
       end
-      xml.to_xml
+      xml.to_xml(encoding: "UTF-8")
     end
 
-    ZH_CHAR = "\\p{Han}|\\p{In CJK Symbols And Punctuation}|"\
+    ZH_CHAR = "\\p{Han}|\\p{In CJK Symbols And Punctuation}|" \
               "\\p{In Halfwidth And Fullwidth Forms}".freeze
 
     # note: we can't differentiate comma from enumeration comma 、
@@ -171,11 +175,11 @@ module IsoDoc
       c = HTMLEntities.new
       if is_xml
         text.split(/([<>])/).each_slice(4).map do |a|
-          a[0] = c.encode(c.decode(a[0]), :hexadecimal)
+          a[0] = c.decode(a[0])
           a
         end.join
       else
-        c.encode(c.decode(text), :hexadecimal)
+        c.decode(text)
       end
     end
 
@@ -195,13 +199,21 @@ module IsoDoc
       num.localize(:en)
     end
 
+    INFLECTIONS = {
+      number: "sg",
+      case: "nom",
+      gender: "masc",
+      person: "3rd",
+      voice: "act",
+      mood: "ind",
+      tense: "pres",
+    }.freeze
+
+    INFLECTION_ORDER = %i(voice mood tense number case gender person).freeze
+
     def ordinal_key(term)
       @labels["ordinal_keys"].each_with_object([]) do |k, m|
-        m << case k
-             when "gender" then term["gender"]
-             when "number" then term["number"] || "sg"
-             when "case" then term["case"] || "nom"
-             end
+        m << (term[k] || INFLECTIONS[k.to_sym])
       end.join(".")
     end
 
@@ -211,5 +223,18 @@ module IsoDoc
       else @lang.to_sym
       end
     end
+
+    # can skip category if not present
+    def inflect(word, options)
+      i = @labels.dig("inflection", word) or return word
+      i.is_a? String and return i
+
+      INFLECTION_ORDER.each do |x|
+        infl = options[x] || INFLECTIONS[x]
+        i = i[infl] if i[infl]
+        i.is_a? String and return i
+      end
+      word
+    end
   end
 end

data/lib/isodoc/version.rb

@@ -1,5 +1,5 @@
 module IsoDoc
   class I18n
-    VERSION = "1.1.1".freeze
+    VERSION = "1.1.3".freeze
   end
 end

data/spec/assets/new.yaml

@@ -5,3 +5,49 @@ arr:
   - arr1
   - arr2
 text: "text2"
+inflection:
+  Fred:
+    sg: Fred
+    pl: Freds
+  Man:
+    dat: viri
+    acc: virem
+  Woman:
+    sg:
+      nom: mulier
+      gen: mulieris
+    pl:
+      nom: mulieres
+      gen: mulierum
+  Good:
+    sg:
+      nom:
+        masc: bonus
+        fem: bona
+        neut: bonum
+      gen:
+        masc: boni
+        fem: bonae
+        neut: boni
+    pl:
+      nom:
+        masc: boni
+        fem: bonae
+        neut: bona
+      gen:
+        masc: bonorum
+        fem: bonarum
+        neut: bonorum
+  Walk:
+    act:
+      ind:
+        pres:
+          sg:
+            1st: ambulo
+            2nd: ambulas
+      subj:
+        pres:
+          pl:
+            1st: ambulemus
+            2nd: ambuletis
+

data/spec/isodoc/base_spec.rb

@@ -90,18 +90,18 @@ RSpec.describe IsoDoc::I18n do
     e = HTMLEntities.new
     c = IsoDoc::I18n.new("fr", "Latn")
     expect(e.encode(c.l10n("Code; «code» and: code!"), :hexadecimal))
-      .to be_equivalent_to "Code ; « code » "\
+      .to be_equivalent_to "Code ; « code » " \
                            "and : code !"
     expect(e.encode(c.l10n("Code; «code» and: code!"), :hexadecimal))
-      .to be_equivalent_to "Code ; « code » "\
+      .to be_equivalent_to "Code ; « code » " \
                            "and : code !"
     c = IsoDoc::I18n.new("fr", "Latn", locale: "FR")
     expect(e.encode(c.l10n("Code; «code» and: code!"), :hexadecimal))
-      .to be_equivalent_to "Code ; « code » "\
+      .to be_equivalent_to "Code ; « code » " \
                            "and : code !"
     c = IsoDoc::I18n.new("fr", "Latn", locale: "CH")
     expect(e.encode(c.l10n("Code; «code» and: code!"), :hexadecimal))
-      .to be_equivalent_to "Code ; « code » "\
+      .to be_equivalent_to "Code ; « code » " \
                            "and : code !"
     expect(e.encode(c.l10n("http://xyz a;b"), :hexadecimal))
       .to be_equivalent_to "http://xyz a;b"
@@ -136,4 +136,23 @@ RSpec.describe IsoDoc::I18n do
     expect(c.inflect_ordinal(5, term, "SpelloutRules"))
       .to eq "fifth"
   end
+
+  it "does inflections" do
+    c = IsoDoc::I18n.new("en", "Latn", i18nyaml: "spec/assets/new.yaml")
+    expect(c.inflect("John", number: "sg")).to eq "John"
+    expect(c.inflect("Fred", number: "sg")).to eq "Fred"
+    expect(c.inflect("Fred", number: "pl")).to eq "Freds"
+    expect(c.inflect("Fred", number: "du")).to eq "Fred"
+    expect(c.inflect("Fred", tense: "pres")).to eq "Fred"
+    expect(c.inflect("Man", case: "dat")).to eq "viri"
+    expect(c.inflect("Man", number: "sg", case: "dat")).to eq "viri"
+    expect(c.inflect("Man", number: "pl", case: "acc")).to eq "virem"
+    expect(c.inflect("Woman", number: "pl", case: "gen")).to eq "mulierum"
+    expect(c.inflect("Good", number: "pl", case: "gen")).to eq "bonorum"
+    expect(c.inflect("Good", number: "pl", case: "gen", gender: "fem"))
+      .to eq "bonarum"
+    expect(c.inflect("Walk", person: "2nd")).to eq "ambulas"
+    expect(c.inflect("Walk", person: "2nd", number: "pl", mood: "subj"))
+      .to eq "ambuletis"
+  end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: isodoc-i18n
 version: !ruby/object:Gem::Version
-  version: 1.1.1
+  version: 1.1.3
 platform: ruby
 authors:
 - Ribose Inc.
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2022-09-27 00:00:00.000000000 Z
+date: 2022-10-13 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: htmlentities