rng 0.3.6 → 0.3.7

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: fdcf95250d1a2a74faf391653627fb81dad5c60eba650a206ad9c9cce4a6b638
-  data.tar.gz: 3ec79ed69ac17cf4a0c7f24c7b4d39b4deaf63c0c2d5868b8579fb29a3ca83ee
+  metadata.gz: 64d22a32278b59bf3f6913561453c616ac39c259f4ffded1b200efddc50461af
+  data.tar.gz: 1fc2453376a3f7f65aa717beb3707e54ff1bdc751ae528ce8132e3f18720cf2b
 SHA512:
-  metadata.gz: af94bb3c3fb3c56877df25ca93a89274664a3e8649524dbc2a6d1b3775541e9c4e6060f5df5eb4851966ad8de23e7fb9e4c70ed8bbf6977bb21578ddaa19dabe
-  data.tar.gz: e29d6b582c39eac4d0c4245a318c2cbae919e178403d44afe5c10fa271ab5391785d02e7b7819ce3beca6e4992d7ab8684e278858fa5e3dbc8326cb4ac527173
+  metadata.gz: dc8735cc2a8a3ae46635100a30ee39e114c762b9bace21955e2d361ad8bcbf65e0a0f10102f81085fea19a13d377ad93e1ac5313c8ae56a3237171712de7f965
+  data.tar.gz: 3ac117095690e1065a10921f4093072f151c84454cc0439fcd4a21764978654b8e5963f548969d92d76efb70c9ecca674984aa38fdc3a67cd3bb72c0924d394c

data/lib/rng/external_ref_resolver.rb

@@ -235,7 +235,8 @@ module Rng
         obj.map { |o| deep_dup(o) }
       when Hash
         obj.each_with_object({}) { |(k, v), h| h[deep_dup(k)] = deep_dup(v) }
-      when NilClass, Symbol, Numeric, TrueClass, FalseClass
+      when NilClass, Symbol, Numeric, TrueClass, FalseClass,
+           Lutaml::Model::UninitializedClass # can't dup instance of singleton
         obj
       else
         obj.dup

data/lib/rng/rnc_parser.rb

@@ -51,8 +51,9 @@ module Rng
     end
 
     # Character escapes for strings: \", \\, \n, \r, \t, and RELAX NG class escapes: \i, \c, \d, \w
+    # Any other \X sequence is preserved literally (e.g., \+, \-, \. in regex patterns)
     rule(:char_escape) do
-      str('\\') >> match('["\\\\ntricdw]').as(:char)
+      str('\\') >> (match('["\\\\ntricdw]') | match('[^x]')).as(:char)
     end
 
     # Identifier can contain regular chars, dots, hex escapes, or backslash escapes
@@ -407,7 +408,8 @@ module Rng
 
     # Choice is handled at content level, not as separate pattern
     rule(:content_item) do
-      annotations.maybe >>
+      (doc_comments >> whitespace).maybe >>
+        annotations.maybe >>
         (element_def | attribute_def |
           # Datatype subtraction: identifier - ( value|identifier|choice|annotated )
           (identifier.as(:datatype_name) >>

data/lib/rng/rnc_to_rng_converter.rb

@@ -394,7 +394,7 @@ module Rng
           elsif part[:char_escape]
             # Map character escape
             char = extract_string(part[:char_escape][:char])
-            CHAR_ESCAPE_MAP[char] || char
+            CHAR_ESCAPE_MAP[char] || "\\#{char}"
           elsif part[:backslash_escape]
             # Backslash escape in identifier: \x -> x
             if part[:backslash_escape][:escaped_backslash]

data/lib/rng/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Rng
-  VERSION = '0.3.6'
+  VERSION = '0.3.7'
 end

data/spec/rng/rnc_parser_spec.rb

@@ -415,6 +415,103 @@ RSpec.describe Rng::RncParser do
       end
     end
 
+    context 'with named pattern after ## comment' do
+      let(:input) do
+        <<~RNC
+          document =
+            element document {
+              ## test 1
+              test
+            }
+
+          test =
+            element sections {
+              attribute test { text }?
+            }
+        RNC
+      end
+
+      let(:grammar) { described_class.parse(input) }
+
+      it 'parses without error' do
+        expect { grammar }.not_to raise_error
+      end
+
+      it 'contains the element document' do
+        expect(grammar.define.map(&:name)).to include('document')
+        document_def = grammar.define.find { |d| d.name == 'document' }
+        expect(document_def.element[0].attr_name).to eq('document')
+      end
+
+      it 'contains the element sections' do
+        expect(grammar.define.map(&:name)).to include('test')
+        test_def = grammar.define.find { |d| d.name == 'test' }
+        expect(test_def.element[0].attr_name).to eq('sections')
+      end
+    end
+
+    context 'with escape sequences in string literals' do
+      it 'parses known escapes (\\n, \\t, \\r, \\", \\\\)' do
+        input = <<~'RNC'
+          start = element value {
+            xsd:string { pattern = "hello\nworld\ttab\r\n" }
+          }
+        RNC
+        expect { described_class.parse(input) }.not_to raise_error
+      end
+
+      it 'parses RELAX NG class escapes (\\i, \\c, \\d, \\w)' do
+        input = <<~'RNC'
+          start = element value {
+            xsd:string { pattern = "\\i\\c\\d\\w" }
+          }
+        RNC
+        expect { described_class.parse(input) }.not_to raise_error
+      end
+
+      it 'parses unknown escape sequences (\\+, \\-, \\s, \\., \\,) preserving backslash' do
+        input = <<~'RNC'
+          start = element value {
+            xsd:string { pattern = "[\\+\\-]?\\d{4}" }
+          }
+        RNC
+        expect { described_class.parse(input) }.not_to raise_error
+      end
+
+      it 'parses ISO 8601 regex pattern with multiple escape sequences' do
+        input = <<~'RNC'
+          ISO8601DateTime = xsd:string { pattern = "([\\+\\-]?\\d{4})((-?)((0[1-9]|1[0-2])((-?)([12]\\d|0[1-9]|3[01]))?|W([0-4]\\d|5[0-2])(-?[1-7])?|(00[1-9]|0[1-9]\\d|[12]\\d{2}|3([0-5]\\d|6[1-6])))([T\\s]((([01]\\d|2[0-3])((:?)[0-5]\\d)?|24:?00)([\\.,]\\d+)?)?((:?)[0-5]\\d([.,]\\d+)?)?([zZ]|([\\+\\-])([01]\\d|2[0-3]):?([0-5]\\d)?)?)?)?)?)?" }
+          elem = element elem { ISO8601DateTime }
+        RNC
+        expect { described_class.parse(input) }.not_to raise_error
+      end
+
+      it 'rejects invalid hex escape \\x{}' do
+        # Single-quoted heredoc: \x{} is literal \x{} (one backslash)
+        input = <<~'RNC'
+          start = element value {
+            xsd:string { pattern = "\x{}" }
+          }
+        RNC
+        expect do
+          described_class.parse(input)
+        end.to raise_error(Parslet::ParseFailed)
+      end
+
+      it 'round-trips unknown escape sequences preserving backslash' do
+        rnc_input = <<~'RNC'
+          start = element value {
+            xsd:string { pattern = "[\\+\\-]?\\d{4}" }
+          }
+        RNC
+        grammar = described_class.parse(rnc_input)
+        rnc_output = described_class.to_rnc(grammar)
+        expect(rnc_output).to include('\\+')
+        expect(rnc_output).to include('\\-')
+        expect(rnc_output).to include('\\d')
+      end
+    end
+
     context 'with except patterns in anyName' do
       let(:input) do
         <<~RNC

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: rng
 version: !ruby/object:Gem::Version
-  version: 0.3.6
+  version: 0.3.7
 platform: ruby
 authors:
 - Ribose
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-05-15 00:00:00.000000000 Z
+date: 2026-05-29 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: lutaml-model