rich/docs/source/live.rst

166 lines
6.2 KiB
ReStructuredText

.. _live:
Live Display
============
Progress bars and status indicators use a *live* display to animate parts of the terminal. You can build custom live displays with the :class:`~rich.live.Live` class.
For a demonstration of a live display, run the following command::
python -m rich.live
.. note::
If you see ellipsis "...", this indicates that the terminal is not tall enough to show the full table.
Basic usage
~~~~~~~~~~~
To create a live display, construct a :class:`~rich.live.Live` object with a renderable and use it as a context manager. The live display will persist for the duration of the context. You can update the renderable to update the display::
import time
from rich.live import Live
from rich.table import Table
table = Table()
table.add_column("Row ID")
table.add_column("Description")
table.add_column("Level")
with Live(table, refresh_per_second=4): # update 4 times a second to feel fluid
for row in range(12):
time.sleep(0.4) # arbitrary delay
# update the renderable internally
table.add_row(f"{row}", f"description {row}", "[red]ERROR")
Updating the renderable
~~~~~~~~~~~~~~~~~~~~~~~
You can also change the renderable on-the-fly by calling the :meth:`~rich.live.Live.update` method. This may be useful if the information you wish to display is too dynamic to generate by updating a single renderable. Here is an example::
import random
import time
from rich.live import Live
from rich.table import Table
def generate_table() -> Table:
"""Make a new table."""
table = Table()
table.add_column("ID")
table.add_column("Value")
table.add_column("Status")
for row in range(random.randint(2, 6)):
value = random.random() * 100
table.add_row(
f"{row}", f"{value:3.2f}", "[red]ERROR" if value < 50 else "[green]SUCCESS"
)
return table
with Live(generate_table(), refresh_per_second=4) as live:
for _ in range(40):
time.sleep(0.4)
live.update(generate_table())
Alternate screen
~~~~~~~~~~~~~~~~
You can opt to show a Live display in the "alternate screen" by setting ``screen=True`` on the constructor. This will allow your live display to go full screen and restore the command prompt on exit.
You can use this feature in combination with :ref:`Layout` to display sophisticated terminal "applications".
Transient display
~~~~~~~~~~~~~~~~~
Normally when you exit live context manager (or call :meth:`~rich.live.Live.stop`) the last refreshed item remains in the terminal with the cursor on the following line.
You can also make the live display disappear on exit by setting ``transient=True`` on the Live constructor.
Auto refresh
~~~~~~~~~~~~
By default, the live display will refresh 4 times a second. You can set the refresh rate with the ``refresh_per_second`` argument on the :class:`~rich.live.Live` constructor.
You should set this to something lower than 4 if you know your updates will not be that frequent or higher for a smoother feeling.
You might want to disable auto-refresh entirely if your updates are not very frequent, which you can do by setting ``auto_refresh=False`` on the constructor.
If you disable auto-refresh you will need to call :meth:`~rich.live.Live.refresh` manually or :meth:`~rich.live.Live.update` with ``refresh=True``.
Vertical overflow
~~~~~~~~~~~~~~~~~
By default, the live display will display ellipsis if the renderable is too large for the terminal. You can adjust this by setting the
``vertical_overflow`` argument on the :class:`~rich.live.Live` constructor.
- "crop" Show renderable up to the terminal height. The rest is hidden.
- "ellipsis" Similar to crop except last line of the terminal is replaced with "...". This is the default behavior.
- "visible" Will allow the whole renderable to be shown. Note that the display cannot be properly cleared in this mode.
.. note::
Once the live display stops on a non-transient renderable, the last frame will render as **visible** since it doesn't have to be cleared.
Print / log
~~~~~~~~~~~
The Live class will create an internal Console object which you can access via ``live.console``. If you print or log to this console, the output will be displayed *above* the live display. Here's an example::
import time
from rich.live import Live
from rich.table import Table
table = Table()
table.add_column("Row ID")
table.add_column("Description")
table.add_column("Level")
with Live(table, refresh_per_second=4) as live: # update 4 times a second to feel fluid
for row in range(12):
live.console.print(f"Working on row #{row}")
time.sleep(0.4)
table.add_row(f"{row}", f"description {row}", "[red]ERROR")
If you have another Console object you want to use, pass it in to the :class:`~rich.live.Live` constructor. Here's an example::
from my_project import my_console
with Live(console=my_console) as live:
my_console.print("[bold blue]Starting work!")
...
.. note::
If you are passing in a file console, the live display only show the last item once the live context is left.
Redirecting stdout / stderr
~~~~~~~~~~~~~~~~~~~~~~~~~~~
To avoid breaking the live display visuals, Rich will redirect ``stdout`` and ``stderr`` so that you can use the builtin ``print`` statement.
This feature is enabled by default, but you can disable by setting ``redirect_stdout`` or ``redirect_stderr`` to ``False``.
Nesting Lives
-------------
Note that only a single live context may be active at any one time. The following will raise a :class:`~rich.errors.LiveError` because status also uses Live::
with Live(table, console=console):
with console.status("working"): # Will not work
do_work()
In practice this is rarely a problem because you can display any combination of renderables in a Live context.
Examples
--------
See `table_movie.py <https://github.com/willmcgugan/rich/blob/master/examples/table_movie.py>`_ and
`top_lite_simulator.py <https://github.com/willmcgugan/rich/blob/master/examples/top_lite_simulator.py>`_
for deeper examples of live displaying.