spark-connect-excel

KO | EN

gitlite — search

#javascript #python #typescript #golang #android #machine-learning #rust #linux #nodejs #macos #kubernetes #awesome-list #mcp #ios

spark-connect-excel

(★ 2)

Run Spark SQL against your own Spark Connect cluster from Excel - results land in a range, refreshable and chartable, with no backend server (powered by pyspark-connect-web).

spark-connect-excel Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Spark Connect for Excel

[![CI](https://github.com/HyukjinKwon/spark-connect-excel/actions/workflows/ci.yml/badge.svg)](https://github.com/HyukjinKwon/spark-connect-excel/actions/workflows/ci.yml)
[![E2E (COI gate)](https://github.com/HyukjinKwon/spark-connect-excel/actions/workflows/e2e.yml/badge.svg)](https://github.com/HyukjinKwon/spark-connect-excel/actions/workflows/e2e.yml)
[![License](https://img.shields.io/badge/license-Apache--2.0-green.svg)](https://raw.githubusercontent.com/HyukjinKwon/spark-connect-excel/main/LICENSE)
[![Docs](https://img.shields.io/badge/docs-GitHub%20Pages-blue)](https://hyukjinkwon.github.io/spark-connect-excel/)

**Power Query, but the engine is your Spark cluster.**

An Excel add-in that runs a Spark SQL query against your own
[Spark Connect](https://spark.apache.org/docs/latest/spark-connect-overview.html)
cluster, lands the result directly in a worksheet range, refreshes it, and
charts it - with **no backend server of its own**. The query client runs
entirely in-browser via
[pyspark-connect-web](https://github.com/HyukjinKwon/pyspark-client-wasm)
(real PySpark, in Pyodide). See the
[full documentation](https://hyukjinkwon.github.io/spark-connect-excel/).

## Features

- **SQL in Excel** - write Spark SQL in the task pane; results land as typed cells.
- **Typed ranges** - Spark's schema drives Excel number formats (dates, decimals, integers).
- **Truncation guard** - row cap (default 10k) with a visual banner when the result is clipped.
- **Refresh** - rebind the query to its range; one click updates stale data.
- **Native charts** - auto-inferred chart type (line for time-series, column for categories, scatter for numeric pairs).
- **No backend** - compute runs on your Spark cluster; the add-in is static HTML/JS.
- **Secure token handling** - bearer tokens never touch a cell or the workbook file.
- **Zero-install web demo** - a standalone page (`/demo`) that runs **SQL _or_ PySpark**
  in the browser, no Excel required - the easiest way to try it.

## Quickstart

### Prerequisites

- Node 20, plus `python3` + `pip`, `git`, `curl`, `tar` (the quickstart vendors
  Pyodide + the wheels the in-browser engine needs - it does this for you).
- Excel **desktop** (Microsoft 365 / 2019, Windows or Mac) to run the engine. (No Excel? The standalone web demo runs the same engine in any Chromium browser.)
- A running Spark Connect server (see below).

### Start a Spark Connect server

The demo and the add-in both talk to a Spark Connect server through an Envoy
grpc-web proxy. Start one first, either way:

```bash
# Docker - Spark Connect + Envoy (grpc-web at :8081) in one command:
docker compose -f deploy/compose.yaml up
```

No Docker? Run a server locally with PySpark (Java 17) and put Envoy in front of
it - see the
[installation guide](https://hyukjinkwon.github.io/spark-connect-excel/installation/).

> The quickstart below vendors the engine assets (Pyodide + wheels) and serves
> the add-in over HTTPS. This exact stack - Pyodide + grpc-web + Spark Connect -
> is exercised end to end by the `e2e-full` CI gate on every push, which runs a
> live Spark SQL query and renders the chart shown above.

### Try it without Excel - the web demo

A standalone page, no Excel and no sideload. One command clones, installs, and
serves the add-in:

```bash
curl -fsSL https://raw.githubusercontent.com/HyukjinKwon/spark-connect-excel/main/scripts/quickstart.sh | bash
```

Open `https://localhost:3000/demo/demo.html`, point it at your Spark Connect
server, and run a query. See the
[usage guide](https://hyukjinkwon.github.io/spark-connect-excel/usage/).

### Try with Excel on Windows / Mac desktop (sideloads) - runs the engine

Excel **desktop** is the host that runs the engine. With a Spark Connect server
running:

```bash
curl -fsSL https://raw.githubusercontent.com/HyukjinKwon/spark-connect-excel/main/scripts/quickstart.sh | bash -s -- desktop
```

This serves the add-in and opens Excel with it sideloaded. The **Apache Spark**
button (group **Spark Connect**) appears on the Home ribbon. Full guides are on
the [documentation site](https://hyukjinkwon.github.io/spark-connect-excel/).

## Compatibility

| Component | Supported |
|-----------|-----------|
| Excel (engine) | Microsoft 365 / 2019 **desktop** - Windows or Mac (Excel on the web is not supported) |
| Excel API requirement | ExcelApi 1.12, DialogApi 1.2 |
| Spark | 4.x (Spark Connect; `apache/spark:4.0.0` in the deploy stack) |
| PySpark | `>=4.0,<4.2` (enforced by `pcw.install()`) |
| Engine WebView | Chromium-based (Windows WebView2) - `COEP: credentialless` is a Chromium feature |
| Node | 20 LTS |
| Python | 3.11+ (for local dev/tests; Pyodide 0.28+ in the browser) |

spark-connect-excel

spark-connect-excel Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

spark-connect-excel

spark-connect-excel Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README