html2rss 0.19.1 → 0.20.0

data/lib/html2rss.rb

@@ -7,18 +7,21 @@ loader.inflector.inflect('cli' => 'CLI')
 loader.setup
 
 require 'logger'
+require 'forwardable'
+require 'html2rss/configuration'
 
 ##
 # The Html2rss namespace.
 module Html2rss
   ##
   # The logger instance.
-  Log = Logger.new($stdout)
+  module Log
+    class << self
+      extend Forwardable
 
-  Log.level = ENV.fetch('LOG_LEVEL', :warn).upcase.to_sym
-
-  Log.formatter = proc do |severity, datetime, _progname, msg|
-    "#{datetime} [#{severity}] #{msg}\n"
+      def_delegator 'Html2rss', :logger
+      def_delegators :logger, :debug, :info, :warn, :error, :fatal, :unknown, :level, :level=, :formatter, :formatter=
+    end
   end
 
   ##
@@ -75,6 +78,50 @@ module Html2rss
     json_feed(build_auto_source_config(url:, strategy:, items_selector:, max_redirects:, max_requests:))
   end
 
+  # rubocop:disable ThreadSafety/ClassInstanceVariable
+  class << self
+    ##
+    # @return [Html2rss::Configuration] the global configuration instance
+    def configuration
+      @configuration ||= Configuration.new.freeze
+    end
+
+    ##
+    # Configures global library defaults.
+    #
+    # @yieldparam config [Html2rss::Configuration]
+    # @return [Html2rss::Configuration] the frozen configuration
+    def configure
+      config = configuration.dup
+      yield config
+      @configuration = config.freeze
+    end
+
+    ##
+    # @return [Object] the logger
+    def logger
+      configuration.logger
+    end
+
+    ##
+    # @param logger [Object] the new logger
+    def logger=(logger)
+      configure { |config| config.logger = logger }
+    end
+
+    private
+
+    ##
+    # Resets the global configuration to defaults (mainly for testing).
+    #
+    # @return [void]
+    def reset_configuration!
+      @configuration = nil
+      logger.level = configuration.log_level if logger.respond_to?(:level=)
+    end
+  end
+  # rubocop:enable ThreadSafety/ClassInstanceVariable
+
   class << self
     private
 
@@ -103,6 +150,8 @@ module Html2rss
       keys
     end
   end
+
+  logger.level = configuration.log_level if logger.respond_to?(:level=)
 end
 
 loader.eager_load

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: html2rss
 version: !ruby/object:Gem::Version
-  version: 0.19.1
+  version: 0.20.0
 platform: ruby
 authors:
 - Gil Desmarais
@@ -293,6 +293,7 @@ files:
 - lib/html2rss/auto_source/scraper.rb
 - lib/html2rss/auto_source/scraper/html.rb
 - lib/html2rss/auto_source/scraper/json_state.rb
+- lib/html2rss/auto_source/scraper/link_heuristics.rb
 - lib/html2rss/auto_source/scraper/microdata.rb
 - lib/html2rss/auto_source/scraper/schema.rb
 - lib/html2rss/auto_source/scraper/schema/category_extractor.rb
@@ -301,6 +302,7 @@ files:
 - lib/html2rss/auto_source/scraper/schema/thing.rb
 - lib/html2rss/auto_source/scraper/semantic_html.rb
 - lib/html2rss/auto_source/scraper/semantic_html/anchor_selector.rb
+- lib/html2rss/auto_source/scraper/semantic_html/deduplicator.rb
 - lib/html2rss/auto_source/scraper/wordpress_api.rb
 - lib/html2rss/auto_source/scraper/wordpress_api/page_scope.rb
 - lib/html2rss/auto_source/scraper/wordpress_api/posts_endpoint.rb
@@ -314,6 +316,7 @@ files:
 - lib/html2rss/config/request_headers.rb
 - lib/html2rss/config/schema.rb
 - lib/html2rss/config/validator.rb
+- lib/html2rss/configuration.rb
 - lib/html2rss/error.rb
 - lib/html2rss/feed_pipeline.rb
 - lib/html2rss/feed_pipeline/auto_fallback.rb
@@ -322,6 +325,9 @@ files:
 - lib/html2rss/html_extractor/date_extractor.rb
 - lib/html2rss/html_extractor/enclosure_extractor.rb
 - lib/html2rss/html_extractor/image_extractor.rb
+- lib/html2rss/html_extractor/list_candidates.rb
+- lib/html2rss/html_extractor/semantic_anchor_candidates.rb
+- lib/html2rss/html_extractor/semantic_containers.rb
 - lib/html2rss/html_navigator.rb
 - lib/html2rss/json_feed_builder.rb
 - lib/html2rss/json_feed_builder/item.rb
@@ -384,7 +390,7 @@ licenses:
 - MIT
 metadata:
   allowed_push_host: https://rubygems.org
-  changelog_uri: https://github.com/html2rss/html2rss/releases/tag/v0.19.1
+  changelog_uri: https://github.com/html2rss/html2rss/releases/tag/v0.20.0
   rubygems_mfa_required: 'true'
 rdoc_options: []
 require_paths:
@@ -400,7 +406,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 4.0.6
+rubygems_version: 4.0.10
 specification_version: 4
 summary: Generates RSS feeds from websites by scraping a URL and using CSS selectors
   to extract item.