Overview
Reboot's programming model enables creating APIs that are easier to use than REST, GraphQL, or legacy gRPC APIs. We include a powerful standard library, as well as easy-to-use integrations for third party APIs.
Because Reboot's method calls compose transactionally, your code can directly call an integration's methods (to send an email, charge a credit card, update your search index, etc) without worrying about order of execution, partial failures, complete failures, or retries. And side-effect isolation makes it easy to write correct integrations.
Just like the rest of Reboot, logic and data are defined together. On the surface an integration exposes a well documented, strongly typed Reboot API that executes transactionally with your own methods. Under the hood, an integration can store data, schedule tasks, and enqueue guaranteed-execution work.
Standard library
The standard library uses Reboot's capabilities to build powerful data structures and APIs for common data-centric tasks, like full-text or vector search, and blob storage.
The currently supported APIs are:
- SortedMap - Standard library collection that may be larger than memory, and which is efficiently mapped to Reboot's underlying storage.
Integrations
Reboot's integrations add clean, surefire APIs to interact with third party services. Because interactions with external services are isolated as side-effects, you can confidently consume an integration's API -- knowing that it will safely compose, and without worrying about whether it might partially fail or retry unsafely.
The currently supported APIs are:
- Mailgun - Integration that supports sending email messages using the Mailgun API.
Coming soon
Many standard library utilities (full text and vector search, blob storage, etc) and third party integrations (Stripe, Twilio, Algolia/Elastic, etc) are planned: if you don't see what you need here yet, please reach out to us, and we can prioritize building them for you. Or we can help you build them, if that's your speed!