godfat-friendly_format 0.6.0 → 0.6.1

data/CHANGES

@@ -1,8 +1,17 @@
 = friendly_format changes history
 
-== friendly_format 0.6.0 / ?
+== friendly_format 0.6.1 / 2009-04-05
 
 * added nokogiri and libxml-ruby support.
+* drop support for hpricot < 0.7
+* you can explicitly choose adapter, or auto-choose:
+  1. hpricot or falls back to:
+  2. nokogiri or falls back to:
+  3. libxml-ruby
+* now internally, we store string instead of symbol
+  for accepted tags to prevent from symbol explosion.
+* ruby 1.9 compliant
+* escape gt(>) in pre tag and forbidden tag as well.
 
 == friendly_format 0.5.1 / 2008-12-11
 

data/README

@@ -1,4 +1,4 @@
-= friendly_format 0.6
+= friendly_format 0.6.1
 by Lin Jen-Shin (a.k.a. godfat-真常[http://godfat.org])
   godfat (XD) godfat.org
 
@@ -24,9 +24,9 @@ by Lin Jen-Shin (a.k.a. godfat-真常[http://godfat.org])
 
 == REQUIREMENTS:
 
- * hpricot     >=0.6 or
- * nokogiri    >=1.1 or
- * libxml-ruby >=0.9
+ * hpricot     >=0.7 or
+ * nokogiri    >=1.2 or
+ * libxml-ruby >=1.1
 
 == INSTALL:
 

data/Rakefile

@@ -1,47 +1,26 @@
 # encoding: utf-8
 
-begin
-  require 'bones'
-  Bones.setup
-rescue LoadError
-  load 'tasks/setup.rb'   # this line should already be there
-end
+require 'bones'
+Bones.setup
 
 PROJ.name = 'friendly_format'
-# supress warnings, there's too many warnings in dm-core
-# PROJ.ruby_opts.delete '-w'
+PROJ.authors = 'Lin Jen-Shin (a.k.a. godfat 真常)'
+PROJ.email = 'godfat (XD) godfat.org'
+PROJ.url = "http://github.com/godfat/#{PROJ.name}"
+PROJ.rubyforge.name = 'ludy'
 
-# PROJ.gem.dependencies << ['hpricot', '>=0.6.0']
-PROJ.gem.development_dependencies << ['minitest',    '>=1.3'] <<
-                                     ['hpricot',     '>=0.6'] <<
+PROJ.gem.development_dependencies << ['hpricot',     '>=0.6'] <<
                                      ['nokogiri',    '>=1.1'] <<
                                      ['libxml-ruby', '>=0.9']
-# PROJ.gem.executables = ["bin/#{PROJ.name}"]
-
-task :default do
-  Rake.application.options.show_task_pattern = /./
-  Rake.application.display_tasks_and_comments
-end
 
-namespace :gem do
-  desc "create #{PROJ.name}.gemspec"
-  task 'gemspec' do
-    puts "rake gem:debug > #{PROJ.name}.gemspec"
-    File.open("#{PROJ.name}.gemspec", 'w'){|spec| spec << `rake gem:debug`.sub(/.*/, '')}
-  end
-end
+# PROJ.ruby_opts.delete '-w'
 
-PROJ.authors = 'Lin Jen-Shin (a.k.a. godfat 真常)'
-PROJ.email = 'godfat (XD) godfat.org'
-PROJ.url = "http://github.com/godfat/#{PROJ.name}"
 PROJ.description = PROJ.summary = paragraphs_of('README', 'description').join("\n\n")
 PROJ.changes = paragraphs_of('CHANGES', 0..1).join("\n\n")
-PROJ.rubyforge.name = 'ludy'
 PROJ.version = File.read("lib/#{PROJ.name}/version.rb").gsub(/.*VERSION = '(.*)'.*/m, '\1')
 
-PROJ.manifest_file = 'Manifest'
-PROJ.exclude += ['^Manifest$', '^tmp', 'tmp$', '^pkg',
-                 '^\.gitignore$', '^ann-', '\.sqlite3$', '\.db$']
+PROJ.exclude += ['^tmp', 'tmp$', '^pkg', '^\.gitignore$',
+                 '^ann-', '\.sqlite3$', '\.db$']
 
 PROJ.rdoc.remote_dir = PROJ.name
 
@@ -49,7 +28,7 @@ PROJ.readme_file = 'README'
 PROJ.rdoc.main = 'README'
 PROJ.rdoc.exclude += ['Rakefile', '^tasks', '^test']
 PROJ.rdoc.include << '\w+'
-PROJ.rdoc.opts << '--diagram' if !WIN32 and `which dot` =~ %r/\/dot/
+# PROJ.rdoc.opts << '--diagram' if !Rake::Win32 and `which dot` =~ %r/\/dot/
 PROJ.rdoc.opts += ['--charset=utf-8', '--inline-source',
                    '--line-numbers', '--promiscuous']
 
@@ -59,3 +38,8 @@ PROJ.ann.file = "ann-#{PROJ.name}-#{PROJ.version}"
 PROJ.ann.paragraphs.concat %w[LINKS SYNOPSIS REQUIREMENTS INSTALL LICENSE]
 
 CLEAN.include Dir['**/*.rbc']
+
+task :default do
+  Rake.application.options.show_task_pattern = /./
+  Rake.application.display_tasks_and_comments
+end

data/TODO

@@ -1,5 +1,4 @@
 = friendly_article todo list
-* add support for nokogiri as well (i wonder if rexml can fix html)
 
 * make format_article take an optional size argument that indicates
   how long should the result be. it should take < as 1 character and

data/friendly_format.gemspec

@@ -1,20 +1,19 @@
-
 # -*- encoding: utf-8 -*-
 
 Gem::Specification.new do |s|
   s.name = %q{friendly_format}
-  s.version = "0.6.0"
+  s.version = "0.6.1"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.authors = ["Lin Jen-Shin (a.k.a. godfat \347\234\237\345\270\270)"]
-  s.date = %q{2009-01-05}
+  s.date = %q{2009-04-05}
   s.description = %q{make user input be valid xhtml and format it with gsub("\n", "<br/>") etc. you can partially allow some tags and don't escape them.}
   s.email = %q{godfat (XD) godfat.org}
   s.extra_rdoc_files = ["CHANGES", "LICENSE", "NOTICE", "README", "TODO", "friendly_format.gemspec"]
-  s.files = ["CHANGES", "LICENSE", "NOTICE", "README", "Rakefile", "TODO", "friendly_format.gemspec", "lib/friendly_format.rb", "lib/friendly_format/adapters/hpricot_adapter.rb", "lib/friendly_format/adapters/libxml_adapter.rb", "lib/friendly_format/adapters/nokogiri_adapter.rb", "lib/friendly_format/set_common.rb", "lib/friendly_format/set_strict.rb", "lib/friendly_format/version.rb", "test/sample/complex_article.txt", "test/sample/complex_article_result.txt", "test/test_friendly_format.rb"]
+  s.files = ["CHANGES", "LICENSE", "NOTICE", "README", "Rakefile", "TODO", "friendly_format.gemspec", "lib/friendly_format.rb", "lib/friendly_format/adapter/abstract.rb", "lib/friendly_format/adapter/hpricot_adapter.rb", "lib/friendly_format/adapter/libxml_adapter.rb", "lib/friendly_format/adapter/nokogiri_adapter.rb", "lib/friendly_format/set_common.rb", "lib/friendly_format/set_strict.rb", "lib/friendly_format/version.rb", "test/sample/complex_article.txt", "test/sample/complex_article_result.txt", "test/test_friendly_format.rb"]
   s.has_rdoc = true
   s.homepage = %q{http://github.com/godfat/friendly_format}
-  s.rdoc_options = ["--diagram", "--charset=utf-8", "--inline-source", "--line-numbers", "--promiscuous", "--main", "README"]
+  s.rdoc_options = ["--charset=utf-8", "--inline-source", "--line-numbers", "--promiscuous", "--main", "README"]
   s.require_paths = ["lib"]
   s.rubyforge_project = %q{ludy}
   s.rubygems_version = %q{1.3.1}
@@ -26,21 +25,18 @@ Gem::Specification.new do |s|
     s.specification_version = 2
 
     if Gem::Version.new(Gem::RubyGemsVersion) >= Gem::Version.new('1.2.0') then
-      s.add_development_dependency(%q<bones>, [">= 2.2.0"])
-      s.add_development_dependency(%q<minitest>, [">= 1.3"])
+      s.add_development_dependency(%q<bones>, [">= 2.4.2"])
       s.add_development_dependency(%q<hpricot>, [">= 0.6"])
       s.add_development_dependency(%q<nokogiri>, [">= 1.1"])
       s.add_development_dependency(%q<libxml-ruby>, [">= 0.9"])
     else
-      s.add_dependency(%q<bones>, [">= 2.2.0"])
-      s.add_dependency(%q<minitest>, [">= 1.3"])
+      s.add_dependency(%q<bones>, [">= 2.4.2"])
       s.add_dependency(%q<hpricot>, [">= 0.6"])
       s.add_dependency(%q<nokogiri>, [">= 1.1"])
       s.add_dependency(%q<libxml-ruby>, [">= 0.9"])
     end
   else
-    s.add_dependency(%q<bones>, [">= 2.2.0"])
-    s.add_dependency(%q<minitest>, [">= 1.3"])
+    s.add_dependency(%q<bones>, [">= 2.4.2"])
     s.add_dependency(%q<hpricot>, [">= 0.6"])
     s.add_dependency(%q<nokogiri>, [">= 1.1"])
     s.add_dependency(%q<libxml-ruby>, [">= 0.9"])

data/lib/friendly_format/adapter/abstract.rb

@@ -0,0 +1,31 @@
+
+module FriendlyFormat
+  module Abstract
+
+    def method_name node
+      # discard body
+      node.children.to_s
+    end
+
+    def to_xhtml node
+      node.to_xhtml
+    end
+
+    def content node
+      node.to_s
+    end
+
+    def element? node
+      node.element?
+    end
+
+    def text? node
+      node.text?
+    end
+
+    def empty? node
+      node.children.empty?
+    end
+
+  end # of Abstract
+end # of FriendlyFormat

data/lib/friendly_format/{adapters → adapter}/hpricot_adapter.rb

@@ -9,10 +9,14 @@ module FriendlyFormat
         Hpricot.parse(html)
       end
 
-      def to_html node
+      def to_xhtml node
         node.to_html
       end
 
+      def content node
+        node.content
+      end
+
       def element? node
         node.kind_of?(Hpricot::Elem)
       end
@@ -25,18 +29,6 @@ module FriendlyFormat
         node.empty?
       end
 
-      def tag_name node
-        node.stag.name
-      end
-
-      def tag_begin node
-        node.stag.inspect
-      end
-
-      def tag_end node
-        (node.etag || Hpricot::ETag.new(node.stag.name)).inspect
-      end
-
     end # of class method for HpricotAdapter
   end # of HpricotAdapter
 end # of FriendlyFormat

data/lib/friendly_format/adapter/libxml_adapter.rb

@@ -0,0 +1,25 @@
+
+require 'friendly_format/adapter/abstract'
+require 'libxml'
+
+module FriendlyFormat
+  class LibxmlAdapter
+    extend Abstract
+
+    def self.parse html
+      parser = LibXML::XML::HTMLParser.string(
+        "<zzz>#{html}</zzz>",
+        :options => LibXML::XML::HTMLParser::Options::RECOVER)
+
+      # root is html, children is [body], first is body
+      # same as nokogiri
+      # drop zzz with .children.first since it would wrap a tag p for the article
+      parser.parse.root.children.first.children.first
+    end
+
+    def self.to_xhtml node
+      node.to_s
+    end
+
+  end # of LibxmlAdapter
+end # of FriendlyFormat

data/lib/friendly_format/adapter/nokogiri_adapter.rb

@@ -0,0 +1,21 @@
+
+require 'friendly_format/adapter/abstract'
+require 'nokogiri'
+
+module FriendlyFormat
+  class NokogiriAdapter
+    extend Abstract
+
+    def self.parse html
+      # root is html, children is [body], first is body
+      # same as libxml
+      # drop zzz with .children.first since it would wrap a tag p for the article
+      Nokogiri::HTML.parse(
+        "<zzz>#{html}</zzz>",
+        nil, # url?
+        html.respond_to?(:encoding) ? html.encoding.name : 'utf-8'
+        ).root.children.first.children.first
+    end
+
+  end # of NokogiriAdapter
+end # of FriendlyFormat

data/lib/friendly_format/set_common.rb

@@ -6,11 +6,11 @@ module FriendlyFormat
   # a few people have permission to post or edit.
   class SetCommon < Set
     def initialize
-      super([ :a, :area, :b, :big, :blockquote, :br,
-              :center, :cite, :code, :del, :div, :em,
-              :font, :h1, :h2, :h3, :h4, :h5, :h6, :hr,
-              :i, :img, :li, :map, :object, :ol, :p,
-              :pre, :small, :span, :strong, :u, :ul     ])
+      super(%w[ a area b big blockquote br
+                center cite code del div em
+                font h1 h2 h3 h4 h5 h6 hr
+                i img li map object ol p
+                pre small span strong u ul  ])
     end
   end
 end

data/lib/friendly_format/set_strict.rb

@@ -5,8 +5,8 @@ module FriendlyFormat
   # to post or edit articles.
   class SetStrict < Set
     def initialize
-      super([ :a, :b, :code, :del, :em, :font,
-              :i, :img, :li, :ol, :strong, :u, :ul ])
+      super(%w[ a b code del em font
+                i img li ol strong u ul ])
     end
   end
 end

data/lib/friendly_format/version.rb

@@ -1,3 +1,3 @@
 module FriendlyFormat
-  VERSION = '0.6.0'
+  VERSION = '0.6.1'
 end

data/lib/friendly_format.rb

@@ -5,14 +5,22 @@ require 'friendly_format/set_strict'
 
 # 2008-05-09 godfat
 module FriendlyFormat
-  autoload(:LibxmlAdapter,   'friendly_format/adapters/libxml_adapter')
-  autoload(:HpricotAdapter,  'friendly_format/adapters/hpricot_adapter')
-  autoload(:NokogiriAdapter, 'friendly_format/adapters/nokogiri_adapter')
+  autoload(:LibxmlAdapter,   'friendly_format/adapter/libxml_adapter')
+  autoload(:HpricotAdapter,  'friendly_format/adapter/hpricot_adapter')
+  autoload(:NokogiriAdapter, 'friendly_format/adapter/nokogiri_adapter')
 
   class << self
     attr_writer(:adapter)
     def adapter
-      @adapter ||= HpricotAdapter
+      @adapter ||= begin
+                     HpricotAdapter
+                   rescue LoadError
+                     begin
+                       NokogiriAdapter
+                     rescue LoadError
+                       LibxmlAdapter
+                     end
+                   end
     end
   end
 
@@ -22,15 +30,20 @@ module FriendlyFormat
   # default was no tags at all, all tags would be escaped.
   # it uses Hpricot to parse input.
   def format_article html, *args
-    FriendlyFormat.format_article_entrance(html,
-      args.inject(Set.new){ |allowed_tags, arg|
-        case arg
-          when Symbol; allowed_tags << arg
-          when Set;    allowed_tags += arg
-          else; raise(TypeError.new("expected Symbol or Set, got #{arg.class}"))
-        end
-        allowed_tags
-      })
+    return html if html.strip == ''
+
+    FriendlyFormat.force_encoding(
+      FriendlyFormat.format_article_entrance(html,
+        args.inject(Set.new){ |allowed_tags, arg|
+          case arg
+            when String; allowed_tags << arg
+            when Symbol; allowed_tags << arg.to_s
+            when Set;    allowed_tags += Set.new(arg.map{|a|a.to_s})
+            else; raise(TypeError.new("expected String|Symbol|Set, got #{arg.class}"))
+          end
+          allowed_tags
+        }),
+      html)
   end
 
   # automaticly add "a href" tag on text starts from
@@ -38,8 +51,12 @@ module FriendlyFormat
   # regexp translated from drupal to find where's the target.
   # it uses simplified regexp to do the task. see format_url.
   def format_autolink html, attrs = {}
-    FriendlyFormat.format_autolink_rec(
-      FriendlyFormat.adapter.parse(html), attrs)
+    return html if html.strip == ''
+
+    FriendlyFormat.force_encoding(
+      FriendlyFormat.format_autolink_rec(
+        FriendlyFormat.adapter.parse(html), attrs),
+      html)
   end
 
   # translated from drupal-6.2/modules/filter/filter.module
@@ -104,13 +121,12 @@ module FriendlyFormat
       text.gsub(
   %r{((http://|https://|ftp://|mailto:|smb://|afp://|file://|gopher://|news://|ssl://|sslv2://|sslv3://|tls://|tcp://|udp://|www\.)([a-zA-Z0-9@:%_+*~#?&=.,/;-]*[a-zA-Z0-9@:%_+*~#&=/;-]))([.,?!]*?)}i){ |match|
         url = $1 # is there any other way to get this variable?
-
         caption = trim(url)
-        attrs = attrs.map{ |k,v| " #{k}=\"#{v}\""}.join
+        html_attrs = attrs.map{ |k,v| " #{k}=\"#{v}\""}.join
 
         # Match www domains/addresses.
         url = "http://#{url}" unless url =~ %r{^http://}
-        "<a href=\"#{url}\" title=\"#{url}\"#{attrs}>#{caption}</a>"
+        "<a href=\"#{url}\" title=\"#{url}\"#{html_attrs}>#{caption}</a>"
       # Match e-mail addresses.
       }.gsub( %r{([A-Za-z0-9._-]+@[A-Za-z0-9._+-]+\.[A-Za-z]{2,4})([.,?!]*?)}i,
               '<a href="mailto:\1">\1</a>')
@@ -118,45 +134,46 @@ module FriendlyFormat
 
     # perhaps we should escape all inside code instead of pre?
     # @api private
-    def escape_all_inside_pre html, allowed_tags
-      return html unless allowed_tags.member? :pre
+    def escape_ltgt_inside_pre html, allowed_tags
+      return html unless allowed_tags.member?('pre')
       # don't bother nested pre, because we escape all tags in pre
       html = html + '</pre>' unless html =~ %r{</pre>}i
       html.gsub(%r{<pre>(.*)</pre>}mi){
         # stop escaping for '>' because drupal's url filter would make &gt; into url...
         # is there any other way to get matched group?
-        "<pre>#{escape_lt(escape_amp($1))}</pre>"
+        "<pre>#{escape_ltgt($1)}</pre>"
       }
     end
 
     # @api private
     def format_autolink_rec elem, attrs = {}
-      elem.children.map{ |c|
-        if adapter.text?(c)
-          format_url(c.content, attrs)
+      elem.children.map{ |e|
+        if adapter.text?(e)
+          format_url(e.content, attrs)
 
-        elsif adapter.element?(c)
-          if adapter.empty?(c)
-            c
+        elsif adapter.element?(e)
+          if adapter.empty?(e)
+            adapter.to_xhtml(e)
           else
-            adapter.tag_begin(c) +
-            format_autolink_rec(c, attrs) +
-            adapter.tag_end(c)
+            "<#{e.name}>" +
+            format_autolink_rec(e, attrs) +
+            "</#{e.name}>"
           end
 
         else
-          c
+          e
 
         end
 
-      }.to_s
+      }.join
     end
 
     # recursion entrance
     # @api private
     def format_article_entrance html, allowed_tags = Set.new
-      format_article_rec(adapter.parse(
-        escape_all_inside_pre(html, allowed_tags)), allowed_tags)
+      format_article_rec(
+        adapter.parse(escape_ltgt_inside_pre(html, allowed_tags)),
+        allowed_tags)
     end
 
     # recursion
@@ -165,38 +182,33 @@ module FriendlyFormat
       elem.children.map{ |e|
         if adapter.text?(e)
           if no_format_newline
-            format_url(e.content)
+            format_url(adapter.content(e))
           else
-            format_newline(format_url(e.content))
+            format_newline(format_url(adapter.content(e)))
           end
 
         elsif adapter.element?(e)
-          if allowed_tags.member?(e.name.to_sym)
+          if allowed_tags.member?(e.name)
             if adapter.empty?(e) || e.name == 'a'
-              e
+              adapter.to_xhtml(e)
             else
-              adapter.tag_begin(e) +
+              "<#{e.name}>" +
               format_article_rec(
-                e, allowed_tags, adapter.tag_name(e) == 'pre') +
-              adapter.tag_end(e)
+                e, allowed_tags, e.name == 'pre') +
+              "</#{e.name}>"
             end
           else
             if adapter.empty?(e)
-              escape_lt(adapter.tag_begin(e))
+              "&lt;#{e.name}&gt;"
             else
-              escape_lt(adapter.tag_begin(e)) +
+              "&lt;#{e.name}&gt;" +
               format_article_rec(e, allowed_tags) +
-              escape_lt(adapter.tag_end(e))
+              "&lt;/#{e.name}&gt;"
             end
           end
 
         end
-      }.to_s
-    end
-
-    # @api private
-    def escape_amp text
-      text.gsub('&', '&amp;')
+      }.join
     end
 
     # i cannot find a way to escape both lt and gt,
@@ -204,8 +216,18 @@ module FriendlyFormat
     # would treat complex lt and gt structure to be a tag
     # wraping content.
     # @api private
-    def escape_lt text
-      text.gsub('<', '&lt;')
+    def escape_ltgt text
+      text.gsub('<', '&lt;').gsub('>', '&gt;')
+    end
+
+    # force encoding for ruby 1.9
+    # @api private
+    def force_encoding output, input
+      if output.respond_to?(:force_encoding)
+        output.force_encoding(input.encoding)
+      else
+        output
+      end
     end
 
   end

data/test/sample/complex_article_result.txt

@@ -2,93 +2,93 @@
  標題  Re: [PSC] variants timer
  時間  2008/03/04 Tue 00:34:54
 <a href="http://phpbb.godfat.org/viewtopic.php?t=873" title="http://phpbb.godfat.org/viewtopic.php?t=873">http://phpbb.godfat.org/viewtopic.php?t=873</a>
-&lt;a href="<a href="http://phpbb.godfat.org/viewtopic.php?t=873" title="http://phpbb.godfat.org/viewtopic.php?t=873">http://phpbb.godfat.org/viewtopic.php?t=873</a>">
-測試&lt;/a>
-&lt;a href="<a href="http://phpbb.godfat.org/viewtopic.php?t=873" title="http://phpbb.godfat.org/viewtopic.php?t=873">http://phpbb.godfat.org/viewtopic.php?t=873</a>">
-&lt;img src="<a href="http://phpbb.godfat.org/templates/subSilver/images/logo_phpBB.gif" title="http://phpbb.godfat.org/templates/subSilver/images/logo_phpBB.gif">http://phpbb.godfat.org/templates/subSilver/images/logo_phpBB.gif</a>"/>&lt;/a>
+&lt;a href="<a href="http://phpbb.godfat.org/viewtopic.php?t=873" title="http://phpbb.godfat.org/viewtopic.php?t=873">http://phpbb.godfat.org/viewtopic.php?t=873</a>"&gt;
+測試&lt;/a&gt;
+&lt;a href="<a href="http://phpbb.godfat.org/viewtopic.php?t=873" title="http://phpbb.godfat.org/viewtopic.php?t=873">http://phpbb.godfat.org/viewtopic.php?t=873</a>"&gt;
+&lt;img src="<a href="http://phpbb.godfat.org/templates/subSilver/images/logo_phpBB.gif" title="http://phpbb.godfat.org/templates/subSilver/images/logo_phpBB.gif">http://phpbb.godfat.org/templates/subSilver/images/logo_phpBB.gif</a>"/&gt;&lt;/a&gt;
 ※ 引述《godfat (godfat 真常)》之銘言：
-> 3. tuple + traits, 也就是一堆 template 技巧了
-> class Dispatcher{
->     template &lt;class T>
->     void sub(shared_ptr&lt;T> p){
->         get_tuple_data_field_by_type&lt;T>(data).push_back(p);
->     }
->     tuple&lt;A, B, C> data;
-> };
-
-以上 A, B, C 是指 timer_subscriber_list&lt;A>, etc.
-
-這個東西會需要從 T => tuple index 的 mapping
+&gt; 3. tuple + traits, 也就是一堆 template 技巧了
+&gt; class Dispatcher{
+&gt;     template &lt;class T&gt;
+&gt;     void sub(shared_ptr&lt;T&gt; p){
+&gt;         get_tuple_data_field_by_type&lt;T&gt;(data).push_back(p);
+&gt;     }
+&gt;     tuple&lt;A, B, C&gt; data;
+&gt; };
+
+以上 A, B, C 是指 timer_subscriber_list&lt;A&gt;, etc.
+
+這個東西會需要從 T =&gt; tuple index 的 mapping
 現在這部份做完了...........（template 沒忘太多啊哈哈）
 
-#include &lt;tr1/tuple>
-#include &lt;tr1/type_traits>
-#include &lt;iostream>
+#include &lt;tr1/tuple&gt;
+#include &lt;tr1/type_traits&gt;
+#include &lt;iostream&gt;
 
 using namespace std::tr1;
 
 // 拿來判斷兩個整數是否相同，這是為了檢查有沒有爆表
-// 像是 tuple_element&lt;3, tuple&lt;int,int,int> >::type
+// 像是 tuple_element&lt;3, tuple&lt;int,int,int&gt; &gt;::type
 // 這樣就是爆表，因為這 tuple 沒這麼大
-template &lt;bool> struct is_equal_type       { typedef true_type type; };
-template &lt;>     struct is_equal_type&lt;false>{ typedef false_type type; };
+template &lt;bool&gt; struct is_equal_type       { typedef true_type type; };
+template &lt;&gt;     struct is_equal_type&lt;false&gt;{ typedef false_type type; };
 
-// 定義相等的話，is_equal&lt;1,1>::value 會是 1
-//               is_equal&lt;1,2>::value 會是 0
+// 定義相等的話，is_equal&lt;1,1&gt;::value 會是 1
+//               is_equal&lt;1,2&gt;::value 會是 0
 //                            ::type 會是 true_type/false_type (from tr1)
-template &lt;int Lhs, int Rhs> struct is_equal{
+template &lt;int Lhs, int Rhs&gt; struct is_equal{
     enum{ value = Lhs == Rhs };
-    typedef typename is_equal_type&lt;value>::type type;
+    typedef typename is_equal_type&lt;value&gt;::type type;
 };
 
-template &lt;class T, class Tuple, int Index> struct type_index_imp;
+template &lt;class T, class Tuple, int Index&gt; struct type_index_imp;
 
 // 如果跑到這裡，表示爆表了，在我的 g++4.3 上會顯示：
 // tuple.cpp:20: error: ‘type_not_found’ is not a member of "T"
-template &lt;class T, class Tuple, int Index, class Bool>
+template &lt;class T, class Tuple, int Index, class Bool&gt;
 struct type_index_check{
     enum{ value = T::type_not_found };
 };
 
 // 沒有爆表就繼續找吧
-template &lt;class T, class Tuple, int Index>
-struct type_index_check&lt;T, Tuple, Index, false_type>{
-        enum{ value = type_index_imp&lt;T, Tuple, Index+1>::value };
+template &lt;class T, class Tuple, int Index&gt;
+struct type_index_check&lt;T, Tuple, Index, false_type&gt;{
+        enum{ value = type_index_imp&lt;T, Tuple, Index+1&gt;::value };
 };
 
 // 找到了！
-template &lt;class T, class Tuple, int Index, class Bool>
+template &lt;class T, class Tuple, int Index, class Bool&gt;
 struct type_index_if{
     enum{ value = Index }; // found, stop recursion
 };
 
 // 不直接呼叫 imp, 呼叫 check 檢查是否爆表
-template &lt;class T, class Tuple, int Index>
-struct type_index_if&lt;T, Tuple, Index, false_type>{
+template &lt;class T, class Tuple, int Index&gt;
+struct type_index_if&lt;T, Tuple, Index, false_type&gt;{
     enum{ value = type_index_check&lt;T, Tuple, Index,
-            typename is_equal&lt;Index+1, tuple_size&lt;Tuple>::value>::type>::value };
+            typename is_equal&lt;Index+1, tuple_size&lt;Tuple&gt;::value&gt;::type&gt;::value };
 };
 
 // 檢查兩個型別是否相同
-template &lt;class T, class Tuple, int Index>
+template &lt;class T, class Tuple, int Index&gt;
 struct type_index_imp{
     enum{ value = type_index_if&lt;T, Tuple, Index,
-            typename is_same&lt;T, typename tuple_element&lt;Index, Tuple>::type>::type>::value };
+            typename is_same&lt;T, typename tuple_element&lt;Index, Tuple&gt;::type&gt;::type&gt;::value };
 };
 
 // simple wrapper
-template &lt;class T, class Tuple>
+template &lt;class T, class Tuple&gt;
 struct type_index{
-    enum{ value = type_index_imp&lt;T, Tuple, 0>::value };
+    enum{ value = type_index_imp&lt;T, Tuple, 0&gt;::value };
 };
 
 int main(){
-    typedef tuple&lt;int, char, void> test_t;
-    std::cout &lt;&lt; type_index&lt;int, test_t>::value &lt;&lt; std::endl;
-    std::cout &lt;&lt; type_index&lt;char, test_t>::value &lt;&lt; std::endl;
-    std::cout &lt;&lt; type_index&lt;void, test_t>::value &lt;&lt; std::endl;
-    std::cout &lt;&lt; tuple_size&lt;test_t>::value &lt;&lt; std::endl;
-    // std::cout &lt;&lt; type_index&lt;int*, test_t>::value &lt;&lt; std::endl;
+    typedef tuple&lt;int, char, void&gt; test_t;
+    std::cout &lt;&lt; type_index&lt;int, test_t&gt;::value &lt;&lt; std::endl;
+    std::cout &lt;&lt; type_index&lt;char, test_t&gt;::value &lt;&lt; std::endl;
+    std::cout &lt;&lt; type_index&lt;void, test_t&gt;::value &lt;&lt; std::endl;
+    std::cout &lt;&lt; tuple_size&lt;test_t&gt;::value &lt;&lt; std::endl;
+    // std::cout &lt;&lt; type_index&lt;int*, test_t&gt;::value &lt;&lt; std::endl;
 }
 
 輸出：

data/test/test_friendly_format.rb

@@ -1,15 +1,19 @@
 # encoding: utf-8
 
 require 'rubygems'
-require 'minitest/unit'
-MiniTest::Unit.autorun
 
-require 'friendly_format'
+TestCase = begin
+             require 'minitest/unit'
+             MiniTest::Unit.autorun
+             MiniTest::Unit::TestCase
+           rescue LoadError
+             require 'test/unit'
+             Test::Unit::TestCase
+           end
 
-# FriendlyFormat.adapter = FriendlyFormat::NokogiriAdapter
+require 'friendly_format'
 
-# 2008-05-09 godfat
-class TestFriendlyFormat < MiniTest::Unit::TestCase
+module TestCases
   include FriendlyFormat
 
   def test_article
@@ -39,7 +43,9 @@ OpenGL, 還有一些 web cgi 之類的東西也有。
 
 現在切入正是時機啊... XD'
 
-    assert_equal \
+    s = format_autolink(str)
+
+    assert(
 ' <a href="http://friends.roodo.com/forum/viewTopic/10170" title="http://friends.roodo.com/forum/viewTopic/10170">http://friends.roodo.com/forum/viewTopic/10170</a>
 用 Haskell 寫成的名軟體？
 
@@ -63,17 +69,19 @@ svn 其實也差不多是要慢慢式微了...
 很多知名 library 都有 haskell binding 了，wxWidgets,
 OpenGL, 還有一些 web cgi 之類的東西也有。
 
-現在切入正是時機啊... XD', s = format_autolink(str)
-    assert_equal s, format_autolink_regexp(str)
+現在切入正是時機啊... XD' == s)
+    assert_equal(s, format_autolink_regexp(str))
   end
   def test_persent
     str =
 'XDDDD
 http://www.amazon.co.jp/%E3%80%8C%E7%84%94~%E3%83%9B%E3%83%A0%E3%83%A9%E3%80%8D~Ar-tonelico2-hymmnos-concert-Side-%E7%B4%85~/dp/B000VKZL30/ref=pd_sbs_sw_img_2 orz'
 
-    assert_equal \
+    s = format_autolink(str)
+
+    assert_equal(
 'XDDDD
-<a href="http://www.amazon.co.jp/%E3%80%8C%E7%84%94~%E3%83%9B%E3%83%A0%E3%83%A9%E3%80%8D~Ar-tonelico2-hymmnos-concert-Side-%E7%B4%85~/dp/B000VKZL30/ref=pd_sbs_sw_img_2" title="http://www.amazon.co.jp/%E3%80%8C%E7%84%94~%E3%83%9B%E3%83%A0%E3%83%A9%E3%80%8D~Ar-tonelico2-hymmnos-concert-Side-%E7%B4%85~/dp/B000VKZL30/ref=pd_sbs_sw_img_2">http://www.amazon.co.jp/%E3%80%8C%E7%84%94~%E3%83%9B%E3%83%A0%E3%83%A9%E...</a> orz', s = format_autolink(str)
+<a href="http://www.amazon.co.jp/%E3%80%8C%E7%84%94~%E3%83%9B%E3%83%A0%E3%83%A9%E3%80%8D~Ar-tonelico2-hymmnos-concert-Side-%E7%B4%85~/dp/B000VKZL30/ref=pd_sbs_sw_img_2" title="http://www.amazon.co.jp/%E3%80%8C%E7%84%94~%E3%83%9B%E3%83%A0%E3%83%A9%E3%80%8D~Ar-tonelico2-hymmnos-concert-Side-%E7%B4%85~/dp/B000VKZL30/ref=pd_sbs_sw_img_2">http://www.amazon.co.jp/%E3%80%8C%E7%84%94~%E3%83%9B%E3%83%A0%E3%83%A9%E...</a> orz', s)
     assert_equal s, format_autolink_regexp(str)
   end
   def test_img_src
@@ -83,9 +91,21 @@ http://www.amazon.co.jp/%E3%80%8C%E7%84%94~%E3%83%9B%E3%83%A0%E3%83%A9%E3%80%8D~
 
 2007年12月14日
 '
-    assert_equal str, s = format_autolink(str)
-    assert_equal s, format_autolink_regexp(str)
+    libxml =
+'Thirst for Knowledge
+<img src="http://friends.roodo.com/images/diary_photos_large/15386/MjMyNjYtdGhpcnN0X2Zvcl9rbm93bGVkZ2U=.jpg"/>
+
+2007年12月14日
+'
+    s = format_autolink(str)
+    if FriendlyFormat.adapter == FriendlyFormat::LibxmlAdapter
+      assert_equal(libxml, s)
+    else
+      assert_equal(str, s)
+    end
+    assert_equal(str, format_autolink_regexp(str))
   end
+
   def test_wikipedia_persent
     str = 'http://en.wikipedia.org/wiki/Haskell_%28programming_language%29'
     assert_equal \
@@ -113,14 +133,20 @@ http://www.amazon.co.jp/%E3%80%8C%E7%84%94~%E3%83%9B%E3%83%A0%E3%83%A9%E3%80%8D~
   def test_escape_html_and_correct_html
     str = 'test<p>if missing end of p'
     assert_equal 'test<p>if missing end of p</p>', format_article(str, :p)
-    assert_equal 'test&lt;p>if missing end of p&lt;/p>', format_article(str, :a)
+    assert_equal 'test&lt;p&gt;if missing end of p&lt;/p&gt;', format_article(str, :a)
+
     str = '<pre>asdasd<a>orz'
-    assert_equal '<pre>asdasd&lt;a>orz</pre>', format_article(str, :a, :pre)
-    assert_equal '&lt;pre>asdasd<a>orz</a>&lt;/pre>', format_article(str, :a)
-    assert_equal '<pre>asdasd&lt;a>orz</pre>', format_article(str, :pre)
+    assert_equal '<pre>asdasd&lt;a&gt;orz</pre>', format_article(str, :a, :pre)
+    assert_equal '&lt;pre&gt;asdasd<a>orz</a>&lt;/pre&gt;', format_article(str, :a)
+    assert_equal '<pre>asdasd&lt;a&gt;orz</pre>', format_article(str, :pre)
+
     str = 'orz<img>asd'
-    assert_equal 'orz<img />asd', format_article(str, :img)
-    assert_equal 'orz&lt;img>asd', format_article(str)
+    if FriendlyFormat.adapter == FriendlyFormat::LibxmlAdapter
+      assert_equal('orz<img/>asd', format_article(str, :img))
+    else
+      assert_equal('orz<img />asd', format_article(str, :img))
+    end
+    assert_equal 'orz&lt;img&gt;asd', format_article(str)
   end
   def test_trim_url
     str = 'test with http://890123456789012345678901234567890123456789012345678901234567890123456789.com'
@@ -129,8 +155,16 @@ http://www.amazon.co.jp/%E3%80%8C%E7%84%94~%E3%83%9B%E3%83%A0%E3%83%A9%E3%80%8D~
     assert_equal s, format_autolink_regexp(str)
   end
   def test_escape_html
-    str = 'a lambda expression is &lambda; x. x+1'
-    assert_equal str, format_article(str)
+       str = 'a lambda expression is &lambda; x. x+1'
+    libxml = 'a lambda expression is λ x. x+1'
+    result = format_article(str)
+
+    if FriendlyFormat.adapter == FriendlyFormat::HpricotAdapter
+      assert_equal(str, result)
+    else
+      assert_equal(libxml, result)
+    end
+
     str = 'as you can see, use &lt;img src="asd"/&gt; to use'
     assert_equal str, format_article(str)
   end
@@ -138,8 +172,8 @@ http://www.amazon.co.jp/%E3%80%8C%E7%84%94~%E3%83%9B%E3%83%A0%E3%83%A9%E3%80%8D~
     result = File.read('test/sample/complex_article_result.txt').chop
     input  = File.read('test/sample/complex_article.txt')
 
-    assert_equal result, format_article(input, :pre)
-    assert_equal result, format_article(input, SetCommon.new)
+    assert_equal(result, format_article(input, :pre))
+    assert_equal(result, format_article(input, SetCommon.new))
   end
   def test_simple_link
     s = '今天是我一歲生日 <a href="http://godfat.org/" title="http://godfat.org/">http://godfat.org/</a> 真的嗎？'
@@ -162,9 +196,39 @@ compilation mode. 非常驚人的開發速度。
 此外，其中一位開發者，<a href="http://blog.headius.com/">Charles Nutter</a> 也經常參與 <a href="http://www.ruby-forum.com/forum/14">ruby-core</a> 的討論，
 對於 Ruby 的開發頗有貢獻。'
 
-    expected = '<img src="http://flolac.iis.sinica.edu.tw/lambdawan/sites/default/files/ruby.png.thumb.jpg" style="float: right;" /><br /><a href="http://www.ruby-forum.com/topic/169911">JRuby 1.1.5 Released</a><br /><a href="http://jruby.codehaus.org/">JRuby</a> 是用 Java 寫成的 Ruby interpreter/compiler.<br />原本 JRuby 只是普通的 open source project, 後來因為 <a href="http://www.sun.com/">Sun Microsystem</a>,<br />也就是 Java 的開發公司，看好 JRuby, 於是僱用 JRuby team,<br />full time 開發 JRuby. 後來 JRuby 在各方面都快速大幅成長，<br />尤其效能有了不可思議的大幅提昇，可能是 Sun 有一些撇步沒有公開吧。<br /><br />效能大幅提昇之後，JRuby 開發沒有停緩，接下來是非常大量的相容性提昇。<br />也從原本僅支援 interpret mode 到後來也支援 just in time 與 ahead of time 的<br />compilation mode. 非常驚人的開發速度。<br /><zzz>&lt;xd><br />此外，其中一位開發者，<a href="http://blog.headius.com/">Charles Nutter</a> 也經常參與 <a href="http://www.ruby-forum.com/forum/14">ruby-core</a> 的討論，<br />對於 Ruby 的開發頗有貢獻。&lt;/xd></zzz>'
+    expected = '<img style="float: right;" src="http://flolac.iis.sinica.edu.tw/lambdawan/sites/default/files/ruby.png.thumb.jpg" /><br /><a href="http://www.ruby-forum.com/topic/169911">JRuby 1.1.5 Released</a><br /><a href="http://jruby.codehaus.org/">JRuby</a> 是用 Java 寫成的 Ruby interpreter/compiler.<br />原本 JRuby 只是普通的 open source project, 後來因為 <a href="http://www.sun.com/">Sun Microsystem</a>,<br />也就是 Java 的開發公司，看好 JRuby, 於是僱用 JRuby team,<br />full time 開發 JRuby. 後來 JRuby 在各方面都快速大幅成長，<br />尤其效能有了不可思議的大幅提昇，可能是 Sun 有一些撇步沒有公開吧。<br /><br />效能大幅提昇之後，JRuby 開發沒有停緩，接下來是非常大量的相容性提昇。<br />也從原本僅支援 interpret mode 到後來也支援 just in time 與 ahead of time 的<br />compilation mode. 非常驚人的開發速度。<br /><zzz>&lt;xd&gt;<br />此外，其中一位開發者，<a href="http://blog.headius.com/">Charles Nutter</a> 也經常參與 <a href="http://www.ruby-forum.com/forum/14">ruby-core</a> 的討論，<br />對於 Ruby 的開發頗有貢獻。&lt;/xd&gt;</zzz>'
 
-    assert_equal expected, format_article(input, SetCommon.new, :zzz)
+    # img style after img src
+    expected_18_hpricot = '<img src="http://flolac.iis.sinica.edu.tw/lambdawan/sites/default/files/ruby.png.thumb.jpg" style="float: right;" /><br /><a href="http://www.ruby-forum.com/topic/169911">JRuby 1.1.5 Released</a><br /><a href="http://jruby.codehaus.org/">JRuby</a> 是用 Java 寫成的 Ruby interpreter/compiler.<br />原本 JRuby 只是普通的 open source project, 後來因為 <a href="http://www.sun.com/">Sun Microsystem</a>,<br />也就是 Java 的開發公司，看好 JRuby, 於是僱用 JRuby team,<br />full time 開發 JRuby. 後來 JRuby 在各方面都快速大幅成長，<br />尤其效能有了不可思議的大幅提昇，可能是 Sun 有一些撇步沒有公開吧。<br /><br />效能大幅提昇之後，JRuby 開發沒有停緩，接下來是非常大量的相容性提昇。<br />也從原本僅支援 interpret mode 到後來也支援 just in time 與 ahead of time 的<br />compilation mode. 非常驚人的開發速度。<br /><zzz>&lt;xd&gt;<br />此外，其中一位開發者，<a href="http://blog.headius.com/">Charles Nutter</a> 也經常參與 <a href="http://www.ruby-forum.com/forum/14">ruby-core</a> 的討論，<br />對於 Ruby 的開發頗有貢獻。&lt;/xd&gt;</zzz>'
 
+    result = format_article(input, SetCommon.new, :zzz)
+    if RUBY_VERSION =~ /^1\.8/ && FriendlyFormat.adapter == FriendlyFormat::HpricotAdapter
+      assert(result == expected_18_hpricot)
+    else
+      assert(result == expected)
+    end
+  end
+
+  def test_empty_article
+    assert_equal '', format_article('')
+    assert_equal ' ', format_article(' ')
+    assert_equal 'a', format_article('a')
+  end
+
+  def test_custom_set
+    s = '<z>zzz</z>'
+    assert_equal s, format_article(s, Set.new << :z)
   end
 end
+
+%w[HpricotAdapter NokogiriAdapter LibxmlAdapter].each{ |adapter|
+  eval <<-RUBY
+    class Test#{adapter} < TestCase
+      include TestCases
+
+      def setup
+        FriendlyFormat.adapter = FriendlyFormat::#{adapter}
+      end
+    end
+  RUBY
+}

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: godfat-friendly_format
 version: !ruby/object:Gem::Version 
-  version: 0.6.0
+  version: 0.6.1
 platform: ruby
 authors: 
 - "Lin Jen-Shin (a.k.a. godfat \xE7\x9C\x9F\xE5\xB8\xB8)"
@@ -9,29 +9,22 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2009-01-05 00:00:00 -08:00
+date: 2009-04-05 00:00:00 -07:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
   name: bones
+  type: :development
   version_requirement: 
   version_requirements: !ruby/object:Gem::Requirement 
     requirements: 
     - - ">="
       - !ruby/object:Gem::Version 
-        version: 2.2.0
-    version: 
-- !ruby/object:Gem::Dependency 
-  name: minitest
-  version_requirement: 
-  version_requirements: !ruby/object:Gem::Requirement 
-    requirements: 
-    - - ">="
-      - !ruby/object:Gem::Version 
-        version: "1.3"
+        version: 2.4.2
     version: 
 - !ruby/object:Gem::Dependency 
   name: hpricot
+  type: :development
   version_requirement: 
   version_requirements: !ruby/object:Gem::Requirement 
     requirements: 
@@ -41,6 +34,7 @@ dependencies:
     version: 
 - !ruby/object:Gem::Dependency 
   name: nokogiri
+  type: :development
   version_requirement: 
   version_requirements: !ruby/object:Gem::Requirement 
     requirements: 
@@ -50,6 +44,7 @@ dependencies:
     version: 
 - !ruby/object:Gem::Dependency 
   name: libxml-ruby
+  type: :development
   version_requirement: 
   version_requirements: !ruby/object:Gem::Requirement 
     requirements: 
@@ -79,9 +74,10 @@ files:
 - TODO
 - friendly_format.gemspec
 - lib/friendly_format.rb
-- lib/friendly_format/adapters/hpricot_adapter.rb
-- lib/friendly_format/adapters/libxml_adapter.rb
-- lib/friendly_format/adapters/nokogiri_adapter.rb
+- lib/friendly_format/adapter/abstract.rb
+- lib/friendly_format/adapter/hpricot_adapter.rb
+- lib/friendly_format/adapter/libxml_adapter.rb
+- lib/friendly_format/adapter/nokogiri_adapter.rb
 - lib/friendly_format/set_common.rb
 - lib/friendly_format/set_strict.rb
 - lib/friendly_format/version.rb
@@ -92,7 +88,6 @@ has_rdoc: true
 homepage: http://github.com/godfat/friendly_format
 post_install_message: 
 rdoc_options: 
-- --diagram
 - --charset=utf-8
 - --inline-source
 - --line-numbers

data/lib/friendly_format/adapters/libxml_adapter.rb

@@ -1,46 +0,0 @@
-
-module FriendlyFormat
-  class LibxmlAdapter
-    class << self
-
-      def parse html
-        require 'libxml'
-        parser = LibXML::XML::HTMLParser.new
-        parser.string = html
-        # root is html, children is [body], first is body
-        # same as nokogiri
-        parser.parse.root.children.first
-      end
-
-      def to_html node
-        # discard body
-        node.children.to_s
-      end
-
-      def element? node
-        node.element?
-      end
-
-      def text? node
-        node.text?
-      end
-
-      def empty? node
-        node.children.empty?
-      end
-
-      def tag_name node
-        node.name
-      end
-
-      def tag_begin node
-        "<#{node.name}#{node.attributes.to_h}>"
-      end
-
-      def tag_end node
-        empty?(node) ? "<#{node.name}/>" : "</#{node.name}>"
-      end
-
-    end # of class method for LibxmlAdapter
-  end # of LibxmlAdapter
-end # of FriendlyFormat

data/lib/friendly_format/adapters/nokogiri_adapter.rb

@@ -1,31 +0,0 @@
-
-require 'friendly_format/adapters/libxml_adapter'
-require 'nokogiri'
-
-module FriendlyFormat
-  class NokogiriAdapter < LibxmlAdapter
-    class << self
-
-      def parse html
-        # root is html, children is [body], first is body
-        # same as libxml
-        Nokogiri::HTML.parse(html).root.children.first
-      end
-
-      def element? node
-        node.kind_of?(Nokogiri::XML::Element)
-      end
-
-      def text? node
-        node.kind_of?(Nokogiri::XML::Text)
-      end
-
-      def tag_begin node
-        attrs = node.attributes.map{ |key_value| key_value.join('="') + '"' }.join(' ')
-        attrs = ' ' + attrs if attrs != ''
-        "<#{node.name}#{attrs}>"
-      end
-
-    end # of class method for NokogiriAdapter
-  end # of NokogiriAdapter
-end # of FriendlyFormat