You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently go-da provides very basic and minimalistic DA interface, and no logic. Every project trying to use go-da has to implement error handling, retry logic and timeouts. This is a perfect candidate for refactoring.
We already identified set of common errors. For some of them, automatic retry logic can be implemented, other has to be reported to upper layers. We can also create isolated tests simulating many DA error scenarios.
This feature can be implemented in various ways:
modify existing proxies (gRPC and JSON-RPC)
create go-da wrapper implementation that can handle retries/timeouts over any other go-da implementation
Regardless of the decision, we should try to end up with single implementation, single configuration, and good tests.
The text was updated successfully, but these errors were encountered:
Implementation ideas
Currently go-da provides very basic and minimalistic DA interface, and no logic. Every project trying to use go-da has to implement error handling, retry logic and timeouts. This is a perfect candidate for refactoring.
We already identified set of common errors. For some of them, automatic retry logic can be implemented, other has to be reported to upper layers. We can also create isolated tests simulating many DA error scenarios.
This feature can be implemented in various ways:
Regardless of the decision, we should try to end up with single implementation, single configuration, and good tests.
The text was updated successfully, but these errors were encountered: