markdown-ruby-china 0.1 → 0.2

data/.gitignore

@@ -0,0 +1,4 @@
+*.gem
+.bundle
+Gemfile.lock
+pkg/*

data/Gemfile

@@ -0,0 +1,5 @@
+# coding: utf-8
+
+source 'https://rubygems.org'
+gemspec
+

data/README.markdown

@@ -0,0 +1,16 @@
+Markdown render from Ruby China
+============================
+
+## 如何使用
+### 添加到Gemfile
+```ruby
+gem 'markdown-ruby-china'
+```
+
+### 调用
+```ruby
+MarkdownTopicConverter.format(content)
+```
+
+## 新功能
+1.   代码高亮 自动识别程序语言，文件名，别名等多种格式。

data/lib/markdown-ruby-china.rb

@@ -0,0 +1,244 @@
+# coding: utf-8
+
+require 'rails'
+require 'rails_autolink'
+require 'redcarpet'
+require 'singleton'
+require 'md_emoji'
+require 'pygments'
+require "nokogiri"
+
+Pygments::Lexer.create :name => "XML", :filenames => ["*.xaml"], :aliases => ["xml"], :mimetypes => ["application/xml+evoque"]
+
+module Redcarpet
+  module Render
+    class HTMLwithSyntaxHighlight < HTML
+
+      def initialize(extensions={})
+        super(extensions.merge(:xhtml => true,
+                               :no_styles => true,
+                               :filter_html => true,
+                               :hard_wrap => true))
+      end
+
+      # 如果换行符没识别，那么code代码块将不会被这里调用
+      def block_code(code, lexer)
+        lexer = begin
+          # 支持程序语言，文件名，别名等多种格式
+          lexer = (Pygments::Lexer.find(lexer) || Pygments::Lexer.find_by_extname(".#{lexer}") || Pygments::Lexer.find_by_alias(lexer.downcase))
+          lexer.aliases[0]
+        rescue
+          lexer.to_s.split('.')[-1] || 'text'
+        end
+
+        opts = {:lexer => lexer, :formatter => 'html', :options => {:encoding => 'utf-8', :linenos => 'table'}}
+        begin
+          Pygments.highlight(code, opts)
+        rescue => e
+          puts :language => lexer, :code => code
+          puts e
+          puts e.backtrace
+          Pygments.highlight(code, opts.merge(:lexer => 'text'))
+        end
+      end
+
+      def autolink(link, link_type)
+        # return link
+        if link_type.to_s == "email"
+          link
+        else
+          begin
+            # 防止 C 的 autolink 出来的内容有编码错误，万一有就直接跳过转换
+            # 比如这句:
+            # 此版本并非线上的http://yavaeye.com的源码.
+            link.match(/.+?/)
+          rescue
+            return link
+          end
+          # Fix Chinese neer the URL
+          bad_text = link.to_s.match(/[^\w:\/\-\,\$\!\.=\?&#+\|\%]+/im).to_s
+          link = link.to_s.gsub(bad_text, '')
+          "<a href=\"#{link}\" rel=\"nofollow\" target=\"_blank\">#{link}</a>#{bad_text}"
+        end
+      end
+    end
+
+    class HTMLwithTopic < HTMLwithSyntaxHighlight
+      # Topic 里面，所有的 head 改为 h4 显示
+      def header(text, header_level)
+        "<h4>#{text}</h4>"
+      end
+    end
+  end
+end
+
+class MarkdownConverter
+  include Singleton
+
+  def self.convert(text)
+    self.instance.convert(text)
+  end
+
+  def convert(text)
+    @converter.render(text)
+  end
+
+  private
+  def initialize
+    @converter = Redcarpet::Markdown.new(Redcarpet::Render::HTMLwithSyntaxHighlight, {
+        :autolink => true,
+        :fenced_code_blocks => true,
+        :no_intra_emphasis => true
+      })
+  end
+
+end
+
+class MarkdownTopicConverter < MarkdownConverter
+  def self.format(raw)
+    self.instance.format(raw)
+  end
+
+  def format(raw)
+    text = raw.clone
+    return '' if text.blank?
+
+    convert_bbcode_img(text)
+    users = nomalize_user_mentions(text)
+
+    # 如果 ``` 在刚刚换行的时候 Redcapter 无法生成正确，需要两个换行
+    text.gsub!("\n```","\n\n```")
+
+    # 调用MarkdownConverter父类来高亮代码
+    result = convert(text)
+
+    doc = Nokogiri::HTML.fragment(result)
+    link_mention_floor(doc)
+    link_mention_user(doc, users)
+    replace_emoji(doc)
+
+    return doc.to_html.strip
+  rescue => e
+    puts "MarkdownTopicConverter.format ERROR: #{e}"
+    return text
+  end
+
+  private
+  # convert bbcode-style image tag [img]url[/img] to markdown syntax ![alt](url)
+  def convert_bbcode_img(text)
+    text.gsub!(/\[img\](.+?)\[\/img\]/i) {"![#{image_alt $1}](#{$1})"}
+  end
+
+  def image_alt(src)
+    File.basename(src, '.*').capitalize
+  end
+
+  # borrow from html-pipeline
+  def has_ancestors?(node, tags)
+    while node = node.parent
+      if tags.include?(node.name.downcase)
+        break true
+      end
+    end
+  end
+
+  # convert '#N楼' to link
+  # Refer to emoji_filter in html-pipeline
+  def link_mention_floor(doc)
+    doc.search('text()').each do |node|
+      content = node.to_html
+      next if !content.include?('#')
+      next if has_ancestors?(node, %w(pre code))
+
+      html = content.gsub(/#(\d+)([楼樓Ff])/) {
+        %(<a href="#reply#{$1}" class="at_floor" data-floor="#{$1}">##{$1}#{$2}</a>)
+      }
+
+      next if html == content
+      node.replace(html)
+    end
+  end
+
+  NOMALIZE_USER_REGEXP = /(^|[^a-zA-Z0-9_!#\$%&*@＠])@([a-zA-Z0-9_]{1,20})/io
+  LINK_USER_REGEXP = /(^|[^a-zA-Z0-9_!#\$%&*@＠])@(user[0-9]{1,6})/io
+
+  # rename user name using incremental id
+  def nomalize_user_mentions(text)
+    users = []
+
+    text.gsub!(NOMALIZE_USER_REGEXP) do
+      prefix = $1
+      user = $2
+      users.push(user)
+      "#{prefix}@user#{users.size}"
+    end
+
+    users
+  end
+
+  # convert '@user' to link
+  # match any user even not exist.
+  def link_mention_user(doc, users)
+    doc.search('text()').each do |node|
+      content = node.to_html
+      next if !content.include?('@')
+      in_code = has_ancestors?(node, %w(pre code))
+      html = content.gsub(LINK_USER_REGEXP) {
+        prefix = $1
+        user_placeholder = $2
+        user_id = user_placeholder.sub(/^user/, '').to_i
+        user = users[user_id - 1] || user_placeholder
+
+        if in_code
+          "#{prefix}@#{user}"
+        else
+          %(#{prefix}<a href="/#{user}" class="at_user" title="@#{user}"><i>@</i>#{user}</a>)
+        end
+      }
+
+      node.replace(html)
+    end
+  end
+
+  def replace_emoji(doc)
+    doc.search('text()').each do |node|
+      content = node.to_html
+      next if !content.include?(':')
+      next if has_ancestors?(node, %w(pre code))
+
+      html = content.gsub(/:(\S+):/) do |emoji|
+
+        emoji_code = emoji #.gsub("|", "_")
+        emoji      = emoji_code.gsub(":", "")
+
+        if MdEmoji::EMOJI.include?(emoji)
+          file_name    = "#{emoji.gsub('+', 'plus')}.png"
+
+          %{<img src="#{upload_url}/assets/emojis/#{file_name}" class="emoji" } +
+            %{title="#{emoji_code}" alt="" />}
+        else
+          emoji_code
+        end
+      end
+
+      next if html == content
+      node.replace(html)
+    end
+  end
+
+  # for testing
+  def upload_url
+    Setting.upload_url
+  end
+
+  def initialize
+    @converter = Redcarpet::Markdown.new(Redcarpet::Render::HTMLwithTopic.new, {
+        :autolink => true,
+        :fenced_code_blocks => true,
+        :strikethrough => true,
+        :space_after_headers => true,
+        :no_intra_emphasis => true
+      })
+    @emoji = MdEmoji::Render.new
+  end
+end

data/markdown-ruby-china.gemspec

@@ -0,0 +1,19 @@
+Gem::Specification.new do |s|
+  s.name        = 'markdown-ruby-china'
+  s.version     = '0.2'
+  s.date        = '2013-04-17'
+  s.summary     = ""
+  s.description = ""
+  s.authors     = ["Ruby-China team", "David Chen"]
+  s.email       = 'mvjome@gmail.com'
+  s.homepage    = 'https://github.com/eoecn/markdown-ruby-china'
+  s.require_paths = ["lib"]
+
+  s.add_dependency "rails_autolink"
+  s.add_dependency 'redcarpet'
+  s.add_dependency 'md_emoji'
+  s.add_dependency 'pygments.rb'
+  s.add_dependency "nokogiri"
+
+  s.files = `git ls-files`.split("\n")
+end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: markdown-ruby-china
 version: !ruby/object:Gem::Version
-  version: '0.1'
+  version: '0.2'
   prerelease: 
 platform: ruby
 authors:
@@ -97,7 +97,12 @@ email: mvjome@gmail.com
 executables: []
 extensions: []
 extra_rdoc_files: []
-files: []
+files:
+- .gitignore
+- Gemfile
+- README.markdown
+- lib/markdown-ruby-china.rb
+- markdown-ruby-china.gemspec
 homepage: https://github.com/eoecn/markdown-ruby-china
 licenses: []
 post_install_message: