pine2py

An open-source Python library that translates TradingView Pine Script (v5/v6) strategies and indicators into executable Python code that runs on pandas, NumPy, and TA-Lib.

Features

• Parse Pine v5/v6 (e.g., //@version=6) and common constructs
• Map built-ins (close, open, high, low, volume, time, bar_index, na)
• Map indicators (ta.sma, ta.ema, ta.rsi, ta.macd, etc.) to TA-Lib
• Translate plot() to matplotlib stubs or console logs
• Translate strategy.entry/exit/close to a Python Strategy class with simple order/position simulation
• Emit a Python class you can execute on a pandas DataFrame with OHLCV columns

Quick Start

1. Install dependencies (ensure TA-Lib is installed in your system):

pip install -r requirements.txt

2. Translate a Pine Script string to Python code:

from pine2py.translator import translate
pine_code = """
//@version=5
indicator(title="SMA Demo", overlay=true)
len = input.int(14)
s = ta.sma(close, len)
plot(s)
"""

code_str = translate(pine_code)
print(code_str)

3. Execute the translated strategy against a DataFrame:

import pandas as pd
from pine2py.executor import execute_translated_code

df = pd.DataFrame({...})  # must include 'open','high','low','close','volume' indexed by datetime
result = execute_translated_code(code_str, df)

App Integration Snippet

In your trading app editor:

from pine2py.translator import translate
from pine2py.executor import execute_translated_code

def on_run_button(pine_script_text: str, df):
    py_code = translate(pine_script_text)
    result = execute_translated_code(py_code, df)
    return result

Project Structure

• pine2py/parser.py: Pine tokenizer/lightweight parser
• pine2py/mapper.py: Mapping dictionaries and helpers
• pine2py/translator.py: translate(pine_code: str) -> str
• pine2py/executor.py: Base Strategy and execution helpers
• pine2py/plotting.py: Plot stubs
• tests/: Unit/integration tests

Status

This is a pragmatic MVP that supports a large subset of Pine commonly used in indicators/strategies. Unknown or unmapped functions raise descriptive errors.

License

MIT