expressir 2.2.0 → 2.3.0

data/benchmark/srl_native_benchmark.rb

@@ -0,0 +1,142 @@
+#!/usr/bin/env ruby
+# frozen_string_literal: true
+
+# SRL Benchmark Script for Expressir - Native Parser Only
+# Tests Rust native parser performance on full STEPmod Resource Library
+
+require 'bundler/setup'
+require 'benchmark'
+require 'fileutils'
+
+# Force loading of native extension first
+require 'parsanol'
+require 'parsanol/native'
+
+# Now require expressir
+require 'expressir'
+
+# Configuration
+SRL_PATH = '/Users/mulgogi/src/mn/iso-10303/schemas/resources'
+
+def find_exp_files
+  Dir.glob("#{SRL_PATH}/*/*.exp").sort
+end
+
+def count_lines(files)
+  files.sum { |f| File.read(f).lines.count }
+end
+
+puts "=" * 70
+puts "EXPRESSIR SRL BENCHMARK: Native Parser (Rust)"
+puts "=" * 70
+puts
+
+# Check native availability
+native_available = Parsanol::Native.available?
+puts "Native parser available: #{native_available ? 'YES' : 'NO'}"
+puts
+
+if !native_available
+  puts "Native parser not available. Run `rake compile` in parsanol-ruby."
+  exit 1
+end
+
+# Find all EXPRESS files
+files = find_exp_files
+total_lines = count_lines(files)
+
+puts "Files to parse: #{files.size}"
+puts "Total lines of code: #{total_lines}"
+puts
+
+if files.empty?
+  puts "ERROR: No EXPRESS files found at #{SRL_PATH}"
+  exit 1
+end
+
+# Warmup - parse first file to ensure everything is loaded
+puts "Warming up..."
+warmup_file = files.first
+begin
+  content = File.read(warmup_file)
+  Expressir::Express::Parser.from_exp(content, skip_references: true, use_native: true)
+  puts "Warmup complete."
+rescue => e
+  puts "Warning: Warmup failed: #{e.message}"
+  puts e.backtrace.first(5)
+end
+puts
+
+# Run Native benchmark
+puts "-" * 70
+puts "NATIVE PARSER BENCHMARK"
+puts "-" * 70
+
+results = { success: 0, failed: 0, errors: [], total_time: 0 }
+file_times = []
+
+time = Benchmark.realtime do
+  files.each_with_index do |file, idx|
+    file_start = Time.now
+    begin
+      content = File.read(file)
+      Expressir::Express::Parser.from_exp(content, skip_references: true, use_native: true)
+      results[:success] += 1
+      file_time = Time.now - file_start
+      file_times << { file: File.basename(file), time: file_time, lines: content.lines.count }
+
+      # Progress every 10 files
+      if (idx + 1) % 10 == 0
+        print "."
+        $stdout.flush
+      end
+    rescue => e
+      results[:failed] += 1
+      results[:errors] << { file: File.basename(file), error: e.message[0..100] }
+    end
+  end
+end
+puts
+
+results[:total_time] = time
+
+# Calculate stats
+lines_per_sec = (total_lines / results[:total_time]).round(1)
+files_per_sec = (files.size / results[:total_time]).round(2)
+
+puts
+puts "Native Parser Results:"
+puts "  Success: #{results[:success]}/#{files.size} files"
+puts "  Failed:  #{results[:failed]} files"
+puts "  Time:    #{results[:total_time].round(2)}s"
+puts "  Speed:   #{lines_per_sec} lines/sec"
+puts "  Speed:   #{files_per_sec} files/sec"
+
+if results[:failed] > 0
+  puts
+  puts "Errors (first 10):"
+  results[:errors].first(10).each do |err|
+    puts "  - #{err[:file]}: #{err[:error]}"
+  end
+end
+
+# Show top 5 slowest files
+puts
+puts "Top 5 slowest files:"
+sorted = file_times.sort_by { |f| -f[:time] }.first(5)
+sorted.each do |f|
+  puts "  #{f[:file]}: #{f[:time].round(2)}s (#{f[:lines]} lines)"
+end
+
+# Show top 5 fastest files
+puts
+puts "Top 5 fastest files:"
+sorted = file_times.sort_by { |f| f[:time] }.first(5)
+sorted.each do |f|
+  puts "  #{f[:file]}: #{f[:time].round(3)}s (#{f[:lines]} lines)"
+end
+
+puts
+puts "=" * 70
+puts "BENCHMARK COMPLETE"
+puts "=" * 70

data/benchmark/srl_ruby_benchmark.rb

@@ -0,0 +1,130 @@
+#!/usr/bin/env ruby
+# frozen_string_literal: true
+
+# SRL Benchmark Script for Expressir - Ruby Parser Only
+# Tests pure Ruby parser performance on full STEPmod Resource Library
+
+require 'bundler/setup'
+require 'benchmark'
+require 'fileutils'
+
+# Load expressir WITHOUT native extension
+require 'expressir'
+
+# Configuration
+SRL_PATH = '/Users/mulgogi/src/mn/iso-10303/schemas/resources'
+
+def find_exp_files
+  Dir.glob("#{SRL_PATH}/*/*.exp").sort
+end
+
+def count_lines(files)
+  files.sum { |f| File.read(f).lines.count }
+end
+
+puts "=" * 70
+puts "EXPRESSIR SRL BENCHMARK: Pure Ruby Parser"
+puts "=" * 70
+puts
+
+# Find all EXPRESS files
+files = find_exp_files
+total_lines = count_lines(files)
+
+puts "Files to parse: #{files.size}"
+puts "Total lines of code: #{total_lines}"
+puts
+
+if files.empty?
+  puts "ERROR: No EXPRESS files found at #{SRL_PATH}"
+  exit 1
+end
+
+# Warmup - parse first file to ensure everything is loaded
+puts "Warming up..."
+warmup_file = files.first
+begin
+  Expressir::Express::Parser.from_file(warmup_file, skip_references: true)
+  puts "Warmup complete."
+rescue => e
+  puts "Warning: Warmup failed: #{e.message}"
+end
+puts
+
+# Run Ruby benchmark
+puts "-" * 70
+puts "RUBY PARSER BENCHMARK"
+puts "-" * 70
+
+results = { success: 0, failed: 0, errors: [], total_time: 0 }
+file_times = []
+
+time = Benchmark.realtime do
+  files.each_with_index do |file, idx|
+    file_start = Time.now
+    begin
+      Expressir::Express::Parser.from_file(file, skip_references: true)
+      results[:success] += 1
+      file_time = Time.now - file_start
+      content = File.read(file)
+      file_times << { file: File.basename(file), time: file_time, lines: content.lines.count }
+
+      # Progress every 10 files
+      if (idx + 1) % 10 == 0
+        print "."
+        $stdout.flush
+      end
+    rescue => e
+      results[:failed] += 1
+      results[:errors] << { file: File.basename(file), error: e.message[0..100] }
+    end
+  end
+end
+puts
+
+results[:total_time] = time
+
+# Calculate stats
+lines_per_sec = (total_lines / results[:total_time]).round(1)
+files_per_sec = (files.size / results[:total_time]).round(2)
+
+puts
+puts "Ruby Parser Results:"
+puts "  Success: #{results[:success]}/#{files.size} files"
+puts "  Failed:  #{results[:failed]} files"
+puts "  Time:    #{results[:total_time].round(2)}s"
+puts "  Speed:   #{lines_per_sec} lines/sec"
+puts "  Speed:   #{files_per_sec} files/sec"
+
+if results[:failed] > 0
+  puts
+  puts "Errors (first 10):"
+  results[:errors].first(10).each do |err|
+    puts "  - #{err[:file]}: #{err[:error]}"
+  end
+end
+
+# Show top 5 slowest files
+puts
+puts "Top 5 slowest files:"
+sorted = file_times.sort_by { |f| -f[:time] }.first(5)
+sorted.each do |f|
+  puts "  #{f[:file]}: #{f[:time].round(2)}s (#{f[:lines]} lines)"
+end
+
+# Show top 5 fastest files
+puts
+puts "Top 5 fastest files:"
+sorted = file_times.sort_by { |f| f[:time] }.first(5)
+sorted.each do |f|
+  puts "  #{f[:file]}: #{f[:time].round(3)}s (#{f[:lines]} lines)"
+end
+
+puts
+puts "=" * 70
+puts "BENCHMARK COMPLETE"
+puts "=" * 70
+puts
+puts "NOTE: Native parser benchmark skipped due to AST format incompatibility."
+puts "The native parser returns a different AST format than Expressir expects."
+puts "This requires further work to implement sequence merging in Rust."

data/expressir.gemspec

@@ -32,13 +32,15 @@ Gem::Specification.new do |spec|
   end
 
   spec.add_dependency "base64"
+  spec.add_dependency "benchmark"
   spec.add_dependency "benchmark-ips"
   spec.add_dependency "csv"
   spec.add_dependency "liquid"
-  spec.add_dependency "lutaml-model"
+  spec.add_dependency "lutaml-model", "~> 0.8.0"
   spec.add_dependency "moxml"
   spec.add_dependency "paint"
-  spec.add_dependency "parsanol", "~> 1.1"
+  spec.add_dependency "parsanol", "~> 1.3.7", ">= 1.3.7"
+  spec.add_dependency "nokogiri"
   spec.add_dependency "ruby-progressbar", "~> 1.11"
   spec.add_dependency "rubyzip", "~> 2.3"
   spec.add_dependency "table_tennis"

data/lib/expressir/benchmark.rb

@@ -268,7 +268,7 @@ module Expressir
             result.each do |r|
               if r.is_a?(Model::Declarations::Schema)
                 schema_count += 1
-                total_objects += 1   # Count the schema itself
+                total_objects += 1 # Count the schema itself
               end
             end
           end

data/lib/expressir/changes/item_change.rb

@@ -1,6 +1,5 @@
 # frozen_string_literal: true
 
-require "lutaml/model"
 
 module Expressir
   module Changes

data/lib/expressir/changes/mapping_change.rb

@@ -1,6 +1,5 @@
 # frozen_string_literal: true
 
-require "lutaml/model"
 
 module Expressir
   module Changes

data/lib/expressir/changes/schema_change.rb

@@ -1,7 +1,5 @@
 # frozen_string_literal: true
 
-require "lutaml/model"
-require_relative "version_change"
 
 module Expressir
   module Changes

data/lib/expressir/changes/version_change.rb

@@ -1,8 +1,5 @@
 # frozen_string_literal: true
 
-require "lutaml/model"
-require_relative "item_change"
-require_relative "mapping_change"
 
 module Expressir
   module Changes

data/lib/expressir/changes.rb

@@ -1,10 +1,9 @@
-# frozen_string_literal: true
-
+# lib/expressir/changes.rb
 module Expressir
-  # Module for EXPRESS schema change tracking and management
   module Changes
+    autoload :SchemaChange, "#{__dir__}/changes/schema_change"
+    autoload :VersionChange, "#{__dir__}/changes/version_change"
+    autoload :ItemChange, "#{__dir__}/changes/item_change"
+    autoload :MappingChange, "#{__dir__}/changes/mapping_change"
   end
 end
-
-require_relative "changes/schema_change"
-require_relative "changes/mapping_change"

data/lib/expressir/cli.rb

@@ -1,20 +1,6 @@
 require "thor"
 require "yaml"
 
-require_relative "commands/base"
-require_relative "commands/format"
-require_relative "commands/clean"
-require_relative "commands/benchmark"
-require_relative "commands/benchmark_cache"
-require_relative "commands/validate"
-require_relative "commands/validate_load"
-require_relative "commands/validate_ascii"
-require_relative "commands/coverage"
-require_relative "commands/changes"
-require_relative "commands/version"
-require_relative "commands/package"
-require_relative "commands/manifest"
-
 module Expressir
   class Cli < Thor
     # Exit with error code on command failures
@@ -33,14 +19,14 @@ module Expressir
                                desc: "Include provenance information (ELF profile only)",
                                default: true
     def format(path)
-      Commands::Format.new(options).run(path)
+      Expressir::Commands::Format.new(options).run(path)
     end
 
     desc "clean PATH", "Strip remarks and prettify EXPRESS schema at PATH"
     method_option :output, type: :string,
                            desc: "Output file path (defaults to stdout)"
     def clean(path)
-      Commands::Clean.new(options).run(path)
+      Expressir::Commands::Clean.new(options).run(path)
     end
 
     desc "benchmark FILE_OR_YAML",
@@ -52,7 +38,7 @@ module Expressir
     method_option :format, type: :string,
                            desc: "Output format (json, csv, default)"
     def benchmark(path)
-      Commands::Benchmark.new(options).run(path)
+      Expressir::Commands::Benchmark.new(options).run(path)
     end
 
     desc "benchmark-cache FILE_OR_YAML", "Benchmark schema loading with caching"
@@ -65,11 +51,11 @@ module Expressir
     method_option :cache_path, type: :string,
                                desc: "Path to store the cache file"
     def benchmark_cache(path)
-      Commands::BenchmarkCache.new(options).run(path)
+      Expressir::Commands::BenchmarkCache.new(options).run(path)
     end
 
     desc "validate SUBCOMMAND", "EXPRESS schema validation commands"
-    subcommand "validate", Commands::Validate
+    subcommand "validate", Expressir::Commands::Validate
 
     desc "coverage *PATH",
          "List EXPRESS entities and check documentation coverage"
@@ -82,21 +68,21 @@ module Expressir
     method_option :ignore_files, type: :string,
                                  desc: "Path to YAML file containing array of files to ignore from overall coverage calculation"
     def coverage(*paths)
-      Commands::Coverage.new(options).run(paths)
+      Expressir::Commands::Coverage.new(options).run(paths)
     end
 
     desc "changes SUBCOMMAND", "Commands for EXPRESS Changes files"
-    subcommand "changes", Commands::Changes
+    subcommand "changes", Expressir::Commands::Changes
 
     desc "manifest SUBCOMMAND", "Schema manifest management commands"
-    subcommand "manifest", Commands::Manifest
+    subcommand "manifest", Expressir::Commands::Manifest
 
     desc "package SUBCOMMAND", "LER package management commands"
-    subcommand "package", Commands::Package
+    subcommand "package", Expressir::Commands::Package
 
     desc "version", "Expressir Version"
     def version
-      Commands::Version.new(options).run
+      Expressir::Commands::Version.new(options).run
     end
   end
 end

data/lib/expressir/commands/base.rb

@@ -17,15 +17,8 @@ module Expressir
         end
       end
 
-      def exit_with_error(message, exit_code = 1)
-        say message
-        # In test mode, raise an exception instead of exiting
-        # This makes it easier to test error cases
-        if defined?(@test_mode) && @test_mode
-          raise message # Just raise the message as an exception
-        else
-          exit exit_code
-        end
+      def exit_with_error(message, _exit_code = 1)
+        raise Expressir::CommandError.new(message)
       end
     end
   end

data/lib/expressir/commands/changes.rb

@@ -16,7 +16,6 @@ module Expressir
       method_option :verbose, type: :boolean,
                               desc: "Show verbose output"
       def validate(path)
-        require_relative "changes_validate"
         ChangesValidate.new(options).run(path)
       end
 
@@ -27,7 +26,6 @@ module Expressir
       method_option :verbose, type: :boolean,
                               desc: "Show verbose output"
       def import_eengine(input_xml, schema_name, version)
-        require_relative "changes_import_eengine"
         ChangesImportEengine.call(input_xml, options[:output], schema_name,
                                   version, **options)
       end

data/lib/expressir/commands/changes_import_eengine.rb

@@ -1,8 +1,5 @@
 # frozen_string_literal: true
 
-require_relative "base"
-require_relative "../eengine/compare_report"
-
 module Expressir
   module Commands
     # Command to import eengine comparison XML to EXPRESS Changes YAML

data/lib/expressir/commands/changes_validate.rb

@@ -1,7 +1,5 @@
 # frozen_string_literal: true
 
-require_relative "base"
-
 module Expressir
   module Commands
     # Command to validate and normalize EXPRESS Changes YAML files

data/lib/expressir/commands/format.rb

@@ -12,9 +12,11 @@ module Expressir
         when "iso"
           format_with_iso_profile(repository)
         else
-          say "Error: Unknown profile '#{profile}'. Valid profiles are: 'iso', 'elf'",
-              :red
-          exit 1
+          raise Expressir::InvalidOptionError.new(
+            "profile",
+            profile,
+            valid_options: ["iso", "elf"],
+          )
         end
       end
 

data/lib/expressir/commands/manifest.rb

@@ -1,8 +1,6 @@
 # frozen_string_literal: true
 
 require "thor"
-require_relative "../schema_manifest"
-require_relative "../model/dependency_resolver"
 
 module Expressir
   module Commands
@@ -124,7 +122,6 @@ module Expressir
         File.write(opts[:output], manifest.to_yaml)
 
         # Use validator for consistent warnings (DRY - single source of truth)
-        require_relative "../manifest/validator"
         validator = Expressir::Manifest::Validator.new(manifest, opts)
         path_warnings = validator.validate_path_completeness
 
@@ -241,7 +238,6 @@ module Expressir
         end
 
         # Create resolver and resolve paths
-        require_relative "../manifest/resolver"
         resolver = Expressir::Manifest::Resolver.new(manifest, resolver_options)
 
         say "Attempting to resolve paths..." if options[:verbose]
@@ -252,7 +248,6 @@ module Expressir
         File.write(options[:output], resolved_manifest.to_yaml)
 
         # Display statistics based on RESOLVED manifest (like manifest create)
-        require_relative "../manifest/validator"
         validator = Expressir::Manifest::Validator.new(resolved_manifest,
                                                        options)
         path_warnings = validator.validate_path_completeness
@@ -323,7 +318,6 @@ module Expressir
         manifest = Expressir::SchemaManifest.from_file(manifest_path)
 
         # Create validator - SINGLE SOURCE OF TRUTH
-        require_relative "../manifest/validator"
         validator = Expressir::Manifest::Validator.new(manifest, opts)
 
         # Run validations using validator (NO INLINE LOGIC)
@@ -352,7 +346,6 @@ module Expressir
       # @param schema_path [String] Path to the schema file
       # @return [String] The schema name declared in the file
       def extract_schema_name(schema_path)
-        require_relative "../express/parser"
         repo = Expressir::Express::Parser.from_file(schema_path)
         schema = repo.schemas.first
         schema&.id || File.basename(schema_path, ".*")