lens-lang 0.1.0

Language for Exploratory & Narrative Statistics - A declarative analytics language

LENS - Language for Exploratory & Narrative Statistics

LENS is an open-source analytics programming language that bridges the gap between business questions and data execution using natural language.

Features

• Declarative Syntax: English-like queries (show net_sales by region)
• Time Intelligence: Built-in FY, YTD, rolling dates
• Auto-Visualization: Generates best-fit charts (Vega-Lite)
• Story Engine: Produces narratives from data trends
• No-Code Studio: Streamlit-based visual interface

Installation

pip install lens-lang

Or install from source:

git clone https://github.com/Anay007-git/lens.git
cd lens
pip install -e .

Quick Start

CLI

python cli/lens_cli.py -q "show net_sales by region for FY2025"

Studio (No-Code UI)

python -m streamlit run studio/app.py

Python API

from lens.compiler.parser import LensParser

parser = LensParser()
ast = parser.parse("show net_sales by region for FY2025")
print(ast)

Example Query

show net_sales by region
for FY2025
compare with previous year

Output:

• SQL query
• Data table
• Auto-generated chart
• Business narrative

Architecture

lens/
├── compiler/        # Lark parser + AST
├── semantics/       # Metric/Dimension registry
├── time_intelligence/  # FY, YTD, relative dates
├── planner/         # Logical plan builder
├── sql_gen/         # DuckDB SQL generator
├── runtime/         # Query execution
├── visualization/   # Altair/Vega-Lite charts
└── story/           # Trend analysis + narratives

License

Apache 2.0

Author

Anay Biswas

---

Built with ❤️ for the analytics community