orthotypo 0.5.1 → 0.6.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: '05389ada5ba288061bd70f8e3170fb77b7d4d9df171b90e55d3c0c6edbf4d810'
-  data.tar.gz: edd5731b28644f5515133994a89185f2da4fc5cffb7c12273f2d1f48985571be
+  metadata.gz: b41b05766be4f11e6494a7e89823a6ab730910aec4a679d22b764ae96fb05f10
+  data.tar.gz: 0c2a072f5c3ebd8583b70c7ee4bf0eafe064c827d4e0af250728bcf5762dc5ae
 SHA512:
-  metadata.gz: 32280cc5abbc8b3d9a728db9943f815821c84c97384f59f6be2f0ba12789e2ef204a8c72ec579a2544cea505b9c2f3d46c51aef5db7006ac74e3e8d568813a85
-  data.tar.gz: b81b849cf147939f7ad2e6fb16ccbc3af8eed2016efa67a288fcc9fc2c7cbd3d9d2d4b78ad4d70588fd75dade26830e4443627024838404f01e4010cbb5f2989
+  metadata.gz: 20ff73ede6bf05d4b2b8842604b09267f36cf2d44fc17c6e4a467e9e703934a20f18616368cc5af88529122166130c8b868612af6caab95ff6f685236b862443
+  data.tar.gz: 55f9a2023f185176d52de197847f92ee61ea0e8ba20d70c920fec7e31b4d86dbae0ab87594adcd7c406fcdbbc2ef189fa8a113bcfeab9f60786c31dd77771f1e

data/Gemfile.lock

@@ -1,8 +1,9 @@
 PATH
   remote: .
   specs:
-    orthotypo (0.5.1)
+    orthotypo (0.6.0)
       htmlentities
+      nokogiri
 
 GEM
   remote: https://rubygems.org/
@@ -14,6 +15,10 @@ GEM
     htmlentities (4.3.4)
     json (2.6.3)
     language_server-protocol (3.17.0.3)
+    nokogiri (1.15.4-arm64-darwin)
+      racc (~> 1.4)
+    nokogiri (1.15.4-x86_64-darwin)
+      racc (~> 1.4)
     parallel (1.23.0)
     parser (3.2.2.4)
       ast (~> 2.4.1)

data/lib/orthotypo/analyzer.rb

@@ -0,0 +1,18 @@
+module Orthotypo
+  class Analyzer
+
+    def self.url?(string)
+      (string =~ /\A#{URI::DEFAULT_PARSER.regexp[:ABS_URI]}\z/) && 
+      (string =~ /\A(http|https):\/\/[a-z0-9]+([\-\.]{1}[a-z0-9]+)*\.[a-z]{2,5}(:[0-9]{1,5})?(\/.*)?\z/ix) ? true : false
+    end
+  
+    def self.email?(string)
+      string =~ /\A#{URI::MailTo::EMAIL_REGEXP}\z/ ? true : false
+    end
+
+    def self.precious?(string)
+      email?(string) || url?(string)
+    end
+    
+  end
+end

data/lib/orthotypo/composer.rb

@@ -5,6 +5,7 @@ module Orthotypo
     SPACE = ' '.freeze
     NBSP = ' '.freeze
     NNBSP = ' '.freeze
+    PRECIOUS_TOKEN = 'orthotypopreciousthing'
 
     def initialize(string, html: nil)
       @string = string
@@ -52,15 +53,18 @@ module Orthotypo
 
     def prepare_ortho
       @ortho = string.dup
-      @ortho = html_entities.decode(@ortho) if contains_html_entities?
+      # @ortho = html_entities.decode(@ortho) if contains_html_entities?
+      @nokogiri = Nokogiri::HTML.fragment @ortho
     end
 
     def clean_ortho
-      @ortho = html_entities.encode(@ortho) if contains_html_entities?
+      @ortho = @nokogiri.to_s
+      # @ortho = html_entities.encode(@ortho) if contains_html_entities?
     end
 
     def parse
       prepare_ortho
+      preserve_precious_things
       # Chars
       parse_chars_with_space_before
       parse_chars_with_space_after
@@ -73,6 +77,30 @@ module Orthotypo
       parse_numbers
       # 
       clean_ortho
+      restore_precious_things
+    end
+
+    def preserve_precious_things
+      @precious_things = []
+      @nokogiri.traverse do |node|
+        next unless node.text?
+        new_content = node.content.split(SPACE).map { |fragment|
+          if Analyzer::precious?(fragment)
+            token = "#{PRECIOUS_TOKEN}#{@precious_things.length}"
+            @precious_things << fragment
+            token
+          else
+            fragment
+          end
+        }.join(SPACE)
+        node.content = new_content
+      end
+    end
+
+    def restore_precious_things
+      @precious_things.each_with_index do |value, index|
+        @ortho.gsub! "#{PRECIOUS_TOKEN}#{index}", value
+      end
     end
 
     def parse_chars_with_space_before
@@ -145,7 +173,10 @@ module Orthotypo
     end
 
     def fix(bad, good)
-      @ortho.gsub! bad, good
+      @nokogiri.traverse do |node|
+        next unless node.text?
+        node.content = node.content.gsub(bad, good)
+      end
     end
   end
 end

data/lib/orthotypo/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Orthotypo
-  VERSION = "0.5.1"
+  VERSION = "0.6.0"
 end

data/lib/orthotypo.rb

@@ -1,6 +1,9 @@
 # frozen_string_literal: true
 
 require "htmlentities"
+require "nokogiri"
+require "uri"
+require_relative "orthotypo/analyzer"
 require_relative "orthotypo/composer"
 require_relative "orthotypo/composer/fr"
 require_relative "orthotypo/composer/fr_fr"

data/orthotypo.gemspec

@@ -21,6 +21,7 @@ Gem::Specification.new do |spec|
   spec.require_paths = "lib"
 
   spec.add_dependency "htmlentities"
+  spec.add_dependency "nokogiri"
 
   spec.add_development_dependency "rspec"
   spec.add_development_dependency "rspec-nc"

data/spec/analyzer_spec.rb

@@ -0,0 +1,8 @@
+require 'spec_helper'
+
+describe Orthotypo::Analyzer do
+  it 'find urls' do
+    expect(Orthotypo::Analyzer.url?("https://unsplash.com/@lusvardi?utm_source=osuny")).to be true
+    expect(Orthotypo::Analyzer.url?("mot:")).to be false
+  end
+end

data/spec/orthotypo_spec.rb

@@ -8,4 +8,14 @@ describe Orthotypo do
   it 'leaves html tags untouched' do
     expect("<b>bold</b>".ortho).to eq "<b>bold</b>"
   end
+
+  it 'leaves URLs untouched' do
+    expect("https://unsplash.com/@lusvardi?utm_source=osuny".ortho).to eq "https://unsplash.com/@lusvardi?utm_source=osuny"
+    expect("<a href=\"https://unsplash.com/@lusvardi?utm_source=osuny\">https://unsplash.com/@lusvardi?utm_source=osuny</a>".ortho).to eq "<a href=\"https://unsplash.com/@lusvardi?utm_source=osuny\">https://unsplash.com/@lusvardi?utm_source=osuny</a>"
+  end
+
+  it 'leaves URLs untouched' do
+    expect("prenom.nom@example.com".ortho).to eq "prenom.nom@example.com"
+    expect("<a href=\"mailto:prenom.nom@example.com\">prenom.nom@example.com</a>".ortho).to eq "<a href=\"mailto:prenom.nom@example.com\">prenom.nom@example.com</a>"
+  end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: orthotypo
 version: !ruby/object:Gem::Version
-  version: 0.5.1
+  version: 0.6.0
 platform: ruby
 authors:
 - Arnaud Levy
@@ -24,6 +24,20 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
+- !ruby/object:Gem::Dependency
+  name: nokogiri
+  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: rspec
   requirement: !ruby/object:Gem::Requirement
@@ -87,6 +101,7 @@ files:
 - bin/console
 - bin/setup
 - lib/orthotypo.rb
+- lib/orthotypo/analyzer.rb
 - lib/orthotypo/composer.rb
 - lib/orthotypo/composer/en.rb
 - lib/orthotypo/composer/en_gb.rb
@@ -98,6 +113,7 @@ files:
 - lib/orthotypo/version.rb
 - orthotypo.gemspec
 - sig/orthotypo.rbs
+- spec/analyzer_spec.rb
 - spec/composer/en_spec.rb
 - spec/composer/fr_spec.rb
 - spec/localizer_spec.rb