A reactive programming library for Python, implementing a signals system similar to @preact/signals
Project description
signals
A signals implementation for Python
install
pip install signals
usage
from signals import Signal, computed, effect
a = Signal(0)
b = Signal(2)
c = computed(lambda: a.value + b.value)
print(c.value) # 2
a.value = 1
print(c.value) # 3
b.value = 3
print(c.value) # 4
# Log the values of a, b, c whenever one changes
@effect
def log_abc():
print(a.value, b.value, c.value)
a.value = 2 # prints (2, 3, 5)
cell magic
we also provide a ipython cell magic %%effect
, which offers a convenient way
re-execute cells that use signals.
In[1]:
%load_ext signals
from signals import Signal
a = Signal(0)
b = Signal(2)
In[2]:
%%effect
a.value + b.value # re-evaluates the cell whenever a or b changes
In[3]:
a.value = 1
what
Signals are a declarative programming model for updating based on fine-grained changes. With signals, application state is represented as a directed graph of relationships between other signals. However, the most important part of signals is that you don't need to manage the graph yourself.
Instead, you declare signals and their relationships, and the signal system automatically tracks dependencies and executes necessary computations when values change. Singal-like constructs have been adopted by popular UI libraries and non-UI contexts (e.g., build systems to avoid uneccessary rebuilds).
why
Signals are an easier way to manage state. We need something in Python other than callbacks and events. This repo is a playground to explore patterns for using a signal-based system in Python.
ideas
Signals for widgets
@anywidget.dataclass
class Counter:
count: int
a = Counter(count=0) # creates a signal internally
shared_count = Signal(0)
b = Counter(count=shared_count) # creates a signal internally
c = Counter(count=shared_count) # creates a signal internally
# behind the scenes, creating Counter creates an effect to update the frontend view
# when the count changes. E.g.,
# @effect
# def update():
# self.comm.send('update', self.count)
#
# Updates b/c views because they share a signal, and have separate effects
shared_count.value += 1
development
this project uses rye
for development.
rye lint # lints code
rye fmt # formats code
rye test # runs tests
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for signals-0.1.0a1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 718b59a79e9172e780c31740c678620a65c72607dfc35eef82b5d69cf4c889b1 |
|
MD5 | a358a75be276479471d0314053b91c27 |
|
BLAKE2b-256 | 993188ce8874916931148d1230674da0e131d1b41d0492dac076be68859ef50c |