Initial version

.gitignore

@@ -0,0 +1,3 @@
+.bundle
+Gemfile.lock
+pkg

.rspec

@@ -0,0 +1 @@
+--color

Gemfile

@@ -0,0 +1 @@
+gemspec

MIT-LICENSE

@@ -0,0 +1,20 @@
+Copyright (c) 2011 Bruno Michel
+
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of this software and associated documentation files (the
+"Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Software, and to
+permit persons to whom the Software is furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

README.md

@@ -0,0 +1,55 @@
+HTML Spellchecker
+=================
+
+Wants to spellcheck an HTML string properly? This gem is for you.
+It's powered by [Nokogiri](http://nokogiri.org/) and
+[hunspell-ffi](https://github.com/ahaller/hunspell-ffi)!
+
+
+How to use it
+-------------
+
+It's very simple. Install it with rubygems:
+
+    gem install html_spellchecker
+
+Or, if you use bundler, add it to your `Gemfile`:
+
+    gem "html_spellchecker", :version => "~>0.1"
+
+Then you can use it in your code:
+
+    require "html_spellchecker"
+	HTML_Spellchecker.english.spellcheck("<p>This is xzqwy.</p>")
+	# => "<p>This is <mark class="misspelled">xzqwy</mark>.</p>"
+
+The HTML_Spellchecker class can be initialized by giving 2 paths:
+the affinity and dictionnary for hunspell. There are helpers to
+create a new instance for english and french dictionnaries.
+
+Then, you can use `spellcheck` method: you give it an HTML string
+and it returns you with the same string with misspelled words
+enclosed in `<mark>` tags (with the `misspelled` class).
+
+HTML_Spellchecker can avoid to check the spelling of special tags
+like `<code>`, by keeping a list of the tags to spellcheck in
+`HTML_Spellchecker.spellcheckable_tags`.
+
+
+Issues or Suggestions
+---------------------
+
+Found an issue or have a suggestion? Please report it on
+[Github's issue tracker](http://github.com/nono/HTML-Spellchecker/issues).
+
+If you wants to make a pull request, please check the specs before:
+
+    rspec spec
+
+
+Credits
+-------
+
+Thanks [Andreas Haller](https://github.com/ahaller) for the hunspell-ffi gem.
+
+Copyright (c) 2011 Bruno Michel <bmichel@menfin.info>, released under the MIT license

html_spellchecker.gemspec

@@ -0,0 +1,17 @@
+Gem::Specification.new do |s|
+  s.name             = "html_spellchecker"
+  s.version          = "0.1.0"
+  s.date             = Time.now.utc.strftime("%Y-%m-%d")
+  s.homepage         = "http://github.com/nono/HTML-Spellchecker"
+  s.authors          = "Bruno Michel"
+  s.email            = "bmichel@menfin.info"
+  s.description      = "Wants to spellcheck an HTML string properly? This gem is for you."
+  s.summary          = "Wants to spellcheck an HTML string properly? This gem is for you."
+  s.extra_rdoc_files = %w(README.md)
+  s.files            = Dir["MIT-LICENSE", "README.md", "Gemfile", "lib/**/*.rb"]
+  s.require_paths    = ["lib"]
+  s.rubygems_version = %q{1.3.7}
+  s.add_dependency "nokogiri", "~>1.4"
+  s.add_dependency "hunspell-ffi", "=0.1.3.alpha2"
+  s.add_development_dependency "rspec", "~>2.4"
+end

lib/html_spellchecker.rb

@@ -0,0 +1,62 @@
+# Encoding: UTF-8
+
+require "hunspell-ffi"
+require "nokogiri"
+require "set"
+
+
+class HTML_Spellchecker
+  def self.english
+    @english ||= self.new("/usr/share/hunspell/en_US.aff", "/usr/share/hunspell/en_US.dic")
+  end
+
+  def self.french
+    @french ||= self.new("/usr/share/hunspell/fr_FR.aff", "/usr/share/hunspell/fr_FR.dic")
+  end
+
+  def initialize(aff, dic)
+    @dict = Hunspell.new(aff, dic)
+  end
+
+  def spellcheck(html)
+    Nokogiri::HTML::DocumentFragment.parse(html).spellcheck(@dict)
+  end
+
+  class <<self
+    attr_accessor :spellcheckable_tags
+  end
+  self.spellcheckable_tags = Set.new(%w(p ol ul li div header article nav section footer aside dd dt dl
+                                        span blockquote cite q mark ins del table td th tr tbody thead tfoot
+                                        a b i s em small strong hgroup h1 h2 h3 h4 h5 h6))
+end
+
+class Nokogiri::HTML::DocumentFragment
+  def spellcheckable?
+    true
+  end
+end
+
+class Nokogiri::XML::Node
+  def spellcheck(dict)
+    if spellcheckable?
+      inner = children.map {|child| child.spellcheck(dict) }.join
+      children.remove
+      add_child Nokogiri::HTML::DocumentFragment.parse(inner)
+    end
+    to_html(:indent => 0)
+  end
+
+  def spellcheckable?
+    HTML_Spellchecker.spellcheckable_tags.include? name
+  end
+end
+
+class Nokogiri::XML::Text
+  WORDS_REGEXP = RUBY_VERSION =~ /^1\.8/ ? /\w+/ : /\p{Word}+/
+
+  def spellcheck(dict)
+    to_xhtml(:encoding => 'UTF-8').gsub(WORDS_REGEXP) do |word|
+      dict.check(word) ? word : "<mark class=\"misspelled\">#{word}</mark>"
+    end
+  end
+end

spec/html_spellchecker_spec.rb

@@ -0,0 +1,35 @@
+# encoding: UTF-8
+path = File.expand_path(File.dirname(__FILE__) + "/../lib/")
+$LOAD_PATH.unshift(path) unless $LOAD_PATH.include?(path)
+require "html_spellchecker"
+
+
+describe HTML_Spellchecker do
+  let(:checker) { HTML_Spellchecker.english }
+
+  it "doesn't modify correct sentences" do
+    correct = "<p>This is a sentence with correct words.</p>"
+    checker.spellcheck(correct).should == correct
+  end
+
+  it "marks spelling errors" do
+    incorrect = "<p>xzqwy is not a word!</p>"
+    checker.spellcheck(incorrect) == "<p><mark class=\"misspelled\">xzqwy</span> is not a word!</p>"
+  end
+
+  it "doesn't try to spellcheck code tags" do
+    txt = "<code>class Foo\ndef hello\nputs 'Hi'\nend\nend</code>"
+    checker.spellcheck(txt).should == txt
+  end
+
+  it "can use different dictionnaries" do
+    french_text = "<p>Ceci est un texte correct, mais xzqwy n'est pas un mot</p>"
+    expected = french_text.gsub('xzqwy', '<mark class="misspelled">xzqwy</mark>')
+    HTML_Spellchecker.french.spellcheck(french_text).should == expected
+  end
+
+  it "can spellcheck nested tags" do
+    txt = "<p>This is <strong>Important and <em>xzqwy</em></strong>!</p>"
+    checker.spellcheck(txt).should == txt.gsub('xzqwy', '<mark class="misspelled">xzqwy</mark>')
+  end
+end