Fetchless A New Take on Data Retrieval
summary
Reimplements the old Scala Fetch as a new library called Fetchless with a tagless final–style DSL and explicit fetch APIs (single, batch, fetchAll/fetchAllDedup). It makes laziness opt‑in and linear execution the default, threads a cache between requests, and adds streaming, doobie and http4s integrations plus debugging. Benchmarks run 1..50k IDs 40× show Fetchless about an order of magnitude faster in the worst case (original Fetch ~6.6e9 ms vs Fetchless ~5.6e8 ms in this test), and the implementation successfully reproduces batching, deduping and parallel runs.
AI-written summary · about
chapters (10)
- 0:00Introduction to fetch
- 1:20Understanding fetch
- 5:26Introducing fetchless
- 12:51The origin of the rebuild
- 20:18Core architecture of fetchless
- 26:28Refining DSL syntax
- 31:05Implementation details
- 35:10Streaming and debugging
- 39:50Benchmark demonstration
- 48:14Q&A session
