json_mend 0.1.2 → 0.1.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 85624e37a002e82e9edcb6d7fffe16799429e75389662d8059e9e89f8b6f14d8
-  data.tar.gz: 19309bc419b6f4481193e140eea31c2d24153ee87265cf2f4e2ecca1f0421bf4
+  metadata.gz: 7aae532930afd3eabb3485f83995b7e1ee22763c4f17f1c7940af3b270f33b60
+  data.tar.gz: 52e364f66106063b8fec6335eff7171031cb4ac2e0f92a55cedae9d4febd205f
 SHA512:
-  metadata.gz: 85aa783092d768f3ff9543e2de2b59b0bbfe7a285e7d5db91bfb3d5c12dab233d2735f6f605ede6cab701b5085bc740332e91d93fddb90f9361b1033e0edf57e
-  data.tar.gz: 75d3dc3b22f72748fe5b21f916ba930a04bd03aaea0d9aca8126ce38b2fdf9cfae9ad38c33f35a7c4e0c39ee7784b79df06d60772bcdd062e529372cc1bbd733
+  metadata.gz: 62803c4814d02e3850e23bb64fba0f31fbbfc5d6fa73b84dd2e6dee7bf670709f7c778506e64a65d0c170e26c54c239fb6f27063183199dc760e18906832630c
+  data.tar.gz: 7657395e2d7fed895bdc2f405ac9cb019497f7445c64e9bc3900b188124910c353d4ee3673bee90009ccc6e133d7d2aad2552971e8e1fdf12e7c3671d31d1ca7

data/.rubocop.yml

@@ -7,19 +7,19 @@ AllCops:
   SuggestExtensions: false
 
 Metrics/AbcSize:
-  Max: 60
+  Max: 65
 
 Metrics/ClassLength:
-  Max: 800
+  Max: 820
 
 Metrics/CyclomaticComplexity:
-  Max: 30
+  Max: 35
 
 Metrics/MethodLength:
-  Max: 70
+  Max: 80
 
 Metrics/PerceivedComplexity:
-  Max: 32
+  Max: 35
 
 Metrics/BlockNesting:
   Max: 5

data/benchmark_comparison.rb

@@ -0,0 +1,144 @@
+# frozen_string_literal: true
+
+# gem install benchmark-ips json-repair json_mend
+# ruby -Ilib benchmark_comparison.rb
+require 'benchmark/ips'
+require 'json'
+
+# --- Load Libraries ---
+begin
+  require 'json_mend'
+rescue LoadError
+  abort "❌ Could not load 'json_mend'. Make sure you are in the gem root or have it installed."
+end
+
+begin
+  require 'json/repair'
+rescue LoadError
+  puts "❌ Could not load 'json-repair'. Benchmarks for it will be skipped."
+end
+
+puts '========================================================='
+puts '  🚀 JSON Repair Benchmark (IPS) '
+puts "     JsonMend (v#{JsonMend::VERSION}) vs json-repair-rb (v#{JSON::Repair::VERSION})"
+puts '========================================================='
+puts
+
+# --- Test Data ---
+
+json_object = '{"id": 1, "name": "Test", "active": true, "tags": ["a", "b"]}'
+
+TEST_CASES = {
+  valid_single: {
+    label: 'Valid Single JSON',
+    input: json_object
+  },
+  concatenated: {
+    label: 'Concatenated JSON (x10)',
+    input: json_object * 10
+  },
+  simple_fix: {
+    label: 'Simple Fix (Missing Quotes)',
+    input: '{name: "Alice", age: 30, city: "Wonderland"}'
+  },
+  trailing_comma: {
+    label: 'Trailing Commas',
+    input: '{"items": [1, 2, 3,], "active": true,}'
+  },
+  comments: {
+    label: 'Comments (// and #)',
+    input: <<~JSON
+      {
+        "key": "value", // This is a comment
+        "config": {
+          "timeout": 100 # Another comment
+        }
+      }
+    JSON
+  },
+  complex: {
+    label: 'Complex & Mixed Errors',
+    input: <<~JSON
+      {
+        name: "Broken",
+        "nested": [
+          {id: 1,},
+          {id: 2}
+        ],
+        "dangling": [1, 2, 3
+    JSON
+  },
+  garbage: {
+    label: 'Heavy Garbage / Hallucinations',
+    input: 'Here is the JSON: ```json {"a": 1} ``` and some other text.'
+  },
+  python_style: {
+    label: 'Python Literals (True/None)',
+    input: '{"is_valid": True, "missing": None, "wrong_bool": False}'
+  },
+  single_quotes: {
+    label: 'Single Quotes (JS Style)',
+    input: "{'id': 123, 'status': 'pending', 'meta': {'active': true}}"
+  },
+  deep_nesting: {
+    label: 'Deeply Nested (Stack Test)',
+    input: "#{'{"a":' * 50}1#{'}' * 50}"
+  },
+  unbalanced: {
+    label: 'Truncated / Unbalanced',
+    input: '{"users": [{"id": 1, "name": "Alice"}, {"id": 2'
+  },
+  unescaped_control: {
+    label: 'Unescaped Newlines/Tabs',
+    input: "{\"bio\": \"This is a \n multi-line string \t with tabs.\"}"
+  },
+  large_array: {
+    label: 'Large Single Array (Throughput)',
+    input: "[#{(1..1000).map { |i| %({"id": #{i}, "val": "item_#{i}"}) }.join(',')}]"
+  },
+  concatenated_complex: {
+    label: 'Concatenated + Broken (LLM Stream)',
+    input: '{"part": 1} {part: 2, "broken": true} {"part": 3}'
+  }
+}.freeze
+
+# Helper to check if a library supports the input before benchmarking
+def supported?(library_proc, input)
+  library_proc.call(input)
+  true
+rescue StandardError
+  false
+end
+
+# --- Run Benchmarks ---
+
+TEST_CASES.each_value do |data|
+  puts "\n\n🔸 Scenario: #{data[:label]}"
+  puts '-' * 40
+
+  Benchmark.ips do |x|
+    x.config(time: 2, warmup: 1) # Short duration for quick checks
+
+    # 1. JsonMend
+    if supported?(->(i) { JsonMend.repair(i) }, data[:input])
+      x.report('JsonMend') do
+        JsonMend.repair(data[:input])
+      end
+    else
+      puts '   JsonMend: ❌ Not Supported'
+    end
+
+    # 2. json-repair
+    if defined?(JSON::Repair)
+      if supported?(->(i) { JSON.repair(i) }, data[:input])
+        x.report('json-repair') do
+          JSON.repair(data[:input])
+        end
+      else
+        puts '   json-repair: ❌ Not Supported'
+      end
+    end
+
+    x.compare!
+  end
+end

data/lib/json_mend/parser.rb

@@ -222,8 +222,10 @@ module JsonMend
     # Parses the key of an object, including the special logic for merging dangling arrays.
     # Returns [key, was_array_merged_flag]
     def parse_object_key(object)
+      char = peek_char
+
       # First, check for and handle the dangling array merge logic.
-      if try_to_merge_dangling_array(object)
+      if char == '[' && try_to_merge_dangling_array(object)
         return [nil, true, false] # Signal that an array was merged.
       end
 
@@ -231,7 +233,7 @@ module JsonMend
       @context.push(:object_key)
       is_bracketed = false
 
-      if peek_char == '['
+      if char == '['
         @scanner.getch # Consume '['
         arr = parse_array
         key = arr.first.to_s
@@ -355,7 +357,7 @@ module JsonMend
       char = prepare_string_parsing
 
       # A valid string can only start with a valid quote or, in our case, with a literal
-      while !@scanner.eos? && !STRING_DELIMITERS.include?(char) && !char.match?(/[\p{L}0-9]/)
+      while !@scanner.eos? && !STRING_DELIMITERS.include?(char) && !char.match?(/[\p{L}0-9$_-]/)
         return '' if TERMINATORS_STRING_GUESSED.include?(char)
 
         @scanner.getch
@@ -434,7 +436,7 @@ module JsonMend
       when '“'
         lstring_delimiter = '“'
         rstring_delimiter = '”'
-      when /[\p{L}0-9]/
+      when /[\p{L}0-9$_-]/
         # Could be a boolean/null, but not if it's an object key.
         if BOOLEAN_OR_NULL_CHARS.include?(char.downcase) && !current_context?(:object_key)
           # parse_literal is non-destructive if it fails to match.
@@ -509,6 +511,17 @@ module JsonMend
       unmatched_delimiter = false
       # --- Main Parsing Loop ---
       while !@scanner.eos? && char != rstring_delimiter
+        # Fast-path for unquoted keys (e.g. { key: val })
+        # consumes a chunk of valid identifier characters at once
+        if missing_quotes && current_context?(:object_key)
+          chunk = @scanner.scan(/[a-zA-Z0-9_$-]+/)
+          if chunk
+            string_parts << chunk
+            char = peek_char
+            next
+          end
+        end
+
         break if context_termination_reached?(
           char:,
           missing_quotes:
@@ -1170,16 +1183,29 @@ module JsonMend
 
     # Peeks the next character without advancing the scanner
     def peek_char(offset = 0)
-      return @scanner.check(/./m) if offset.zero?
+      # Handle the common 0-offset case
+      if offset.zero?
+        # peek(1) returns the next BYTE, not character
+        byte_str = @scanner.peek(1)
+        return nil if byte_str.empty?
+
+        # Fast path: If it's a standard ASCII char (0-127), return it directly.
+        # This avoids the regex overhead for standard JSON characters ({, [, ", etc).
+        return byte_str if byte_str.getbyte(0) < 128
+
+        # Slow path: If it's a multibyte char (e.g. “), use regex to match the full character.
+        return @scanner.check(/./m)
+      end
 
+      # For offsets > 0, we must scan to skip correctly (as characters can be variable width)
       saved_pos = @scanner.pos
-      c = nil
+      res = nil
       (offset + 1).times do
-        c = @scanner.getch
-        break if c.nil?
+        res = @scanner.getch
+        break if res.nil?
       end
       @scanner.pos = saved_pos
-      c
+      res
     end
 
     def current_context?(value)
@@ -1192,7 +1218,7 @@ module JsonMend
 
     # Checks if the character signifies the start of a string or literal
     def string_start?(char)
-      STRING_DELIMITERS.include?(char) || char&.match?(/\p{L}/)
+      STRING_DELIMITERS.include?(char) || char&.match?(/[\p{L}$_]/)
     end
 
     # Checks if the character signifies the start of a number

data/lib/json_mend/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module JsonMend
-  VERSION = '0.1.2'
+  VERSION = '0.1.4'
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: json_mend
 version: !ruby/object:Gem::Version
-  version: 0.1.2
+  version: 0.1.4
 platform: ruby
 authors:
 - Oleksii Vasyliev
@@ -15,28 +15,28 @@ dependencies:
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: '0'
+        version: '2.3'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: '0'
+        version: '2.3'
 - !ruby/object:Gem::Dependency
   name: strscan
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: '0'
+        version: '3'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: '0'
+        version: '3'
 description: JsonMend is a robust Ruby gem designed to repair broken or malformed
   JSON strings. It is specifically optimized to handle common errors found in JSON
   generated by Large Language Models (LLMs), such as missing quotes, trailing commas,
@@ -54,6 +54,7 @@ files:
 - LICENSE
 - README.md
 - Rakefile
+- benchmark_comparison.rb
 - lib/json_mend.rb
 - lib/json_mend/parser.rb
 - lib/json_mend/version.rb