A reactive notebook for Python — run reproducible experiments, query with SQL, execute as a script, deploy as an app, and version with git. Stored as pure Python. All in a modern, AI-native editor.
Python artificial-intelligencedagdata-sciencedata-visualizationdataflowdeveloper-toolsmachine-learningnotebookspipelinepythonreactivesqlweb-app
from __future__ import annotations
from marimo._plugins import ui
from marimo._runtime.commands import UpdateUIElementCommand
from marimo._runtime.runtime import Kernel
from tests.conftest import ExecReqProvider
def test_refresh_initial_value_is_empty_string():
refresh = ui.refresh(options=["1s", "5s"], default_interval="1s")
assert refresh.value == ""
async def test_refresh_value_updates_to_frontend_string(
any_kernel: Kernel, exec_req: ExecReqProvider
):
k = any_kernel
await k.run(
[
exec_req.get("import marimo as mo"),
exec_req.get("r = mo.ui.refresh(options=['1s', '5s'])"),
]
)
refresh = k.globals["r"]
assert refresh.value == ""
await k.set_ui_element_value(
UpdateUIElementCommand([refresh._id], ["1s (0)"]),
notify_frontend=False,
)
assert refresh.value == "1s (0)"
def test_refresh_on_change_receives_string():
seen: list[str] = []
refresh = ui.refresh(
options=["1s"], default_interval="1s", on_change=seen.append
)
refresh._update("1s (0)")
refresh._update("1s (1)")
assert seen == ["1s (0)", "1s (1)"]