Advanced LLDB Formatters for C++ Data Structures

A powerful, modern, and extensible Python-based formatter package for LLDB that provides rich, interactive visualizations and summaries for common C++ data structures. Designed to work seamlessly with IDEs like VS Code (via CodeLLDB), this package goes beyond simple text summaries, offering in-IDE web-based visualizers.

---

Features

• Clean & Informative Summaries: Get concise, readable one-line summaries for your data structures directly in the debugger's variable panel.
• Interactive Visualizers: Go beyond text with rich, interactive visualizers for Lists, Trees, and Graphs that open directly inside VS Code.
• Console Pretty-Printing: Use custom commands like pptree to print a visual representation of your tree directly in the LLDB console.
• Graphviz Export: Export Trees and Graphs to .dot files for offline analysis and documentation using export_tree and export_graph.
• Highly configurable: Customize formatter behavior at runtime with the formatter_config command.
• Extensible architecture: Built with an advanced Strategy and Registry pattern, making it incredibly easy to add support for new data structures or visualization strategies.

Supported Data Structures

The formatters are designed to be generic and will automatically detect and format classes with common member names.

• Linear Containers: LinkedList, Stack, Queue (and custom variants).
  • Recognizes members like head, top, next, count, size.
• STL Containers: std::vector (libc++ layout).
  • Shows size, capacity, data pointer, and values preview.
• Trees: Binary Search Trees and other node-based trees.
  • Recognizes root, left, right, children, value.
• Graphs: Adjacency-list based graphs.
  • Recognizes nodes, adj, neighbors, value.

---

Installation

1. Clone the Repository:

   git clone https://github.com/XtremeXSPC/Pretty_LLDB.git /path/to/LLDB_Formatters

   A common location is ~/.lldb/, but any path will work.

2. Load the Formatters in LLDB: Add the following command to your global LLDB initialization file (~/.lldbinit):

   command script import /path/to/LLDB_Formatters
   

   This ensures the formatters are loaded every time you start a debug session.

3. (Optional) For VS Code users: You can also load the formatters per-project by adding it to your launch.json configuration:

   "configurations": [
       {
           "name": "Debug C++",
           "type": "lldb",
           "request": "launch",
           // ... your other settings ...
           "initCommands": [
               "command script import /path/to/LLDB_Formatters"
           ]
       }
   ]

---

Usage

Once installed, the formatters work automatically. Simply inspect your variables in the debugger GUI or use print in the LLDB console.

Custom Commands

This package adds several powerful commands to your LLDB console. Use fhelp to see them all.

Command	Alias	Description
formatter_help	fhelp	Displays a detailed list of all custom commands.
formatter_config	fconf	View or change global settings (e.g., formatter_config summary_max_items 50).
weblist <var>	-	Opens an interactive visualizer for a list.
webtree <var>	webt	Opens an interactive visualizer for a tree.
webgraph <var>	webg	Opens an interactive visualizer for a graph.
pptree <var>	-	Pretty-prints a tree structure in the console.
export_tree <var>	-	Exports a tree to a Graphviz .dot file.
export_graph <var>	-	Exports a graph to a Graphviz .dot file.

---

Architecture Overview

This project uses an advanced software architecture to ensure it is robust and easy to extend.

• Registry Pattern: Formatters are automatically discovered using Python decorators. Adding support for a new data structure is as simple as creating a new class—no need to edit central initialization files.
• Strategy Pattern: The logic for traversing a data structure (e.g., "pre-order traversal" for a tree) is separated from the presentation logic. This makes it easy to add new traversal algorithms without changing the core formatter code.
• Configuration Object: All settings are managed in a single, clean configuration object, providing a centralized point of control.

Contributing

Contributions are welcome! Whether it's adding a new visualizer, supporting a new data structure, or improving the documentation, your help is appreciated.

1. Fork the repository.
2. Create a new branch (git checkout -b feature/your-feature-name).
3. Make your changes. Follow the existing code style and architecture.
4. Submit a Pull Request.

---

License

This project is licensed under the MIT License. See the LICENSE file for details.