brickflowui 0.1.8

A Databricks-first Python UI framework for building secure dashboards, portals, and web apps with pure Python.

BrickflowUI

Databricks-first Python UI framework for building dashboards, portals, chatbot workspaces, landing pages, and internal web apps with pure Python.

Canonical package name: brickflowui

Install:

pip install brickflowui

Import:

import brickflowui as db

Documentation:

• Docs site
• Learn BrickflowUI
• Component Gallery
• Architecture

What BrickflowUI is for

Use BrickflowUI when you want to stay in Python and still ship a serious interactive app.

Good fits:

• executive dashboards
• data engineering pipeline portals
• Databricks App experiences
• internal ops tools
• chatbot and copilot UIs
• product-style landing pages

Why teams use it

• Python-first authoring model
• session-scoped reactive state
• multi-page routing and app shell support
• built-in tables, forms, overlays, charts, pipeline graphs, workflow boards, and chat patterns
• packaged frontend assets that work in stricter CSP environments
• branding and theme tokens for enterprise rollout

Quick start

Scaffold a new app:

brickflowui new my_app
cd my_app
brickflowui dev

Or write one manually:

import brickflowui as db

app = db.App(title="Hello BrickflowUI")

@app.page("/", title="Home")
def home():
    count, set_count = db.use_state(0)
    return db.Column(
        [
            db.Text("Hello BrickflowUI", variant="h1"),
            db.Text(f"Count: {count}"),
            db.Button("Increment", on_click=lambda: set_count(count + 1)),
        ],
        gap=4,
        padding=6,
    )

if __name__ == "__main__":
    app.run()

Core concepts

BrickflowUI works as:

1. Python page functions produce a VNode tree.
2. The server serializes that tree and sends it over WebSocket.
3. The frontend renders it and returns interaction events.
4. Python handlers update state and trigger rerenders.

This gives you a React-style interaction model without requiring users to write frontend code.

Current component surface

Layout and app structure

• Column
• Row
• Grid
• Card
• Divider
• Spacer
• Hero
• SectionHeader
• StatusStrip

Inputs and overlays

• Button
• Input
• Select
• Checkbox
• Toggle
• Slider
• DateRangePicker
• MultiSelect
• Form
• Modal
• Drawer
• Popup

Data and workflow

• Table
• Timeline
• SparklineStat
• Stepper
• KanbanBoard
• ChatMessage
• ChatInput
• PipelineGraph

Charts

• Plot
• AreaChart
• BarChart
• LineChart
• DonutChart
• ScatterChart
• ComposedChart
• GaugeChart
• RadarChart
• Heatmap
• FunnelChart
• TreeMap

Databricks Apps

Use this minimum setup:

requirements.txt

brickflowui>=0.1.6

Install from GitHub instead:

brickflowui @ git+https://github.com/AjayAJ2000/brickflowUI.git@main

app.yaml

command:
  - python
  - app.py

Important packaging rule:

The installed package must contain:

brickflowui/frontend/dist/index.html
brickflowui/frontend/dist/assets/*

If those files are missing, Databricks Apps often stop at the loading shell.

Local development

Framework tests:

python -m pytest -q

Frontend build:

cd frontend
npm run build

Docs build:

python -m mkdocs build

Package build:

python -m build

Repo map

brickflowui/
  app.py
  components.py
  server.py
  state.py
  vdom.py
  frontend/dist/
  cli/
  databricks/

frontend/
  src/
  vite.config.ts

docs/
  learning/
  components/
  ARCHITECTURE.md

License

MIT