indieweb-endpoints 0.4.0 → 0.5.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 1788ffc5ea1dd7ec1a6d4c1839e19c4b208b38044a05aef7d618bad2485b7638
-  data.tar.gz: 4458a1d611d045035916ec9fc6f69c752f628a6c65776fbfdc0fdfc4bb823c83
+  metadata.gz: 9cb1b27367683210b63651d8195e6d12013f994be97d8c216b2e2165429c4adb
+  data.tar.gz: 587e29ea23c02d6de575cefbc6a5f9a808c5e6b47bf662ee138ceaadba03c57f
 SHA512:
-  metadata.gz: cd530923575c34d392253a090fb9458c0709f6049ba055e9665879c067cc9571814b6079806773b3080c1b63def074b9331d8e9415855c91987acc56679090bb
-  data.tar.gz: 4f2836abdae4e424d2981b654d1d9a8b07cf84a970ee47571dca2235f85905c788136cc63ecde32d3ffbcc632e6f77e96d90aa6f3fc2c4f4bcb1d556d07ad587
+  metadata.gz: 1ca06c752aa3ff624d1ab3dbcad733d69bd5ab701ccdac754e7f5bc4e4d7c70ee8b42b23dd911448fb85fab44264e2e7f3c9c136cb23f5b4aaad282a21555ec7
+  data.tar.gz: adbb98ffc16e4664d8ea47a94bdfb32b5fbd552155716a1928bf3b62fbdd4bc545768d00ce046101db454b8a71752acdb14330c99e57c1597eca73b8b0e71b93

data/.rubocop.yml

@@ -7,6 +7,7 @@ Layout/AlignHash:
 
 Metrics/BlockLength:
   Exclude:
+    - indieweb-endpoints.gemspec
     - spec/**/*.rb
 
 Metrics/LineLength:

data/CHANGELOG.md

@@ -1,5 +1,14 @@
 # Changelog
 
+## 0.5.0 / 2019-05-09
+
+- Add support for Microsub endpoint discovery ([5e81d9f](https://github.com/indieweb/indieweb-endpoints-ruby/commit/5e81d9f)).
+- Refactor parsers to ignore URLs with fragments ([797b376](https://github.com/indieweb/indieweb-endpoints-ruby/commit/797b376)).
+- Rescue `NoMethodError` (for `nil`) and `TypeError` (for non-`String`) ([e33522e](https://github.com/indieweb/indieweb-endpoints-ruby/commit/e33522e)).
+- Raise `ArgumentError` if url scheme is not `http` or `https` ([8eb1b1a](https://github.com/indieweb/indieweb-endpoints-ruby/commit/8eb1b1a)).
+- Shorten up User Agent string ([f9717b4](https://github.com/indieweb/indieweb-endpoints-ruby/commit/f9717b4)).
+- Refactor `HTTPRequest` class using specification defaults ([feef2ba](https://github.com/indieweb/indieweb-endpoints-ruby/commit/feef2ba)).
+
 ## 0.4.0 / 2019-05-01
 
 - Add `IndieWeb::Endpoints.client` method ([c4d42d0](https://github.com/indieweb/indieweb-endpoints-ruby/commit/c4d42d0)).

data/README.md

@@ -1,6 +1,6 @@
 # indieweb-endpoints-ruby
 
-**A Ruby gem for discovering a URL's [IndieAuth](https://indieweb.org/IndieAuth), [Micropub](https://indieweb.org/Micropub), and [Webmention](https://indieweb.org/Webmention) endpoints.**
+**A Ruby gem for discovering a URL's [IndieAuth](https://indieweb.org/IndieAuth), [Micropub](https://indieweb.org/Micropub), [Microsub](https://indieweb.org/Microsub), and [Webmention](https://indieweb.org/Webmention) endpoints.**
 
 [![Gem](https://img.shields.io/gem/v/indieweb-endpoints.svg?style=for-the-badge)](https://rubygems.org/gems/indieweb-endpoints)
 [![Downloads](https://img.shields.io/gem/dt/indieweb-endpoints.svg?style=for-the-badge)](https://rubygems.org/gems/indieweb-endpoints)
@@ -56,6 +56,7 @@ This example will search `https://aaronparecki.com` for valid IndieAuth, Micropu
 {
   authorization_endpoint: 'https://aaronparecki.com/auth',
   micropub: 'https://aaronparecki.com/micropub',
+  microsub: 'https://aperture.p3k.io/microsub/1',
   redirect_uri: nil,
   token_endpoint: 'https://aaronparecki.com/auth/token',
   webmention: 'https://webmention.io/aaronpk/webmention'

data/indieweb-endpoints.gemspec

@@ -3,7 +3,6 @@ $LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
 
 require 'indieweb/endpoints/version'
 
-# rubocop:disable Metrics/BlockLength
 Gem::Specification.new do |spec|
   spec.required_ruby_version = ['>= 2.4', '< 2.7']
 
@@ -12,7 +11,7 @@ Gem::Specification.new do |spec|
   spec.authors       = ['Jason Garber']
   spec.email         = ['jason@sixtwothree.org']
 
-  spec.summary       = 'Discover a URL’s IndieAuth, Micropub, and Webmention endpoints.'
+  spec.summary       = 'Discover a URL’s IndieAuth, Micropub, Microsub, and Webmention endpoints.'
   spec.description   = spec.summary
   spec.homepage      = 'https://github.com/indieweb/indieweb-endpoints-ruby'
   spec.license       = 'MIT'
@@ -41,4 +40,3 @@ Gem::Specification.new do |spec|
   spec.add_runtime_dependency 'http', '~> 5.0.0.pre'
   spec.add_runtime_dependency 'nokogiri', '~> 1.10'
 end
-# rubocop:enable Metrics/BlockLength

data/lib/indieweb/endpoints/client.rb

@@ -2,13 +2,13 @@ module IndieWeb
   module Endpoints
     class Client
       def initialize(url)
-        raise ArgumentError, "url must be a String (given #{url.class.name})" unless url.is_a?(String)
-
         @uri = Addressable::URI.parse(url)
 
-        raise ArgumentError, 'url must be an absolute URL (e.g. https://example.com)' unless @uri.absolute?
+        raise ArgumentError, 'url must be an absolute URL (e.g. https://example.com)' unless @uri.absolute? && @uri.scheme.match?(/^https?$/)
       rescue Addressable::URI::InvalidURIError => exception
         raise InvalidURIError, exception
+      rescue NoMethodError, TypeError
+        raise ArgumentError, "url must be a String (given #{url.class})"
       end
 
       def endpoints

data/lib/indieweb/endpoints/http_request.rb

@@ -1,13 +1,24 @@
 module IndieWeb
   module Endpoints
     class HttpRequest
-      HTTP_HEADERS_OPTS = {
-        accept: '*/*',
-        user_agent: 'IndieAuth, Micropub, and Webmention Endpoint Discovery (https://rubygems.org/gems/indieweb-endpoints)'
+      # Defaults derived from Webmention specification examples
+      # https://www.w3.org/TR/webmention/#limits-on-get-requests
+      HTTP_CLIENT_OPTS = {
+        follow: {
+          max_hops: 20
+        },
+        headers: {
+          accept: '*/*',
+          user_agent: 'IndieWeb Endpoint Discovery (https://rubygems.org/gems/indieweb-endpoints)'
+        },
+        timeout_options: {
+          connect_timeout: 5,
+          read_timeout: 5
+        }
       }.freeze
 
       def self.get(uri)
-        HTTP.follow.headers(HTTP_HEADERS_OPTS).timeout(connect: 10, read: 10).get(uri)
+        HTTP::Client.new(HTTP_CLIENT_OPTS).request(:get, uri)
       rescue HTTP::ConnectionError,
              HTTP::TimeoutError,
              HTTP::Redirector::TooManyRedirectsError => exception

data/lib/indieweb/endpoints/parsers/microsub_parser.rb

@@ -0,0 +1,13 @@
+module IndieWeb
+  module Endpoints
+    module Parsers
+      class MicrosubParser < BaseParser
+        def self.identifier
+          :microsub
+        end
+
+        Parsers.register(self)
+      end
+    end
+  end
+end

data/lib/indieweb/endpoints/parsers/redirect_uri_parser.rb

@@ -11,7 +11,7 @@ module IndieWeb
         def results
           return unless results_from_http_request.any?
 
-          @results ||= results_from_http_request.map { |endpoint| Absolutely.to_abs(base: @response.uri.to_s, relative: endpoint) }.uniq.sort
+          @results ||= results_from_http_request.map { |endpoint| Absolutely.to_abs(base: response.uri.to_s, relative: endpoint) }.uniq.sort
         rescue Absolutely::InvalidURIError => exception
           raise InvalidURIError, exception
         end
@@ -19,10 +19,26 @@ module IndieWeb
         private
 
         def results_from_body
-          link_elements.map { |element| element['href'] } if response_is_html && link_elements.any?
+          RedirectUriLinkElementParser.new(response, self.class.identifier).results
         end
 
         def results_from_headers
+          RedirectUriLinkHeaderParser.new(response, self.class.identifier).results
+        end
+
+        def results_from_http_request
+          @results_from_http_request ||= [results_from_headers, results_from_body].flatten.compact
+        end
+      end
+
+      class RedirectUriLinkElementParser < LinkElementParser
+        def results
+          link_elements.map { |element| element['href'] } if response_is_html && link_elements.any?
+        end
+      end
+
+      class RedirectUriLinkHeaderParser < LinkHeaderParser
+        def results
           return unless link_headers.any?
 
           link_headers.map do |header|
@@ -31,10 +47,6 @@ module IndieWeb
             endpoint_match_data[1] if endpoint_match_data
           end
         end
-
-        def results_from_http_request
-          @results_from_http_request ||= [results_from_headers, results_from_body].flatten.compact
-        end
       end
     end
   end

data/lib/indieweb/endpoints/parsers/webmention_parser.rb

@@ -10,6 +10,14 @@ module IndieWeb
 
         private
 
+        def results_from_body
+          WebmentionLinkElementParser.new(response, self.class.identifier).results
+        end
+      end
+
+      class WebmentionLinkElementParser < LinkElementParser
+        private
+
         def link_element
           # Return first `a` or `link` element with valid `rel` attribute
           # https://www.w3.org/TR/webmention/#sender-discovers-receiver-webmention-endpoint
@@ -17,7 +25,7 @@ module IndieWeb
         end
 
         def link_elements_css_selector
-          @link_elements_css_selector ||= %([rel~="#{self.class.identifier}"][href])
+          @link_elements_css_selector ||= %([rel~="#{identifier}"][href]:not([href*="#"]))
         end
       end
     end

data/lib/indieweb/endpoints/parsers.rb

@@ -4,13 +4,7 @@ module IndieWeb
       extend Registerable
 
       class BaseParser
-        # Ultra-orthodox pattern matching allowed values in Link header `rel` parameter
-        # https://tools.ietf.org/html/rfc8288#section-3.3
-        REGEXP_REG_REL_TYPE_PATTERN = '[a-z\d][a-z\d\-\.]*'.freeze
-
-        # Liberal pattern matching a string of text between angle brackets
-        # https://tools.ietf.org/html/rfc5988#section-5.1
-        REGEXP_TARGET_URI_PATTERN = /^<(.*)>;/.freeze
+        attr_reader :response
 
         def initialize(response)
           raise ArgumentError, "response must be an HTTP::Response (given #{response.class.name})" unless response.is_a?(HTTP::Response)
@@ -21,21 +15,42 @@ module IndieWeb
         def results
           return unless results_from_http_request
 
-          @results ||= Absolutely.to_abs(base: @response.uri.to_s, relative: results_from_http_request)
+          @results ||= Absolutely.to_abs(base: response.uri.to_s, relative: results_from_http_request)
         rescue Absolutely::InvalidURIError => exception
           raise InvalidURIError, exception
         end
 
         private
 
-        def discrete_link_headers
-          # Split Link headers with multiple values, flatten the resulting array, and strip whitespace
-          # https://webmention.rocks/test/19
-          @discrete_link_headers ||= @response.headers.get('link').map { |header| header.split(',') }.flatten.map(&:strip)
+        def results_from_body
+          LinkElementParser.new(response, self.class.identifier).results
+        end
+
+        def results_from_headers
+          LinkHeaderParser.new(response, self.class.identifier).results
+        end
+
+        def results_from_http_request
+          @results_from_http_request ||= results_from_headers || results_from_body || nil
+        end
+      end
+
+      class LinkElementParser
+        attr_reader :identifier, :response
+
+        def initialize(response, identifier)
+          @response = response
+          @identifier = identifier
         end
 
+        def results
+          link_element['href'] if response_is_html && link_element
+        end
+
+        private
+
         def doc
-          @doc ||= Nokogiri::HTML(@response.body.to_s)
+          @doc ||= Nokogiri::HTML(response.body.to_s)
         end
 
         def link_element
@@ -50,42 +65,59 @@ module IndieWeb
         end
 
         def link_elements_css_selector
-          @link_elements_css_selector ||= %(link[rel~="#{self.class.identifier}"][href])
+          @link_elements_css_selector ||= %(link[rel~="#{identifier}"][href]:not([href*="#"]))
         end
 
-        def link_header
-          @link_header ||= link_headers.shift
+        def response_is_html
+          @response_is_html ||= response.mime_type == 'text/html'
         end
+      end
 
-        def link_headers
-          # Reduce Link headers to those with valid `rel` attribute
-          @link_headers ||= discrete_link_headers.find_all { |header| header.match?(regexp_rel_paramater_pattern) }
-        end
+      class LinkHeaderParser
+        # Ultra-orthodox pattern matching allowed values in Link header `rel` parameter
+        # https://tools.ietf.org/html/rfc8288#section-3.3
+        REGEXP_REG_REL_TYPE_PATTERN = '[a-z\d][a-z\d\-\.]*'.freeze
 
-        def regexp_rel_paramater_pattern
-          # Ultra-orthodox pattern matching Link header `rel` parameter including a matching identifier value
-          # https://www.w3.org/TR/webmention/#sender-discovers-receiver-webmention-endpoint
-          @regexp_rel_paramater_pattern ||= /(?:;|\s)rel="?(?:#{REGEXP_REG_REL_TYPE_PATTERN}+\s)?#{self.class.identifier}(?:\s#{REGEXP_REG_REL_TYPE_PATTERN})?"?/
-        end
+        # Liberal pattern capturing a string of text (excepting the octothorp) between angle brackets
+        # https://tools.ietf.org/html/rfc5988#section-5.1
+        REGEXP_TARGET_URI_PATTERN = '^<(.[^#]*)>;'.freeze
 
-        def results_from_body
-          link_element['href'] if response_is_html && link_element
+        attr_reader :identifier, :response
+
+        def initialize(response, identifier)
+          @response = response
+          @identifier = identifier
         end
 
-        def results_from_headers
+        def results
           return unless link_header
 
-          endpoint_match_data = link_header.match(REGEXP_TARGET_URI_PATTERN)
+          endpoint_match_data = link_header.match(/#{REGEXP_TARGET_URI_PATTERN}/)
 
           return endpoint_match_data[1] if endpoint_match_data
         end
 
-        def response_is_html
-          @response_is_html ||= @response.mime_type == 'text/html'
+        private
+
+        def discrete_link_headers
+          # Split Link headers with multiple values, flatten the resulting array, and strip whitespace
+          # https://webmention.rocks/test/19
+          @discrete_link_headers ||= response.headers.get('link').map { |header| header.split(',') }.flatten.map(&:strip)
         end
 
-        def results_from_http_request
-          @results_from_http_request ||= results_from_headers || results_from_body || nil
+        def link_header
+          @link_header ||= link_headers.shift
+        end
+
+        def link_headers
+          # Reduce Link headers to those with valid `rel` attribute
+          @link_headers ||= discrete_link_headers.find_all { |header| header.match?(/#{REGEXP_TARGET_URI_PATTERN}\s*#{regexp_rel_paramater_pattern}/) }
+        end
+
+        def regexp_rel_paramater_pattern
+          # Ultra-orthodox pattern matching Link header `rel` parameter including a matching identifier value
+          # https://www.w3.org/TR/webmention/#sender-discovers-receiver-webmention-endpoint
+          @regexp_rel_paramater_pattern ||= %(rel="?(?:#{REGEXP_REG_REL_TYPE_PATTERN}+\s)?#{identifier}(?:\s#{REGEXP_REG_REL_TYPE_PATTERN})?"?)
         end
       end
     end

data/lib/indieweb/endpoints/version.rb

@@ -1,5 +1,5 @@
 module IndieWeb
   module Endpoints
-    VERSION = '0.4.0'.freeze
+    VERSION = '0.5.0'.freeze
   end
 end

data/lib/indieweb/endpoints.rb

@@ -13,6 +13,7 @@ require 'indieweb/endpoints/registerable'
 require 'indieweb/endpoints/parsers'
 require 'indieweb/endpoints/parsers/authorization_endpoint_parser'
 require 'indieweb/endpoints/parsers/micropub_parser'
+require 'indieweb/endpoints/parsers/microsub_parser'
 require 'indieweb/endpoints/parsers/redirect_uri_parser'
 require 'indieweb/endpoints/parsers/token_endpoint_parser'
 require 'indieweb/endpoints/parsers/webmention_parser'

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: indieweb-endpoints
 version: !ruby/object:Gem::Version
-  version: 0.4.0
+  version: 0.5.0
 platform: ruby
 authors:
 - Jason Garber
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2019-05-02 00:00:00.000000000 Z
+date: 2019-05-10 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
@@ -192,7 +192,7 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '1.10'
-description: Discover a URL’s IndieAuth, Micropub, and Webmention endpoints.
+description: Discover a URL’s IndieAuth, Micropub, Microsub, and Webmention endpoints.
 email:
 - jason@sixtwothree.org
 executables: []
@@ -224,6 +224,7 @@ files:
 - lib/indieweb/endpoints/parsers.rb
 - lib/indieweb/endpoints/parsers/authorization_endpoint_parser.rb
 - lib/indieweb/endpoints/parsers/micropub_parser.rb
+- lib/indieweb/endpoints/parsers/microsub_parser.rb
 - lib/indieweb/endpoints/parsers/redirect_uri_parser.rb
 - lib/indieweb/endpoints/parsers/token_endpoint_parser.rb
 - lib/indieweb/endpoints/parsers/webmention_parser.rb
@@ -234,7 +235,7 @@ licenses:
 - MIT
 metadata:
   bug_tracker_uri: https://github.com/indieweb/indieweb-endpoints-ruby/issues
-  changelog_uri: https://github.com/indieweb/indieweb-endpoints-ruby/blob/v0.4.0/CHANGELOG.md
+  changelog_uri: https://github.com/indieweb/indieweb-endpoints-ruby/blob/v0.5.0/CHANGELOG.md
 post_install_message: 
 rdoc_options: []
 require_paths:
@@ -256,5 +257,5 @@ requirements: []
 rubygems_version: 3.0.3
 signing_key: 
 specification_version: 4
-summary: Discover a URL’s IndieAuth, Micropub, and Webmention endpoints.
+summary: Discover a URL’s IndieAuth, Micropub, Microsub, and Webmention endpoints.
 test_files: []