isodoc-i18n 1.0.6 → 1.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 40bf798e8180b8b3014f6822a1ac8e5336728231d1436dbfe74c49fd54326551
-  data.tar.gz: d41acbc509af8a536d02fc3e4fd0511972b61e8730eb47d43d1520b0efd9e000
+  metadata.gz: 71a7b4e2dcb8b00587cc92988c85a99c411444fb8698f8bd067c92a4f90751e2
+  data.tar.gz: 6304cf921c57b555ff259b9c1dfad4d7306e726c7f56f08291e4959a188a6181
 SHA512:
-  metadata.gz: 16aa8550bf43b723bd10576d86be8a371d7fe88cb4967ccd3a88f295f695a6fce19830b250ad587b6a968fee2f8bd4c11cfa37c34c6d49ff7a15ef4544426fe3
-  data.tar.gz: 6e0ddd03856d58eed0655f1c978adaec56442eca6c378da15b81c21fd4177be1daf573d157f63da82d21364b7b10c59df726a0cf10cbfe2c87104e25a0e88adb
+  metadata.gz: 9de07fcd21fdfd8ba2ab4d758f76952cbfe84075caad216543cd0a6fa2bcf409fab8e71bc71309aa98ed50342727be9a46c5ffc2625c27f987ede80c65e29b18
+  data.tar.gz: fff4809002eef23c8d550baed4fb1185f134b99fb7a61733d3413045134b11b1e18eaab54a8787e0e4817601ff9051d67ffc1efc97cfaa589a73320188e70d98

data/.github/workflows/rake.yml

@@ -10,6 +10,6 @@ on:
 
 jobs:
   rake:
-    uses: metanorma/metanorma-build-scripts/.github/workflows/generic-rake.yml@main
+    uses: metanorma/ci/.github/workflows/generic-rake.yml@main
     secrets:
       pat_token: ${{ secrets.METANORMA_CI_PAT_TOKEN }}

data/.github/workflows/release.yml

@@ -0,0 +1,24 @@
+# Auto-generated by Cimas: Do not edit it manually!
+# See https://github.com/metanorma/cimas
+name: release
+
+on:
+  workflow_dispatch:
+    inputs:
+      next_version:
+        description: |
+          Next release version. Possible values: x.y.z, major, minor, patch or pre|rc|etc
+        required: true
+        default: 'skip'
+  push:
+    tags: [ v* ]
+
+jobs:
+  release:
+    uses: metanorma/ci/.github/workflows/rubygems-release.yml@main
+    with:
+      next_version: ${{ github.event.inputs.next_version }}
+    secrets:
+      rubygems-api-key: ${{ secrets.METANORMA_CI_RUBYGEMS_API_KEY }}
+      pat_token: ${{ secrets.METANORMA_CI_PAT_TOKEN }}
+

data/isodoc-i18n.gemspec

@@ -23,7 +23,7 @@ Gem::Specification.new do |spec|
   spec.required_ruby_version = Gem::Requirement.new(">= 2.5.0")
 
   spec.add_dependency "htmlentities", "~> 4.3.4"
-  spec.add_dependency "metanorma-utils", "~> 1.3.0"
+  spec.add_dependency "metanorma-utils", "~> 1.4.0"
   spec.add_dependency "twitter_cldr"
 
   spec.add_development_dependency "debug"

data/lib/isodoc/i18n.rb

@@ -54,9 +54,10 @@ module IsoDoc
       @labels[key] = val
     end
 
-    def initialize(lang, script, i18nyaml: nil, i18nhash: nil)
+    def initialize(lang, script, locale: nil, i18nyaml: nil, i18nhash: nil)
       @lang = lang
       @script = script
+      @locale = locale
       y = load_yaml(lang, script, i18nyaml, i18nhash)
       @labels = y
       @labels["language"] = @lang
@@ -66,17 +67,16 @@ module IsoDoc
       end
     end
 
-    def self.l10n(text, lang = @lang, script = @script)
-      l10n(text, lang, script)
+    def self.l10n(text, lang = @lang, script = @script, locale = @locale)
+      l10n(text, lang, script, locale)
     end
 
-    # TODO: move to localization file
     # function localising spaces and punctuation.
     # Not clear if period needs to be localised for zh
-    def l10n(text, lang = @lang, script = @script)
-      if lang == "zh" && script == "Hans" then l10n_zh(text)
-      else bidiwrap(text, lang, script)
-      end
+    def l10n(text, lang = @lang, script = @script, locale = @locale)
+      lang == "zh" && script == "Hans" and text = l10n_zh(text)
+      lang == "fr" && text = l10n_fr(text, locale || "FR")
+      bidiwrap(text, lang, script)
     end
 
     def bidiwrap(text, lang, script)
@@ -107,23 +107,51 @@ module IsoDoc
       xml.to_xml.gsub(/<b>/, "").gsub("</b>", "").gsub(/<\?[^>]+>/, "")
     end
 
+    def l10n_fr(text, locale)
+      xml = Nokogiri::HTML::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
+    end
+
     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 、
     def l10_zh1(text)
+      l10n_zh_remove_space(l10n_zh_punct(text))
+    end
+
+    def l10n_zh_punct(text)
       [":：", ",，", ".。", ")）", "]】", ":：", ";；", "?？", "!！"].each do |m|
         text = text.gsub(/(?<=#{ZH_CHAR})#{Regexp.quote m[0]}/, m[1])
+        text = text.gsub(/^#{Regexp.quote m[0]}/, m[1])
       end
       ["(（", "[【"].each do |m|
         text = text.gsub(/#{Regexp.quote m[0]}(?=#{ZH_CHAR})/, m[1])
       end
+      text
+    end
+
+    def l10n_zh_remove_space(text)
       text.gsub(/(?<=#{ZH_CHAR}) (?=#{ZH_CHAR})/o, "")
         .gsub(/(?<=\d) (?=#{ZH_CHAR})/o, "")
         .gsub(/(?<=#{ZH_CHAR}) (?=\d)/o, "")
         .gsub(/(?<=#{ZH_CHAR}) (?=[A-Za-z](#{ZH_CHAR}|$))/o, "")
     end
 
+    def l10n_fr1(text, locale)
+      text = text.gsub(/(?<=\p{Alnum})([»›;?!])/, "\u202f\\1")
+      text = text.gsub(/^([»›;?!])/, "\u202f\\1")
+      text = text.gsub(/([«‹])/, "\\1\u202f")
+      colonsp = locale == "CH" ? "\u202f" : "\u00a0"
+      text = text.gsub(/(?<=\p{Alnum})(:)/, "#{colonsp}\\1")
+      text.gsub(/^(:)/, "#{colonsp}\\1")
+    end
+
     def boolean_conj(list, conn)
       case list.size
       when 0 then ""

data/lib/isodoc/version.rb

@@ -1,5 +1,5 @@
 module IsoDoc
   class I18n
-    VERSION = "1.0.6".freeze
+    VERSION = "1.1.0".freeze
   end
 end

data/spec/isodoc/base_spec.rb

@@ -38,7 +38,8 @@ RSpec.describe IsoDoc::I18n do
   end
 
   it "loads language hash overrides" do
-    c = IsoDoc::I18n.new("en", "Latn", i18nhash: YAML.load_file("spec/assets/new.yaml"))
+    c = IsoDoc::I18n.new("en", "Latn",
+                         i18nhash: YAML.load_file("spec/assets/new.yaml"))
     expect(c.text).to eq "text2"
     expect(c.at).to eq "at"
     expect(c.hash.to_s).to be_equivalent_to '{"key1"=>"val1", "key2"=>"val2"}'
@@ -85,6 +86,25 @@ RSpec.describe IsoDoc::I18n do
       .to be_equivalent_to "؜Code (hello, world.)؜"
   end
 
+  it "does French localisation" 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 » "\
+                           "and : code !"
+    expect(e.encode(c.l10n("Code; «code» and: code!"), :hexadecimal))
+      .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 » "\
+                           "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 » "\
+                           "and : code !"
+  end
+
   it "does boolean conjunctions" do
     c = IsoDoc::I18n.new("en", "Latn")
     expect(c.boolean_conj([], "and")).to eq ""

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: isodoc-i18n
 version: !ruby/object:Gem::Version
-  version: 1.0.6
+  version: 1.1.0
 platform: ruby
 authors:
 - Ribose Inc.
-autorequire:
+autorequire: 
 bindir: bin
 cert_chain: []
-date: 2022-08-16 00:00:00.000000000 Z
+date: 2022-09-25 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: htmlentities
@@ -30,14 +30,14 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 1.3.0
+        version: 1.4.0
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 1.3.0
+        version: 1.4.0
 - !ruby/object:Gem::Dependency
   name: twitter_cldr
   requirement: !ruby/object:Gem::Requirement
@@ -202,6 +202,7 @@ extensions: []
 extra_rdoc_files: []
 files:
 - ".github/workflows/rake.yml"
+- ".github/workflows/release.yml"
 - ".hound.yml"
 - ".rubocop.yml"
 - CODE_OF_CONDUCT.md
@@ -224,7 +225,7 @@ homepage: https://github.com/metanorma/isodoc-i18n
 licenses:
 - BSD-2-Clause
 metadata: {}
-post_install_message:
+post_install_message: 
 rdoc_options: []
 require_paths:
 - lib
@@ -239,8 +240,8 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.3.16
-signing_key:
+rubygems_version: 3.1.6
+signing_key: 
 specification_version: 4
 summary: isodoc-i18n
 test_files: []