webmention 2.2.0 → 3.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: b7e2e981cfe2dea4eec4e8e38fb9d948e14dd9b50c52f7455ca3583fbf4969aa
-  data.tar.gz: 1f9f50af537f60c4fa705690d7553563c54a74ff09c4d653b2a1ab6940b71d47
+  metadata.gz: 39599574ee4962a000832e7a47c48fd78d58797181a12a554ca26065b2d860c1
+  data.tar.gz: aaf047a87adf3b2c171622725aabb9a5f5fc64267a98077c7f2eb8b8e5db0c55
 SHA512:
-  metadata.gz: 3a45c8cfe715e165a81927e37af2ad020562a6af4c33ad59d15bed148e8a219623ac459dafb70c6ceae4c2cbb56f7e62406790ff1db7a89e8f34400cb6e3ee18
-  data.tar.gz: db933998d46aac3dfdfd28202de070438062e8e7c2f3c18b03a160c97f242c6b926bf24fb81a4804f68360736b60c196825ecc2547fc11751527c5fdb0de0959
+  metadata.gz: db45f488f6e2f527b075764ad580817294ac8971248365270102ee244820c04a7f044612eebe6b7ebc52dbc1fe754b7b997e0da01f2a7b9994d4d51a77e92f82
+  data.tar.gz: 254fe4bda676161657ff674d75a0866f7711dca89d5f7b5f0f961178efbbd9915a3b89ac94497442906b73ee572ababc90fedf3909c85b8e39c2ed53b897689b

data/.rubocop.yml

@@ -6,6 +6,7 @@ AllCops:
 
 Layout/HashAlignment:
   EnforcedColonStyle: table
+  EnforcedHashRocketStyle: table
 
 Layout/LineLength:
   Enabled: false

data/CHANGELOG.md

@@ -1,5 +1,10 @@
 # Changelog
 
+## v3.0.0 / 2020-05-19
+
+- Reject "internal" URLs when sending webmentions (#24) (ccc82c8)
+- Select only HTTP/HTTPS URLs when sending webmentions (#22) (39e5852)
+
 ## 2.2.0 / 2020-05-18
 
 - Update absolutely and indieweb-endpoints gems (350d2ed)

data/lib/webmention.rb

@@ -14,16 +14,25 @@ require 'webmention/parsers/base_parser'
 require 'webmention/parsers/html_parser'
 
 require 'webmention/services/http_request_service'
-require 'webmention/services/node_parser_service'
 
 module Webmention
-  class << self
-    def client(source)
-      Client.new(source)
-    end
+  # Create a new Webmention::Client
+  # Convenience method for Webmention::Client.new
+  #
+  #   client = Webmention.client('https://source.example.com/post/100')
+  #
+  # @param source [String] An absolute URL representing the source document
+  # @return [Webmention::Client]
+  def self.client(source)
+    Client.new(source)
+  end
 
-    def send_mention(source, target)
-      client(source).send_mention(target)
-    end
+  # Send a webmention from the source URL to the target URL
+  #
+  # @param source [String] An absolute URL representing the source document
+  # @param target [String] An absolute URL representing the target document
+  # @return [HTTP::Response, nil]
+  def self.send_mention(source, target)
+    client(source).send_mention(target)
   end
 end

data/lib/webmention/client.rb

@@ -1,5 +1,10 @@
 module Webmention
   class Client
+    # Create a new Webmention::Client
+    #
+    #   client = Webmention::Client.new('https://source.example.com/post/100')
+    #
+    # @param source [String] An absolute URL representing the source document
     def initialize(source)
       raise ArgumentError, "source must be a String (given #{source.class.name})" unless source.is_a?(String)
 
@@ -8,16 +13,28 @@ module Webmention
       raise ArgumentError, 'source must be an absolute URL (e.g. https://example.com)' unless source_uri.absolute?
     end
 
+    # Send webmentions to all mentioned URLs in this client's source document
+    #
+    # @return [Hash{String => HTTP::Response, nil}]
     def send_all_mentions
       mentioned_urls.each_with_object({}) { |url, hash| hash[url] = send_mention(url) }
     end
 
+    # Extract mentioned URLs from this client's source document
+    #
+    # @return [Array<String>]
+    # @raise [Webmention::UnsupportedMimeTypeError]
     def mentioned_urls
       raise UnsupportedMimeTypeError, "Unsupported MIME Type: #{source_response.mime_type}" unless parser_for_mime_type
 
       @mentioned_urls ||= parser_for_mime_type.new(source_response).results
     end
 
+    # Send a webmention from this client's source URL to the target URL
+    #
+    # @param target [String] An absolute URL representing the target document
+    # @return [HTTP::Response, nil]
+    # @raise [Webmention::ArgumentError, Webmention::ConnectionError, Webmention::InvalidURIError, Webmention::TimeoutError, Webmention::TooManyRedirectsError]
     def send_mention(target)
       endpoint = IndieWeb::Endpoints.get(target).webmention
 

data/lib/webmention/parsers/base_parser.rb

@@ -13,10 +13,6 @@ module Webmention
         raise UnsupportedMimeTypeError, "Unsupported MIME Type: #{response.mime_type}" unless self.class.mime_types.include?(response.mime_type)
       end
 
-      def results
-        @results ||= parse_response_body
-      end
-
       private
 
       def response_body

data/lib/webmention/parsers/html_parser.rb

@@ -5,17 +5,22 @@ module Webmention
 
       Parsers.register(self)
 
-      HTML_ATTRIBUTE_MAP = {
-        cite:   %w[blockquote del ins q],
-        data:   %w[object],
-        href:   %w[a area],
-        poster: %w[video],
-        src:    %w[audio embed img source track video],
-        srcset: %w[img source]
+      HTML_ATTRIBUTES_MAP = {
+        'cite'   => %w[blockquote del ins q],
+        'data'   => %w[object],
+        'href'   => %w[a area],
+        'poster' => %w[video],
+        'src'    => %w[audio embed img source track video],
+        'srcset' => %w[img source]
       }.freeze
 
-      CSS_SELECTORS_MAP = HTML_ATTRIBUTE_MAP.each_with_object({}) do |(attribute, elements), hash|
-        hash[attribute] = elements.map { |element| "#{element}[#{attribute}]" }
+      CSS_SELECTORS_ARRAY = HTML_ATTRIBUTES_MAP.flat_map { |attribute, names| names.map { |name| "#{name}[#{attribute}]" } }.freeze
+
+      # Parse an HTML string for URLs
+      #
+      # @return [Array<String>] Unique external URLs whose scheme matches http/https
+      def results
+        @results ||= resolved_urls.uniq.select { |url| url.match?(%r{https?://}) }.reject { |url| url.match(/^#{response_url}(?:#.*)?$/) }
       end
 
       private
@@ -24,23 +29,36 @@ module Webmention
         @doc ||= Nokogiri::HTML(response_body)
       end
 
-      # Parse an HTML string for URLs
-      #
-      # @return [Array] the URLs
-      def parse_response_body
-        CSS_SELECTORS_MAP
-          .each_with_object([]) { |(*args), array| array << search_node(*args) }
-          .flatten
-          .map { |url| Absolutely.to_abs(base: response_url, relative: url) }
-          .uniq
+      def resolved_urls
+        UrlAttributesParser.parse(*url_attributes).map { |url| Absolutely.to_abs(base: response_url, relative: url) }
       end
 
       def root_node
-        @root_node ||= doc.at_css('.h-entry .e-content') || doc.at_css('.h-entry') || doc.css('body')
+        doc.at_css('.h-entry .e-content', '.h-entry') || doc.css('body')
       end
 
-      def search_node(attribute, selectors)
-        Services::NodeParserService.nodes_from(root_node, selectors).map { |node| Services::NodeParserService.values_from(node, attribute) }.reject(&:empty?)
+      def url_attributes
+        url_nodes.flat_map(&:attribute_nodes).select { |attribute| HTML_ATTRIBUTES_MAP.key?(attribute.name) }
+      end
+
+      def url_nodes
+        root_node.css(*CSS_SELECTORS_ARRAY)
+      end
+
+      module UrlAttributesParser
+        # @param attributes [Array<Nokogiri::XML::Attr>]
+        # @return [Array<String>]
+        def self.parse(*attributes)
+          attributes.flat_map { |attribute| value_from(attribute) }
+        end
+
+        # @param attribute [Nokogiri::XML::Attr]
+        # @return [String, Array<String>]
+        def self.value_from(attribute)
+          return attribute.value unless attribute.name == 'srcset'
+
+          attribute.value.split(',').map { |value| value.strip.match(/^\S+/).to_s }
+        end
       end
     end
   end

data/lib/webmention/version.rb

@@ -1,3 +1,3 @@
 module Webmention
-  VERSION = '2.2.0'.freeze
+  VERSION = '3.0.0'.freeze
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: webmention
 version: !ruby/object:Gem::Version
-  version: 2.2.0
+  version: 3.0.0
 platform: ruby
 authors:
 - Aaron Parecki
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2020-05-18 00:00:00.000000000 Z
+date: 2020-05-19 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: absolutely
@@ -109,7 +109,6 @@ files:
 - lib/webmention/parsers/base_parser.rb
 - lib/webmention/parsers/html_parser.rb
 - lib/webmention/services/http_request_service.rb
-- lib/webmention/services/node_parser_service.rb
 - lib/webmention/version.rb
 - webmention.gemspec
 homepage: https://github.com/indieweb/webmention-client-ruby
@@ -117,7 +116,7 @@ licenses:
 - Apache-2.0
 metadata:
   bug_tracker_uri: https://github.com/indieweb/webmention-client-ruby/issues
-  changelog_uri: https://github.com/indieweb/webmention-client-ruby/blob/v2.2.0/CHANGELOG.md
+  changelog_uri: https://github.com/indieweb/webmention-client-ruby/blob/v3.0.0/CHANGELOG.md
 post_install_message: 
 rdoc_options: []
 require_paths:

data/lib/webmention/services/node_parser_service.rb

@@ -1,25 +0,0 @@
-module Webmention
-  module Services
-    module NodeParserService
-      # Search a node for matching elements
-      #
-      # @param node [Nokogiri::XML::Element]
-      # @param selectors [Array]
-      # @return [Nokogiri::XML::NodeSet]
-      def self.nodes_from(node, selectors)
-        node.css(*selectors)
-      end
-
-      # Derive attribute values from a single node
-      #
-      # @param node [Nokogiri::XML::Element]
-      # @param attribute [Symbol]
-      # @return [Array] the HTML attribute values
-      def self.values_from(node, attribute)
-        return Array(node[attribute]) unless attribute == :srcset
-
-        node[attribute].split(',').map { |value| value.strip.match(/^\S+/).to_s }
-      end
-    end
-  end
-end