parsanol 1.3.8-aarch64-linux → 1.3.10-aarch64-linux

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 11fa862698ade99f509fc95b23031680cbe951a0a16cde95d0e1766108741479
-  data.tar.gz: 0e32431822dc551ea6ed7b6f4642c64481ffb9b61c011e870007d5212fc52d47
+  metadata.gz: 69ee119824b358519e2833e4d77b3c58a8921e4dede01700148c95024e17b5a2
+  data.tar.gz: 46fd8493a20d6277510345bb5c692bfcc7e9227ad8ce3b934d788d9bb120db3a
 SHA512:
-  metadata.gz: 3a2ceb99dc545d14fa8d84aad4913aa72354365a81af405b446af665c5a3d6605f52916b902bfd2b68f201a6089816a022e826c5226cd313c4b387913c89bb31
-  data.tar.gz: bc39f6141c521d4212224306f5bca82adaf83a78123890a99d20f1830b7ce47179622a90d29ab7b7982b94990f22d62bb79c6ff1bff240f5b82761d3584f4911
+  metadata.gz: 253408a7fc6923ff834c46c6dcd80466b1b1bc6a354d1b0d5cf895250d0b84552687b14d0405c95ea1501d5aca88aa6a1ac4d5e7cded56118e352875050566a9
+  data.tar.gz: 41938dced4ae4395a1cdb71aed258c7c209eb8173b912e84869887ffce83e37950e01a576eaebef21caf62f920d2139240a129cb31b1fa39f84315b12f7a190a

data/lib/parsanol/grammar_builder.rb

@@ -115,7 +115,7 @@ module Parsanol
     # Convert to JSON for native parser
     #
     # @return [String] JSON representation
-    def to_json(*_args)
+    def to_json(*)
       build.to_json
     end
 

data/lib/parsanol/native/batch_decoder.rb

@@ -57,9 +57,8 @@ module Parsanol
         #   pre-decoded Ruby value from _parse_raw
         # @param input [String] Original input string (for Slice references)
         # @param slice_class [Class] The Slice class to use
-        # @param grammar_atom [Parsanol::Atoms::Base] The grammar atom (unused, kept for API compat)
         # @return [Object] Transformed Ruby AST
-        def decode_and_flatten(data, input, slice_class, _grammar_atom)
+        def decode_and_flatten(data, input, slice_class)
           # Check if data is batch data (flat u64 array) or already a Ruby value
           if data.is_a?(Integer) || (data.is_a?(Array) && data.first.is_a?(Integer))
             # Batch data (flat u64 array) - decode first, then transform

data/lib/parsanol/native/transformer.rb

@@ -323,9 +323,26 @@ module Parsanol
               all_items_are_hashes = non_hash_items.empty?
 
               if all_items_are_hashes
-                # Merge all inner hashes into merged_hash
-                item.each do |sub_item|
-                  merged_hash.merge!(sub_item) if sub_item.is_a?(Hash)
+                # Check if merging would overwrite existing keys in merged_hash.
+                # If so, this is a repetition pattern (item >> (sep >> item).repeat)
+                # and should be kept as array, not merged.
+                # Example: merged_hash={namedTypeOrRename: A}, array=[{namedTypeOrRename: B}]
+                # → should produce [{namedTypeOrRename: A}, {namedTypeOrRename: B}]
+                existing_keys = merged_hash.keys
+                shares_keys = item.any? do |sub_item|
+                  sub_item.is_a?(Hash) && sub_item.keys.intersect?(existing_keys)
+                end
+
+                if shares_keys
+                  has_non_empty_array = true
+                  item.each do |sub_item|
+                    hash_count += 1 if sub_item.is_a?(Hash)
+                  end
+                  total_items += 1
+                else
+                  item.each do |sub_item|
+                    merged_hash.merge!(sub_item) if sub_item.is_a?(Hash)
+                  end
                 end
               else
                 # Non-empty repetition with non-hash items - mark that we should keep as array

data/lib/parsanol/native/types.rb

@@ -25,18 +25,5 @@ module Parsanol
       EMPTY_ARRAY = [].freeze
       EMPTY_HASH = {}.freeze
     end
-
-    # Symbol cache to avoid repeated string-to-symbol conversions
-    # This is a class variable to share across all transformations
-    @@symbol_cache = {}
-
-    # Convert string key to symbol with caching
-    # @param key [String, Symbol] The key to convert
-    # @return [Symbol] The symbol version of the key
-    def self.cached_symbol(key)
-      return key if key.is_a?(Symbol)
-
-      @@symbol_cache[key] ||= key.to_sym
-    end
   end
 end

data/lib/parsanol/native.rb

@@ -10,8 +10,6 @@ require "parsanol/native/batch_decoder"
 
 module Parsanol
   module Native
-    VERSION = "0.1.0"
-
     class << self
       # Check if native extension is available
       def available?
@@ -41,20 +39,38 @@ module Parsanol
         raise LoadError, "Native parser not available" unless available?
 
         # Handle both grammar atoms and pre-serialized JSON strings
-        if grammar.is_a?(String)
-          grammar_json = grammar
-          grammar_atom = nil
-        else
-          grammar_json = Parser.serialize_grammar(grammar)
-          grammar_atom = grammar
-        end
+        grammar_json = if grammar.is_a?(String)
+                         grammar
+                       else
+                         Parser.serialize_grammar(grammar)
+                       end
 
         # Use _parse_raw which returns properly tagged Ruby arrays via transform_ast.
         # The batch format doesn't preserve :repetition/:sequence tags, so we use
         # the direct FFI path. Apply the Ruby transformer to handle tags correctly.
         raw_ast = _parse_raw(grammar_json, input)
-        BatchDecoder.decode_and_flatten(raw_ast, input, Parsanol::Slice,
-                                        grammar_atom)
+        BatchDecoder.decode_and_flatten(raw_ast, input, Parsanol::Slice)
+      end
+
+      # Memory-bounded parsing without packrat cache.
+      #
+      # This creates a fresh arena and empty cache per call, bounding memory
+      # to AST size rather than input × atoms. Use for large files.
+      #
+      # @param grammar [Parsanol::Atoms::Base] Ruby grammar definition
+      # @param input [String] Input string to parse
+      # @return [Hash, Array, Parsanol::Slice] Transformed AST
+      def parse_fresh(grammar, input)
+        raise LoadError, "Native parser not available" unless available?
+
+        grammar_json = if grammar.is_a?(String)
+                         grammar
+                       else
+                         Parser.serialize_grammar(grammar)
+                       end
+
+        raw_ast = _parse_fresh_raw(grammar_json, input)
+        BatchDecoder.decode_and_flatten(raw_ast, input, Parsanol::Slice)
       end
 
       # Parse and return RAW AST without transformation.
@@ -166,13 +182,6 @@ module Parsanol
         end
         stats
       end
-
-      private
-
-      # Get the Slice class
-      def get_slice_class
-        Parsanol::Slice
-      end
     end
   end
 end

data/lib/parsanol/options/zero_copy.rb

@@ -89,11 +89,10 @@ module Parsanol
     # Parse input and return direct Ruby objects (no serialization)
     #
     # @param input [String] The input string to parse
-    # @param options [Hash] Parse options (ignored for zero-copy)
     # @return [Object] Direct Ruby object (type depends on grammar)
     # @raise [LoadError] If native extension not available
     # @raise [Parsanol::ParseFailed] If parsing fails
-    def parse(input, _options = {})
+    def parse(input)
       unless Parsanol::Native.available?
         raise LoadError,
               "ZeroCopy mode requires native extension for direct FFI object construction. " \

data/lib/parsanol/slice.rb

@@ -67,6 +67,10 @@ module Parsanol
 
     alias length size
 
+    def empty?
+      content.empty?
+    end
+
     def +(other)
       self.class.new(@byte_position, content + other.to_s, @input)
     end
@@ -110,7 +114,7 @@ module Parsanol
       as_json.to_json(*)
     end
 
-    def as_json(_options = {})
+    def as_json
       result = { "value" => content, "offset" => offset, "length" => length }
       if @input
         line, column = line_and_column

data/lib/parsanol/source/line_cache.rb

@@ -97,13 +97,6 @@ module Parsanol
           return hi if hi <= lo
         end
       end
-
-      # Legacy method name for backward compatibility
-      alias find_mid midpoint_index
-      alias lbound lower_bound_index
     end
-
-    # Legacy constant name for backward compatibility
-    RangeSearch = IntervalLookup
   end
 end

data/lib/parsanol/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Parsanol
-  VERSION = "1.3.8"
+  VERSION = "1.3.10"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: parsanol
 version: !ruby/object:Gem::Version
-  version: 1.3.8
+  version: 1.3.10
 platform: aarch64-linux
 authors:
 - Ribose Inc.
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2026-03-26 00:00:00.000000000 Z
+date: 2026-04-12 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
@@ -151,7 +151,6 @@ files:
 - README.adoc
 - Rakefile
 - lib/parsanol.rb
-- lib/parsanol/3.2/parsanol_native.so
 - lib/parsanol/3.3/parsanol_native.so
 - lib/parsanol/3.4/parsanol_native.so
 - lib/parsanol/4.0/parsanol_native.so
@@ -262,7 +261,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: '3.2'
+      version: '3.3'
   - - "<"
     - !ruby/object:Gem::Version
       version: 4.1.dev