html-pipeline 0.0.8 → 0.0.10

data/.gitignore

@@ -15,5 +15,5 @@ spec/reports
 test/tmp
 test/version_tmp
 tmp
-bin/*
+exec/*
 vendor/gems

data/CHANGELOG.md

@@ -1,5 +1,15 @@
 # CHANGELOG
 
+## 0.0.10
+
+  * add bin/html-pipeline util indirect #44
+  * add result[:mentioned_usernames] for MentionFilter fachen #42
+
+## 0.0.9
+
+  * bump escape_utils ~> 0.3, github-linguist ~> 2.6.2 brianmario #41
+  * remove nokogiri monkey patch for ruby >= 1.9 defunkt #40
+
 ## 0.0.8
 
   * raise LoadError instead of printing to stderr if linguist is missing. gjtorikian #36

data/README.md

@@ -85,7 +85,8 @@ filter.call
 ## Filters
 
 * `MentionFilter` - replace `@user` mentions with links
-* `AutolinkFilter` - auto_linking urls in HTML
+* `AbsoluteSourceFilter` - replace relative image urls with fully qualified versions
+* `AutoLinkFilter` - auto_linking urls in HTML
 * `CamoFilter` - replace http image urls with [camo-fied](https://github.com/atmos/camo) https versions
 * `EmailReplyFilter` - util filter for working with emails
 * `EmojiFilter` - everyone loves [emoji](http://www.emoji-cheat-sheet.com/)!

data/bin/html-pipeline

@@ -0,0 +1,80 @@
+#!/usr/bin/env ruby
+require 'html/pipeline'
+
+require 'optparse'
+
+# Accept "help", too
+ARGV.map!{|a| a == "help" ? "--help" : a }
+
+OptionParser.new do |opts|
+  opts.banner = <<-HELP.gsub(/^    /, '')
+    Usage: html-pipeline [-h] [-f]
+           html-pipeline [FILTER [FILTER [...]]] < file.md
+           cat file.md | html-pipeline [FILTER [FILTER [...]]]
+  HELP
+
+  opts.separator "Options:"
+
+  opts.on("-f", "--filters", "List the available filters") do
+    filters = HTML::Pipeline.constants.grep(/\w+Filter$/).
+      map{|f| f.to_s.gsub(/Filter$/,'') }
+
+    # Text filter doesn't work, no call method
+    filters -= ["Text"]
+
+    abort <<-HELP.gsub(/^      /, '')
+      Available filters:
+        #{filters.join("\n        ")}
+    HELP
+  end
+end.parse!
+
+# Default to a GitHub-ish pipeline
+if ARGV.empty?
+
+  filters = [
+    HTML::Pipeline::MarkdownFilter,
+    HTML::Pipeline::SanitizationFilter,
+    HTML::Pipeline::ImageMaxWidthFilter,
+    HTML::Pipeline::EmojiFilter,
+    HTML::Pipeline::AutolinkFilter,
+    HTML::Pipeline::TableOfContentsFilter,
+  ]
+
+  # Add syntax highlighting if linguist is present
+  begin
+    require 'linguist'
+    filters << HTML::Pipeline::SyntaxHighlightFilter
+  rescue LoadError
+  end
+
+else
+
+  def filter_named(name)
+    case name
+    when "Text"
+      raise NameError # Text filter doesn't work, no call method
+    when "Textile"
+      require "RedCloth" # Textile filter doesn't require RedCloth
+    end
+
+    HTML::Pipeline.const_get("#{name}Filter")
+  rescue NameError => e
+    abort "Unknown filter '#{name}'. List filters with the -f option."
+  end
+
+  filters = []
+  until ARGV.empty?
+    name = ARGV.shift
+    filters << filter_named(name)
+  end
+
+end
+
+context = {
+  :asset_root => "/assets",
+  :base_url   => "/",
+  :gfm        => true
+}
+
+puts HTML::Pipeline.new(filters, context).call(ARGF.read)[:output]

data/html-pipeline.gemspec

@@ -20,8 +20,8 @@ Gem::Specification.new do |gem|
   gem.add_dependency "github-markdown", "~> 0.5"
   gem.add_dependency "sanitize",        "~> 2.0"
   gem.add_dependency "rinku",           "~> 1.7"
-  gem.add_dependency "escape_utils",    "~> 0.2"
+  gem.add_dependency "escape_utils",    "~> 0.3"
   gem.add_dependency "activesupport",   ">= 2"
 
-  gem.add_development_dependency "github-linguist", "~> 2.1"
+  gem.add_development_dependency "github-linguist", "~> 2.6.2"
 end

data/lib/html/pipeline.rb

@@ -27,6 +27,7 @@ module HTML
     autoload :VERSION,               'html/pipeline/version'
     autoload :Pipeline,              'html/pipeline/pipeline'
     autoload :Filter,                'html/pipeline/filter'
+    autoload :AbsoluteSourceFilter,  'html/pipeline/absolute_source_filter'
     autoload :BodyContent,           'html/pipeline/body_content'
     autoload :AutolinkFilter,        'html/pipeline/autolink_filter'
     autoload :CamoFilter,            'html/pipeline/camo_filter'
@@ -108,23 +109,25 @@ module HTML
   end
 end
 
-# XXX nokogiri monkey patches
-class Nokogiri::XML::Node
-  # Work around an issue with utf-8 encoded data being erroneously converted to
-  # ... some other shit when replacing text nodes. See 'utf-8 output 2' in
-  # user_content_test.rb for details.
-  def replace_with_encoding_fix(replacement)
-    if replacement.respond_to?(:to_str)
-      replacement = document.fragment("<div>#{replacement}</div>").children.first.children
+# XXX nokogiri monkey patches for 1.8
+if not ''.respond_to?(:force_encoding)
+  class Nokogiri::XML::Node
+    # Work around an issue with utf-8 encoded data being erroneously converted to
+    # ... some other shit when replacing text nodes. See 'utf-8 output 2' in
+    # user_content_test.rb for details.
+    def replace_with_encoding_fix(replacement)
+      if replacement.respond_to?(:to_str)
+        replacement = document.fragment("<div>#{replacement}</div>").children.first.children
+      end
+      replace_without_encoding_fix(replacement)
     end
-    replace_without_encoding_fix(replacement)
-  end
 
-  alias_method :replace_without_encoding_fix, :replace
-  alias_method :replace, :replace_with_encoding_fix
+    alias_method :replace_without_encoding_fix, :replace
+    alias_method :replace, :replace_with_encoding_fix
 
-  def swap(replacement)
-    replace(replacement)
-    self
+    def swap(replacement)
+      replace(replacement)
+      self
+    end
   end
 end

data/lib/html/pipeline/@mention_filter.rb

@@ -60,6 +60,8 @@ module HTML
       IGNORE_PARENTS = %w(pre code a).to_set
 
       def call
+        result[:mentioned_usernames] ||= []
+
         doc.search('text()').each do |node|
           content = node.to_html
           next if !content.include?('@')
@@ -108,6 +110,7 @@ module HTML
       end
 
       def link_to_mentioned_user(login)
+        result[:mentioned_usernames] |= [login]
         url = File.join(base_url, login)
         "<a href='#{url}' class='user-mention'>" +
         "@#{login}" +

data/lib/html/pipeline/absolute_source_filter.rb

@@ -0,0 +1,48 @@
+require 'uri'
+
+module HTML
+  class Pipeline
+
+    class AbsoluteSourceFilter < Filter
+      # HTML Filter for replacing relative and root relative image URLs with
+      # fully qualified URLs
+      #
+      # This is useful if an image is root relative but should really be going
+      # through a cdn, or if the content for the page assumes the host is known
+      # i.e. scraped webpages and some RSS feeds.
+      #
+      # Context options:
+      #   :image_base_url - Base URL for image host for root relative src.
+      #   :image_subpage_url - For relative src.
+      #
+      # This filter does not write additional information to the context.
+      # This filter would need to be run before CamoFilter.
+      def call
+        doc.search("img").each do |element| 
+          next if element['src'].nil? || element['src'].empty?
+          src = element['src'].strip
+          unless src.start_with? 'http'
+            if src.start_with? '/'
+              base = image_base_url
+            else
+              base = image_subpage_url
+            end
+            element["src"] = URI.join(base, src).to_s
+          end
+        end
+        doc
+      end
+      
+      # Private: the base url you want to use
+      def image_base_url
+        context[:image_base_url] or raise "Missing context :image_base_url for #{self.class.name}"
+      end
+
+      # Private: the relative url you want to use
+      def image_subpage_url
+        context[:image_subpage_url] or raise "Missing context :image_subpage_url for #{self.class.name}"
+      end
+    
+    end
+  end
+end

data/lib/html/pipeline/version.rb

@@ -1,5 +1,5 @@
 module HTML
   class Pipeline
-    VERSION = "0.0.8"
+    VERSION = "0.0.10"
   end
 end

data/test/html/pipeline/absolute_source_filter_test.rb

@@ -0,0 +1,56 @@
+require "test_helper"
+
+class HTML::Pipeline::AbsoluteSourceFilterTest < Test::Unit::TestCase
+  AbsoluteSourceFilter = HTML::Pipeline::AbsoluteSourceFilter
+
+  def setup
+    @image_base_url = 'http://assets.example.com'
+    @image_subpage_url = 'http://blog.example.com/a/post'
+    @options = {
+      :image_base_url    => @image_base_url,
+      :image_subpage_url => @image_subpage_url
+    }
+  end
+
+  def test_rewrites_root_relative_urls
+    orig = %(<p><img src="/img.png"></p>)
+    puts AbsoluteSourceFilter.call(orig, @options).to_s
+    assert_equal "<p><img src=\"#{@image_base_url}/img.png\"></p>",
+      AbsoluteSourceFilter.call(orig, @options).to_s
+  end
+
+  def test_rewrites_root_relative_urls
+    orig = %(<p><img src="post/img.png"></p>)
+    assert_equal "<p><img src=\"#{@image_subpage_url}/img.png\"></p>",
+      AbsoluteSourceFilter.call(orig, @options).to_s
+  end
+
+  def test_does_not_rewrite_absolute_urls
+    orig = %(<p><img src="http://other.example.com/img.png"></p>)
+    result = AbsoluteSourceFilter.call(orig, @options).to_s
+    assert_no_match /@image_base_url/, result
+    assert_no_match /@image_subpage_url/, result
+  end
+
+  def test_fails_when_context_is_missing
+    assert_raise RuntimeError do
+      AbsoluteSourceFilter.call("<img src=\"img.png\">", {})
+    end
+    assert_raise RuntimeError do
+      AbsoluteSourceFilter.call("<img src=\"/img.png\">", {})
+    end
+  end
+  
+  def test_tells_you_where_context_is_required
+    exception = assert_raise(RuntimeError) { 
+      AbsoluteSourceFilter.call("<img src=\"img.png\">", {}) 
+    }
+    assert_match 'HTML::Pipeline::AbsoluteSourceFilter', exception.message
+
+    exception = assert_raise(RuntimeError) { 
+      AbsoluteSourceFilter.call("<img src=\"/img.png\">", {}) 
+    }
+    assert_match 'HTML::Pipeline::AbsoluteSourceFilter', exception.message
+  end
+
+end

data/test/html/pipeline/mention_filter_test.rb

@@ -76,9 +76,7 @@ class HTML::Pipeline::MentionFilterTest < Test::Unit::TestCase
   def mentioned_usernames
     result = {}
     MarkdownPipeline.call(@body, {}, result)
-    html = result[:output].to_html
-    users = html.scan(/user-mention">@(.+?)</)
-    users ? users.flatten.uniq : []
+    result[:mentioned_usernames]
   end
 
   def test_matches_usernames_in_body

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: html-pipeline
 version: !ruby/object:Gem::Version
-  version: 0.0.8
+  version: 0.0.10
   prerelease: 
 platform: ruby
 authors:
@@ -10,7 +10,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-02-07 00:00:00.000000000 Z
+date: 2013-03-21 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: gemoji
@@ -99,7 +99,7 @@ dependencies:
     requirements:
     - - ~>
       - !ruby/object:Gem::Version
-        version: '0.2'
+        version: '0.3'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
@@ -107,7 +107,7 @@ dependencies:
     requirements:
     - - ~>
       - !ruby/object:Gem::Version
-        version: '0.2'
+        version: '0.3'
 - !ruby/object:Gem::Dependency
   name: activesupport
   requirement: !ruby/object:Gem::Requirement
@@ -131,7 +131,7 @@ dependencies:
     requirements:
     - - ~>
       - !ruby/object:Gem::Version
-        version: '2.1'
+        version: 2.6.2
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
@@ -139,7 +139,7 @@ dependencies:
     requirements:
     - - ~>
       - !ruby/object:Gem::Version
-        version: '2.1'
+        version: 2.6.2
 description: GitHub HTML processing filters and utilities
 email:
 - ryan@github.com
@@ -155,9 +155,11 @@ files:
 - LICENSE
 - README.md
 - Rakefile
+- bin/html-pipeline
 - html-pipeline.gemspec
 - lib/html/pipeline.rb
 - lib/html/pipeline/@mention_filter.rb
+- lib/html/pipeline/absolute_source_filter.rb
 - lib/html/pipeline/autolink_filter.rb
 - lib/html/pipeline/body_content.rb
 - lib/html/pipeline/camo_filter.rb
@@ -174,6 +176,7 @@ files:
 - lib/html/pipeline/textile_filter.rb
 - lib/html/pipeline/toc_filter.rb
 - lib/html/pipeline/version.rb
+- test/html/pipeline/absolute_source_filter_test.rb
 - test/html/pipeline/autolink_filter_test.rb
 - test/html/pipeline/camo_filter_test.rb
 - test/html/pipeline/emoji_filter_test.rb
@@ -210,6 +213,7 @@ signing_key:
 specification_version: 3
 summary: Helpers for processing content through a chain of filters
 test_files:
+- test/html/pipeline/absolute_source_filter_test.rb
 - test/html/pipeline/autolink_filter_test.rb
 - test/html/pipeline/camo_filter_test.rb
 - test/html/pipeline/emoji_filter_test.rb