feedjira 2.2.0 → 3.1.2

data/Rakefile

@@ -1,17 +1,19 @@
-require 'rspec/core/rake_task'
-require 'rubocop/rake_task'
-require 'yard'
+# frozen_string_literal: true
+
+require "rspec/core/rake_task"
+require "rubocop/rake_task"
+require "yard"
 
 RSpec::Core::RakeTask.new(:spec) do |t|
   t.verbose = false
 end
 
 RuboCop::RakeTask.new(:rubocop) do |t|
-  t.options = ['--display-cop-names']
+  t.options = ["--display-cop-names"]
 end
 
 YARD::Rake::YardocTask.new do |t|
-  t.files   = ['lib/**/*.rb', '-', 'LICENSE']
+  t.files   = ["lib/**/*.rb", "-", "LICENSE"]
 end
 
-task default: [:spec, :rubocop]
+task default: %i[spec rubocop]

data/feedjira.gemspec

@@ -1,31 +1,42 @@
-# -*- encoding: utf-8 -*-
-require File.expand_path('../lib/feedjira/version', __FILE__)
+# frozen_string_literal: true
+
+require File.expand_path("lib/feedjira/version", __dir__)
 
 Gem::Specification.new do |s|
-  s.authors  = ['Paul Dix', 'Julien Kirch', 'Ezekiel Templin', 'Jon Allured']
-  s.email    = 'feedjira@gmail.com'
-  s.homepage = 'http://feedjira.com'
-  s.license  = 'MIT'
-  s.name     = 'feedjira'
+  s.authors = [
+    "Adam Hess",
+    "Akinori Musha",
+    "Ezekiel Templin",
+    "Jon Allured",
+    "Julien Kirch",
+    "Michael Stock",
+    "Paul Dix"
+  ]
+  s.homepage = "https://github.com/feedjira/feedjira"
+  s.license  = "MIT"
+  s.name     = "feedjira"
   s.platform = Gem::Platform::RUBY
-  s.summary  = 'A feed fetching and parsing library'
+  s.summary  = "A feed parsing library"
   s.version  = Feedjira::VERSION
 
+  s.metadata = {
+    "homepage_uri" => "http://feedjira.com",
+    "source_code_uri" => "https://github.com/feedjira/feedjira",
+    "changelog_uri" => "https://github.com/feedjira/feedjira/blob/master/CHANGELOG.md"
+  }
+
   s.files         = `git ls-files`.split("\n")
-  s.require_paths = ['lib']
+  s.require_paths = ["lib"]
   s.test_files    = `git ls-files -- {test,spec,features}/*`.split("\n")
 
-  s.required_ruby_version = '>=1.9.3'
+  s.required_ruby_version = ">=2.2"
 
-  s.add_dependency 'faraday',            '>= 0.9'
-  s.add_dependency 'faraday_middleware', '>= 0.9'
-  s.add_dependency 'loofah',             '>= 2.0'
-  s.add_dependency 'sax-machine',        '>= 1.0'
+  s.add_dependency "loofah",             ">= 2.3.1"
+  s.add_dependency "sax-machine",        ">= 1.0"
 
-  s.add_development_dependency 'danger'
-  s.add_development_dependency 'danger-commit_lint'
-  s.add_development_dependency 'rspec'
-  s.add_development_dependency 'rubocop', '0.46'
-  s.add_development_dependency 'vcr'
-  s.add_development_dependency 'yard', '~> 0.9'
+  s.add_development_dependency "faraday"
+  s.add_development_dependency "rspec"
+  s.add_development_dependency "rubocop"
+  s.add_development_dependency "vcr"
+  s.add_development_dependency "yard"
 end

data/lib/feedjira.rb

@@ -1,47 +1,81 @@
-require 'zlib'
-require 'faraday'
-require 'faraday_middleware'
-require 'sax-machine'
-require 'loofah'
-require 'logger'
-
-require 'feedjira/core_ext'
-require 'feedjira/configuration'
-require 'feedjira/date_time_utilities/date_time_epoch_parser'
-require 'feedjira/date_time_utilities/date_time_language_parser'
-require 'feedjira/date_time_utilities/date_time_pattern_parser'
-require 'feedjira/date_time_utilities'
-require 'feedjira/date_time_utilities'
-require 'feedjira/feed_entry_utilities'
-require 'feedjira/feed_utilities'
-require 'feedjira/feed'
-require 'feedjira/parser'
-require 'feedjira/parser/rss_entry'
-require 'feedjira/parser/rss_image'
-require 'feedjira/parser/rss'
-require 'feedjira/parser/atom_entry'
-require 'feedjira/parser/atom'
-require 'feedjira/preprocessor'
-require 'feedjira/version'
-
-require 'feedjira/parser/rss_feed_burner_entry'
-require 'feedjira/parser/rss_feed_burner'
-require 'feedjira/parser/podlove_chapter'
-require 'feedjira/parser/itunes_rss_owner'
-require 'feedjira/parser/itunes_rss_category'
-require 'feedjira/parser/itunes_rss_item'
-require 'feedjira/parser/itunes_rss'
-require 'feedjira/parser/atom_feed_burner_entry'
-require 'feedjira/parser/atom_feed_burner'
-require 'feedjira/parser/google_docs_atom_entry'
-require 'feedjira/parser/google_docs_atom'
-require 'feedjira/parser/atom_youtube_entry'
-require 'feedjira/parser/atom_youtube'
+# frozen_string_literal: true
+
+require "zlib"
+require "sax-machine"
+require "loofah"
+require "logger"
+require "json"
+
+require "feedjira/core_ext"
+require "feedjira/configuration"
+require "feedjira/date_time_utilities/date_time_epoch_parser"
+require "feedjira/date_time_utilities/date_time_language_parser"
+require "feedjira/date_time_utilities/date_time_pattern_parser"
+require "feedjira/date_time_utilities"
+require "feedjira/feed_entry_utilities"
+require "feedjira/feed_utilities"
+require "feedjira/feed"
+require "feedjira/rss_entry_utilities"
+require "feedjira/atom_entry_utilities"
+require "feedjira/parser"
+require "feedjira/parser/globally_unique_identifier"
+require "feedjira/parser/rss_entry"
+require "feedjira/parser/rss_image"
+require "feedjira/parser/rss"
+require "feedjira/parser/atom_entry"
+require "feedjira/parser/atom"
+require "feedjira/preprocessor"
+require "feedjira/version"
+
+require "feedjira/parser/rss_feed_burner_entry"
+require "feedjira/parser/rss_feed_burner"
+require "feedjira/parser/podlove_chapter"
+require "feedjira/parser/itunes_rss_owner"
+require "feedjira/parser/itunes_rss_category"
+require "feedjira/parser/itunes_rss_item"
+require "feedjira/parser/itunes_rss"
+require "feedjira/parser/atom_feed_burner_entry"
+require "feedjira/parser/atom_feed_burner"
+require "feedjira/parser/atom_google_alerts_entry"
+require "feedjira/parser/atom_google_alerts"
+require "feedjira/parser/google_docs_atom_entry"
+require "feedjira/parser/google_docs_atom"
+require "feedjira/parser/atom_youtube_entry"
+require "feedjira/parser/atom_youtube"
+require "feedjira/parser/json_feed"
+require "feedjira/parser/json_feed_item"
 
 # Feedjira
 module Feedjira
-  class NoParserAvailable < StandardError; end
-  class FetchFailure < StandardError; end
+  NoParserAvailable = Class.new(StandardError)
 
   extend Configuration
+
+  # Parse XML with first compatible parser
+  #
+  # @example
+  #   xml = HTTParty.get("http://example.com").body
+  #   Feedjira.parse(xml)
+  def parse(xml, parser: nil, &block)
+    parser ||= parser_for_xml(xml)
+
+    if parser.nil?
+      raise NoParserAvailable, "No valid parser for XML."
+    end
+
+    parser.parse(xml, &block)
+  end
+  module_function :parse # rubocop:disable Style/AccessModifierDeclarations
+
+  # Find compatible parser for given XML
+  #
+  # @example
+  #   xml = HTTParty.get("http://example.com").body
+  #   parser = Feedjira.parser_for_xml(xml)
+  #   parser.parse(xml)
+  def parser_for_xml(xml)
+    start_of_doc = xml.slice(0, 2000)
+    Feedjira.parsers.detect { |klass| klass.able_to_parse?(start_of_doc) }
+  end
+  module_function :parser_for_xml # rubocop:disable Style/AccessModifierDeclarations
 end

data/lib/feedjira/atom_entry_utilities.rb

@@ -0,0 +1,51 @@
+# frozen_string_literal: true
+
+module Feedjira
+  module AtomEntryUtilities
+    def self.included(mod)
+      mod.class_exec do
+        element :title, as: :raw_title, with: { type: "html" }
+        element :title, as: :raw_title, with: { type: "xhtml" }
+        element :title, as: :raw_title, with: { type: "xml" }
+        element :title, as: :title, with: { type: "text" }
+        element :title, as: :title, with: { type: nil }
+        element :title, as: :title_type, value: :type
+
+        element :name, as: :author
+        element :content
+        element :summary
+        element :enclosure, as: :image, value: :href
+
+        element :published
+        element :id, as: :entry_id
+        element :created, as: :published
+        element :issued, as: :published
+        element :updated
+        element :modified, as: :updated
+
+        elements :category, as: :categories, value: :term
+
+        element :link, as: :url, value: :href, with: {
+          type: "text/html",
+          rel: "alternate"
+        }
+
+        elements :link, as: :links, value: :href
+      end
+    end
+
+    def title
+      @title ||=
+        case @raw_title
+        when String
+          Loofah.fragment(@raw_title).xpath("normalize-space(.)")
+        else
+          @title
+        end
+    end
+
+    def url
+      @url ||= links.first
+    end
+  end
+end

data/lib/feedjira/configuration.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 # Feedjira::Configuration
 module Feedjira
   # Provides global configuration options for Feedjira
@@ -8,12 +10,9 @@ module Feedjira
   #   end
   module Configuration
     attr_accessor(
-      :follow_redirect_limit,
       :logger,
       :parsers,
-      :request_timeout,
-      :strip_whitespace,
-      :user_agent
+      :strip_whitespace
     )
 
     # Modify Feedjira's current configuration
@@ -42,20 +41,17 @@ module Feedjira
 
     # @private
     def set_default_configuration
-      self.follow_redirect_limit = 3
       self.logger = default_logger
       self.parsers = default_parsers
-      self.request_timeout = 30
       self.strip_whitespace = false
-      self.user_agent = "Feedjira #{Feedjira::VERSION}"
     end
 
     private
 
     # @private
     def default_logger
-      Logger.new(STDOUT).tap do |logger|
-        logger.progname = 'Feedjira'
+      Logger.new($stdout).tap do |logger|
+        logger.progname = "Feedjira"
         logger.level = Logger::WARN
       end
     end
@@ -67,9 +63,11 @@ module Feedjira
         Feedjira::Parser::GoogleDocsAtom,
         Feedjira::Parser::AtomYoutube,
         Feedjira::Parser::AtomFeedBurner,
+        Feedjira::Parser::AtomGoogleAlerts,
         Feedjira::Parser::Atom,
         Feedjira::Parser::ITunesRSS,
-        Feedjira::Parser::RSS
+        Feedjira::Parser::RSS,
+        Feedjira::Parser::JSONFeed
       ]
     end
   end

data/lib/feedjira/core_ext.rb

@@ -1,3 +1,5 @@
-require 'feedjira/core_ext/time'
-require 'feedjira/core_ext/date'
-require 'feedjira/core_ext/string'
+# frozen_string_literal: true
+
+require "feedjira/core_ext/time"
+require "feedjira/core_ext/date"
+require "feedjira/core_ext/string"

data/lib/feedjira/core_ext/date.rb

@@ -1,8 +1,9 @@
+# frozen_string_literal: true
+
 # Date code pulled and adapted from:
 # Ruby Cookbook by Lucas Carlson and Leonard Richardson
 # Published by O'Reilly
 # ISBN: 0-596-52369-6
-# rubocop:disable Style/DocumentationMethod
 class Date
   def feed_utils_to_gm_time
     feed_utils_to_time(new_offset, :gm)

data/lib/feedjira/core_ext/string.rb

@@ -1,4 +1,5 @@
-# rubocop:disable Style/DocumentationMethod Style/Documentation
+# frozen_string_literal: true
+
 class String
   def sanitize!
     replace(sanitize)

data/lib/feedjira/core_ext/time.rb

@@ -1,7 +1,8 @@
-require 'time'
-require 'date'
+# frozen_string_literal: true
+
+require "time"
+require "date"
 
-# rubocop:disable Style/DocumentationMethod
 class Time
   # Parse a time string and convert it to UTC without raising errors.
   # Parses a flattened 14-digit time (YYYYmmddHHMMMSS) as UTC.
@@ -11,17 +12,16 @@ class Time
   #
   # === Returns
   # A Time instance in UTC or nil if there were errors while parsing.
-  # rubocop:disable Metrics/MethodLength
-  def self.parse_safely(dt)
-    if dt.is_a?(Time)
-      dt.utc
-    elsif dt.respond_to?(:to_datetime)
-      dt.to_datetime.utc
-    elsif dt.respond_to? :to_s
-      parse_string_safely dt.to_s
+  def self.parse_safely(datetime)
+    if datetime.is_a?(Time)
+      datetime.utc
+    elsif datetime.respond_to?(:to_datetime)
+      datetime.to_datetime.utc
+    elsif datetime.respond_to? :to_s
+      parse_string_safely datetime.to_s
     end
   rescue StandardError => e
-    Feedjira.logger.debug { "Failed to parse time #{dt}" }
+    Feedjira.logger.debug { "Failed to parse time #{datetime}" }
     Feedjira.logger.debug(e)
     nil
   end

data/lib/feedjira/date_time_utilities.rb

@@ -1,4 +1,5 @@
-# rubocop:disable Style/Documentation
+# frozen_string_literal: true
+
 module Feedjira
   module DateTimeUtilities
     # This is our date parsing heuristic.
@@ -12,16 +13,13 @@ module Feedjira
 
     # Parse the given string starting with the most common parser (default ruby)
     # and going over all other available parsers
-    # rubocop:disable Metrics/MethodLength
     def parse_datetime(string)
-      res = DATE_PARSERS.find do |parser|
-        begin
-          return parser.parse(string).feed_utils_to_gm_time
-        rescue StandardError => e
-          Feedjira.logger.debug { "Failed to parse date #{string}" }
-          Feedjira.logger.debug(e)
-          nil
-        end
+      res = DATE_PARSERS.detect do |parser|
+        return parser.parse(string).feed_utils_to_gm_time
+      rescue StandardError => e
+        Feedjira.logger.debug { "Failed to parse date #{string}" }
+        Feedjira.logger.debug(e)
+        nil
       end
 
       Feedjira.logger.warn { "Failed to parse date #{string}" } if res.nil?

data/lib/feedjira/date_time_utilities/date_time_epoch_parser.rb

@@ -1,11 +1,12 @@
-# rubocop:disable Style/Documentation
-# rubocop:disable Style/DocumentationMethod
+# frozen_string_literal: true
+
 module Feedjira
   module DateTimeUtilities
     class DateTimeEpochParser
       def self.parse(string)
         epoch_time = string.to_i
         return Time.at(epoch_time).to_datetime if epoch_time.to_s == string
+
         raise "#{string} is not a valid epoch time"
       end
     end

data/lib/feedjira/date_time_utilities/date_time_language_parser.rb

@@ -1,12 +1,12 @@
-# rubocop:disable Style/Documentation
-# rubocop:disable Style/DocumentationMethod
+# frozen_string_literal: true
+
 module Feedjira
   module DateTimeUtilities
     class DateTimeLanguageParser
       MONTHS_ENGLISH =
-        %w(Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec).freeze
+        %w[Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec].freeze
       MONTHS_SPANISH =
-        %w(Ene Feb Mar Abr May Jun Jul Ago Sep Oct Nov Dic).freeze
+        %w[Ene Feb Mar Abr May Jun Jul Ago Sep Oct Nov Dic].freeze
 
       def self.parse(string)
         DateTime.parse(translate(string))