A Full-Stack, Self-Contained Shopping Demo App With Svelte, Golang and SQLite

This is a simple shopping demo app, based on the same Angular/React/Vue.js examples in Essential Typescript by Adam Freeman:

• Front-end: (/src)
  • TypeScript
  • Component framework: Svelte 5 (upgraded to Svelte 5 syntax)
  • Application framework: SvelteKit 2
  • Static site generation: @sveltejs/adapter-static
  • CSS styles: Bootstrap 5
• Back-end: (/backend)
  • Golang
  • Static web server/RESTful web services: gin
  • SQLite driver: go-sqlite3
• Database: (/db)
  • SQLite

The project comes with a Dockerfile that can create a single small container with multi-stage builds (image size ~25 MB) and also support to be opened in DevContainer/CodeSpace.

A Github Action workflow is created to run build test for dependabot's pull-requests for dependency updates.

The purpose of project is a demonstration to build a small and self-contained full-stack monolithic application with modern frameworks, but not meant to be a practical template for any real world applications. For example, error handlings between front-end and authentication are mostly ignored.

A similar version using Vue.js, Express, MongoDB and Docker Compose can be found here (no longer maintained).

Routes

The Svelte app has the following routes:

The Application Flow

1. When the app is opened at /, it will immediately redirect to /products.
2. /products loads the list of products from an API service and display the categories and items.
3. The user can filter items using the category buttons.
4. When the user clicks Add To Cart on any items, it will be added to the cart (global state array orderLines).
5. When the user clicks Submit Order on /products, the app will redirect to /order, whcih will show the detail of the order.
6. When the user clicks Submit Order on /order, the app will send the order to an API service and get the order ID.
7. After the order is "sent", the app will redirect to /summary/[id] as the result.

Backend APIs

The backend creates two RESTful-like APIs:

Adam Freeman's original projects use json-server on an Express server as mock API services. I keep the input/output spec of the services for the sake of demonstration. Right now, like all the original examples, the app only reads product lists and write order data. The Axios package used in the original examples is also replaced with fetch.

SvelteKit also has a feature to create "backend APIs". However the Golang server here is enough, so we don't really need to create duplicated APIs.

/api/products

Return a list of products from the database. The category will be used to create category filter buttons on the app.

Example request body:

(none)

Example response body:

[
    {
        "id": 1,
        "name": "Kayak",
        "category": "Watersports",
        "description": "A boat for one person",
        "price": 275.0
    },
    {
        "id": 2,
        "name": "Lifejacket",
        "category": "Watersports",
        "description": "Protective and fashionable",
        "price": 48.95
    }
]

/api/orders

Write ordered item and quantities. The service will create a new order ID and associate it with the ordered products.

Example request body:

{
    "lines": [
        {
            "productId": 1,
            "productName": "Kayak",
            "quantity": 2
        },
        {
            "productId": 2,
            "productName": "Lifejacket",
            "quantity": 4
        }
    ]
}

Example response body:

{
    "id": 42
}

Database Schemes and Example Data

The SQLite database (./db/data.sqlite3) in this repo already contains the table products with 9 product records (which can be found in many Adam Freeman's books) and an empty table orders. You can use DB Browser for SQLite to read the database. There is also a backup file in case you need to restore the database.

Here's the SQL statements to recreate them:

CREATE TABLE "products" (
    "id"    INTEGER NOT NULL UNIQUE, -- product ID
    "name"    TEXT NOT NULL,
    "category"    TEXT NOT NULL,
    "description"    TEXT,
    "price"    REAL NOT NULL,
    PRIMARY KEY("id" AUTOINCREMENT)
);

CREATE TABLE "orders" (
    "id"    INTEGER NOT NULL, -- order ID
    "product_id"    INTEGER NOT NULL, -- product ID
    "quantity"    INTEGER NOT NULL
);

INSERT INTO "main"."products" (
    "id",
    "name",
    "category",
    "description",
    "price"
)
VALUES
    ('1', 'Kayak', 'Watersports', 'A boat for one person', '275.0'),
    ('2', 'Lifejacket', 'Watersports', 'Protective and fashionable', '48.95'),
    ('3', 'Soccer Ball', 'Soccer', 'FIFA-approved size and weight', '19.5'),
    ('4', 'Corner Flags', 'Soccer', 'Give your playing field a professional touch', '34.95'),
    ('5', 'Stadium', 'Soccer', 'Flat-packed 35,000-seat stadium', '79500.0'),
    ('6', 'Thinking Cap', 'Chess', 'Improve brain efficiency by 75%', '16.0'),
    ('7', 'Unsteady Chair', 'Chess', 'Secretly give your opponent a disadvantage', '29.95'),
    ('8', 'Human Chess Board', 'Chess', 'A fun game for the family', '75.0'),
    ('9', 'Bling Bling King', 'Chess', 'Gold-plated, diamond-studded King', '1200.0');

Setup Local Project

For local development you'll need

• Git
• Node.js (for dev or production)
• Golang (for production)
• Docker (only required for generating the container)

Note: the go-sqlite3 package requires GCC to compile with the environment variable CGO_ENABLED=1.

For Windows users, it can be installed with MinGW (unzip and add \mingw64\bin to $PATH, then restart VS Code). On Linux it can be installed with the package build-essential.

Clone Repository

git clone https://github.com/alankrantas/svelteapp-typescript-go.git
cd svelteapp-typescript-go
npm i -g yarn@latest
yarn setup-full

And install/upgrade yarn:

npm i -g yarn@latest

Serve Frontend in Dev Mode

Run the Svelte app in development mode. The app will not call any backend APIs, instead it returns mock product data and the returned order number is always 42.

yarn dev

The app will be open at http://localhost:3000.

Download Dependencies

# download frontend dependencies
yarn

# download backend dependencies
yarn setup-server

# download both dependencies
yarn setup-full
# or
# yarn setup-all

Upgrade Dependencies

# upgrade frontend dependencies
yarn upgrade-app

# upgrade backend dependencies
yarn upgrade-server

# upgrade both dependencies
yarn upgrade-full
# or
# yarn upgrade-all

Build Production

Install dependencies, build both front-end and back-end apps and run the local server:

# build frontend app
yarn build-app

# build backend server (which will set CGO_ENABLED=1)
yarn build-server

# build both
yarn build-full
# or
# yarn build
# yarn build-all

Serve Production

# serve in macOS or Linux
yarn serve

# serve in Windows
yarn serve-win

The app would open at http://localhost:8080.

Build and Run as a Docker Container

# build container
yarn docker

# run container
yarn docker-run

The app would open at http://localhost:8080.