Skip to main content

Persist state across multiple pages in streamlit.

Project description

Streamlit fire state

Streamlit App pypi version

In a multipage streamlit app, one of the most headache issues is that your state is not preserved if you switch between pages.

That's why fire-state is here for you.

Installation

pip install fire-state

Quick Start

Persist state in Form

import streamlit as st
from fire_state import create_store, form_update

# Register state with initiate values in a slot
slot = "home_page"
key1, key2 = create_store(slot, [
    ("state1", 5),
    ("state2", 12),
])

# Now create a form using the generated keys
with st.form("my form"):
    st.slider("State 1", 1, 10, step=1, key=key1)
    st.slider("State 1", 10, 20, step=1, key=key2)
    st.form_submit_button(label="Submit", on_click=form_update, args=(slot,))

When you switch between pages, the states are preserved.

Persist state in any place

You need to control the state by yourself, using the set_state function.

import streamlit as st
from fire_state import create_store, get_state, set_state


slot = "home_page"
create_store(slot, [
    ("state1", 0),
])

def increment():
    prev = get_state(slot, "state1")
    set_state(slot, ("state1", prev + 1))


st.button("+1", on_click=increment)
st.text(f"Value: {get_state(slot, 'state1')}")

Advanced Usage

Persist state after form submission

In this example, we have a form to control how the line chart is drawn.

Let's do some setup first

import numpy as np
import pandas as pd
import streamlit as st

np.random.seed(0)

@st.cache
def chart_data(line_count, data_size):
    return pd.DataFrame(
        np.random.randn(data_size, line_count),
        columns=np.random.choice(list('abcdefghijlkmn'), line_count))

Now we can use fire state to record the user action.

The idea is that the first time user opens the page, they never click the run button, so the number of times they click a button is 0. When it no longer is 0, which means user clicked it. Therefore, the plot is rendered or updated (if chart data is changed).

from fire_state import create_store, get_state, set_state, form_update

PAGE_SLOT = "Home_Page"
key1, key2, key3 = create_store(PAGE_SLOT, [
    ("line_count", 2),
    ("data_size", 13),
    ("run", 0)
])

with st.form(key="a form"):
    line_count = st.slider("Line Count", 1, 10, step=1, key=key1)
    data_size = st.slider("Data Size", 10, 20, step=1, key=key2)
    run = st.form_submit_button(label="Run", on_click=form_update, args=(PAGE_SLOT,))

prev_run_state = get_state(PAGE_SLOT, 'run')
if (prev_run_state != 0) or run:
    data = chart_data(line_count, data_size)
    st.line_chart(data)
    # increase by 1 every time user click it
    set_state(PAGE_SLOT, ("run", prev_run_state + 1))

Reset the state

Use the set_store function to update states in a batch:

import streamlit as st
from fire_state import create_store, \
    get_store, set_store, \
    get_state, set_state

slot = "page"
init_state = [
    ("state1", 1),
    ("state2", 2),
    ("state3", 3),
]
create_store(slot, init_state)

def reset():
    set_store(slot, init_state)

st.button("Reset", on_click=reset)

The set_store and get_store functions allow you to modify and get your state in a batch.

The Life Cycle of State

The state persists if you don't close or refresh the page. The state instance is only destroyed if you stop your app.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

fire_state-0.1.2.tar.gz (5.3 kB view hashes)

Uploaded Source

Built Distribution

fire_state-0.1.2-py3-none-any.whl (4.9 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page