Testing: ContentSwitcher behaving differently under testing

[kraanzu]

Have you checked closed issues? https://github.com/Textualize/textual/issues?q=is%3Aissue+is%3Aclosed

Yes

Please give a brief but clear explanation of the issue. If you can, include a complete working example that demonstrates the bug. Check it can run without modifications.

Thanks for the built in testing mechanism! I've noticed some weird behaviour while testing with ContentSwitcher

Take this example for instance:

from textual.app import App, ComposeResult
from textual.widgets import ContentSwitcher, Label


class MyApp(App):
    def compose(self) -> ComposeResult:
        self.c = ContentSwitcher()
        yield self.c


async def test_basic():
    async with MyApp().run_test() as pilot:
        app = pilot.app

        assert isinstance(app, MyApp)

        app.c.add_content(Label("Hello", id="test"), set_current=True)

        assert app.c.visible_content.id == "test"

When run, this throws an error:

::test_basic - AttributeError: 'NoneType' object has no attribute 'id'

which means, it doesn't even get mounted
Note: If I pass an initial parameter, that gets recognized

from textual.app import App, ComposeResult
from textual.widgets import ContentSwitcher, Label


class MyApp(App):
    def compose(self) -> ComposeResult:
        self.c = ContentSwitcher(Label("Default", id="default"), initial = "default")
        yield self.c


async def test_basic():
    async with MyApp().run_test() as pilot:
        app = pilot.app

        assert isinstance(app, MyApp)

        app.c.add_content(Label("Hello", id="test"), set_current=True)

        assert app.c.visible_content.id == "default"

It will be helpful if you run the following command and paste the results:

textual diagnose

Textual Diagnostics

Versions

Name	Value
Textual	0.83.0
Rich	13.9.2

Python

Name	Value
Version	3.12.5
Implementation	CPython
Compiler	GCC 13.3.0
Executable	/home/kraanzu/Projects/dooit/.venv/bin/python3

Operating System

Name	Value
System	Linux
Release	6.6.54
Version	#1-NixOS SMP PREEMPT_DYNAMIC Fri Oct 4 14:30:05 UTC 2024

Terminal

Name	Value
Terminal Application	WezTerm (20240203-110809-5046fc22)
TERM	xterm-256color
COLORTERM	truecolor
FORCE_COLOR	Not set
NO_COLOR	Not set

Rich Console options

Name	Value
size	width=145, height=36
legacy_windows	False
min_width	1
max_width	145
is_terminal	True
encoding	utf-8
max_height	36
justify	None
overflow	None
no_wrap	False
highlight	None
markup	None
height	None

If you don't have the textual command on your path, you may have forgotten to install the textual-dev package.

Feel free to add screenshots and / or videos. These can be very helpful!

[github-actions]

We found the following entries in the FAQ which you may find helpful:

• Does Textual support images?
• How do I center a widget in a screen?

Feel free to close this issue if you found an answer in the FAQ. Otherwise, please give us a little time to review.

This is an automated reply, generated by FAQtory

[TomJGooding]

I think you might just need to add an await to ensure the new content is mounted before your assert?

[kraanzu]

Hmm.. That worked. Pretty weird. I didn't think about this because all the places where I used this function, I did not use the await keyword and that works :)

[github-actions]

Don't forget to star the repository!

Follow @textualizeio for Textual updates.

[willmcgugan]

It will still work without the await, but it won't have completed on the very next line. Adding new content will do a little work in the background.

[kraanzu]

That makes a lot more sense. Thanks Will !

[kraanzu]

Question @willmcgugan , how should I test this then ideally?
For example, if I have a reactive variable that posts a message (OptionList's highlighted variable in this case) and I handle it by changing some portion in the UI, how do I verify that via tests?

Thank you!

[willmcgugan]

If you call await pilot.pause() it will wait for your messages to be processed. Then you can confirm the state is what you expect it to be.

[kraanzu]

That works. Thank you