regexp_parser 2.0.1 → 2.0.3

data/lib/regexp_parser/scanner/scanner.rl

@@ -37,7 +37,7 @@
   class_posix           = ('[:' . '^'? . class_name_posix . ':]');
 
 
-  # these are not supported in ruby, and need verification
+  # these are not supported in ruby at the moment
   collating_sequence    = '[.' . (alpha | [\-])+ . '.]';
   character_equivalent  = '[=' . alpha . '=]';
 
@@ -228,13 +228,13 @@
       emit(type, class_name.to_sym, text)
     };
 
-    collating_sequence >(open_bracket, 1) @set_closed @eof(premature_end_error)  {
-      emit(:set, :collation, copy(data, ts, te))
-    };
-
-    character_equivalent >(open_bracket, 1) @set_closed @eof(premature_end_error)  {
-      emit(:set, :equivalent, copy(data, ts, te))
-    };
+    # These are not supported in ruby at the moment. Enable them if they are.
+    # collating_sequence >(open_bracket, 1) @set_closed @eof(premature_end_error)  {
+    #   emit(:set, :collation, copy(data, ts, te))
+    # };
+    # character_equivalent >(open_bracket, 1) @set_closed @eof(premature_end_error)  {
+    #   emit(:set, :equivalent, copy(data, ts, te))
+    # };
 
     meta_char > (set_meta, 1) {
       emit(:literal, :literal, copy(data, ts, te))

data/lib/regexp_parser/syntax.rb

@@ -1,9 +1,9 @@
+module Regexp::Syntax
+  class SyntaxError < StandardError; end
+end
+
 require File.expand_path('../syntax/tokens', __FILE__)
 require File.expand_path('../syntax/base', __FILE__)
 require File.expand_path('../syntax/any', __FILE__)
 require File.expand_path('../syntax/version_lookup', __FILE__)
 require File.expand_path('../syntax/versions', __FILE__)
-
-module Regexp::Syntax
-  class SyntaxError < StandardError; end
-end

data/lib/regexp_parser/syntax/any.rb

@@ -8,8 +8,8 @@ module Regexp::Syntax
       @implements = { :* => [:*] }
     end
 
-    def implements?(type, token) true end
-    def implements!(type, token) true end
+    def implements?(_type, _token) true end
+    def implements!(_type, _token) true end
   end
 
 end

data/lib/regexp_parser/syntax/base.rb

@@ -1,7 +1,7 @@
 require 'set'
 
 module Regexp::Syntax
-  class NotImplementedError < SyntaxError
+  class NotImplementedError < Regexp::Syntax::SyntaxError
     def initialize(syntax, type, token)
       super "#{syntax.class.name} does not implement: [#{type}:#{token}]"
     end

data/lib/regexp_parser/syntax/version_lookup.rb

@@ -3,13 +3,13 @@ module Regexp::Syntax
   VERSION_REGEXP = /#{VERSION_FORMAT}/
   VERSION_CONST_REGEXP = /\AV\d+_\d+(?:_\d+)?\z/
 
-  class InvalidVersionNameError < SyntaxError
+  class InvalidVersionNameError < Regexp::Syntax::SyntaxError
     def initialize(name)
       super "Invalid version name '#{name}'. Expected format is '#{VERSION_FORMAT}'"
     end
   end
 
-  class UnknownSyntaxNameError < SyntaxError
+  class UnknownSyntaxNameError < Regexp::Syntax::SyntaxError
     def initialize(name)
       super "Unknown syntax name '#{name}'."
     end

data/lib/regexp_parser/version.rb

@@ -1,5 +1,5 @@
 class Regexp
   class Parser
-    VERSION = '2.0.1'
+    VERSION = '2.0.3'
   end
 end

data/spec/expression/subexpression_spec.rb

@@ -32,7 +32,7 @@ RSpec.describe(Regexp::Expression::Subexpression) do
     }
 
     root.each_expression do |exp|
-      next unless expected_nesting_level = tests.delete(exp.to_s)
+      next unless (expected_nesting_level = tests.delete(exp.to_s))
       expect(expected_nesting_level).to eq exp.nesting_level
     end
 

data/spec/expression/to_s_spec.rb

@@ -1,58 +1,50 @@
 require 'spec_helper'
 
 RSpec.describe('Expression#to_s') do
-  specify('literal alternation') do
-    pattern = 'abcd|ghij|klmn|pqur'
+  def parse_frozen(pattern, ruby_version = nil)
+    IceNine.deep_freeze(RP.parse(pattern, *ruby_version))
+  end
 
-    expect(RP.parse(pattern).to_s).to eq pattern
+  def expect_round_trip(pattern, ruby_version = nil)
+    parsed = parse_frozen(pattern, ruby_version)
+
+    expect(parsed.to_s).to eql(pattern)
   end
 
-  specify('quantified alternations') do
-    pattern = '(?:a?[b]+(c){2}|d+[e]*(f)?)|(?:g+[h]?(i){2,3}|j*[k]{3,5}(l)?)'
+  specify('literal alternation') do
+    expect_round_trip('abcd|ghij|klmn|pqur')
+  end
 
-    expect(RP.parse(pattern).to_s).to eq pattern
+  specify('quantified alternations') do
+    expect_round_trip('(?:a?[b]+(c){2}|d+[e]*(f)?)|(?:g+[h]?(i){2,3}|j*[k]{3,5}(l)?)')
   end
 
   specify('quantified sets') do
-    pattern = '[abc]+|[^def]{3,6}'
-
-    expect(RP.parse(pattern).to_s).to eq pattern
+    expect_round_trip('[abc]+|[^def]{3,6}')
   end
 
   specify('property sets') do
-    pattern = '[\\a\\b\\p{Lu}\\P{Z}\\c\\d]+'
-
-    expect(RP.parse(pattern, 'ruby/1.9').to_s).to eq pattern
+    expect_round_trip('[\\a\\b\\p{Lu}\\P{Z}\\c\\d]+', 'ruby/1.9')
   end
 
   specify('groups') do
-    pattern = "(a(?>b(?:c(?<n>d(?'N'e)??f)+g)*+h)*i)++"
-
-    expect(RP.parse(pattern, 'ruby/1.9').to_s).to eq pattern
+    expect_round_trip("(a(?>b(?:c(?<n>d(?'N'e)??f)+g)*+h)*i)++", 'ruby/1.9')
   end
 
   specify('assertions') do
-    pattern = '(a+(?=b+(?!c+(?<=d+(?<!e+)?f+)?g+)?h+)?i+)?'
-
-    expect(RP.parse(pattern, 'ruby/1.9').to_s).to eq pattern
+    expect_round_trip('(a+(?=b+(?!c+(?<=d+(?<!e+)?f+)?g+)?h+)?i+)?', 'ruby/1.9')
   end
 
   specify('comments') do
-    pattern = '(?#start)a(?#middle)b(?#end)'
-
-    expect(RP.parse(pattern).to_s).to eq pattern
+    expect_round_trip('(?#start)a(?#middle)b(?#end)')
   end
 
   specify('options') do
-    pattern = '(?mix:start)a(?-mix:middle)b(?i-mx:end)'
-
-    expect(RP.parse(pattern).to_s).to eq pattern
+    expect_round_trip('(?mix:start)a(?-mix:middle)b(?i-mx:end)')
   end
 
   specify('url') do
-    pattern = ('(^$)|(^(http|https):\\/\\/[a-z0-9]+([\\-\\.]{1}[a-z0-9]+)*' + '\\.[a-z]{2,5}(([0-9]{1,5})?\\/.*)?$)')
-
-    expect(RP.parse(pattern).to_s).to eq pattern
+    expect_round_trip('(^$)|(^(http|https):\\/\\/[a-z0-9]+([\\-\\.]{1}[a-z0-9]+)*' + '\\.[a-z]{2,5}(([0-9]{1,5})?\\/.*)?$)')
   end
 
   specify('multiline source') do
@@ -64,7 +56,7 @@ RSpec.describe('Expression#to_s') do
           \z
         /x
 
-    expect(RP.parse(multiline).to_s).to eq multiline.source
+    expect(parse_frozen(multiline).to_s).to eql(multiline.source)
   end
 
   specify('multiline #to_s') do
@@ -76,7 +68,7 @@ RSpec.describe('Expression#to_s') do
           \z
         /x
 
-    expect(RP.parse(multiline.to_s).to_s).to eq multiline.to_s
+    expect_round_trip(multiline.to_s)
   end
 
   # Free spacing expressions that use spaces between quantifiers and their
@@ -93,24 +85,24 @@ RSpec.describe('Expression#to_s') do
         /x
 
     str = 'bbbcged'
-    root = RP.parse(multiline)
+    root = parse_frozen(multiline)
 
-    expect(Regexp.new(root.to_s, Regexp::EXTENDED).match(str)[0]).to eq multiline.match(str)[0]
+    expect(Regexp.new(root.to_s, Regexp::EXTENDED).match(str)[0]).to eql(multiline.match(str)[0])
   end
 
   # special case: implicit groups used for chained quantifiers produce no parens
   specify 'chained quantifiers #to_s' do
     pattern = /a+{1}{2}/
-    root = RP.parse(pattern)
-    expect(root.to_s).to eq 'a+{1}{2}'
+    root = parse_frozen(pattern)
+    expect(root.to_s).to eql('a+{1}{2}')
   end
 
   # regression test for https://github.com/ammar/regexp_parser/issues/74
   specify('non-ascii comment') do
     pattern = '(?x) 😋 # 😋'
     root = RP.parse(pattern)
-    expect(root.last).to be_a Regexp::Expression::Comment
-    expect(root.last.to_s).to eq '# 😋'
-    expect(root.to_s).to eq pattern
+    expect(root.last).to be_a(Regexp::Expression::Comment)
+    expect(root.last.to_s).to eql('# 😋')
+    expect(root.to_s).to eql(pattern)
   end
 end

data/spec/parser/errors_spec.rb

@@ -9,7 +9,7 @@ RSpec.describe('Parsing errors') do
       .to raise_error(Regexp::Parser::UnknownTokenTypeError)
   end
 
-  RSpec.shared_examples 'UnknownTokenError' do |type, token|
+  RSpec.shared_examples 'UnknownTokenError' do |type|
     it "raises for unkown tokens of type #{type}" do
       expect { parser.send(:parse_token, Regexp::Token.new(type, :foo)) }
         .to raise_error(Regexp::Parser::UnknownTokenError)

data/spec/parser/quantifiers_spec.rb

@@ -11,6 +11,7 @@ RSpec.describe('Quantifier parsing') do
       expect(exp.quantifier.min).to eq min
       expect(exp.quantifier.max).to eq max
       expect(exp.quantifier.mode).to eq mode
+      expect(exp.quantifier.text).to eq text
     end
   end
 

data/spec/scanner/sets_spec.rb

@@ -61,9 +61,6 @@ RSpec.describe('Set scanning') do
   include_examples 'scan', /[[:digit:][:space:]]/,  2 => [:posixclass,    :space,    '[:space:]', 10, 19]
   include_examples 'scan', /[[:^digit:]]/,          1 => [:nonposixclass, :digit,    '[:^digit:]', 1, 11]
 
-  include_examples 'scan', /[a[.a-b.]c]/,           2 => [:set,    :collation,       '[.a-b.]',    2,  9]
-  include_examples 'scan', /[a[=e=]c]/,             2 => [:set,    :equivalent,      '[=e=]',      2,  7]
-
   include_examples 'scan', /[a-d&&g-h]/,            4 => [:set,    :intersection,    '&&',         4, 6]
   include_examples 'scan', /[a&&]/,                 2 => [:set,    :intersection,    '&&',         2, 4]
   include_examples 'scan', /[&&z]/,                 1 => [:set,    :intersection,    '&&',         1, 3]
@@ -88,6 +85,21 @@ RSpec.describe('Set scanning') do
     8 => [:set,    :range,           '-',          9, 10],
     10=> [:set,    :close,           ']',          11, 12]
 
+  # Collations/collating sequences and character equivalents are not enabled
+  # in Ruby at the moment. If they ever are, enable them in the scanner,
+  # add them to a new syntax version, and handle them in the parser. Until then,
+  # expect them to be scanned as regular subsets containing literals.
+  # include_examples 'scan', /[a[.a-b.]c]/,           2 => [:set,    :collation,       '[.a-b.]',    2,  9]
+  # include_examples 'scan', /[a[=e=]c]/,             2 => [:set,    :equivalent,      '[=e=]',      2,  7]
+  include_examples 'scan', '[a[.a-b.]c]',
+    2 => [:set,     :open,           '[',          2,  3],
+    3 => [:literal, :literal,        '.',          3,  4],
+    4 => [:literal, :literal,        'a',          4,  5]
+  include_examples 'scan', '[a[=e=]c]',
+    2 => [:set,     :open,           '[',          2,  3],
+    3 => [:literal, :literal,        '=',          3,  4],
+    4 => [:literal, :literal,        'e',          4,  5]
+
   # multi-byte characters should not affect indices
   include_examples 'scan', /[れます]/,
     0 => [:set,     :open,           '[',          0, 1],

data/spec/spec_helper.rb

@@ -1,3 +1,4 @@
+require 'ice_nine'
 require 'regexp_parser'
 require 'regexp_property_values'
 require_relative 'support/shared_examples'

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: regexp_parser
 version: !ruby/object:Gem::Version
-  version: 2.0.1
+  version: 2.0.3
 platform: ruby
 authors:
 - Ammar Ali
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2020-12-20 00:00:00.000000000 Z
+date: 2020-12-28 00:00:00.000000000 Z
 dependencies: []
 description: A library for tokenizing, lexing, and parsing Ruby regular expressions.
 email: