Download the PHP package brezzhnev/atlassian-connect-core without Composer

Atlassian Connect Core

The easiest way to create an add-on for JIRA and Confluence.

Version Compatibility

Getting Started

Installing

Install dependency via Composer

Register route middleware jwt by adding to app\Http\Kernel.php the following line:

Set the authentication driver to jwt in config/auth.php:

Set the model class in config/auth.php providers section:

Register the subscriber in the app/Providers/EventServiceProvider.php:

Configure the database and run:

The command php artisan plugin:install will publish config, views and resources that you can change for your needs.

Also, it will create "dummy" tenant needed for local testing and developing without needing of installing the add-on on real JIRA or Confluence instances.

Publishing as an add-on

If your application returns the add-on descriptor on the request to URL http://localhost:8000/atlassian-connect.json it means you are close to happiness and you can install the add-on.

Step 1. Make your application accessible

To install the add-on in the instance, you should be visible through the internet. For testing purposes the easiest way is to use ngrok.

Then you have it accessible, put your actual website URL to environment variable PLUGIN_URL.

Make sure your add-on is accessible via HTTPS, it is a requirement.

Step 2. Configure your JIRA / Confluence instance

Configure your add-on using config/plugin.php. Most values may be overwritten using env vars.

Step 3. Upload your add-on

Finally, you need to upload the add-on. Click "Upload add-on" and paste your public URL with descriptor path, eg. https://d1ea31ce.ngrok.io/atlassian-connect.json or https://yourplugindomain.com/atlassian-connect.json

Step 4. Testing the successfulness

After the successful installation, on JIRA instance you may see "Your add-on" top menu item. You also can go to the add-on general page by direct link :product_base_url/plugins/servlet/ac/sample-plugin/hello-page

Instead of :product_base_url you should put your JIRA or Cofluence instance URL (eg. https://google-dev.atlassian.net).

If you see page working, the application configured and add-on installed correctly.

Publish resources

Instead of using plugin:install you can perform actions manually.

To copy all the publishes you should use the following command:

To copy only specific publish you must call this command with option --tag. The value can be public (to copy assets), views and config.

Usage

Default routes

The following routes are registered by default:

• GET /atlassian-connect.json descriptor contents
• POST /installed add-on installed callback
• POST /uninstalled add-on uninstalled callback
• POST /enabled add-on enabled callback
• POST /disabled add-on disabled callback
• GET /hello sample page to persuade all working correctly

You may disable them by setting the config value plugin.loadRoutes to false.

Descriptor

You can use Descriptor facade to customize or create from scratch your own descriptor contents.

For example, you may customize it by adding to the app\Providers\AppServiceProvider in boot section the following:

Warning: if you are using route helper in the AppServiceProvider you should have RouteServiceProvider defined above AppServiceProvider in your app.php config.

API requests

In most cases of add-on development for Atlassian Product you need to perform requests to the instance.

For this case you can use JWTClient. It uses GuzzleHttp as HTTP client.

If you want to have custom handling (middlewares etc.) you can pass client instance to the constructor.

Pagination

If you want to send a request to an endpoint with pagination you should use JWTClient::paginate method. In most cases you don't need to pass paginator instance to the JWTClient constructor because it will be instantiated automatically by resolving your Tenant product type (JIRA or Confluence), but you always can use the specific paginator.

There are two paginators supported by default:

• JiraPaginator
• ConfluencePaginator

You're always able to extend Paginator class and create your own.

Examples

Get a Confluence page content

Get a JIRA issue

Webhooks

The plugin provides a convenient way to handle incoming webhooks, based on habitual Laravel Events.

If you don't familiar with Laravel Events, please take a look at Laravel Docs

There are two ways to define webhook listeners:

1. Define listeners in the config/plugin.php

2. Define listeners using the Webhook facade, for example:

As you can see, you can define event listener as a closure or as a string in Laravel-like syntax:

You don't need to define the webhooks within your add-on descriptor, they will be described automatically.

Example listener

Your event listeners may also type-hint any dependencies they need on their constructors. All event listeners are resolved via the Laravel service container, so dependencies will be injected automatically.

Console commands

• plugin:install is a helper command that creates "dummy" tenant with the fake data and publishes package resources (config, views, assets)
• plugin:dummy provides interactive way to set a tenant as "dummy" without manually editing database

Tests

Run the following in the package folder:

Security

If you discover any security related issues, please email [email protected] instead of using the issue tracker.

Credits

• brezzhnev
• All Contributors

License

The MIT License (MIT). Please see License File for more information.