A line-based programming language optimized for accurate LLM code generation

日本語版 README

Sui (粋) is a programming language named after the Japanese aesthetic concept meaning "refined" and "elimination of excess." It is designed with LLM (Large Language Model) code generation accuracy as the top priority.

1. Line Independence - Each line is completely self-contained
2. Minimal Bracket Matching - Nesting only for function blocks {}
3. Single-Character Instructions - Maximum token efficiency
4. Sequential Variables - No meaningful names needed (v0, v1, g0, a0)
5. Explicit Control Flow - Labels and jumps

# PyPI
pip install sui-lang

# Homebrew (macOS/Linux)
brew tap TakatoHonda/sui
brew install sui-lang

# From source
git clone https://github.com/TakatoHonda/sui-lang.git
cd sui-lang

# Start REPL
sui

# Example session
>>> = v0 10
>>> + v1 v0 5
>>> . v1
>>>
15
>>> .exit

Commands: .exit / .quit (exit), .reset (reset state)

# Run file
sui examples/fibonacci.sui

# Run with arguments
sui examples/fib_args.sui 15

# Validate
sui --validate examples/fibonacci.sui

# Show help
sui --help

# Show converted code
sui2py examples/fibonacci.sui

# Output to file
sui2py examples/fibonacci.sui -o fib.py

# Convert and execute
sui2py examples/fib_args.sui --run 15

# Show converted code
py2sui your_code.py

# Output to file
py2sui your_code.py -o output.sui

# Using python directly
python sui.py examples/fibonacci.sui
python sui2py.py examples/fibonacci.sui
python py2sui.py your_code.py

# 0 1 {
< v0 a0 2
! v1 v0
? v1 1
^ a0
: 1
- v2 a0 1
$ v3 0 v2
- v4 a0 2
$ v5 0 v4
+ v6 v3 v5
^ v6
}
= g0 10
$ g1 0 g0
. g1

Output: 55

; Math functions
P g0 "math.sqrt" 16
. g0

; Random number
P g1 "random.randint" 1 100
. g1

; Type conversion
P g2 "int" "123"
+ g3 g2 1
. g3

Output: 4.0, random number, 124

= v0 1
: 0
> v1 v0 100
? v1 9
% v2 v0 15
~ v3 v2 0
? v3 1
% v4 v0 3
~ v5 v4 0
? v5 2
% v6 v0 5
~ v7 v6 0
? v7 3
. v0
@ 4
: 1
. "FizzBuzz"
@ 4
: 2
. "Fizz"
@ 4
: 3
. "Buzz"
@ 4
: 4
+ v0 v0 1
@ 0
: 9

sui/
├── README.md           # This file (English)
├── README_ja.md        # Japanese README
├── LICENSE             # MIT License
├── sui.py              # Interpreter
├── sui2py.py           # Sui → Python transpiler
├── py2sui.py           # Python → Sui transpiler (for humans)
├── examples/
│   ├── fibonacci.sui
│   ├── fib_args.sui
│   ├── fizzbuzz.sui
│   ├── list_sum.sui
│   ├── args_demo.sui
│   └── ffi_demo.sui
└── prompts/
    ├── system_prompt.md  # LLM system prompts
    └── examples.md       # Generation examples

Sui is designed for LLM code generation. Use the prompts in prompts/ directory:

1. Copy the system prompt from prompts/system_prompt.md
2. Paste it into ChatGPT / Claude / Gemini / etc.
3. Ask to generate Sui code for your task
4. Run with sui your_code.sui

See prompts/examples.md for prompt templates and expected outputs.

Sui (粋) - A Japanese word meaning "refined," "sophisticated," or "the essence." It represents the aesthetic of eliminating excess and keeping only what is essential.

• Transpiler (Python output)
• Transpiler (Python input, for humans)
• Interactive mode (REPL)
• Transpiler (JavaScript output)
• Type annotations (optional)
• LLVM IR output
• WebAssembly output

MIT License