html2rss 0.30.0 → 0.31.0

data/lib/html2rss/defaults.rb

@@ -4,13 +4,20 @@ module Html2rss
   ##
   # Global defaults for the Html2rss gem.
   class Defaults
-    # The valid symbol log levels.
-    VALID_LOG_LEVELS = %i[debug info warn error fatal unknown].freeze
-
-    # @return [Object] the logger
+    # Named Logger severities accepted by {#log_level=}.
+    LOG_LEVELS = {
+      debug: Logger::DEBUG,
+      info: Logger::INFO,
+      warn: Logger::WARN,
+      error: Logger::ERROR,
+      fatal: Logger::FATAL,
+      unknown: Logger::UNKNOWN
+    }.freeze
+
+    # @return [Logger] the logger
     attr_reader :logger
 
-    # @return [Proc, nil] the logger formatter
+    # @return [Proc, Logger::Formatter, nil] the logger formatter
     attr_reader :logger_formatter
 
     # @return [Symbol, Integer] the current log level
@@ -46,12 +53,15 @@ module Html2rss
     ##
     # Sets the logger.
     #
-    # @param logger [Object]
-    # @return [Object] the logger
+    # @param logger [Logger]
+    # @return [Logger] the logger
+    # @raise [ArgumentError] if logger is not a Logger
     def logger=(logger)
+      raise ArgumentError, 'logger must be a Logger' unless logger.is_a?(::Logger)
+
       @logger = logger
-      @logger.level = @log_level if @logger.respond_to?(:level=)
-      @logger.formatter = @logger_formatter if @logger_formatter && @logger.respond_to?(:formatter=)
+      @logger.level = @log_level
+      @logger.formatter = @logger_formatter if @logger_formatter
     end
 
     ##
@@ -62,34 +72,39 @@ module Html2rss
     # @raise [ArgumentError] if the log level is invalid
     def log_level=(level)
       @log_level = normalize_log_level(level)
-      @logger.level = @log_level if @logger.respond_to?(:level=)
+      @logger.level = @log_level
     end
 
     ##
     # Sets the logger formatter.
     #
-    # @param formatter [Proc, #call, nil] the new logger formatter
-    # @return [Proc, #call, nil] the new logger formatter
-    # @raise [ArgumentError] if formatter does not respond to #call
+    # @param formatter [Proc, Logger::Formatter, nil] the new logger formatter
+    # @return [Proc, Logger::Formatter, nil] the new logger formatter
+    # @raise [ArgumentError] if formatter is not a Proc, Logger::Formatter, or nil
     def logger_formatter=(formatter)
-      raise ArgumentError, 'formatter must respond to #call or be nil' if formatter && !formatter.respond_to?(:call)
-
-      @logger_formatter = formatter
-      @logger.formatter = @logger_formatter if @logger.respond_to?(:formatter=)
+      @logger_formatter = case formatter
+                          when nil, Proc, Logger::Formatter
+                            formatter
+                          else
+                            raise ArgumentError, 'formatter must be a Proc, Logger::Formatter, or nil'
+                          end
+      @logger.formatter = @logger_formatter
     end
 
     ##
     # Sets the global request headers.
     #
-    # @param headers [Hash, Proc, #call, nil] the HTTP request headers to globally apply
-    # @return [Hash, Proc, #call, nil] the assigned headers
-    # @raise [ArgumentError] if headers is not a Hash or callable
+    # @param headers [Hash, Proc, nil] the HTTP request headers to globally apply
+    # @return [Hash, Proc, nil] the assigned headers
+    # @raise [ArgumentError] if headers is not a Hash, Proc, or nil
     def headers=(headers)
-      if headers && !headers.is_a?(Hash) && !headers.respond_to?(:call)
-        raise ArgumentError, 'headers must be a Hash or respond to #call'
-      end
-
-      @headers = headers.is_a?(Hash) ? headers.dup.freeze : headers
+      @headers = case headers
+                 when nil then nil
+                 when Hash then headers.dup.freeze
+                 when Proc then headers
+                 else
+                   raise ArgumentError, 'headers must be a Hash or Proc'
+                 end
     end
 
     ##
@@ -164,14 +179,11 @@ module Html2rss
 
     def normalize_log_level(level)
       if level.is_a?(Integer)
-        raise ArgumentError, "invalid log level: #{level}" unless level.between?(0, 5)
+        raise ArgumentError, "invalid log level: #{level}" unless LOG_LEVELS.value?(level)
 
         level
       else
-        sym = level.to_s.downcase.to_sym
-        raise ArgumentError, "invalid log level: #{level}" unless VALID_LOG_LEVELS.include?(sym)
-
-        Logger.const_get(sym.upcase)
+        LOG_LEVELS.fetch(level.to_s.downcase.to_sym) { raise ArgumentError, "invalid log level: #{level}" }
       end
     end
   end

data/lib/html2rss/doctor/botasaurus.rb

@@ -4,15 +4,15 @@ require 'httpx'
 require 'json'
 
 module Html2rss
-  ##
-  # CLI runtime health checks for optional scrape dependencies.
   module Doctor
     ##
     # Preflight checks for Botasaurus scrape-api connectivity.
     module Botasaurus
+      Check = Data.define(:name, :ok, :detail)
+
       ##
       # One named doctor check outcome.
-      Check = Data.define(:name, :ok, :detail) do
+      class Check
         ##
         # @return [Hash{Symbol => Object}]
         def to_h
@@ -20,9 +20,11 @@ module Html2rss
         end
       end
 
+      Result = Data.define(:ok, :checks, :message)
+
       ##
       # Aggregate doctor command outcome.
-      Result = Data.define(:ok, :checks, :message) do
+      class Result
         ##
         # @return [Hash{Symbol => Object}]
         def to_h

data/lib/html2rss/doctor.rb

@@ -0,0 +1,8 @@
+# frozen_string_literal: true
+
+module Html2rss
+  ##
+  # CLI runtime health checks for optional scrape dependencies.
+  module Doctor
+  end
+end

data/lib/html2rss/error.rb

@@ -7,7 +7,7 @@ module Html2rss
   # Raised when auto fallback exhausts all concrete tiers and extractors find no feed items.
   class NoFeedItemsExtracted < Error
     # Categories that append shared surface guidance to the empty-feed message.
-    SURFACE_HINT_CATEGORIES = %i[app_shell blocked_surface high_entropy_surface].freeze
+    SURFACE_HINT_CATEGORIES = Set[:app_shell, :blocked_surface, :high_entropy_surface].freeze
 
     ##
     # @param attempts [Array<Hash{Symbol => Object}>] tier attempt diagnostics

data/lib/html2rss/feed_builder/rss.rb

@@ -34,16 +34,10 @@ module Html2rss
         private
 
         def add_item_string_values(article, item_maker)
-          {
-            title: article.title,
-            description: ItemPresentation.description_for(article),
-            author: article.author
-          }.each do |attr, value|
-            next if value.nil? || value.empty?
-
-            item_maker.public_send(:"#{attr}=", value)
-          end
-
+          item_maker.title = article.title
+          item_maker.author = article.author
+          description = ItemPresentation.description_for(article)
+          item_maker.description = description if description && !description.empty?
           item_maker.link = article.url.to_s if article.url
           item_maker.pubDate = article.published_at&.rfc2822
         end
@@ -96,12 +90,13 @@ module Html2rss
         @stylesheets.map { |style| Stylesheet.new(**style) }
       end
 
-      # rubocop:disable-next Metrics/AbcSize
+      # rubocop:disable-next Metrics/AbcSize, Metrics/MethodLength
       def make_channel(maker)
         channel_maker = maker.channel
-        %i[language title description ttl].each do |key|
-          channel_maker.public_send(:"#{key}=", channel.public_send(key))
-        end
+        channel_maker.language = channel.language
+        channel_maker.title = channel.title
+        channel_maker.description = channel.description
+        channel_maker.ttl = channel.ttl
 
         channel_maker.managingEditor = channel.author if channel.author
         channel_maker.author = channel.author if channel.author

data/lib/html2rss/feed_pipeline/runtime_policy.rb

@@ -45,6 +45,20 @@ module Html2rss
       # @return [Html2rss::RequestService::Budget] shared budget for the feed build
       def self.budget_for(config) = resources_for(config).budget
 
+      ##
+      # Reserve enough HTTP request slots for the initial request plus predictable
+      # follow-ups.
+      #
+      # @param config [Html2rss::Config] validated feed config
+      # @return [Integer] number of reserved baseline request slots
+      def self.baseline_request_budget_for(config)
+        1 +
+          RequestSession::Pager.request_slots_for(config.selectors&.dig(:items, :pagination)) +
+          AutoSource.request_slots_for(config.auto_source) +
+          FeedResolution.request_slots_for(config.auto_source) +
+          StrategyPlan.resolve(config.strategy).request_slots
+      end
+
       class << self
         private
 
@@ -53,16 +67,6 @@ module Html2rss
 
           [baseline_request_budget_for(config), config.max_requests].max
         end
-
-        # Reserve enough HTTP request slots for the initial request plus predictable
-        # follow-ups.
-        def baseline_request_budget_for(config)
-          1 +
-            RequestSession::Pager.request_slots_for(config.selectors&.dig(:items, :pagination)) +
-            AutoSource.request_slots_for(config.auto_source) +
-            FeedResolution.request_slots_for(config.auto_source) +
-            StrategyPlan.resolve(config.strategy).request_slots
-        end
       end
     end
   end

data/lib/html2rss/feed_pipeline/strategy_plan.rb

@@ -8,16 +8,20 @@ module Html2rss
     # {RequestService} executes concrete adapters only. +:auto+ is resolved here before
     # any session or adapter call — plan vs transport locality.
     class StrategyPlan
+      Auto = Data.define
+
       # Feed-level auto plan: try {AutoFallback::CHAIN} until items are extracted.
-      Auto = Data.define do
+      class Auto
         # @return [Integer]
         def request_slots
           [AutoFallback::CHAIN.size - 1, 0].max
         end
       end
 
+      Concrete = Data.define(:strategy)
+
       # Concrete transport strategy name for {RequestService}.
-      Concrete = Data.define(:strategy) do
+      class Concrete
         # @return [Integer]
         def request_slots = 0
       end

data/lib/html2rss/feed_resolution/candidate_generator.rb

@@ -99,11 +99,8 @@ module Html2rss
 
       def segment_urls_for(sst, strategy, link_resolver)
         segmenter = AutoSource::Segmenter.new(sst, base_url: entry_url, strategy:, link_resolver:)
-        primary = AutoSource::Segmenter::PrimaryLink.new(segmenter)
-
         segmenter.call.filter_map do |segment|
-          link = segment.primary_link || primary.select(segment.root_node)
-          absolute_same_origin(link&.attrs&.href)
+          absolute_same_origin(segment.primary_link&.attrs&.href)
         end.uniq
       end
 

data/lib/html2rss/feed_resolution/diag.rb

@@ -2,9 +2,11 @@
 
 module Html2rss
   module FeedResolution
+    Diag = Data.define(:applied, :probe_count, :reason, :winner_score)
+
     ##
     # Wire-safe diagnostics for one entry-resolution attempt (Status / Marshal edge).
-    Diag = Data.define(:applied, :probe_count, :reason, :winner_score) do
+    class Diag
       ##
       # @param result [Html2rss::FeedResolution::Result]
       # @return [Diag]

data/lib/html2rss/feed_resolution/options.rb

@@ -2,9 +2,11 @@
 
 module Html2rss
   module FeedResolution
+    Options = Data.define(:enabled, :max_probes)
+
     ##
     # Typed +auto_source.entry_resolution+ options (one expansion from config Hash).
-    Options = Data.define(:enabled, :max_probes) do
+    class Options
       ##
       # @param auto_source [Hash, nil]
       # @return [Options]

data/lib/html2rss/feed_resolution/policy.rb

@@ -13,13 +13,13 @@ module Html2rss
       ##
       # @param config [Html2rss::Config]
       # @param articles [Array<Html2rss::Article>] typed extract results (count derived)
-      # @param surface_category [Html2rss::SurfaceCategory, Symbol, nil]
+      # @param surface_category [Html2rss::SurfaceCategory, Symbol]
       # @return [Boolean]
       def resolve?(config:, articles:, surface_category:)
         raise ArgumentError, 'articles must be an Array' unless articles.is_a?(Array)
         return false unless eligible_config?(config)
 
-        category = SurfaceCategory.coerce(surface_category)
+        category = SurfaceCategory[surface_category]
         return false if category.blocked?
 
         articles.size < ARTICLE_FLOOR || category.weak? || native_feed_majority?(articles)

data/lib/html2rss/html/article_extractor/date_extractor.rb

@@ -10,7 +10,7 @@ module Html2rss
         # @param time_zone [String] channel time zone for naive values
         # @return [DateTime, nil]
         def self.call(article_tag, leftover_lines: [], time_zone: 'UTC')
-          datetimes = article_tag.respond_to?(:css) ? article_tag.css('[datetime]').map { |tag| tag['datetime'] } : []
+          datetimes = article_tag.css('[datetime]').map { |tag| tag['datetime'] }
           ArticleRules::Date.earliest(datetimes, leftover_lines:, time_zone:)
         end
       end

data/lib/html2rss/html/article_extractor.rb

@@ -124,8 +124,7 @@ module Html2rss
       end
 
       def leftover_exclude_nodes_for(node)
-        times = node.respond_to?(:css) ? node.css('time') : []
-        [heading, selected_anchor, kicker_node, *times].compact
+        [heading, selected_anchor, kicker_node, *node.css('time')].compact
       end
 
       def heading_or_anchor_item?
@@ -156,8 +155,7 @@ module Html2rss
       end
 
       def immediate_card_parent
-        parent = article_tag.respond_to?(:parent) ? article_tag.parent : nil
-        Navigator.parent_until_condition(parent, method(:usable_walk_parent?))
+        Navigator.parent_until_condition(article_tag.parent, ->(node) { usable_walk_parent?(node) })
       end
 
       def usable_walk_parent?(node)

data/lib/html2rss/html/feed_link.rb

@@ -2,10 +2,12 @@
 
 module Html2rss
   module Html
+    FeedLink = Data.define(:href, :mime_type)
+
     # A native RSS/Atom hint from a document head +link[rel=alternate]+.
     #
     # Does not guess +/feed+ or +/rss.xml+ paths — that stays in configs +probe_rss+.
-    FeedLink = Data.define(:href, :mime_type) do
+    class FeedLink
       class << self
         ##
         # Collects RSS/Atom +rel=alternate+ links from +doc+'s +head+.

data/lib/html2rss/html/navigator/text_extractor.rb

@@ -19,7 +19,7 @@ module Html2rss
           # @param exclude_nodes [Array<Nokogiri::XML::Node>, nil] nodes to exclude from extraction
           # @return [String, nil] the concatenated visible text, or nil if none is found
           def call(tag, separator: ' ', exclude_nodes: nil)
-            return tag.text.gsub(/\s+/, ' ').strip if tag.respond_to?(:text?) && tag.text?
+            return tag.text.gsub(/\s+/, ' ').strip if tag.is_a?(Nokogiri::XML::Text)
 
             parts = iterate_children(tag, separator, exclude_nodes)
             return if parts.empty?

data/lib/html2rss/html/navigator.rb

@@ -6,16 +6,16 @@ module Html2rss
     # Navigator owns DOM chrome recognition and node traversal helpers.
     class Navigator
       # Heading tags used to prioritize title extraction and container assessment.
-      HEADING_TAGS = %w[h1 h2 h3 h4 h5 h6].freeze
+      HEADING_TAGS = Set['h1', 'h2', 'h3', 'h4', 'h5', 'h6'].freeze
 
       # Element tags that indicate ignored DOM chrome when found in a container path.
-      IGNORED_CONTAINER_TAGS = %w[nav footer header svg script style].to_set.freeze
+      IGNORED_CONTAINER_TAGS = Set['nav', 'footer', 'header', 'svg', 'script', 'style'].freeze
 
       # Ancestor tags that usually indicate navigation/utility regions inside a content container.
-      UTILITY_LANDMARK_TAGS = %w[nav aside footer menu].to_set.freeze
+      UTILITY_LANDMARK_TAGS = Set['nav', 'aside', 'footer', 'menu'].freeze
 
       # Immediate parent walk stops here — not a usable article card.
-      CARD_WALK_STOP_TAGS = (UTILITY_LANDMARK_TAGS | IGNORED_CONTAINER_TAGS | %w[html body]).freeze
+      CARD_WALK_STOP_TAGS = (UTILITY_LANDMARK_TAGS | IGNORED_CONTAINER_TAGS | Set['html', 'body']).freeze
 
       # Inner tags that mean a wrapping <a> is a card, not a span-styled list link.
       WRAPPING_ANCHOR_CHILD_TAGS = (HEADING_TAGS + %w[p]).freeze
@@ -58,7 +58,7 @@ module Html2rss
         # @param condition [Proc] The condition to be met.
         # @return [Nokogiri::XML::Node, nil] The first parent that satisfies the condition.
         def parent_until_condition(node, condition)
-          while node && !node.document? && Probe.tag(node) != 'html'
+          while node && !node.is_a?(Nokogiri::XML::Document) && Probe.tag(node) != 'html'
             return node if condition.call(node)
 
             node = node.parent
@@ -72,7 +72,7 @@ module Html2rss
         # @return [Boolean]
         def usable_card_parent?(node)
           return false unless node
-          return false if node.respond_to?(:document?) && node.document?
+          return false if node.is_a?(Nokogiri::XML::Document)
 
           !CARD_WALK_STOP_TAGS.include?(Probe.tag(node))
         end
@@ -88,8 +88,7 @@ module Html2rss
           while current_tag
             found = current_tag.at_css(selector)
             return found if found
-
-            return nil unless current_tag.respond_to?(:parent)
+            break if current_tag.is_a?(Nokogiri::XML::Document)
 
             current_tag = current_tag.parent
           end
@@ -115,11 +114,12 @@ module Html2rss
         # @param parent_node [Nokogiri::XML::Node] potential ancestor
         # @return [Boolean] true when child_node is a descendant of parent_node
         def descendant_of?(child_node, parent_node)
-          curr = child_node.respond_to?(:parent) ? child_node.parent : nil
+          curr = child_node.is_a?(Nokogiri::XML::Document) ? nil : child_node.parent
           while curr
             return true if curr == parent_node
+            break if curr.is_a?(Nokogiri::XML::Document)
 
-            curr = curr.respond_to?(:parent) ? curr.parent : nil
+            curr = curr.parent
           end
           false
         end

data/lib/html2rss/html/rendering/description_builder.rb

@@ -81,7 +81,7 @@ module Html2rss
 
         def processed_base_description
           text = self.class.remove_pattern_from_start(@base, @title)
-          Html2rss::Selectors::PostProcessors::SanitizeHtml.get(text, @url)
+          Html2rss::Selectors::PostProcessors::SanitizeHtml.call(text, @url)
         end
       end
     end

data/lib/html2rss/html/sst_article_extractor.rb

@@ -12,9 +12,9 @@ module Html2rss
       # CSS class tokens that mark kicker / eyebrow text (excluded from titles).
       KICKER_CLASS_PATTERN = /kicker|eyebrow|pre-title|pretitle|overline/i
       # Inline emphasis tags used as title fallbacks when no heading exists.
-      FALLBACK_HEADING_NAMES = %i[strong b].freeze
+      FALLBACK_HEADING_NAMES = Set[:strong, :b].freeze
       # Nested blocks that mean a category node is actually a content container.
-      CATEGORY_CONTAINER_NAMES = %i[p article section].to_set.freeze
+      CATEGORY_CONTAINER_NAMES = Set[:p, :article, :section].freeze
 
       class << self
         ##
@@ -175,10 +175,13 @@ module Html2rss
       end
 
       def descendant_of?(child, ancestor)
-        # Walk via root children since Extractor may not hold Index; use tag walk on tree.
         return true if child.equal?(ancestor)
 
-        ancestor.descendants.any? { |d| d.equal?(child) }
+        if (index = SST::Index.for_node(child))
+          return index.descendant_of?(child, ancestor)
+        end
+
+        !ancestor.find { |d| d.equal?(child) }.nil?
       end
 
       def leftover_lines
@@ -220,7 +223,7 @@ module Html2rss
       def immediate_card_parent
         parent = sst_parent
         index = SST::Index.for_node(@root)
-        index&.parent_until(parent, method(:usable_walk_parent?))
+        index&.parent_until(parent, ->(node) { usable_walk_parent?(node) })
       end
 
       def usable_walk_parent?(node)

data/lib/html2rss/link_destination/destination_facts.rb

@@ -2,7 +2,6 @@
 
 module Html2rss
   module LinkDestination
-    # Normalized URL plus reusable route-classification facts for one link.
     DestinationFacts = Data.define(
       :url,
       :destination,
@@ -15,7 +14,10 @@ module Html2rss
       :strong_post_suffix,
       :high_confidence_junk_path,
       :high_confidence_utility_destination
-    ) do
+    )
+
+    # Normalized URL plus reusable route-classification facts for one link.
+    class DestinationFacts
       # @param url [Html2rss::Url] normalized destination URL
       # @return [DestinationFacts] route facts for downstream link scoring
       def self.build(url) # rubocop:disable Metrics/MethodLength

data/lib/html2rss/link_destination/noise_policy.rb

@@ -42,7 +42,16 @@ module Html2rss
         destination_facts.utility_path &&
           !destination_facts.content_path &&
           !destination_facts.strong_post_suffix &&
-          text.to_s.scan(/\p{Alnum}+/).size <= 3
+          word_count_at_most?(text, 3)
+      end
+
+      def word_count_at_most?(text, max)
+        count = 0
+        text.to_s.scan(/\p{Alnum}+/) do
+          count += 1
+          return false if count > max
+        end
+        true
       end
 
       def recommended_chrome?(text, destination_facts, heading_anchor:)