js_regex 3.11.1 → 3.13.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 4152eef358fa4cdee93ac92b3a2b16590e49be305e81b53e88732ba11eade1f7
-  data.tar.gz: 7e43bd8ec3ab16a2512769e49508e655f53960353646937b840777c71406af3f
+  metadata.gz: 054de046ea7e8e7244c80e8c8385270ffd821f0fdd690b8a15a79ea1c1d2a6f4
+  data.tar.gz: d6d3dbd4d191e16008bfd586ff22a10a8240275ac328c0a080af910d99b650a1
 SHA512:
-  metadata.gz: e98d601411d6afd3b2ae10df1b18389109a647d4c2ca5e548cb12831badf20ca31de72a20c4fdc93c971aebf3b694623a586e9d1bdc6741084d46bc05a492bfd
-  data.tar.gz: fd337def96701bc627d09bceb921ba50f341ace408533128b8f8fe4a806d1757805ac2437bc864c3a6c82567190f88c760924a1f0f2df23863cbc15c781ca0e1
+  metadata.gz: 93e7c0a574cadfb867c598a9adbcbd9b18585204912865e589691246ce879bec7b7d48a02872c599447db3f2feeda80236d08b1030aecd1d6bf0568792ac859a
+  data.tar.gz: e766f2ce0305ed0e6b10372d58f07fb0137e51a4a62a607b1f0272ac4b431c820fdb9ca2b8d2ce18de30d4bd33ca6cd8c49846d5c8efa79c081324342b990e87

data/lib/js_regex/conversion.rb

@@ -37,7 +37,7 @@ class JsRegex
       end
 
       def convert_options(input, custom_options, required_options)
-        options = custom_options.to_s.scan(/[gimsuy]/) + required_options
+        options = custom_options.to_s.scan(/[dgimsuvy]/) + required_options
         if input.is_a?(Regexp) && (input.options & Regexp::IGNORECASE).nonzero?
           options << 'i'
         end

data/lib/js_regex/converter/anchor_converter.rb

@@ -10,8 +10,10 @@ class JsRegex
 
       def convert_data
         case subtype
-        when :bol, :bos then '^'
-        when :eol, :eos then '$'
+        when :bol then convert_bol
+        when :bos then '^'
+        when :eol then '(?=$|\n)'
+        when :eos then '$'
         when :eos_ob_eol then '(?=\n?$)'
         when :word_boundary then convert_boundary
         when :nonword_boundary then convert_nonboundary
@@ -20,6 +22,15 @@ class JsRegex
         end
       end
 
+      def convert_bol
+        if context.es_2018_or_higher?
+          '(?<=^|\n(?!$))'
+        else
+          # TODO: warn in v4.0.0, or drop ES2009 & ES2015 support
+          '^'
+        end
+      end
+
       def convert_boundary
         if context.es_2018_or_higher? && context.enable_u_option
           BOUNDARY_EXPANSION

data/lib/js_regex/converter/backreference_converter.rb

@@ -29,7 +29,8 @@ class JsRegex
 
       def convert_to_plain_num_ref
         position = new_position
-        Node.new("\\#{position}", reference: position, type: :backref)
+        text = "\\#{position}#{'(?:)' if expression.x?}"
+        Node.new(text, reference: position, type: :backref)
       end
 
       def new_position
@@ -43,7 +44,7 @@ class JsRegex
       def convert_call
         if context.recursions(expression) >= 5
           warn_of("Recursion for '#{expression}' curtailed at 5 levels")
-          return ''
+          return drop
         end
 
         context.count_recursion(expression)
@@ -51,7 +52,9 @@ class JsRegex
         target_copy = expression.referenced_expression.unquantified_clone
         # avoid "Duplicate capture group name" error in JS
         target_copy.token = :capture if target_copy.is?(:named, :group)
+        context.start_subexp_recursion
         result = convert_expression(target_copy)
+        context.end_subexp_recursion
         # wrap in group if it is a full-pattern recursion
         expression.reference == 0 ? Node.new('(?:', result, ')') : result
       end

data/lib/js_regex/converter/base.rb

@@ -79,8 +79,9 @@ class JsRegex
         number = context.capturing_group_count + 1
         backref_node = Node.new("\\#{number}", reference: number, type: :backref)
         context.increment_local_capturing_group_count
-        # an empty passive group (?:) is appended as literal digits may follow
-        Node.new('(?=(', *content, '))', backref_node, '(?:)')
+        # The surrounding group is added so that quantifiers apply to the whole.
+        # Without it, `(?:)` would need to be appended as literal digits may follow.
+        Node.new('(?:(?=(', *content, '))', backref_node, ')')
       end
     end
   end

data/lib/js_regex/converter/context.rb

@@ -10,6 +10,7 @@ class JsRegex
                   :case_insensitive_root,
                   :fail_fast,
                   :in_atomic_group,
+                  :in_subexp_recursion,
                   :warnings
 
       def initialize(case_insensitive_root: false, fail_fast: false, target: nil)
@@ -81,6 +82,14 @@ class JsRegex
         [exp.class, exp.starts_at]
       end
 
+      def start_subexp_recursion
+        self.in_subexp_recursion = true
+      end
+
+      def end_subexp_recursion
+        self.in_subexp_recursion = false
+      end
+
       # takes and returns 1-indexed group positions.
       # new is different from old if capturing groups were added in between.
       def new_capturing_group_position(old_position)
@@ -106,6 +115,7 @@ class JsRegex
                   :case_insensitive_root,
                   :fail_fast,
                   :in_atomic_group,
+                  :in_subexp_recursion,
                   :warnings
 
       def total_added_capturing_groups

data/lib/js_regex/converter/escape_converter.rb

@@ -42,10 +42,10 @@ class JsRegex
           unicode_escape_codepoint
         when :literal
           LiteralConverter.convert_data(expression.char, context)
+        when :bell, :escape, :hex, :octal
+          hex_escape_codepoint
         when *ESCAPES_SHARED_BY_RUBY_AND_JS
           pass_through
-        when :bell, :escape, :octal
-          hex_escape_codepoint
         else
           warn_of_unsupported_feature
         end

data/lib/js_regex/converter/group_converter.rb

@@ -84,7 +84,7 @@ class JsRegex
         tail = opts[:tail] || ')'
         return Node.new(*wrap(head, tail)) if opts[:capturing].equal?(false)
 
-        context.capture_group
+        context.capture_group unless context.in_subexp_recursion
         ref = expression.number
         Node.new(*wrap(head, tail), reference: ref, type: :captured_group)
       end

data/lib/js_regex/converter/set_converter.rb

@@ -41,11 +41,7 @@ class JsRegex
       def simple_convert_child(exp)
         case exp.type
         when :literal
-          return false if !context.u? &&
-            exp.text =~ LiteralConverter::ASTRAL_PLANE_CODEPOINT_PATTERN &&
-            !context.enable_u_option
-
-          LiteralConverter.escape_incompatible_bmp_literals(exp.text)
+          simple_convert_literal_child(exp)
         when :set
           # full conversion is needed for nested sets and intersections
           exp.token.equal?(:range) && exp.expressions.map do |op|
@@ -67,6 +63,19 @@ class JsRegex
         end
       end
 
+      def simple_convert_literal_child(exp)
+        if !context.u? &&
+           exp.text =~ LiteralConverter::ASTRAL_PLANE_CODEPOINT_PATTERN &&
+           !context.enable_u_option
+          false
+        elsif SET_LITERALS_REQUIRING_ESCAPE_PATTERN.match?(exp.text)
+          "\\#{exp.text}"
+        else
+          LiteralConverter.escape_incompatible_bmp_literals(exp.text)
+        end
+      end
+
+      SET_LITERALS_REQUIRING_ESCAPE_PATTERN = Regexp.union(%w<( ) [ ] { } / - |>)
       SET_SPECIFIC_ESCAPES_PATTERN = /[\^\-]/
       CONVERTIBLE_ESCAPE_TOKENS = %i[control meta_sequence bell escape octal] +
         EscapeConverter::ESCAPES_SHARED_BY_RUBY_AND_JS

data/lib/js_regex/converter/type_converter.rb

@@ -8,6 +8,8 @@ class JsRegex
     class TypeConverter < JsRegex::Converter::Base
       HEX_EXPANSION              = '[0-9A-Fa-f]'
       NONHEX_EXPANSION           = '[^0-9A-Fa-f]'
+      I_MODE_HEX_EXPANSION       = '[0-9A-F]'
+      I_MODE_NONHEX_EXPANSION    = '[^0-9A-F]'
       ES2018_HEX_EXPANSION       = '\p{AHex}'
       ES2018_NONHEX_EXPANSION    = '\P{AHex}'
       ES2018_XGRAPHEME_EXPANSION = '[\P{M}\P{Lm}](?:(?:[\u035C\u0361]\P{M}\p{M}*)|\u200d|\p{M}|\p{Lm}|\p{Emoji_Modifier})*'
@@ -28,7 +30,7 @@ class JsRegex
         case subtype
         when :hex then hex_expansion
         when :nonhex then nonhex_expansion
-        when :linebreak then LINEBREAK_EXPANSION
+        when :linebreak then linebreak_expansion
         when :xgrapheme then xgrapheme
         when :digit, :space, :word
           return pass_through if self.class.directly_compatible?(expression)
@@ -44,6 +46,8 @@ class JsRegex
       def hex_expansion
         if context.es_2018_or_higher? && context.enable_u_option
           ES2018_HEX_EXPANSION
+        elsif context.case_insensitive_root
+          I_MODE_HEX_EXPANSION
         else
           HEX_EXPANSION
         end
@@ -52,11 +56,17 @@ class JsRegex
       def nonhex_expansion
         if context.es_2018_or_higher? && context.enable_u_option
           ES2018_NONHEX_EXPANSION
+        elsif context.case_insensitive_root
+          I_MODE_NONHEX_EXPANSION
         else
           NONHEX_EXPANSION
         end
       end
 
+      def linebreak_expansion
+        wrap_in_backrefed_lookahead(LINEBREAK_EXPANSION)
+      end
+
       def negative_set_substitution
         # ::of_expression returns an inverted set for negative expressions,
         # so we need to un-invert before wrapping in [^ and ]. Kinda lame.
@@ -73,7 +83,7 @@ class JsRegex
 
       def xgrapheme
         if context.es_2018_or_higher? && context.enable_u_option
-          ES2018_XGRAPHEME_EXPANSION
+          wrap_in_backrefed_lookahead(ES2018_XGRAPHEME_EXPANSION)
         else
           warn_of_unsupported_feature
         end

data/lib/js_regex/version.rb

@@ -1,3 +1,3 @@
 class JsRegex
-  VERSION = '3.11.1'
+  VERSION = '3.13.0'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: js_regex
 version: !ruby/object:Gem::Version
-  version: 3.11.1
+  version: 3.13.0
 platform: ruby
 authors:
 - Janosch Müller
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-01-07 00:00:00.000000000 Z
+date: 2025-01-27 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: character_set
@@ -28,22 +28,16 @@ dependencies:
   name: regexp_parser
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: 2.6.2
-    - - "<"
+    - - "~>"
       - !ruby/object:Gem::Version
-        version: 3.0.0
+        version: '2.10'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: 2.6.2
-    - - "<"
+    - - "~>"
       - !ruby/object:Gem::Version
-        version: 3.0.0
+        version: '2.10'
 - !ruby/object:Gem::Dependency
   name: regexp_property_values
   requirement: !ruby/object:Gem::Requirement
@@ -111,7 +105,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.5.0.dev
+rubygems_version: 3.5.22
 signing_key:
 specification_version: 4
 summary: Converts Ruby regexes to JavaScript regexes.