There are benefits in having a separate Yard runner app. This app should live in Yard so it supports the Yard release cycle and has better reusability between different users.
Reasons why it is good to have a separately contained app for this.
- You can write any code into code fragments. This makes allows accessing the underlaying operating system and makes calling external resources possible. One concern is attacking a database if the application where Yard already has permissions to access one. The Yard runner should be only accessed by the specific REST API calls and every other access in or out needs to be blocked.
- You can write code that drains all the resources by accident. This will freeze the operating system running it
- Since the Yard file contains the data about the version that is used. Picking the correct external app based on the need is possible. Yard runner has it's own release cycle so the software using it does not need to do a release or update when Yard moves forward
- Since it is one file per request and preferably a full reset after each one. Two files can not manipulate each other
- 100 Yard file runs can be ran in parallel if there are resources for it.
There are benefits in having a separate Yard runner app. This app should live in Yard so it supports the Yard release cycle and has better reusability between different users.
Reasons why it is good to have a separately contained app for this.