reading 0.7.0 → 0.9.0

data/lib/reading/parsing/rows/blank.rb

@@ -0,0 +1,23 @@
+module Reading
+  module Parsing
+    module Rows
+      # A row that is a blank line.
+      module Blank
+        using Util::HashArrayDeepFetch
+
+        # No columns.
+        # @return [Array]
+        def self.column_classes
+          []
+        end
+
+        # Is a blank line.
+        # @param row_string [String]
+        # @return [Boolean]
+        def self.match?(row_string)
+          row_string == "\n"
+        end
+      end
+    end
+  end
+end

data/lib/reading/parsing/rows/comment.rb

@@ -5,20 +5,19 @@ module Reading
       module Comment
         using Util::HashArrayDeepFetch
 
-        # No columns; comments are parsed as if the row were blank.
+        # No columns; parsed as if the row were blank.
         # @return [Array]
         def self.column_classes
           []
         end
 
-        # Starts with a comment character and does not include any format emojis.
-        # (Commented rows that DO include format emojis are matched as compact
-        # planned rows.)
+        # Starts with a comment character. Note: this must be called *after*
+        # calling ::match? on Rows::CompactPlanned and Rows::CustomConfig,
+        # because those check for starting with a comment character too.
         # @param row_string [String]
-        # @param config [Hash]
         # @return [Boolean]
-        def self.match?(row_string, config)
-          row_string.lstrip.start_with?(config.fetch(:comment_character))
+        def self.match?(row_string)
+          row_string.lstrip.start_with?(Config.hash.fetch(:comment_character))
         end
       end
     end

data/lib/reading/parsing/rows/compact_planned.rb

@@ -1,6 +1,7 @@
-require_relative "column"
-require_relative "compact_planned_columns/head"
-require_relative "regular_columns/sources"
+require_relative 'column'
+require_relative 'compact_planned_columns/head'
+require_relative 'regular_columns/sources'
+require_relative 'regular_columns/length'
 
 module Reading
   module Parsing
@@ -12,17 +13,16 @@ module Reading
         # The columns that are possible in this type of row.
         # @return [Array<Class>]
         def self.column_classes
-          [CompactPlanned::Head, Regular::Sources]
+          [CompactPlanned::Head, Regular::Sources, Regular::Length]
         end
 
         # Starts with a comment character and includes one or more format emojis.
         # @param row_string [String]
-        # @param config [Hash]
         # @return [Boolean]
-        def self.match?(row_string, config)
-          row_string.lstrip.start_with?(config.fetch(:comment_character)) &&
-            row_string.match?(config.deep_fetch(:regex, :formats)) &&
-            row_string.count(config.fetch(:column_separator)) <= column_classes.count - 1
+        def self.match?(row_string)
+          row_string.lstrip.start_with?(Config.hash.fetch(:comment_character)) &&
+            row_string.match?(Config.hash.deep_fetch(:regex, :formats)) &&
+            row_string.count(Config.hash.fetch(:column_separator)) <= column_classes.count - 1
         end
       end
     end

data/lib/reading/parsing/rows/compact_planned_columns/head.rb

@@ -15,7 +15,7 @@ module Reading
                 \\ # comment character
                 \s*
                 (
-                  (?<genres>[^a-z]+)?
+                  (?<genres>[^a-z@]+)?
                   \s*
                   (?<sources>@.+)?
                   \s*:
@@ -49,7 +49,7 @@ module Reading
                 )?
                 (?<title>[^@]+)
                 (?<sources>@.+)?
-              \z}x if  segment_index.zero?),
+              \z}x if segment_index.zero?),
               *Column::SHARED_REGEXES[:series_and_extra_info],
             ].compact
           end

data/lib/reading/parsing/rows/custom_config.rb

@@ -0,0 +1,42 @@
+module Reading
+  module Parsing
+    module Rows
+      # A row that declares custom config.
+      module CustomConfig
+        using Util::HashArrayDeepFetch
+
+        # No columns; parsed as if the row were blank.
+        # @return [Array]
+        def self.column_classes
+          []
+        end
+
+        # Starts with a comment character and opening curly brace, and ends with
+        # a closing curly brace.
+        # @param row_string [String]
+        # @return [Boolean]
+        def self.match?(row_string)
+          row_string.match?(
+            %r{\A
+              \s*
+              #{Regexp.escape(Config.hash.fetch(:comment_character))}
+              \s*
+              \{.+\}
+              \s*
+            \z}x
+          )
+        end
+
+        # Adds this row's custom config to the singleton config.
+        # @param row_string [String]
+        # @param config [Hash] an entire config.
+        def self.merge_custom_config!(row_string)
+          stripped_row = row_string.strip.delete_prefix(Config.hash.fetch(:comment_character))
+          custom_config = eval(stripped_row)
+
+          Config.hash.merge!(custom_config)
+        end
+      end
+    end
+  end
+end

data/lib/reading/parsing/rows/regular.rb

@@ -1,13 +1,13 @@
-require_relative "column"
-require_relative "regular_columns/rating"
-require_relative "regular_columns/head"
-require_relative "regular_columns/sources"
-require_relative "regular_columns/start_dates"
-require_relative "regular_columns/end_dates"
-require_relative "regular_columns/genres"
-require_relative "regular_columns/length"
-require_relative "regular_columns/notes"
-require_relative "regular_columns/history"
+require_relative 'column'
+require_relative 'regular_columns/rating'
+require_relative 'regular_columns/head'
+require_relative 'regular_columns/sources'
+require_relative 'regular_columns/start_dates'
+require_relative 'regular_columns/end_dates'
+require_relative 'regular_columns/genres'
+require_relative 'regular_columns/length'
+require_relative 'regular_columns/notes'
+require_relative 'regular_columns/history'
 
 module Reading
   module Parsing
@@ -20,12 +20,13 @@ module Reading
           [Rating, Head, Sources, StartDates, EndDates, Genres, Length, Notes, History]
         end
 
-        # Does not start with a comment character.
+        # Does not start with a comment character. Note: this must be called
+        # *after* calling ::match? on Rows::Blank, because that one catches
+        # blank lines.
         # @param row_string [String]
-        # @param config [Hash]
         # @return [Boolean]
-        def self.match?(row_string, config)
-          !row_string.lstrip.start_with?(config.fetch(:comment_character))
+        def self.match?(row_string)
+          !row_string.lstrip.start_with?(Config.hash.fetch(:comment_character))
         end
       end
     end

data/lib/reading/parsing/rows/regular_columns/length.rb

@@ -15,16 +15,16 @@ module Reading
                 )
                 (\s+|\z)
               )
-              # each or repetitions, used in conjunction with the History column
+              # each and repetitions are used in conjunction with the History column
+              # each
               (
-                # each
                 (?<each>each)
-                |
-                # repetitions
-                (
-                  x
-                  (?<repetitions>\d+)
-                )
+                (\s+|\z)
+              )?
+              # repetitions
+              (
+                x
+                (?<repetitions>\d+)
               )?
             \z}x]
           end

data/lib/reading/parsing/rows/regular_columns/sources.rb

@@ -6,17 +6,12 @@ module Reading
         # and https://github.com/fpsvogel/reading/blob/main/doc/csv-format.md#sources-column-variants
         class Sources < Column
           SOURCES_PARSING_ERRORS = {
-            "Missing comma before URL(s) in the Sources column" =>
-              ->(source) {
-                source.match?(/\shttps?:\/\//) || source.scan(/https?:\/\//).count > 1
-              },
-            "The ISBN/ASIN must be placed after sources in the Sources column" =>
+            "The ISBN/ASIN must be placed last in the Sources column" =>
               ->(source) {
                 source.match?(/\A#{ISBN_REGEX}/o) || source.match(/\A#{ASIN_REGEX}/o)
               },
           }
 
-
           def self.split_by_format?
             true
           end
@@ -32,7 +27,18 @@ module Reading
           def self.tweaks
             {
               sources: -> {
-                sources = _1.split(/\s*,\s*/)
+                comma = /\s*,\s*/
+                space_before_url = / (?=https?:\/\/)/
+                sources = _1.split(Regexp.union(comma, space_before_url))
+
+                # Split by space after URL.
+                sources = sources.flat_map { |src|
+                  if src.match?(/\Ahttps?:\/\//)
+                    src.split(" ", 2)
+                  else
+                    src
+                  end
+                }
 
                 SOURCES_PARSING_ERRORS.each do |message, check|
                   if sources.any? { |source| check.call(source) }
@@ -58,7 +64,7 @@ module Reading
                   |
                   (?<length_time>\d+:\d\d)
                 )?
-              \z}x if  segment_index.zero?),
+              \z}x if segment_index.zero?),
               # sources, ISBN/ASIN, length
               (%r{\A
                 (
@@ -78,7 +84,7 @@ module Reading
                   |
                   (?<length_time>\d+:\d\d)
                 )?
-              \z}xo if  segment_index.zero?),
+              \z}xo if segment_index.zero?),
               *Column::SHARED_REGEXES[:series_and_extra_info],
             ].compact
           end

data/lib/reading/parsing/transformer.rb

@@ -1,12 +1,12 @@
-require_relative "attributes/shared"
-require_relative "attributes/attribute"
-require_relative "attributes/rating"
-require_relative "attributes/author"
-require_relative "attributes/title"
-require_relative "attributes/genres"
-require_relative "attributes/variants"
-require_relative "attributes/experiences"
-require_relative "attributes/notes"
+require_relative 'attributes/shared'
+require_relative 'attributes/attribute'
+require_relative 'attributes/rating'
+require_relative 'attributes/author'
+require_relative 'attributes/title'
+require_relative 'attributes/genres'
+require_relative 'attributes/variants'
+require_relative 'attributes/experiences'
+require_relative 'attributes/notes'
 
 module Reading
   module Parsing
@@ -14,19 +14,15 @@ module Reading
     # Transforms an intermediate hash (parsed from a CSV row) into item data.
     # While the intermediate hash mirrors the structure of a row, the output of
     # Transformer is based around item attributes, which are listed in
-    # Config#default_config[:item_template] and in the files in parsing/attributes.
+    # Config#default_config[:item][:template] and in the files in parsing/attributes.
     #
     class Transformer
       using Util::HashArrayDeepFetch
       using Util::HashCompactByTemplate
 
-      attr_reader :config
       private attr_reader :attributes
 
-      # @param config [Hash] an entire config.
-      def initialize(config)
-        @config = config
-
+      def initialize
         set_attributes
       end
 
@@ -34,13 +30,13 @@ module Reading
       # @param parsed_row [Hash{Symbol => Hash, Array}] output from
       #   Parsing::Parser#parse_row_to_intermediate_hash.
       # @return [Array<Hash>] an array of Hashes like the template in
-      #   Config#default_config[:item_template].
+      #   Config#default_config[:item][:template].
       def transform_intermediate_hash_to_item_hashes(parsed_row)
         if parsed_row[:head].blank?
           raise InvalidHeadError, "Blank or missing Head column"
         end
 
-        template = config.fetch(:item_template)
+        template = Config.hash.deep_fetch(:item, :template)
 
         parsed_row[:head].map.with_index { |_head, head_index|
           template.map { |attribute_name, default_value|
@@ -58,11 +54,11 @@ module Reading
       # Sets the attributes classes which do all the transforming work.
       # See parsing/attributes/*.
       def set_attributes
-        @attributes ||= config.fetch(:item_template).map { |attribute_name, _default|
+        @attributes ||= Config.hash.deep_fetch(:item, :template).map { |attribute_name, _default|
           attribute_name_camelcase = attribute_name.to_s.split("_").map(&:capitalize).join
           attribute_class = Attributes.const_get(attribute_name_camelcase)
 
-          [attribute_name, attribute_class.new(config)]
+          [attribute_name, attribute_class.new]
         }.to_h
       end
     end