SPAship API

REST API for deploying SPAs to the SPAship platform.

Global install

npm install -g @spaship/api
spaship-api

Hacking

[e-mail]:spaship/api.git
npm install

From here, you can npm start to launch the service, or npm run dev to launch the service with auto-restart when source files are changed.

Testing

Command	Purpose
npm test	Run tests.
npm test -- --watch	Run tests, and re-run them when files change.

Pull requests and commit messages

This repo follows Conventional Commits, a standard format for writing commit messages. Each commit message becomes an entry in CHANGELOG.md, and the commit messages are also used to determine what version bump to apply.

Read more about Conventional Commits for a description and examples!

If you are working on a pull request, don't worry about commit message format. Commit early and often.

When your pull request is merged, "squash and merge" should be used, and a Conventional Commit message written at that point. In this way, your pull request will become a single commit in the master branch and one entry in the CHANGELOG will be created.

Configuration

Configuration can be provided by CLI flags, environment variables, or a configuration file. Arguments are processed in that order, so CLI flags take precedence over environment variables, which take precedence over the configuration file.

Option	Description	CLI	Env	config.json	Default
config file	Where to find the config file.	--config-file	SPASHIP_API_CONFIG_FILE	N/A	none
upload dir	Directory to upload SPA archives.	--upload-dir	SPASHIP_UPLOAD_DIR	"upload_dir"	/tmp/spaship_uploads
webroot	Directory to extract/deploy SPAs.	--webroot	SPASHIP_WEBROOT	"webroot"	/var/www
host	Hostname to run on.	--host	SPASHIP_HOST	"host"	localhost
port	Port to run on.	--port	SPASHIP_API_PORT	"port"	8008
log-level	Granularity of log messages to print. Options are: fatal, error, warn, info, debug, trace or silent.	--log-level	SPASHIP_LOG_LEVEL	"log_level"	info
log-format	pretty for human-friendly logs, json for machine-friendly logs.	--log-format	SPASHIP_LOG_FORMAT	"log_format"	pretty
mongo_url	The hosts of your mongodb instance.	--db:mongo:url	SPASHIP_DB__MONGO__URL	"db.mongo.url"	"localhost:27017"
mongo_user	(Optional) The username of your mongodb instance.	--db:mongo:user	SPASHIP_DB__MONGO__USER	"db.mongo.user"	null
mongo_password	(Optional) The password of your mongodb instance.	--db:mongo:password	SPASHIP_DB__MONGO__PASSWORD	"db.mongo.password"	null
mongo_db	The mongodb database name.	--db:mongo:db_name	SPASHIP_DB__MONGO__DB_NAME	"db.mongo.db_name"	"spaship"
mock_db	Whether to use a mock database (mongo-mock).	--db:mongo:mock	SPASHIP_DB__MONGO__MOCK	"db.mongo.mock"	true for dev, false when NODE_ENV == "production"
Keycloak URL	The URL to a Keycloak instance you wish to use for authentication.2	--auth:keycloak:url	SPASHIP_AUTH__KEYCLOAK__URL	auth.keycloak.url	none
Keycloak REALM	The Keycloak Realm under which your SPAship Manager client is registered.	--auth:keycloak:realm	SPASHIP_AUTH__KEYCLOAK__REALM	auth.keycloak.realm	none
Keycloak client id	The Keycloak client id for your SPAship Manager instance.	--auth:keycloak:client-id	SPASHIP_AUTH__KEYCLOAK__CLIENTID	auth.keycloak.clientid	none
JWT user UUID prop	The JWT property to treat as a UUID.3	--auth:keycloak:id_prop	SPASHIP_AUTH__KEYCLOAK__ID_PROP	auth.keycloak.id_prop	"sub" 4
Keycloak public key	Your Keycloak realm's public key.	--auth:keycloak:pubkey	SPASHIP_AUTH__KEYCLOAK__PUBKEY	auth.keycloak.pubkey	none
Keycloak public key file	A file path to your Keycloak realm's public key.	--auth:keycloak:pubkey_file	SPASHIP_AUTH__KEYCLOAK__PUBKEY_FILE	auth.keycloak.pubkey_file	none
LDAP admin group name	LDAP group name used for identifying SPAship users with admin privileges, i.e. access to create API Keys.	--auth:ldap:admin_group	SPASHIP_AUTH__LDAP__ADMIN_GROUP	auth.ldap.admin_group	none
LDAP user group name	LDAP group name used for identifying SPAship users with view only access.	--auth:ldap:user_group	SPASHIP_AUTH__LDAP__USER_GROUP	auth.ldap.user_group	none

Note: the filepath configurations (config file, upload dir, and webroot) must be absolute paths when defined in an environment variable or config file. When defined in CLI options like, they can be written relative to CWD. Example: --config-file=../config.json.

Note: When mock_db is on, the mocked database will be persisted as a flat file named mock-db.js right here in the same directory as this README.

SPA metadata

Each deployed SPA gets a hidden directory inside webroot which houses two files that contain some SPA metadata, ref and name.

For example, a SPA deployed with name "My App", path /my-app, and ref v1.0.0 would result in a webroot that looks like this:

www
├── .my-app
│   ├── name
│   └── ref
└── my-app
    └── index.html

API

/deploy

Deploy a SPA to SPAship. A very simple Web UI is provided, or the deployment can be automated with an HTTP request.

With cURL

NAME="My Awesome Application"
SPA_PATH="/my-app"
REF="v1.0.0"
curl -v -F upload=@test.zip -F name="$NAME" -F path="$SPA_PATH" -F ref="$REF" localhost:8008/deploy

Change localhost:8008 to the host and port the service is running on. Change APP_NAME to your preferred app name, and test.zip to the archive you want to upload.

With the web UI

When you run the deployment service, it will print something like:

Listening on http://localhost:8008

Open that URL in a browser and you'll see a form for uploading apps.

/list

Returns a list of all deployed SPAs. If they have a metadata directory, that metadata will be included.

Here's an example response from /list. Two apps are listed in the example, "SPAnom" has metadata and "SPAnonymous" does not.

[
  {
    "name": "SPAnominal",
    "path": "/spanominal",
    "ref": "v1.0.0"
  },
  {
    "name": null,
    "path": "/spanonymous",
    "ref": null
  }
]

A few notes about the response.

• path is not stored in the metadata directory; it's inferred from the SPA's directory in the webroot.
• The "SPAnonymous" app has null values for name and ref because it was not deployed with /deploy, but it's included so that /list provides a complete report of what paths are being made available.

Scripts

During development, this repo includes a handy script (npm run get-pubkey) for downloading your Keycloak server's public key. It accepts the same auth:keycloak:url and auth:keycloak:realm options as the API itself. You can run it as follows:

npm run get-pubkey -- --auth:keycloak:url https://auth.spaship.io --auth:keycloak:realm SPAshipUsers

(Just like the API, it also accepts environment variables or a config file instead of CLI options.)

After running the script, a .key file will be saved in your current directory. You can reference that file in the auth:keycloak:pubkey_file configuration option.

/apikeys

HTTP Method	Endpoint	Description
POST	/apiKeys	Creates and returns an API key object. It expects the body of the request body to contain an user id and a label name as { user: "username", label: "labelname" }
GET	/apiKeys	Returns a list of API key objects
DELETE	/apiKeys/<label-name>	Deletes a key by it's label name

cURL Examples

Create New API Key

curl --location --request POST 'https://<hostname>/api/v1/apiKeys' \
--header 'Host: <hostname>' \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '{"label": "<labelname>", "expiredDate": "<expirationDateValue>"}'

List API Keys

curl --location --request GET 'https://<hostname>/api/v1/apiKeys' \
--header 'Host: <hostname>' \
--header 'Authorization: Bearer <token>'

Delete an API Key

curl --location --request DELETE 'https://<hostname>/api/v1/apiKeys/<apiKey-label>' \
--header 'Host: <hostname>' \
--header 'Authorization: Bearer <token>'