html2rss 0.11.0 → 0.12.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 3c8df87caea7cc1dd9855993e9dd4fcb4285b37a5c1f87fb48a3d85e75ce0a51
-  data.tar.gz: 8ab432f13d76f45e30ea65316669c324bcba46b5234a082a6f0670cc445a6958
+  metadata.gz: ebe536d8051a64c6e2adf9fa8e1d9d1f9fa3743541c44ca85022d0603f9032b2
+  data.tar.gz: 7b3aaa213aaf6a37fb6e94fa72c9936ffd2391322297553b253b097edea300cc
 SHA512:
-  metadata.gz: 7115ee36e47ac65d19b13bb0f2ef0b19ebdb387d2e6fa53128e804b8d22d6a834517b88642f5c3e31e8850e344a2f94b6cb0de6619504603cfeacae52e47f3e2
-  data.tar.gz: b2c9f342cff237d9cafd4817f2570d18803fa9ba844ab306681ac02d12fb7cf147f3b40117ab21c78937325d80623ad27e25c6df4c8f8150fe81be2a07d12817
+  metadata.gz: '03985002d050b996c1dc315cbe8e3fc79b6619447a048ad3d2dca86f792eab5c2356716cf6198a24efc61de7e7ddceba2780da49c3e68a3c9efe895eb7cf0cf1'
+  data.tar.gz: 8315473528f46a5ba28297af296b879a66ac00f86ba9eb117b4e6c9ec61c285e4090cfd999ff712368f5b988b1cbda460e268aa3ea8928912bcdb1960ae25a4a

data/lib/html2rss/attribute_post_processors/base.rb

@@ -0,0 +1,71 @@
+# frozen_string_literal: true
+
+module Html2rss
+  ##
+  # Provides a namespace for attribute post processors.
+  module AttributePostProcessors
+    ##
+    # All post processors must inherit from this base class and implement `self.validate_args!` and `#get`.
+    class Base
+      # Validates the presence of required options in the context
+      #
+      # @param keys [Array<Symbol>] the keys to check for presence
+      # @param context [Hash] the context containing options
+      # @raise [MissingOption] if any key is missing
+      def self.expect_options(keys, context)
+        keys.each do |key|
+          unless (options = context[:options]).key?(key)
+            raise MissingOption, "The `#{key}` option is missing in: #{options.inspect}", [],
+                  cause: nil
+          end
+        end
+      end
+
+      # Asserts that the value is of the expected type(s)
+      #
+      # @param value [Object] the value to check
+      # @param types [Array<Class>, Class] the expected type(s)
+      # @param name [String] the name of the option being checked
+      # @raise [InvalidType] if the value is not of the expected type(s)
+      def self.assert_type(value, types = [], name)
+        types = [types] unless types.is_a?(Array)
+
+        return if types.any? { |type| value.is_a?(type) }
+
+        error_message_template = 'The type of `%s` must be %s, but is: %s'
+        raise InvalidType, format(error_message_template, name, types.join(' or '), value.class), [], cause: nil
+      end
+
+      # private_class_method :expect_options, :assert_type
+
+      ##
+      # This method validates the arguments passed to the post processor. Must be implemented by subclasses.
+      def self.validate_args!(_value, _context)
+        raise NotImplementedError, 'You must implement the `validate_args!` method in the post processor'
+      end
+
+      # Initializes the post processor
+      #
+      # @param value [Object] the value to be processed
+      # @param context [Item::Context] the context
+      def initialize(value, context)
+        klass = self.class
+        # TODO: get rid of Hash
+        klass.assert_type(context, [Item::Context, Hash], 'context')
+        klass.validate_args!(value, context)
+
+        @value = value
+        @context = context
+      end
+
+      attr_reader :value, :context
+
+      # Abstract method to be implemented by subclasses
+      #
+      # @raise [NotImplementedError] if not implemented in subclass
+      def get
+        raise NotImplementedError, 'You must implement the `get` method in the post processor'
+      end
+    end
+  end
+end

data/lib/html2rss/attribute_post_processors/gsub.rb

@@ -25,39 +25,39 @@ module Html2rss
     # `replacement` can be a String or a Hash.
     #
     # See the doc on [String#gsub](https://ruby-doc.org/core/String.html#method-i-gsub) for more info.
-    class Gsub
+    class Gsub < Base
+      def self.validate_args!(value, context)
+        assert_type value, String, :value
+        expect_options(%i[replacement pattern], context)
+        assert_type context.dig(:options, :replacement), [String, Hash], :replacement
+      end
+
       ##
       # @param value [String]
       # @param context [Item::Context]
       def initialize(value, context)
-        @value = value
-        @options = context[:options]
+        super
+
+        options = context[:options]
+
+        @replacement = options[:replacement]
+        @pattern = options[:pattern]
       end
 
       ##
       # @return [String]
       def get
-        @value.to_s.gsub(pattern, replacement)
+        value.to_s.gsub(pattern, replacement)
       end
 
       private
 
+      attr_accessor :replacement
+
       ##
       # @return [Regexp]
       def pattern
-        pattern = @options[:pattern]
-        raise ArgumentError, 'The `pattern` option is missing' unless pattern
-
-        pattern.is_a?(String) ? Utils.build_regexp_from_string(pattern) : pattern
-      end
-
-      ##
-      # @return [Hash, String]
-      def replacement
-        replacement = @options[:replacement]
-        return replacement if replacement.is_a?(String) || replacement.is_a?(Hash)
-
-        raise ArgumentError, 'The `replacement` option must be a String or Hash'
+        @pattern.is_a?(String) ? Utils.build_regexp_from_string(@pattern) : @pattern
       end
     end
   end

data/lib/html2rss/attribute_post_processors/html_to_markdown.rb

@@ -26,18 +26,17 @@ module Html2rss
     #
     # Would return:
     #    'Lorem **ipsum** dolor'
-    class HtmlToMarkdown
-      ##
-      # @param value [String]
-      # @param env [Item::Context]
-      def initialize(value, env)
-        @sanitized_value = SanitizeHtml.new(value, env).get
+    class HtmlToMarkdown < Base
+      def self.validate_args!(value, _context)
+        assert_type value, String, :value
       end
 
       ##
       # @return [String] formatted in Markdown
       def get
-        ReverseMarkdown.convert(@sanitized_value)
+        sanitized_value = SanitizeHtml.new(value, context).get
+
+        ReverseMarkdown.convert(sanitized_value)
       end
     end
   end

data/lib/html2rss/attribute_post_processors/markdown_to_html.rb

@@ -32,13 +32,9 @@ module Html2rss
     #    <h1>Section</h1>
     #
     #    <p>Price: 12.34</p>
-    class MarkdownToHtml
-      ##
-      # @param value [String] Markdown content to convert to HTML
-      # @param env [Item::Context] Context object providing additional environment details
-      def initialize(value, env)
-        @value = value
-        @env = env
+    class MarkdownToHtml < Base
+      def self.validate_args!(value, _context)
+        assert_type value, String, :value
       end
 
       ##
@@ -46,8 +42,8 @@ module Html2rss
       #
       # @return [String] Sanitized HTML content
       def get
-        html_content = Kramdown::Document.new(@value).to_html
-        SanitizeHtml.new(html_content, @env).get
+        html_content = Kramdown::Document.new(value).to_html
+        SanitizeHtml.new(html_content, context).get
       end
     end
   end

data/lib/html2rss/attribute_post_processors/parse_time.rb

@@ -24,22 +24,22 @@ module Html2rss
     # Would return:
     #    "Tue, 02 Jul 2019 00:00:00 +0200"
     #
-    # It uses {https://ruby-doc.org/stdlib-2.5.3/libdoc/time/rdoc/Time.html#method-c-parse Time.parse}.
-    class ParseTime
-      ##
-      # @param value [String] the time to parse
-      # @param env [Item::Context] Context object providing additional environment details
-      def initialize(value, env)
-        @value = value.to_s
-        @time_zone = env[:config].time_zone
+    # It uses `Time.parse`.
+    class ParseTime < Base
+      def self.validate_args!(value, context)
+        assert_type value, String, :value
+        assert_type context[:config].time_zone, String, :time_zone
       end
 
       ##
-      # Converts the provided time string to RFC822 format, taking into account the configured time zone.
+      # Converts the provided time string to RFC822 format, taking into account the time_zone.
       #
       # @return [String] RFC822 formatted time
+      # @raise [TZInfo::InvalidTimezoneIdentifier] if the configured time zone is invalid
       def get
-        Utils.use_zone(@time_zone) { Time.parse(@value).rfc822 }
+        time_zone = context[:config].time_zone
+
+        Utils.use_zone(time_zone) { Time.parse(value).rfc822 }
       end
     end
   end

data/lib/html2rss/attribute_post_processors/parse_uri.rb

@@ -21,21 +21,24 @@ module Html2rss
     #
     # Would return:
     #    'http://why-not-use-a-link.uh'
-    class ParseUri
-      ##
-      # @param value [String]
-      # @param context [Item::Context]
-      def initialize(value, context)
-        @value = value
-        @config_url = context.config.url
+    class ParseUri < Base
+      def self.validate_args!(value, context)
+        url_types = [String, URI::HTTP, Addressable::URI].freeze
+
+        assert_type(value, url_types, :value)
+        assert_type(context.config.url, url_types, :url)
+
+        raise ArgumentError, 'The `value` option is missing or empty.' if value.to_s.empty?
       end
 
       ##
       # @return [String]
       def get
+        config_url = context.config.url
+
         Html2rss::Utils.build_absolute_url_from_relative(
-          Html2rss::Utils.sanitize_url(@value),
-          @config_url
+          Html2rss::Utils.sanitize_url(value),
+          config_url
         ).to_s
       end
     end

data/lib/html2rss/attribute_post_processors/sanitize_html.rb

@@ -38,19 +38,15 @@ module Html2rss
     #
     # Would return:
     #    '<p>Lorem <b>ipsum</b> dolor ...</p>'
-    class SanitizeHtml
-      ##
-      # @param value [String]
-      # @param env [Item::Context]
-      def initialize(value, env)
-        @value = value
-        @channel_url = env[:config].url
+    class SanitizeHtml < Base
+      def self.validate_args!(value, _context)
+        assert_type value, String, :value
       end
 
       ##
       # @return [String]
       def get
-        sanitized_html = Sanitize.fragment(@value, sanitize_config)
+        sanitized_html = Sanitize.fragment(value, sanitize_config)
         sanitized_html.to_s.gsub(/\s+/, ' ').strip
       end
 
@@ -77,13 +73,15 @@ module Html2rss
         }
       end
 
+      def channel_url = context[:config].url
+
       ##
       # Wrapper for transform_urls_to_absolute_ones to pass the channel_url.
       #
       # @param env [Hash]
       # @return [nil]
       def transform_urls_to_absolute_ones(env)
-        HtmlTransformers::TransformUrlsToAbsoluteOnes.new(@channel_url).call(**env)
+        HtmlTransformers::TransformUrlsToAbsoluteOnes.new(channel_url).call(**env)
       end
 
       ##

data/lib/html2rss/attribute_post_processors/substring.rb

@@ -28,13 +28,15 @@ module Html2rss
     #
     # Would return:
     #    'bar'
-    class Substring
-      ##
-      # @param value [String] The original string to extract a substring from.
-      # @param env [Item::Context] Context object providing additional environment details.
-      def initialize(value, env)
-        @value = value
-        @options = env[:options]
+    class Substring < Base
+      def self.validate_args!(value, context)
+        assert_type value, String, :value
+
+        options = context[:options]
+        assert_type options[:start], Integer, :start
+
+        end_index = options[:end]
+        assert_type end_index, Integer, :end if end_index
       end
 
       ##
@@ -42,11 +44,29 @@ module Html2rss
       #
       # @return [String] The extracted substring.
       def get
-        start_index = @options[:start].to_i
-        end_index = @options[:end]&.to_i || @value.length
+        value[range]
+      end
+
+      ##
+      # Determines the range for the substring extraction based on the provided start and end indices.
+      #
+      # @return [Range] The range object representing the start and end/Infinity (integers).
+      def range
+        return (start_index..) unless end_index?
+
+        if start_index == end_index
+          raise ArgumentError,
+                'The `start` value must be unequal to the `end` value.'
+        end
 
-        @value[start_index..end_index]
+        (start_index..end_index)
       end
+
+      private
+
+      def end_index?  = !context[:options][:end].to_s.empty?
+      def end_index   = context[:options][:end].to_i
+      def start_index = context[:options][:start].to_i
     end
   end
 end

data/lib/html2rss/attribute_post_processors/template.rb

@@ -31,15 +31,23 @@ module Html2rss
     #
     # Would return:
     #    'Product (23,42€)'
-    class Template
+    class Template < Base
+      def self.validate_args!(value, context)
+        assert_type value, String, :value
+
+        string = context[:options]&.dig(:string).to_s
+        raise InvalidType, 'The `string` template is absent.' if string.empty?
+      end
+
       ##
       # @param value [String]
-      # @param env [Item::Context]
-      def initialize(value, env)
-        @value = value
-        @options = env[:options]
-        @item = env[:item]
-        @string = @options[:string]
+      # @param context [Item::Context]
+      def initialize(value, context)
+        super
+
+        @options = context[:options] || {}
+        @item = context[:item]
+        @string = @options[:string].to_s
       end
 
       ##
@@ -86,7 +94,7 @@ module Html2rss
       # @param method_name [String, Symbol]
       # @return [String]
       def item_value(method_name)
-        method_name.to_sym == :self ? @value.to_s : @item.public_send(method_name).to_s
+        method_name.to_sym == :self ? value : @item.public_send(method_name).to_s
       end
     end
   end

data/lib/html2rss/attribute_post_processors.rb

@@ -8,6 +8,14 @@ module Html2rss
     # Error raised when an unknown post processor name is requested.
     class UnknownPostProcessorName < Html2rss::Error; end
 
+    ##
+    # Error raised when a required option is missing.
+    class MissingOption < Html2rss::Error; end
+
+    ##
+    # Error raised when an invalid type is provided.
+    class InvalidType < Html2rss::Error; end
+
     ##
     # Maps the post processor name to the class implementing the post processor.
     #

data/lib/html2rss/config/selectors.rb

@@ -10,6 +10,9 @@ module Html2rss
       # Struct to represent a selector with associated attributes for extraction and processing.
       Selector = Struct.new(:selector, :attribute, :extractor, :post_process, :order, :static, keyword_init: true)
 
+      # raised when an invalid selector name is used
+      class InvalidSelectorName < Html2rss::Error; end
+
       ##
       # @param config [Hash<Symbol, Object>]
       def initialize(config)
@@ -28,9 +31,15 @@ module Html2rss
       # @param name [Symbol]
       # @return [Selector]
       def selector(name)
-        raise ArgumentError, "invalid item's selector name: #{name}" unless selector?(name)
+        raise InvalidSelectorName, "invalid selector name: #{name}" unless selector?(name)
+
+        keywords = config[name].slice(*available_keys)
 
-        Selector.new(config[name])
+        if (additional_keys = available_keys - keywords.keys).any?
+          warn "additional keys (#{additional_keys.join(', ')}) present in selector #{name}"
+        end
+
+        Selector.new(keywords)
       end
 
       ##
@@ -86,6 +95,8 @@ module Html2rss
           array.map!(&:to_sym)
         end.to_set
       end
+
+      def available_keys = @available_keys ||= Selector.members
     end
   end
 end

data/lib/html2rss/version.rb

@@ -3,6 +3,6 @@
 ##
 # The Html2rss namespace.
 module Html2rss
-  VERSION = '0.11.0'
+  VERSION = '0.12.0'
   public_constant :VERSION
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: html2rss
 version: !ruby/object:Gem::Version
-  version: 0.11.0
+  version: 0.12.0
 platform: ruby
 authors:
 - Gil Desmarais
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2024-08-09 00:00:00.000000000 Z
+date: 2024-08-10 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: addressable
@@ -219,6 +219,7 @@ files:
 - html2rss.gemspec
 - lib/html2rss.rb
 - lib/html2rss/attribute_post_processors.rb
+- lib/html2rss/attribute_post_processors/base.rb
 - lib/html2rss/attribute_post_processors/gsub.rb
 - lib/html2rss/attribute_post_processors/html_to_markdown.rb
 - lib/html2rss/attribute_post_processors/html_transformers/transform_urls_to_absolute_ones.rb
@@ -252,7 +253,7 @@ licenses:
 - MIT
 metadata:
   allowed_push_host: https://rubygems.org
-  changelog_uri: https://github.com/html2rss/html2rss/releases/tag/v0.11.0
+  changelog_uri: https://github.com/html2rss/html2rss/releases/tag/v0.12.0
   rubygems_mfa_required: 'true'
 post_install_message: 
 rdoc_options: []