v0.19.3 is live

Reactive State.
Zero Chaos.

A declarative state management library for Python.

True lazy evaluation. Fine-grained reactivity.
Updates are pushed to invalidate. Data is pulled on demand. Synchronous. Glitch-free.

Star on GitHub Docs
Fresh
Stale (Dirty)
Live Code

Declare Relationships, Not Steps.

Define how your data is related, and reaktiv handles the updates automatically.

1. The Signal 
from reaktiv import Signal

# 1. Create a Signal
count = Signal(0)

# 2. Update it
# Writing is cheap (just invalidates)
count.set(5)
2. The Computed 
from reaktiv import Signal, Computed

count = Signal(0)

# No calculation happens here!
doubled = Computed(
    lambda: count() * 2
)

# Calculation happens NOW (Pull)
print(doubled())
3. The Effect 
from reaktiv import Signal, Computed, Effect

count = Signal(0)
doubled = Computed(lambda: count() * 2)

# ⚠️ Important: Keep a reference!
# Otherwise GC deletes the listener.
watcher = Effect(
    lambda: print(doubled())
)

# Updates trigger the effect
count.set(6)
4. LinkedSignal 
from reaktiv import Signal, LinkedSignal

count = Signal(0)

# Writable, but auto-resets when
# dependency changes.
sync = LinkedSignal(
    lambda: count() * 10
)

sync.set(999) # Override manually
count.set(2)  # sync resets to 20!
lazy evaluation linked signals type safe fine-grained batch updates memoization

Try it yourself

Running live in your browser with PyScript.

main.py
OutputLoading PyScript...