Lesson 10: Copying Code and Setting the Working Directory

To build an image for a real application, we need to copy our code into the image filesystem and define where our application should run.

1. WORKDIR (Working Directory)

The WORKDIR instruction sets the current working directory for subsequent instructions (RUN, CMD, ENTRYPOINT, COPY). If the directory doesn't exist, Docker will create it.

dockerfile

Good practice: Create a dedicated directory for the app

WORKDIR /app

Subsequent RUN commands will now execute inside /app

RUN pwd

Output: /app

2. COPY (Copying Files)

The COPY instruction copies files or directories from the build context (your local machine) into the image filesystem.

Syntax: COPY <source> <destination>

Example: Containerizing a Node.js Application

Assume you have a package.json and a server file (server.js).

dockerfile FROM node:20-alpine

1. Set the working directory

WORKDIR /usr/src/app

2. Copy dependency files first (for layer caching optimization!)

We copy package*.json separately so if only the code changes,

the expensive 'npm install' step is still cached.

COPY package*.json ./

3. Install dependencies

RUN npm install

4. Copy the rest of the application code

COPY . .

5. Define the application entry point

CMD ["node", "server.js"]

Difference between COPY and ADD:

• COPY is generally preferred as it is simpler and clearer.
• ADD has additional functionality: it can handle remote URLs and automatically extract compressed archives (tar, gzip). Use COPY unless you explicitly need the archive extraction feature.