asynchronous deduplicator with optional LRU caching
If you have "slow", "expensive" or "flaky" tasks for which you'd like to provide de-duplication and, optionally, result caching, then this may be the crate you are looking for.
The crate provides a trait Retriever which should be implemented for your task. Once implemented, you can provide an instance of a Retriever to a Deduplicate which will then handle task de-duplication for you.
toml
[dependencies]
deduplicate = "0.1"
Apache 2.0 licensed. See LICENSE for details.