reverse_markdown 0.4.7 → 0.5.0

data/lib/reverse_markdown/converters.rb

@@ -0,0 +1,22 @@
+module ReverseMarkdown
+  module Converters
+    def self.register(tag_name, converter)
+      @@converters ||= {}
+      @@converters[tag_name.to_sym] = converter
+    end
+
+    def self.lookup(tag_name)
+      @@converters[tag_name.to_sym] or default_converter(tag_name)
+    end
+
+    private
+
+    def self.default_converter(tag_name)
+      if ReverseMarkdown.config.ignore_unknown_tags
+        ReverseMarkdown::Converters::Dump.new
+      else
+        raise UnknownTagError, "unknown tag: #{tag_name}"
+      end
+    end
+  end
+end

data/lib/reverse_markdown/errors.rb

@@ -1,3 +1,7 @@
 module ReverseMarkdown
-  class ParserError < StandardError; end
-end
+  class Error < StandardError
+  end
+
+  class UnknownTagError < Error
+  end
+end

data/lib/reverse_markdown/version.rb

@@ -1,3 +1,3 @@
 module ReverseMarkdown
-  VERSION = "0.4.7"
+  VERSION = "0.5.0"
 end

data/lib/reverse_markdown.rb

@@ -1,24 +1,56 @@
 require 'digest'
+require 'nokogiri'
 require 'reverse_markdown/version'
-require 'reverse_markdown/mapper'
 require 'reverse_markdown/errors'
-require 'nokogiri'
+require 'reverse_markdown/cleaner'
+require 'reverse_markdown/config'
+require 'reverse_markdown/converters'
+require 'reverse_markdown/converters/base'
+require 'reverse_markdown/converters/a'
+require 'reverse_markdown/converters/blockquote'
+require 'reverse_markdown/converters/br'
+require 'reverse_markdown/converters/code'
+require 'reverse_markdown/converters/div'
+require 'reverse_markdown/converters/dump'
+require 'reverse_markdown/converters/em'
+require 'reverse_markdown/converters/h'
+require 'reverse_markdown/converters/hr'
+require 'reverse_markdown/converters/img'
+require 'reverse_markdown/converters/li'
+require 'reverse_markdown/converters/ol'
+require 'reverse_markdown/converters/p'
+require 'reverse_markdown/converters/pass_through'
+require 'reverse_markdown/converters/pre'
+require 'reverse_markdown/converters/strong'
+require 'reverse_markdown/converters/table'
+require 'reverse_markdown/converters/td'
+require 'reverse_markdown/converters/tr'
+require 'reverse_markdown/converters/text'
 
 module ReverseMarkdown
 
-  def self.parse(input, opts={})
+  def self.convert(input, options = {})
     root = case input
       when String                  then Nokogiri::HTML(input).root
       when Nokogiri::XML::Document then input.root
       when Nokogiri::XML::Node     then input
     end
 
-    ReverseMarkdown::Mapper.new(opts).process_root(root)
+    root or return ''
+    config.apply(options)
+    result = ReverseMarkdown::Converters.lookup(root.name).convert(root)
+    config.reset
+    cleaner.tidy(result)
   end
 
-  # 2012/08/11 joe: possibly deprecate in favour of #parse
-  class << self
-    alias parse_string  parse
-    alias parse_element parse
+  def self.config
+    @config ||= Config.new
+    yield @config if block_given?
+    @config
   end
+
+  def self.cleaner
+    @cleaner ||= Cleaner.new
+  end
+
 end

data/reverse_markdown.gemspec

@@ -8,7 +8,7 @@ Gem::Specification.new do |s|
   s.authors     = ["Johannes Opper"]
   s.email       = ["xijo@gmx.de"]
   s.homepage    = "http://github.com/xijo/reverse_markdown"
-  s.summary     = %q{Transform html code into markdown.}
+  s.summary     = %q{Convert html code into markdown.}
   s.description = %q{Map simple html back into markdown, e.g. if you want to import existing html data in your application.}
 
   s.rubyforge_project = "reverse_markdown"
@@ -24,4 +24,6 @@ Gem::Specification.new do |s|
   s.add_development_dependency 'simplecov'
   s.add_development_dependency 'rake'
   s.add_development_dependency 'redcarpet'
+  s.add_development_dependency 'byebug'
+  s.add_development_dependency 'codeclimate-test-reporter'
 end

data/spec/assets/lists.html

@@ -52,5 +52,29 @@
       <li><p>li 2, p 1</p></li>
     </ul>
 
+    <ol>
+      <li>
+        one
+        <ol>
+          <li>one one</li>
+          <li>one two</li>
+        </ol>
+      </li>
+      <li>
+        two
+        <ol>
+          <li>
+            two one
+            <ol>
+              <li>two one one</li>
+              <li>two one two</li>
+            </ol>
+          </li>
+          <li>two two</li>
+        </ol>
+      </li>
+      <li>three</li>
+    </ol>
+
   </body>
 </html>

data/spec/assets/unknown_tags.html

@@ -0,0 +1,9 @@
+<html>
+  <body>
+    <p>
+      <foo>
+        <bar>Foo with bar</bar>
+      </foo>
+    </p>
+  </body>
+</html>

data/spec/components/anchors_spec.rb

@@ -1,13 +1,13 @@
 require 'spec_helper'
 
-describe ReverseMarkdown::Mapper do
+describe ReverseMarkdown do
 
   let(:input)    { File.read('spec/assets/anchors.html') }
   let(:document) { Nokogiri::HTML(input) }
-  subject { ReverseMarkdown.parse_string(input) }
+  subject { ReverseMarkdown.convert(input) }
 
   it { should include ' [Foobar](http://foobar.com) ' }
-  it { should include ' [Fubar](http://foobar.com "f***** up beyond all recognition") ' }
+  it { should include ' [Fubar](http://foobar.com "f\*\*\*\*\* up beyond all recognition") ' }
   it { should include ' [**Strong foobar**](http://strong.foobar.com) ' }
 
   it { should include ' ![](http://foobar.com/logo.png) ' }
@@ -18,9 +18,9 @@ describe ReverseMarkdown::Mapper do
 
   context "links to ignore" do
     it { should include ' ignore anchor tags with no link text ' }
-    it { should include ' not ignore [ ![An Image](image.png) ](foo.html) anchor tags with images' }
+    it { should include ' not ignore [![An Image](image.png)](foo.html) anchor tags with images' }
     it { should include ' pass through the text of internal jumplinks without treating them as links ' }
     it { should include ' pass through the text of anchor tags with no href without treating them as links ' }
   end
 
-end
+end

data/spec/components/basic_spec.rb

@@ -1,10 +1,10 @@
 require 'spec_helper'
 
-describe ReverseMarkdown::Mapper do
+describe ReverseMarkdown do
 
   let(:input)    { File.read('spec/assets/basic.html') }
   let(:document) { Nokogiri::HTML(input) }
-  subject { ReverseMarkdown.parse_string(input) }
+  subject { ReverseMarkdown.convert(input) }
 
   it { should match /plain text ?\n/ }
   it { should match /# h1\n/ }
@@ -25,7 +25,7 @@ describe ReverseMarkdown::Mapper do
   it { should match /before and after strong tags containing whitespace/ }
   it { should match /\*\*double strong tags\*\*/ }
   it { should match /\*\*double strong tags in p tag\*\*/ }
-  it { should match /before \*\* double strong tags containing whitespace \*\* after/ }
+  it { should match /before \*\*double strong tags containing whitespace\*\* after/ }
 
   it { should match /_i tag content_/ }
   it { should match /\*\*b tag content\*\*/ }
@@ -33,7 +33,7 @@ describe ReverseMarkdown::Mapper do
   it { should match /br tags become double space followed by newline  \n/ }
   #it { should match /br tags XXX  \n/ }
 
-  it { should match /\nbefore hr ?\n\* \* \*\n ?after hr\n/ }
+  it { should match /before hr \n\* \* \*\n after hr/ }
 
   it { should match /section 1\n ?\nsection 2/ }
-end
+end

data/spec/components/code_spec.rb

@@ -1,24 +1,24 @@
 require 'spec_helper'
 
-describe ReverseMarkdown::Mapper do
+describe ReverseMarkdown do
 
   let(:input)    { File.read('spec/assets/code.html') }
   let(:document) { Nokogiri::HTML(input) }
-  subject { ReverseMarkdown.parse_string(input) }
+  subject { ReverseMarkdown.convert(input) }
 
   it { should match /inline `code` block/ }
   it { should match /\    var this\;\n    this\.is/ }
   it { should match /block"\)\n    console/ }
 
   context "with github style code blocks" do
-    subject { ReverseMarkdown.parse_string(input, :github_style_code_blocks => true) }
+    subject { ReverseMarkdown.convert(input, github_flavored: true) }
     it { should match /inline `code` block/ }
     it { should match /```\nvar this\;\nthis/ }
-    it { should match /it is"\) ?\n\t\n```/ }
+    it { should match /it is"\) ?\n```/ }
   end
 
   context "code with indentation" do
-    subject { ReverseMarkdown.parse_string(input) }
+    subject { ReverseMarkdown.convert(input) }
     it { should match(/^    tell application "Foo"\n/) }
     it { should match(/^        beep\n/) }
     it { should match(/^    end tell\n/) }

data/spec/components/escapables_spec.rb

@@ -1,10 +1,10 @@
 require 'spec_helper'
 
-describe ReverseMarkdown::Mapper do
+describe ReverseMarkdown do
 
   let(:input)    { File.read('spec/assets/escapables.html') }
   let(:document) { Nokogiri::HTML(input) }
-  subject { ReverseMarkdown.parse_string(input) }
+  subject { ReverseMarkdown.convert(input) }
 
   context "multiple asterisks" do
     it { should include ' \*\*two asterisks\*\* ' }
@@ -19,4 +19,4 @@ describe ReverseMarkdown::Mapper do
   context "underscores within words in code blocks" do
     it { should include '    var theoretical_max_infin = 1.0;' }
   end
-end
+end

data/spec/components/from_the_wild_spec.rb

@@ -1,16 +1,16 @@
 require 'spec_helper'
 
-describe ReverseMarkdown::Mapper do
+describe ReverseMarkdown do
 
   let(:input)    { File.read('spec/assets/from_the_wild.html') }
   let(:document) { Nokogiri::HTML(input) }
-  subject { ReverseMarkdown.parse_string(input) }
+  subject { ReverseMarkdown.convert(input) }
 
   it "should make sense of strong-crazy markup (as seen in the wild)" do
     subject.should ==
       '** .' + "  \n" +
-      '\*\*\* intentcast ** : logo design' + "  \n" +
-      '** . **'
+      ' \*\*\* intentcast ** : logo design' + "   \n" +
+      "**.**\n\n"
   end
 
-end
+end

data/spec/components/html_fragment_spec.rb

@@ -1,10 +1,10 @@
 require 'spec_helper'
 
-describe ReverseMarkdown::Mapper do
+describe ReverseMarkdown do
 
   let(:input)    { File.read('spec/assets/html_fragment.html') }
   let(:document) { Nokogiri::HTML(input) }
-  subject { ReverseMarkdown.parse_string(input) }
+  subject { ReverseMarkdown.convert(input) }
 
   it { should == "naked text 1\n\nparagraph text\n\nnaked text 2" }
 end

data/spec/components/lists_spec.rb

@@ -1,10 +1,10 @@
 require 'spec_helper'
 
-describe ReverseMarkdown::Mapper do
+describe ReverseMarkdown do
 
   let(:input)    { File.read('spec/assets/lists.html') }
   let(:document) { Nokogiri::HTML(input) }
-  subject { ReverseMarkdown.parse_string(input) }
+  subject { ReverseMarkdown.convert(input) }
 
   it { should match /\n- unordered list entry\n/ }
   it { should match /\n- unordered list entry 2\n/ }
@@ -22,10 +22,10 @@ describe ReverseMarkdown::Mapper do
   end
 
   context "nested list with lots of whitespace" do
-    it { should match /\n- item wa\n/ }
-    it { should match /\n- item wb\n/ }
-    it { should match /\n {2}- item wbb\n/ }
-    it { should match /\n {2}- item wbc\n/ }
+    it { should match /\n- item wa \n/ }
+    it { should match /\n- item wb \n/ }
+    it { should match /\n  - item wbb \n/ }
+    it { should match /\n  - item wbc \n/ }
   end
 
   context "lists containing links" do
@@ -42,4 +42,15 @@ describe ReverseMarkdown::Mapper do
     xit { should match /\n- li 1, p 1\n\n- li 1, p 2\n/ }
   end
 
-end
+  context 'it produces correct numbering' do
+    it { should include "1. one" }
+    it { should include "  1. one one" }
+    it { should include "  2. one two" }
+    it { should include "2. two" }
+    it { should include "  1. two one" }
+    it { should include "    1. two one one" }
+    it { should include "    2. two one two" }
+    it { should include "  2. two two" }
+    it { should include "3. three" }
+  end
+end

data/spec/components/paragraphs_spec.rb

@@ -1,10 +1,10 @@
 require 'spec_helper'
 
-describe ReverseMarkdown::Mapper do
+describe ReverseMarkdown do
 
   let(:input)    { File.read('spec/assets/paragraphs.html') }
   let(:document) { Nokogiri::HTML(input) }
-  subject { ReverseMarkdown.parse_string(input) }
+  subject { ReverseMarkdown.convert(input) }
 
   it { should_not start_with "\n\n" }
   it { should start_with "First content\n\nSecond content\n\n" }

data/spec/components/quotation_spec.rb

@@ -1,12 +1,12 @@
 require 'spec_helper'
 
-describe ReverseMarkdown::Mapper do
+describe ReverseMarkdown do
 
   let(:input)    { File.read('spec/assets/quotation.html') }
   let(:document) { Nokogiri::HTML(input) }
-  subject { ReverseMarkdown.parse_string(input) }
+  subject { ReverseMarkdown.convert(input) }
 
-  it { should include "\n    Block of code" }
-  it { should include "\n> First quoted paragraph\n\n> Second quoted paragraph" }
+  it { should match /^    Block of code$/ }
+  it { should include "\n> First quoted paragraph\n> \n> Second quoted paragraph" }
 
-end
+end

data/spec/components/tables_spec.rb

@@ -1,10 +1,10 @@
 require 'spec_helper'
 
-describe ReverseMarkdown::Mapper do
+describe ReverseMarkdown do
 
   let(:input)    { File.read('spec/assets/tables.html') }
   let(:document) { Nokogiri::HTML(input) }
-  subject { ReverseMarkdown.parse_string(input) }
+  subject { ReverseMarkdown.convert(input) }
 
   it { should match /\n\| header 1 \| header 2 \| header 3 \|\n\| --- \| --- \| --- \|\n/ }
   it { should match /\n\| data 1-1 \| data 2-1 \| data 3-1 \|\n/ }

data/spec/components/unknown_tags_spec.rb

@@ -0,0 +1,21 @@
+require 'spec_helper'
+
+describe ReverseMarkdown do
+
+  let(:input)    { File.read('spec/assets/unknown_tags.html') }
+  let(:document) { Nokogiri::HTML(input) }
+  let(:result)   { ReverseMarkdown.convert(input) }
+
+  context 'with ignore_unknown_tags enabled' do
+    before { ReverseMarkdown.config.ignore_unknown_tags = true }
+
+    it { result.should include "<foo><bar>Foo with bar</bar></foo>" }
+  end
+
+  context 'with ignore_unknown_tags disabled' do
+    before { ReverseMarkdown.config.ignore_unknown_tags = false }
+
+    it { expect { result }.to raise_error(ReverseMarkdown::UnknownTagError) }
+  end
+end
+

data/spec/html_to_markdown_to_html_spec.rb

@@ -12,7 +12,7 @@ describe 'Round trip: HTML to markdown (via reverse_markdown) to HTML (via redca
   end
 
   def html2markdown2html(orig_html)
-    markdown = ReverseMarkdown.parse_string orig_html
+    markdown = ReverseMarkdown.convert orig_html
     new_html = Redcarpet::Markdown.new(Redcarpet::Render::HTML).render(markdown)
     new_html
   end
@@ -60,7 +60,6 @@ describe 'Round trip: HTML to markdown (via reverse_markdown) to HTML (via redca
   end
 
   it "should preserve links inside <strong> tags" do
-    pending
     roundtrip_should_preserve(%{<p><strong><a href="/wiki/Western_philosophy" title="Western philosophy">Western philosophy</a></strong></p>})
   end
 

data/spec/lib/reverse_markdown/cleaner_spec.rb

@@ -0,0 +1,76 @@
+require 'spec_helper'
+
+describe ReverseMarkdown::Cleaner do
+  let(:cleaner) { ReverseMarkdown::Cleaner.new }
+
+  describe '#remove_newlines' do
+    it 'removes more than 2 subsequent newlines' do
+      result = cleaner.remove_newlines("foo\n\n\nbar")
+      result.should eq "foo\n\nbar"
+    end
+
+    it 'skips single and double newlines' do
+      result = cleaner.remove_newlines("foo\nbar\n\nbaz")
+      result.should eq "foo\nbar\n\nbaz"
+    end
+  end
+
+  describe '#remove_inner_whitespaces' do
+    it 'removes duplicate whitespaces from the string' do
+      result = cleaner.remove_inner_whitespaces('foo  bar')
+      result.should eq "foo bar"
+    end
+
+    it 'performs changes for multiple lines' do
+      result = cleaner.remove_inner_whitespaces("foo  bar\nbar  foo")
+      result.should eq "foo bar\nbar foo"
+    end
+
+    it 'keeps leading whitespaces' do
+      result = cleaner.remove_inner_whitespaces("    foo  bar\n    bar  foo")
+      result.should eq "    foo bar\n    bar foo"
+    end
+
+    it 'keeps trailing whitespaces' do
+      result = cleaner.remove_inner_whitespaces("foo  \n")
+      result.should eq "foo  \n"
+    end
+
+    it 'keeps trailing newlines' do
+      result = cleaner.remove_inner_whitespaces("foo\n")
+      result.should eq "foo\n"
+    end
+
+    it 'removes tabs as well' do
+      result = cleaner.remove_inner_whitespaces("foo\t \tbar")
+      result.should eq "foo bar"
+    end
+  end
+
+  describe '#clean_tag_borders' do
+    it 'removes not needed whitespaces from strong tags' do
+      input = "foo ** foobar ** bar"
+      result = cleaner.clean_tag_borders(input)
+      result.should eq "foo **foobar** bar"
+    end
+
+    it 'remotes leading or trailing whitespaces independently' do
+      input = "1 **fat ** 2 ** fat** 3"
+      result = cleaner.clean_tag_borders(input)
+      result.should eq "1 **fat** 2 **fat** 3"
+    end
+
+    it 'adds whitespaces if there are none' do
+      input = "1**fat**2"
+      result = cleaner.clean_tag_borders(input)
+      result.should eq "1 **fat** 2"
+    end
+
+    it 'cleans italic stuff as well' do
+      input = "1 __italic __ 2 __ italic__ 3__italic __4"
+      result = cleaner.clean_tag_borders(input)
+      result.should eq "1 __italic__ 2 __italic__ 3 __italic__ 4"
+    end
+  end
+
+end

data/spec/lib/reverse_markdown/converters/blockquote_spec.rb

@@ -0,0 +1,18 @@
+require 'spec_helper'
+
+describe ReverseMarkdown::Converters::Blockquote do
+
+  let(:converter) { ReverseMarkdown::Converters::Blockquote.new }
+
+  it 'converts nested elements as well' do
+    input = Nokogiri::XML.parse("<blockquote><ul><li>foo</li></ul></blockquote>").root
+    result = converter.convert(input)
+    result.should eq "> - foo"
+  end
+
+  it 'can deal with paragraphs inside' do
+    input = Nokogiri::XML.parse("<blockquote><p>Some text.</p><p>Some more text.</p></blockquote>").root
+    result = converter.convert(input)
+    result.should eq "> Some text.\n> \n> Some more text."
+  end
+end

data/spec/lib/reverse_markdown/converters/br_spec.rb

@@ -0,0 +1,9 @@
+require 'spec_helper'
+
+describe ReverseMarkdown::Converters::Br do
+  let(:converter) { ReverseMarkdown::Converters::Br.new }
+
+  it 'just converts into two spaces and a newline' do
+    converter.convert(:anything).should eq "  \n"
+  end
+end

data/spec/lib/reverse_markdown_spec.rb

@@ -0,0 +1,37 @@
+require 'spec_helper'
+
+describe ReverseMarkdown do
+  let(:input)    { File.read('spec/assets/minimum.html') }
+  let(:document) { Nokogiri::HTML(input) }
+
+  it "parses nokogiri documents" do
+    lambda { ReverseMarkdown.convert(document) }.should_not raise_error
+  end
+
+  it "parses nokogiri elements" do
+    lambda { ReverseMarkdown.convert(document.root) }.should_not raise_error
+  end
+
+  it "parses string input" do
+    lambda { ReverseMarkdown.convert(input) }.should_not raise_error
+  end
+
+  it "behaves in a sane way when root element is nil" do
+    ReverseMarkdown.convert(nil).should == ''
+  end
+
+  describe '#config' do
+    it 'stores a given configuration option' do
+      ReverseMarkdown.config.github_flavored = true
+      ReverseMarkdown.config.github_flavored.should be_true
+    end
+
+    it 'can be used as a block configurator as well' do
+      ReverseMarkdown.config do |config|
+        config.github_flavored.should be_false
+        config.github_flavored = true
+      end
+      ReverseMarkdown.config.github_flavored.should be_true
+    end
+  end
+end

data/spec/spec_helper.rb

@@ -1,4 +1,8 @@
+require 'codeclimate-test-reporter'
+CodeClimate::TestReporter.start
+
 require 'simplecov'
+require 'byebug'
 
 SimpleCov.adapters.define 'gem' do
   add_filter '/spec/'
@@ -11,4 +15,8 @@ require 'reverse_markdown'
 
 RSpec.configure do |config|
   config.color_enabled = true
-end
+
+  config.after(:each) do
+    ReverseMarkdown.config.reset
+  end
+end