dry-transaction is a business transaction DSL. It provides a simple way to define a complex business transaction that includes processing by many different objects. It makes error handling a primary concern by using a “Railway Oriented Programming” approach for capturing and returning errors from any step in the transaction.
dry-transaction is based on the following ideas:
- A business transaction is a series of operations where any can fail and stop the processing.
- A business transaction may resolve its operations using an external container.
- A business transaction can describe its steps on an abstract level without being coupled to any details about how individual operations work.
- A business transaction doesn’t have any state.
- Each operation shouldn’t accumulate state, instead it should receive an input and return an output without causing any side-effects.
- The only interface of an operation is
- Each operation provides a meaningful piece of functionality and can be reused.
- Errors in any operation can be easily caught and handled as part of the normal application flow.
Requiring a business transaction’s steps to be independent operations directly addressable via a container means that they can be tested in isolation and easily reused throughout your application. The business transaction can then become a simple series of declarative steps, which ensures that it’s easy to understand at a glance.
The output of each step is a dry-monads
Result object (either a
Failure). This allows the steps to be chained together and ensures that processing stops in the case of a failure. Returning a
Result from the overall transaction also allows for error handling to remain a primary concern without it getting in the way of tidy, straightforward operation logic.
View the full API documentation on RubyDoc.info.
dry-transaction’s error handling is based on Scott Wlaschin’s Railway Oriented Programming, found via Zohaib Rauf’s Railway Oriented Programming in Elixir blog post. dry-transaction’s behavior as a business transaction library draws heavy inspiration from Piotr Solnica’s Transflow and Gilbert B Garza’s Solid Use Case. Thank you all!