A reactive notebook for Python — run reproducible experiments, query with SQL, execute as a script, deploy as an app, and version with git. Stored as pure Python. All in a modern, AI-native editor.
Python artificial-intelligencedagdata-sciencedata-visualizationdataflowdeveloper-toolsmachine-learningnotebookspipelinepythonreactivesqlweb-app
/* Copyright 2026 Marimo. All rights reserved. */
import { type Atom, atom, useAtom } from "jotai";
import React from "react";
import { ReactFlowProvider } from "reactflow";
import type { CellId } from "@/core/cells/ids";
import type { CellData } from "@/core/cells/types";
import type { Variables } from "@/core/variables/types";
import { DependencyGraphTree } from "./dependency-graph-tree";
import { GraphToolbar } from "./panels";
import type { GraphSettings, LayoutDirection } from "./types";
import "reactflow/dist/style.css";
import "./dependency-graph.css";
interface Props {
cellIds: CellId[];
variables: Variables;
cellAtoms: Atom<CellData>[];
children?: React.ReactNode;
}
const graphViewAtom = atom<LayoutDirection>("TB");
const graphViewSettings = atom<GraphSettings>({
hidePureMarkdown: true,
hideReusableFunctions: false,
});
export const DependencyGraph: React.FC<Props> = (props) => {
const [layoutDirection, setLayoutDirection] = useAtom(graphViewAtom);
const [settings, setSettings] = useAtom(graphViewSettings);
return (
<ReactFlowProvider key={layoutDirection}>
<DependencyGraphTree
{...props}
settings={settings}
layoutDirection={layoutDirection}
>
<GraphToolbar
settings={settings}
onSettingsChange={setSettings}
view={layoutDirection}
onChange={setLayoutDirection}
/>
</DependencyGraphTree>
</ReactFlowProvider>
);
};