js_regex 3.3.0 → 3.7.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 0f418197d1c8fea37bd549b701049f1130bfca486c5a11ba25e56d7d0cf990f8
-  data.tar.gz: 8679277fb51aca528933f4727cd0e70f1e1204c78acaf4e20da48d6b790ea7be
+  metadata.gz: c133491821c27a1ce87f58e59c828e92cef7e1f7f9be99b015e79e55ded6874e
+  data.tar.gz: 7ea0268f1e0d8cf48c6a56b635d8189dfb1c2f0ad969902eb0d2a9380d01f03a
 SHA512:
-  metadata.gz: ddeadbe690e2928c068d8c2962d80e88c50f5fef46304e46fa98e8dc215c6c57c7790e9ef7ba8b1f640e71c3ca834b7233449c9b7d68466ac76581de73c7ffe3
-  data.tar.gz: 9c10d4e6a529ec54385ed55b34e4a910d354d7665f0831bc65b65bc20f176f4e46f9aa90cff2b36ea8f1306dfa65f7e42f4a93516ab228f206e74db7d708b779
+  metadata.gz: 2f31af53ed6fd3213839df7ece799dc951a0f979cf9858cce47ae8151c167de5a5c5464de3d7355782a8bd56f6d142c5d152d4e381ffed9e26f7f51b13c445b1
+  data.tar.gz: 46525b2bb74b11189495f831fa860980291c9be05c5d546e7fc1ed1bee335d4cb29059b6ba1ffb008528c78309c4e75e2f9088570d7691565180821b912ffc96

data/lib/js_regex.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 # JsRegex converts ::Regexp instances to JavaScript.
 #
 # Usage:
@@ -10,6 +8,7 @@
 #
 class JsRegex
   require_relative File.join('js_regex', 'conversion')
+  require_relative File.join('js_regex', 'error')
   require_relative File.join('js_regex', 'version')
   require 'json'
 
@@ -30,4 +29,15 @@ class JsRegex
   def to_s
     "/#{source.empty? ? '(?:)' : source}/#{options}"
   end
+
+  def self.new!(ruby_regex, options: nil)
+    js_regex = new(ruby_regex, options: options)
+    if js_regex.warnings.any?
+      raise StandardError.new(
+        "Could not fully convert the given regex #{ruby_regex.inspect}:\n" +
+        js_regex.warnings.join("\n")
+      ).extend(JsRegex::Error)
+    end
+    js_regex
+  end
 end

data/lib/js_regex/conversion.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 class JsRegex
   #
   # This class acts as a facade, passing a regex to the converters.
@@ -9,29 +7,34 @@ class JsRegex
   class Conversion
     require 'regexp_parser'
     require_relative 'converter'
+    require_relative 'error'
     require_relative 'node'
     require_relative 'second_pass'
 
     class << self
-      def of(ruby_regex, options: nil)
-        source, warnings = convert_source(ruby_regex)
-        options_string   = convert_options(ruby_regex, options)
+      def of(input, options: nil)
+        source, warnings = convert_source(input)
+        options_string   = convert_options(input, options)
         [source, options_string, warnings]
       end
 
       private
 
-      def convert_source(ruby_regex)
-        tree = Regexp::Parser.parse(ruby_regex)
+      def convert_source(input)
+        tree = Regexp::Parser.parse(input)
         context = Converter::Context.new(case_insensitive_root: tree.i?)
         converted_tree = Converter.convert(tree, context)
         final_tree = SecondPass.call(converted_tree)
         [final_tree.to_s, context.warnings]
+      rescue Regexp::Parser::Error => e
+        raise e.extend(JsRegex::Error)
       end
 
-      def convert_options(ruby_regex, custom_options)
+      def convert_options(input, custom_options)
         options = custom_options.to_s.scan(/[gimuy]/)
-        options << 'i' if (ruby_regex.options & Regexp::IGNORECASE).nonzero?
+        if input.is_a?(Regexp) && (input.options & Regexp::IGNORECASE).nonzero?
+          options << 'i'
+        end
         options.uniq.sort.join
       end
     end

data/lib/js_regex/converter.rb

@@ -1,8 +1,6 @@
-# frozen_string_literal: true
-
 class JsRegex
   module Converter
-    Dir[File.join(File.dirname(__FILE__), 'converter', '*.rb')].each do |file|
+    Dir[File.join(__dir__, 'converter', '*.rb')].sort.each do |file|
       require file
     end
 
@@ -32,18 +30,10 @@ class JsRegex
         MAP[expression.type].new
       end
 
-      # Limit the number of generated surrogate pairs, else the output might
-      # get to large for certain applications. The chosen number is somewhat
-      # arbitrary. 100 pairs make for about 1 KB, uncompressed. The median char
-      # count of all properties supported by Ruby is 92. 75% are below 300 chars.
-      #
-      # Set this to nil if you need full unicode matches and size doesn't matter.
-      attr_writer :surrogate_pair_limit
-
-      def in_surrogate_pair_limit?(&pair_count)
-        @surrogate_pair_limit.nil? || @surrogate_pair_limit >= pair_count.call
+      # Legacy method. Remove in v4.0.0.
+      def surrogate_pair_limit=(_arg)
+        warn '#surrogate_pair_limit= is deprecated and has no effect anymore.'
       end
     end
-    self.surrogate_pair_limit = 300
   end
 end

data/lib/js_regex/converter/anchor_converter.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 require_relative 'base'
 
 class JsRegex
@@ -23,9 +21,8 @@ class JsRegex
       end
 
       def pass_boundary_with_warning(boundary)
-        warn("The boundary '#{boundary}' at index #{expression.ts} "\
-             'is not unicode-aware in JavaScript, '\
-             'so it might act differently than in Ruby.')
+        warn_of("The anchor '#{boundary}' at index #{expression.ts} "\
+                'only works at ASCII word boundaries in JavaScript.')
         boundary
       end
     end

data/lib/js_regex/converter/assertion_converter.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 require_relative 'base'
 require_relative 'group_converter'
 

data/lib/js_regex/converter/backreference_converter.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 require_relative 'base'
 
 class JsRegex

data/lib/js_regex/converter/base.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 class JsRegex
   module Converter
     #
@@ -54,11 +52,11 @@ class JsRegex
       def warn_of_unsupported_feature(description = nil)
         description ||= "#{subtype} #{expression.type}".tr('_', ' ')
         full_desc = "#{description} '#{expression}'"
-        warn("Dropped unsupported #{full_desc} at index #{expression.ts}")
+        warn_of("Dropped unsupported #{full_desc} at index #{expression.ts}")
         drop
       end
 
-      def warn(text)
+      def warn_of(text)
         context.warnings << text
       end
 

data/lib/js_regex/converter/conditional_converter.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 require_relative 'base'
 
 class JsRegex

data/lib/js_regex/converter/context.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 class JsRegex
   module Converter
     #

data/lib/js_regex/converter/escape_converter.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 require_relative 'base'
 require_relative 'literal_converter'
 
@@ -44,7 +42,7 @@ class JsRegex
         when :control, :meta_sequence
           unicode_escape_codepoint
         when :literal
-          LiteralConverter.convert_data(data)
+          LiteralConverter.convert_data(expression.char)
         when *ESCAPES_SHARED_BY_RUBY_AND_JS
           pass_through
         when :bell, :escape

data/lib/js_regex/converter/freespace_converter.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 require_relative 'base'
 
 class JsRegex

data/lib/js_regex/converter/group_converter.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 require_relative 'base'
 
 class JsRegex

data/lib/js_regex/converter/literal_converter.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 require_relative 'base'
 
 class JsRegex
@@ -15,42 +13,28 @@ class JsRegex
           if data =~ ASTRAL_PLANE_CODEPOINT_PATTERN
             convert_astral_data(data)
           else
-            convert_bmp_data(data)
+            escape_incompatible_bmp_literals(data)
           end
         end
 
         def convert_astral_data(data)
-          data.each_char.each_with_object(Node.new) do |chr, node|
-            if chr =~ ASTRAL_PLANE_CODEPOINT_PATTERN
-              node << surrogate_pair_for(chr)
+          data.each_char.each_with_object(Node.new) do |char, node|
+            if char =~ ASTRAL_PLANE_CODEPOINT_PATTERN
+              node << surrogate_substitution_for(char)
             else
-              node << convert_bmp_data(chr)
+              node << escape_incompatible_bmp_literals(char)
             end
           end
         end
 
-        def convert_bmp_data(data)
-          ensure_json_compatibility(
-            ensure_forward_slashes_are_escaped(data)
-          )
+        def escape_incompatible_bmp_literals(data)
+          data.gsub('/', '\\/').gsub(/[\f\n\r\t]/) { |lit| Regexp.escape(lit) }
         end
 
         private
 
-        def surrogate_pair_for(astral_char)
-          base = astral_char.codepoints.first - 65_536
-          high = ((base / 1024).floor + 55_296).to_s(16)
-          low  = (base % 1024 + 56_320).to_s(16)
-          "(?:\\u#{high}\\u#{low})"
-        end
-
-        def ensure_forward_slashes_are_escaped(data)
-          # literal slashes would signify the pattern end in JsRegex#to_s
-          data.gsub(%r{\\?/}, '\\/')
-        end
-
-        def ensure_json_compatibility(data)
-          data.gsub(%r{\\?([\f\n\r\t])}) { Regexp.escape($1) }
+        def surrogate_substitution_for(char)
+          CharacterSet::Writer.write_surrogate_ranges([], [char.codepoints])
         end
       end
 

data/lib/js_regex/converter/meta_converter.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 require_relative 'base'
 
 class JsRegex
@@ -8,6 +6,9 @@ class JsRegex
     # Template class implementation.
     #
     class MetaConverter < JsRegex::Converter::Base
+      DOT_EXPANSION    = '(?:[\uD800-\uDBFF][\uDC00-\uDFFF]|[^\n\uD800-\uDFFF])'
+      ML_DOT_EXPANSION = '(?:[\uD800-\uDBFF][\uDC00-\uDFFF]|[^\uD800-\uDFFF])'
+
       private
 
       def convert_data
@@ -15,7 +16,7 @@ class JsRegex
         when :alternation
           convert_alternatives
         when :dot
-          expression.multiline? ? '(?:.|\n)' : '.'
+          expression.multiline? ? ML_DOT_EXPANSION : DOT_EXPANSION
         else
           warn_of_unsupported_feature
         end

data/lib/js_regex/converter/property_converter.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 require_relative 'base'
 require 'character_set'
 
@@ -15,35 +13,15 @@ class JsRegex
       private
 
       def convert_data
-        content = character_set_of_property
-
-        if expression.negative?
-          if content.astral_part?
-            warn_of_unsupported_feature('astral plane negation by property')
-          end
-        elsif Converter.in_surrogate_pair_limit? { content.astral_part.size }
-          return content.to_s_with_surrogate_alternation
-        else
-          warn_of_unsupported_feature('large astral plane match of property')
-        end
+        content = CharacterSet.of_expression(expression)
 
-        limit_to_bmp_part(content)
-      end
-
-      def character_set_of_property
-        character_set = CharacterSet.of_property(subtype)
         if expression.case_insensitive? && !context.case_insensitive_root
-          character_set.case_insensitive
-        else
-          character_set
+          content = content.case_insensitive
+        elsif !expression.case_insensitive? && context.case_insensitive_root
+          warn_of_unsupported_feature('nested case-sensitive property')
         end
-      end
-
-      def limit_to_bmp_part(content)
-        bmp_part = content.bmp_part
-        return drop if bmp_part.empty?
 
-        "[#{'^' if expression.negative?}#{bmp_part}]"
+        content.to_s_with_surrogate_ranges
       end
     end
   end

data/lib/js_regex/converter/set_converter.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 require_relative 'base'
 require_relative 'escape_converter'
 require_relative 'type_converter'
@@ -28,11 +26,7 @@ class JsRegex
           warn_of_unsupported_feature('nested case-sensitive set')
         end
 
-        if Converter.in_surrogate_pair_limit? { content.astral_part.size }
-          content.to_s_with_surrogate_alternation
-        else
-          limit_to_bmp_part_with_warning(content)
-        end
+        content.to_s_with_surrogate_ranges
       end
 
       def directly_compatible?
@@ -41,9 +35,7 @@ class JsRegex
 
       def all_children_directly_compatible?
         # note that #each_expression is recursive
-        expression.each_expression do |exp|
-          return unless child_directly_compatible?(exp)
-        end
+        expression.each_expression.all? { |ch| child_directly_compatible?(ch) }
       end
 
       def child_directly_compatible?(exp)
@@ -66,13 +58,8 @@ class JsRegex
       end
 
       def pass_through_with_escaping
-        expression.to_s(:base).gsub(%r{\\?([\f\n\r\t])}) { Regexp.escape($1) }
-      end
-
-      def limit_to_bmp_part_with_warning(content)
-        warn_of_unsupported_feature('large astral plane match of set')
-        bmp_part = content.bmp_part
-        bmp_part.empty? ? drop : bmp_part.to_s(in_brackets: true)
+        string = expression.to_s(:base)
+        LiteralConverter.escape_incompatible_bmp_literals(string)
       end
     end
   end

data/lib/js_regex/converter/subexpression_converter.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 require_relative 'base'
 
 class JsRegex

data/lib/js_regex/converter/type_converter.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 require_relative 'base'
 
 class JsRegex

data/lib/js_regex/converter/unsupported_token_converter.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 require_relative 'base'
 
 class JsRegex

data/lib/js_regex/error.rb

@@ -0,0 +1,5 @@
+class JsRegex
+  # This is mixed into errors, e.g. those thrown by the parser,
+  # allowing to `rescue JsRegex::Error`.
+  module Error; end
+end

data/lib/js_regex/node.rb

@@ -1,11 +1,11 @@
-# frozen_string_literal: true
-
 class JsRegex
   #
   # Converter#convert result. Represents a branch or leaf node with an optional
   # quantifier as well as type and reference annotations for SecondPass.
   #
   class Node
+    require_relative 'error'
+
     attr_reader :children, :quantifier, :reference, :type
 
     TYPES = %i[
@@ -49,7 +49,9 @@ class JsRegex
       when :backref_num, :captured_group, :plain
         children.join << quantifier.to_s
       else
-        raise TypeError, "#{type} must be substituted before stringification"
+        raise TypeError.new(
+          "#{type} must be substituted before stringification"
+        ).extend(JsRegex::Error)
       end
     end
 

data/lib/js_regex/second_pass.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 class JsRegex
   #
   # After conversion of a full Regexp::Expression tree, this
@@ -61,7 +59,7 @@ class JsRegex
 
       def condition_permutations(conditions)
         (0..(conditions.length)).inject([]) do |arr, n|
-          arr += conditions.combination(n).to_a
+          arr + conditions.combination(n).to_a
         end
       end
 

data/lib/js_regex/version.rb

@@ -1,5 +1,3 @@
-# frozen_string_literal: true
-
 class JsRegex
-  VERSION = '3.3.0'
+  VERSION = '3.7.0'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: js_regex
 version: !ruby/object:Gem::Version
-  version: 3.3.0
+  version: 3.7.0
 platform: ruby
 authors:
 - Janosch Müller
-autorequire: 
+autorequire:
 bindir: bin
 cert_chain: []
-date: 2019-05-26 00:00:00.000000000 Z
+date: 2021-03-03 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: character_set
@@ -16,37 +16,51 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.3'
+        version: '1.4'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.3'
+        version: '1.4'
 - !ruby/object:Gem::Dependency
   name: regexp_parser
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.5'
+        version: '2.1'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.5'
+        version: '2.1'
 - !ruby/object:Gem::Dependency
   name: regexp_property_values
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.3'
+        version: '1.0'
   type: :runtime
   prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.0'
+- !ruby/object:Gem::Dependency
+  name: mini_racer
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.3'
+  type: :development
+  prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
@@ -58,70 +72,70 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '12.3'
+        version: '13.0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '12.3'
+        version: '13.0'
 - !ruby/object:Gem::Dependency
-  name: rspec-core
+  name: rspec
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '3.8'
+        version: '3.10'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '3.8'
+        version: '3.10'
 - !ruby/object:Gem::Dependency
-  name: rspec-expectations
+  name: codecov
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '3.8'
+        version: 0.2.12
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '3.8'
+        version: 0.2.12
 - !ruby/object:Gem::Dependency
-  name: rspec-mocks
+  name: gouteur
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '3.8'
+        version: '1.0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '3.8'
+        version: '1.0'
 - !ruby/object:Gem::Dependency
-  name: therubyracer
+  name: rubocop
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.12'
+        version: '1.8'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.12'
+        version: '1.8'
 description: JsRegex converts Ruby's native regular expressions for JavaScript, taking
   care of various incompatibilities and returning warnings for unsolvable differences.
 email:
@@ -149,6 +163,7 @@ files:
 - lib/js_regex/converter/subexpression_converter.rb
 - lib/js_regex/converter/type_converter.rb
 - lib/js_regex/converter/unsupported_token_converter.rb
+- lib/js_regex/error.rb
 - lib/js_regex/node.rb
 - lib/js_regex/second_pass.rb
 - lib/js_regex/version.rb
@@ -156,7 +171,7 @@ homepage: https://github.com/jaynetics/js_regex
 licenses:
 - MIT
 metadata: {}
-post_install_message: 
+post_install_message:
 rdoc_options: []
 require_paths:
 - lib
@@ -171,8 +186,8 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.0.3
-signing_key: 
+rubygems_version: 3.2.3
+signing_key:
 specification_version: 4
 summary: Converts Ruby regexes to JavaScript regexes.
 test_files: []