Overview

📊 Financial Charts - Candlestick, line, area with crosshair and time-scale synchronization across multiple charts.

✨ Data Visualization - Charts, Tables, JSON tree viewer, gauge and more to come...

📚 Notebook-Native: Visualizations render inline as notebook cell outputs. They are fully embedded in the document, making them ideal for creating reproducible analysis reports, interactive dashboards, and storytelling with data.

⚡ Lightweight - CDN-based dependencies for fast loading and no build step.

🔧 TypeScript: Designed specifically for TS kernels. Get full type checking, autocompletion (IntelliSense) for chart configurations, and safe refactoring as you explore data.

Setup:

• pip3 install jupyterlab
• Install tslab
• npm i -S tslab chartslab tslib lightweight-charts

Screenshots

Widget list:

• Chart - Financial charts powered by lightweight-charts

  • Series types: Candlestick, Line, Area, Bar, Baseline, Histogram
  • Advanced features: tooltips, trend lines, volume profiles, infinite history for large datasets
  • Multi-chart synchronization for crosshair and time scale alignment
  • Efficient handling of large datasets (3+ years of timeseries data) with lazy loading

• Barchart - Versatile bar charts powered by recharts

  • 3 variations: Simple, Stacked, Mix
  • Ideal for categorical data comparison and trends
  • Supports custom colors and labels

• Pie - Interactive pie and donut charts powered by recharts

  • Single or two-level nested pie charts
  • Configurable donut mode with inner radius control
  • Custom labels with percentage display
  • Supports custom colors per segment

• Scatter - Scatter plot visualization powered by ag-charts

  • Multi-series support with different shapes (circle, square, diamond, etc.)
  • Customizable axes with titles and formatters
  • Interactive tooltips and legend
  • Perfect for correlation analysis and data distribution

• Table - Responsive data grid powered by ag-grid-community

  • Built-in sorting, filtering, and column resizing
  • Auto-generates columns from data structure
  • Balham dark theme integration

• Gauge - Circular KPI indicators powered by recharts

  • Animated needle with smooth transitions
  • Three size presets (S, M, L)
  • Configurable color sections and ranges

• JSON - Collapsible tree viewer powered by react-json-tree

  • Expandable/collapsible object inspection
  • Syntax highlighting for different data types
  • Scrollable container for large objects

• Layout - Composition utilities (not a widget):

  • row() - Arrange widgets horizontally with flex controls
  • Use .html() method to get HTML string without rendering (Note: chart widget doesn't support .html())

Review the example notebooks

CDN Strategy

chartslab uses a CDN-based approach for runtime dependencies (React, lightweight-charts, etc.). This means:

• ✅ No bundling overhead - Dependencies loaded directly from esm.sh
• ✅ Faster notebook loading - Browser caches shared dependencies
• ✅ Smaller package size - No need to bundle React into the library
• ✅ Centralized version management - All CDN versions managed in src/config/versions.ts

Configure VSCode to use Typescript Kernel

Go to Select Kernel -> Jupyter Kernel... -> Typescript