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 { z } from "zod";
import { arrayMove } from "@/utils/arrays";
import { NotebookScopedLocalStorage } from "@/utils/storage/typed";
const BASE_KEY = "marimo:notebook-col-sizes";
interface ColumnSizes {
widths: (number | "contentWidth")[];
}
function initialState(): ColumnSizes {
return { widths: [] };
}
const storage = new NotebookScopedLocalStorage<ColumnSizes>(
BASE_KEY,
z.object({
widths: z.array(
z.union([z.number(), z.literal("contentWidth")]).default("contentWidth"),
),
}),
initialState,
);
export const storageFn = {
// Default to "contentWidth" if the column width is not set.
getColumnWidth: (index: number) => {
const widths = storage.get(BASE_KEY).widths;
return widths[index] ?? "contentWidth";
},
saveColumnWidth: (index: number, width: number | "contentWidth") => {
const widths = storage.get(BASE_KEY).widths;
if (widths[index]) {
widths[index] = width;
} else {
// If the index is out of bounds, add "contentWidth" until we reach the index
while (widths.length <= index) {
widths.push("contentWidth");
}
widths[index] = width;
}
storage.set(BASE_KEY, { widths });
},
clearStorage: () => {
storage.remove(BASE_KEY);
},
};
// When a column is reordered, we need to update the storage to reflect the new order.
export function reorderColumnSizes(fromIdx: number, toIdx: number) {
const widths = storage.get(BASE_KEY).widths;
const newWidths = arrayMove(widths, fromIdx, toIdx);
storage.set(BASE_KEY, { widths: newWidths });
}