Towards Introducing Asynchronous Tasks to an FRP Language for Small-Scale Embedded Systems
summary
Emfrp currently enforces termination per update and static memory bounds, which makes long-running work like parsing or graph search block input and ruin responsiveness. The paper shows a concrete design that embeds futures and promises into Emfrp: examples move expensive graph-construction and remote-data exploration into asynchronous tasks while keeping the reactive update loop synchronous and bounded. A runtime prototype implements cancellable async tasks and integration with the scheduler, and discusses how the design preserves Emfrp’s resource guarantees.
AI-written summary · about
chapters (10)
- 0:00RxJS debugging challenges
- 1:47Debugging type-ahead app
- 5:02RxJS operators and stateful subs
- 8:54RxJS semantic model
- 12:22RxJS event propagation rules
- 16:19Debugging ideas from semantics
- 20:36Runtime invariants and errors
- 23:11RxJS semantics implementation
- 24:36Summary and challenges
- 25:05Q&A: Other debugging methods
related talks
36mPeriodic and Aperiodic Task Description Mechanisms in an FRP Language for Small-...
29mWhat is Functional Reactive Programming?
30mAsynchronous Modal FRP
31mThe Beauty and Elegance of Functional Reactive Animation
49mAre promises our only choice?
36mAsynchronous Reactive Programming with Modal Types in Haskell
29mModal FRP for all: Functional reactive programming without space leaks in Haskel...
23mEmbrace or ditch the async?