Intro
The Medplum CLI (Command Line Interface) is a set of command line tools to quickly deploy Medplum web applications and Medplum bots.
Installation
Add globally:
npm install --global @medplum/cli
Or add as a package dependency:
npm install @medplum/cli
Authentication
Use one of these authentication options:
- Stored credentials in
~/.medplum/credentials. You can use themedplum logincommand (see below) to automatically create this file. - Client credentials in environment variables
MEDPLUM_CLIENT_IDandMEDPLUM_CLIENT_SECRET.dotenvis enabled, so you can store them in a.envfile.
Usage
If installed globally, you can use the medplum command directly:
medplum <command> <args>
If installed as a package dependency, you can use the medplum command via npx:
npx medplum <command> <args>
By default, the medplum command uses the Medplum hosted API at "https://api.medplum.com". If you want to use the medplum command against your own self-hosted server, you can use the MEDPLUM_BASE_URL environment variable. dotenv is enabled, so you can store this value in a .env file.
Auth
login
The login command opens a web browser to a Medplum authentication page.
On successful login, the command writes credentials to disk at ~/.medplum/credentials.
The medplum command will then load those credentials on all future runs.
Example:
medplum login
whoami
The whoami command displays whether the client is authenticated, and, if so, the name of the current user and current Medplum project.
medplum whoami
RESTful Operations
The medplum command can be used as a convenient tool for basic Medplum CRUD and RESTful operations.
While all API endpoints are available to any command line HTTP client such as curl or wget, there are a few advantages to using the medplum command:
- Authentication and credentials - login once using the
logincommand, and theAuthorizationheader will be set automatically. - URL prefixes - adds the base URL (i.e., "https://api.medplum.com") and FHIR path prefix (i.e., "fhir/R4/").
- Pretty print - formats the JSON with spaces and newlines.
- Medplum extended mode - adds the
X-MedplumHTTP header for private Medplum fields.
get
Makes an HTTP GET request.
medplum get <url>
Example: Search for patients:
medplum get 'Patient?name=homer'
Example: Read patient by ID:
medplum get Patient/$id
flags
--as-transaction- Convert the output response to a transaction bundle
post
Makes an HTTP POST request.
medplum post <url> <body>
Example: Create a patient:
medplum post Patient '{"resourceType":"Patient","name":[{"family":"Simpson"}]}'
Example: Invoke a FHIR operation:
medplum post 'Patient/$validate' '{"resourceType":"Patient","name":[{"family":"Simpson"}]}'
put
Makes an HTTP PUT request.
medplum put <url> <body>
Example: Update a patient:
medplum put Patient/$id '{"resourceType":"Patient","name":[{"family":"Simpson"}]}'
patch
Makes an HTTP PATCH request.
medplum patch <url> <body>
Example: Update a patient with JSONPatch:
medplum patch Patient/$id '[{"op":"add","path":"/active","value":[true]}]'
delete
Makes an HTTP DELETE request.
medplum delete <url>
Example: Delete patient by ID:
medplum delete Patient/$id
Project
project will have administration commands for visibilty and management
current
Sees your current project
medplum project current
list
Sees your list of project ids and names
medplum project list
switch
Switching to another project from the current one
medplum project switch <projectId>
AWS
aws includes commands for creating and managing AWS resources.
The AWS commands require AWS authentication and access credentials. Please make sure your credentials are configured before using the aws commands.
The aws commands are in beta, and likely to change.
list
List your Medplum deployments. This command lists AWS CloudFormation stacks with the medplum:environment tag.
medplum aws list
describe
Describe a Medplum deployment. Displays select AWS resources such as ECS Cluster, ECS Service, and S3 buckets from the AWS CloudFormation stack with the corresponding medplum:environment tag.
medplum aws describe <name>
update-server
Updates the ECS Service in a Medplum deployment to the latest version.
medplum aws describe <name>
Bots
Bots Config file
Create a Medplum config file called medplum.config.json:
{
"bots": [
{
"name": "hello-world",
"id": "f0465c2e-11d4-4c36-b834-8e86f7472b4b",
"source": "src/index.ts",
"dist": "dist/index.js"
}
]
}
The name property is a friendly name you can use to reference the Bot in commands.
The id property refers to the Bot ID in your Medplum project.
The source property is the file path to the original source. When you "save" the Bot, the contents of this file will be saved to the Bot code property. This file can be JavaScript or TypeScript.
The dist property is the optional file path to the compiled source. If omitted, the command falls back to using the source property. When you "deploy" the Bot, the contents of this file will be deployed to the Bot runtime. This file must be JavaScript.
bot save
Updates the code value on a Bot resource
Syntax:
npx medplum bot save <bot name>
Example:
npx medplum bot save hello-world
bot deploy
Deploys the Bot code
Syntax:
npx medplum bot deploy <bot name>
Example:
npx medplum bot deploy hello-world
bot create
Creates a bot and saves it
Syntax:
npx medplum bot create <bot name> <project id> <source file> <dist file>
Bots Example
Create a Medplum config file medplum.config.json:
{
"bots": [
{
"name": "hello-world",
"id": "f0465c2e-11d4-4c36-b834-8e86f7472b4b",
"source": "src/hello-world.ts",
"dist": "dist/hello-world.js"
}
]
}
Replace the sample id with your Bot's ID.
Write your bot in src/hello-world.ts. This can be TypeScript. It can reference @medplum/core and node-fetch:
import { MedplumClient } from '@medplum/core';
import { Resource } from '@medplum/fhirtypes';
export async function handler(medplum: MedplumClient, event: BotEvent): Promise<any> {
console.log('Hello world');
}
You can use the Medplum CLI to save it:
npx medplum bot save hello-world
Compile with vanilla tsc (no bundler required)
npx tsc
The result will be JavaScript output in dist/hello-world.js:
export async function handler(medplum, input) {
console.log('Hello world');
}
You can then use the Medplum CLI to deploy it.
npx medplum bot deploy hello-world