html2rss 0.32.4 → 0.33.0

data/lib/html2rss/url.rb

@@ -26,6 +26,9 @@ module Html2rss
   class Url
     include Comparable
 
+    # Raised when a string cannot be parsed or resolved to a valid URL.
+    class InvalidUrlError < Html2rss::Error; end
+
     # Regular expression for basic URI format validation
     URI_REGEXP = Addressable::URI::URIREGEX
     # Schemes accepted by channel URL validation.
@@ -37,22 +40,21 @@ module Html2rss
     # @param relative_url [String, Html2rss::Url] the relative URL to resolve
     # @param base_url [String, Html2rss::Url] the base URL to resolve against
     # @return [Url] the resolved absolute URL
-    # @raise [ArgumentError] if the URL cannot be parsed
+    # @raise [InvalidUrlError] if the URL cannot be parsed
     def self.from_relative(relative_url, base_url)
       url = Addressable::URI.parse(relative_url.to_s.strip)
       return new(url) if url.absolute?
 
-      base_url.is_a?(self) ? base_url.join_uri(url) : join_parsed_base(base_url, url)
+      join_parsed_base(base_url, url)
     rescue Addressable::URI::InvalidURIError
-      raise ArgumentError, 'URL could not be parsed'
+      raise InvalidUrlError, 'URL could not be parsed'
     end
 
-    # @param base_url [String]
+    # @param base_url [String, Url]
     # @param url [Addressable::URI]
     # @return [Url]
     def self.join_parsed_base(base_url, url)
       uri = Addressable::URI.parse(base_url.to_s)
-      uri.path = '/' if uri.path.empty?
       new(uri.join(url).normalize)
     end
     private_class_method :join_parsed_base
@@ -64,10 +66,10 @@ module Html2rss
     # @param raw_url [String] the raw URL string to sanitize
     # @return [Url, nil] the sanitized URL, or nil if no valid URL found
     def self.sanitize(raw_url)
-      match = raw_url.to_s.match(%r{(?:(?:https?|ftp|mailto)://|mailto:)[^\s<>"]+})
+      match = raw_url.to_s.match(%r{(?:(?:https?|ftp)://|mailto:)[^\s<>"]+})
       return unless match
 
-      new(Addressable::URI.parse(match[0].strip).normalize)
+      new(Addressable::URI.parse(match[0]).normalize)
     end
 
     ##
@@ -75,15 +77,15 @@ module Html2rss
     #
     # @param url_string [String, Html2rss::Url] the absolute URL to parse
     # @return [Url] the parsed and normalized URL
-    # @raise [ArgumentError] if the URL is not absolute or cannot be parsed
+    # @raise [InvalidUrlError] if the URL is not absolute or cannot be parsed
     def self.from_absolute(url_string)
       return url_string if url_string.is_a?(self)
 
-      new(Addressable::URI.parse(url_string.to_s.strip).normalize).tap do |url|
-        raise ArgumentError, 'URL must be absolute' unless url.absolute?
+      new(Addressable::URI.parse(url_string.to_s.lstrip).normalize).tap do |url|
+        raise InvalidUrlError, 'URL must be absolute' unless url.absolute?
       end
     rescue Addressable::URI::InvalidURIError
-      raise ArgumentError, 'URL must be absolute'
+      raise InvalidUrlError, 'URL must be absolute'
     end
 
     ##
@@ -92,20 +94,30 @@ module Html2rss
     #
     # @param url_string [String] the URL string to validate and parse
     # @return [Url] the validated and parsed URL
-    # @raise [ArgumentError] if the URL doesn't meet channel requirements
+    # @raise [InvalidUrlError] if the URL doesn't meet channel requirements
     # @example Creating a channel URL
     #   Url.for_channel('https://example.com')
     #   # => #<Html2rss::Url:... @uri=#<Addressable::URI:... URI:https://example.com>>
     # @example Invalid channel URL
     #   Url.for_channel('/relative/path')
-    #   # => raises ArgumentError: "URL must be absolute"
+    #   # => raises InvalidUrlError: "URL must be absolute"
     def self.for_channel(url_string)
-      stripped = url_string.to_s.strip
+      stripped = url_string.to_s.lstrip
       return nil if stripped.empty?
 
       from_absolute(stripped).tap { validate_channel_url(_1) }
     end
 
+    ##
+    # Creates an absolute http(s) URL, for example for a channel image.
+    #
+    # @param url_string [String, Html2rss::Url] the URL to validate and parse
+    # @return [Url] the validated and parsed URL
+    # @raise [InvalidUrlError] if the URL is not absolute or not http(s)
+    def self.from_absolute_web(url_string)
+      from_absolute(url_string).tap { validate_scheme(_1) }
+    end
+
     ##
     # Extracts a base URL from HTML metadata tags.
     #
@@ -116,10 +128,10 @@ module Html2rss
       tags = { 'link[rel="canonical"]' => 'href', 'meta[property="og:url"]' => 'content',
                'meta[name="twitter:url"]' => 'content', 'base[href]' => 'href' }
       tags.each do |sel, attr|
-        val = doc.at_css(sel)&.[](attr).to_s.strip
-        return from_absolute(val) unless val.empty?
-      rescue ArgumentError
-        next
+        val = doc.at_css(sel)&.[](attr)
+        return from_absolute(val)
+      rescue InvalidUrlError
+        nil
       end
       nil
     end
@@ -128,31 +140,32 @@ module Html2rss
     # Validates that a URL meets channel requirements.
     #
     # @param url [Url] the URL to validate
-    # @raise [ArgumentError] if the URL doesn't meet channel requirements
+    # @raise [InvalidUrlError] if the URL doesn't meet channel requirements
     def self.validate_channel_url(url)
-      raise ArgumentError, 'URL must be absolute' unless url.absolute?
+      has_forbidden_at = url.user
+      has_forbidden_at ||= [url.query, url.fragment].compact.any? { |value| value.include?('@') }
+      raise InvalidUrlError, 'URL must not contain an @ character' if has_forbidden_at
 
-      uri = Addressable::URI.parse(url.to_s)
-      has_forbidden_at = uri.user || uri.password
-      has_forbidden_at ||= [uri.query, uri.fragment].compact.any? { |value| value.include?('@') }
-      raise ArgumentError, 'URL must not contain an @ character' if has_forbidden_at
+      validate_scheme(url)
+    end
 
+    def self.validate_scheme(url)
       scheme = url.scheme
-      raise ArgumentError, "URL scheme '#{scheme}' is not supported" unless SUPPORTED_SCHEMES.include?(scheme)
+      raise InvalidUrlError, "URL scheme '#{scheme}' is not supported" unless SUPPORTED_SCHEMES.include?(scheme)
     end
 
-    private_class_method :validate_channel_url
+    private_class_method :validate_channel_url, :validate_scheme
 
     ##
     # @param uri [Addressable::URI] the underlying Addressable::URI object (internal use only)
     def initialize(uri)
       @uri = uri.freeze
-      @path_segments = @uri.path.to_s.split('/').reject(&:empty?).freeze
+      @path_segments = path.split('/').reject(&:empty?).freeze
       freeze
     end
 
     # @return [String] normalized URL string
-    def to_s = @uri.to_s
+    def to_s = @uri.to_str
 
     # @return [String, nil] URI scheme, for example `http` or `https`
     def scheme = @uri.scheme
@@ -160,6 +173,9 @@ module Html2rss
     # @return [String, nil] URI host component
     def host = @uri.host
 
+    # @return [String, nil] userinfo user, without the password
+    def user = @uri.user
+
     # Registrable domain (eTLD+1). Same-site hosts like www.wp.pl and wiadomosci.wp.pl
     # share `wp.pl`. IP literals stay the host so 1.2.3.4 is not treated as `3.4`.
     #
@@ -189,7 +205,7 @@ module Html2rss
     # Returns the URL query string as a hash of string keys and values.
     #
     # @return [Hash{String => String}] normalized query parameters
-    def query_values = @uri.query_values(Hash) || {}
+    def query_values = @uri.query_values || {}
 
     ##
     # Returns the URL path split into non-empty segments.
@@ -205,7 +221,7 @@ module Html2rss
     def with_path(path)
       uri = @uri.dup
       uri.path = path
-      self.class.from_absolute(uri.normalize.to_s)
+      self.class.from_absolute(uri)
     end
 
     ##
@@ -213,7 +229,7 @@ module Html2rss
     #
     # @return [Url] this URL when already fragment-free; otherwise a new URL
     def without_fragment
-      @uri.fragment ? self.class.from_absolute(@uri.omit(:fragment).normalize.to_s) : self
+      fragment ? self.class.from_absolute(@uri.omit(:fragment)) : self
     end
 
     ##
@@ -224,7 +240,7 @@ module Html2rss
     def with_query_values(values)
       uri = @uri.dup
       uri.query_values = values.transform_keys(&:to_s).transform_values(&:to_s)
-      self.class.from_absolute(uri.normalize.to_s)
+      self.class.from_absolute(uri)
     end
 
     ##
@@ -240,15 +256,7 @@ module Html2rss
     #   url = Url.from_absolute('https://example.com/hello%20world/article.html')
     #   url.titleized # => "Hello World Article"
     def titleized
-      path = @uri.path
-      return '' if path.empty?
-
-      nicer_path = CGI.unescapeURIComponent(path)
-                      .split('/')
-                      .flat_map do |part|
-                        part.gsub(/[^a-zA-Z0-9.]/, ' ').gsub(/\s+/, ' ').split
-                      end
-
+      nicer_path = CGI.unescapeURIComponent(path).gsub(/[^a-zA-Z0-9.]/, ' ').split
       nicer_path.map!(&:capitalize)
       File.basename(nicer_path.join(' '), '.*')
     end
@@ -266,8 +274,7 @@ module Html2rss
     #   url = Url.from_absolute('https://example.com')
     #   url.channel_titleized # => "example.com"
     def channel_titleized
-      nicer_path = CGI.unescapeURIComponent(@uri.path).split('/').reject(&:empty?)
-      host = @uri.host
+      nicer_path = CGI.unescapeURIComponent(path).split('/').reject(&:empty?)
 
       nicer_path.any? ? "#{host}: #{nicer_path.map(&:capitalize).join(' ')}" : host
     end
@@ -318,12 +325,7 @@ module Html2rss
     # @param relative_uri [Addressable::URI]
     # @return [Url]
     def join_uri(relative_uri)
-      base = @uri
-      if base.path.empty?
-        base = base.dup
-        base.path = '/'
-      end
-      self.class.new(base.join(relative_uri).normalize)
+      self.class.new(@uri.join(relative_uri).normalize)
     end
 
     protected
@@ -332,15 +334,13 @@ module Html2rss
     #
     # @return [String]
     def document_identity
-      uri = @uri.omit(:fragment)
-      "#{uri.scheme}\0#{uri.authority}\0#{uri.path.to_s.chomp('/')}\0#{uri.query}"
+      "#{scheme}\0#{@uri.authority}\0#{path.chomp('/')}\0#{query}"
     end
 
     private
 
     def ip_address_host?
-      IPAddr.new(host.to_s)
-      true
+      !IPAddr.new(host).nil?
     rescue IPAddr::Error
       false
     end

data/lib/html2rss/version.rb

@@ -2,6 +2,6 @@
 
 module Html2rss
   # Current application version.
-  VERSION = '0.32.4'
+  VERSION = '0.33.0'
   public_constant :VERSION
 end

data/lib/html2rss.rb

@@ -101,7 +101,7 @@ module Html2rss # rubocop:disable Metrics/ModuleLength
   #
   # @param config_input [Hash, String] config hash, YAML string, or file path
   # @param feed_name [String, nil] optional feed name in multi-feed file
-  # @param min_items [Integer] minimum required items (default: 1)
+  # @param min_items [Integer] minimum required items, >= 0 (default: 1)
   # @param params [Hash] dynamic feed params
   # @param strategy [Symbol, nil] optional strategy override
   # @param strict_quality [Boolean] when true, fail on ship-quality audit thresholds
@@ -109,6 +109,7 @@ module Html2rss # rubocop:disable Metrics/ModuleLength
   # @return [Html2rss::Test::Result]
   def self.test(config_input, feed_name = nil, min_items: 1, params: {}, strategy: nil, strict_quality: false, # rubocop:disable Metrics/ParameterLists
                 compare_enhance: false)
+    Option.integer(:min_items, min_items, min: 0)
     Test.call(config_input, feed_name, min_items:, params:, strategy:, strict_quality:, compare_enhance:)
   end
 

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: html2rss
 version: !ruby/object:Gem::Version
-  version: 0.32.4
+  version: 0.33.0
 platform: ruby
 authors:
 - Gil Desmarais
@@ -391,6 +391,7 @@ files:
 - lib/html2rss/mcp/runtime.rb
 - lib/html2rss/mcp/server.rb
 - lib/html2rss/mcp/server/tools.rb
+- lib/html2rss/option.rb
 - lib/html2rss/page_recon.rb
 - lib/html2rss/page_recon/README.md
 - lib/html2rss/page_recon/diagnostics.rb
@@ -463,6 +464,7 @@ files:
 - lib/html2rss/surface_category.rb
 - lib/html2rss/syndication.rb
 - lib/html2rss/syndication/README.md
+- lib/html2rss/syndication/alternate_links.rb
 - lib/html2rss/syndication/candidate_catalog.rb
 - lib/html2rss/syndication/discovery.rb
 - lib/html2rss/syndication/parser.rb
@@ -478,7 +480,7 @@ licenses:
 - MIT
 metadata:
   allowed_push_host: https://rubygems.org
-  changelog_uri: https://github.com/html2rss/html2rss/releases/tag/v0.32.4
+  changelog_uri: https://github.com/html2rss/html2rss/releases/tag/v0.33.0
   rubygems_mfa_required: 'true'
 rdoc_options: []
 require_paths: