ColumnLens

ColumnLens is a high-performance desktop data viewer for exploring structured data files using SQL. Open CSV, Parquet, JSONL, TSV, SQLite, and DuckDB files — no server, no cloud, no setup. Your data stays on your Mac. KEY FEATURES - SQL-Powered Exploration: Write SQL queries against any supported file. Powered by DuckDB, an embedded analytical database engine. - Instant Table View: Browse millions of rows with a fast, scrollable data grid. Filter, sort, and search across all columns. - Interactive Charts: Create line, bar, and scatter charts directly from your data. Select axes, configure chart types, and export as PNG. - 3D City Visualization: See your data as an interactive 3D cityscape where building height, color, and position map to data columns. - Lua Scripting: Automate workflows with built-in Lua scripting. Query data, configure charts, and process results programmatically. - Export: Export filtered results as CSV or TSV for use in other tools. - S3 Support: Connect to Amazon S3 buckets and query remote Parquet and CSV files directly. - Text Toolkit: Built-in tools for common text transformations including Markdown to HTML conversion. SUPPORTED FILE FORMATS - CSV and TSV (including gzipped) - Apache Parquet - JSONL / JSON Lines (including gzipped) - SQLite databases - DuckDB databases - Excel (.xlsx) LIGHTWEIGHT AND MEMORY-EFFICIENT ColumnLens is engineered to run great on every Mac — including 8 GB machines like the MacBook Air and MacBook Neo. The app itself uses minimal memory, and the embedded DuckDB engine handles datasets larger than available RAM by intelligently spilling to disk. No bloated runtime eating into your limited memory. No Electron, no web views, no Java — just a lean, native C++ binary built for Apple Silicon. DESIGNED FOR DATA PROFESSIONALS ColumnLens is built for data engineers, analysts, students, and developers who need to quickly inspect and explore data files without spinning up a database or writing a Python script. It launches instantly and handles large files efficiently.