How yoshi-782/variousvcmconv works
Main Components
• Language Implementations: Within each version, include the language-specific components:
• New Version: C#, Dart, JavaScript, Python.
• Old Version: C#, Java, Python.
• Core Conversion Logic: Although each implementation might be coded differently, conceptually they wrap around the “conversion engine” which handles the logic of transforming text using either:
• The Unicode normalization strategy (for the New Version).
• The array substitution approach (for the Old Version).
• Shared Architectural Concepts: Indicate that for each language there is an entry point (e.g., main file in C# “Program.cs” or Dart’s “dakuten.dart”) that triggers the conversion process.
Picture the Relationships and Interactions
• For each language-specific component, show an entry point that invokes core conversion logic.
• Within each version’s section, illustrate that the conversion engine is applied per character:
• In the New Version: Steps include normalizing each character (using Unicode normalization to NFD), checking the character-length (1 or 2 characters), and then applying the appropriate logic.
• In the Old Version: Use array lookups to decide if a character gets replaced or concatenated with a diacritic.
Architectural Patterns and Design Principles
• Modular Architecture: Emphasize that each language’s implementation is self-contained. Even though they implement the same “conversion” function, they are independently maintained.
• Cross-Version Comparison: Optionally, show a shared conceptual space for “Conversion Engine” with branches into “New Version” and “Old Version” strategies.
Relevant Technologies and Frameworks
• C#: .NET Framework / .NET Core (as per the Program.cs and ConsoleApp1.csproj).
• Dart: Dart SDK (highlighting files like bin/dakuten.dart and configuration files such as analysis_options.yaml and pubspec.yaml).
• JavaScript: A standalone JS file (dakutenConv.js).
• Python: Python scripts (dakutenConv.py and OldDakuhen.py).
• Java in the Old Version.
• That external libraries or frameworks might be used for Unicode normalization (in languages that support it).
Final Detailed Architectural Elements
• Top-Level: Project repository split into “NewVersion” and “OldVersion”.
• For Each Version:
• Submodules for each language.
• That each language submodule contains an “Entry Point” which calls the “Conversion Routine”.
• General Flow: Input → Conversion Routine → Output.
• Extra Annotations: Add any notes/callouts that explain the rationale behind two distinct conversion strategies between versions.