test app init of examples

[Corvince]

First stab at testing the exmamples app.

This already tests if there are any errors in app.py itself and if solara starts without any exceptions by calling app.page. We still don't know if anything is displayed and if the buttons etc. do anything, but initial errors are caught.

[Corvince]

@quaquel this seems to already catch an error in WolfSheep related to the ABMSimulator?

[github-actions]

Performance benchmarks:

Model	Size	Init time [95% CI]	Run time [95% CI]
BoltzmannWealth	small	🔴 +6.1% [+3.7%, +8.5%]	🔵 -0.4% [-0.6%, -0.1%]
BoltzmannWealth	large	🔵 +0.3% [-0.7%, +1.2%]	🔵 +0.2% [-1.4%, +1.5%]
Schelling	small	🔵 +1.1% [+0.7%, +1.4%]	🔵 +0.3% [+0.1%, +0.5%]
Schelling	large	🔵 -0.1% [-0.6%, +0.4%]	🔵 +3.2% [+1.8%, +4.8%]
WolfSheep	small	🔵 +0.5% [+0.1%, +0.9%]	🔵 -2.1% [-2.3%, -1.8%]
WolfSheep	large	🔵 -0.8% [-1.5%, -0.2%]	🟢 -6.5% [-9.0%, -4.1%]
BoidFlockers	small	🔵 -0.8% [-1.3%, -0.3%]	🔵 +0.0% [-0.8%, +0.8%]
BoidFlockers	large	🔵 +0.5% [-0.0%, +1.1%]	🔵 -0.3% [-0.9%, +0.5%]

[EwoutH]

I did so way more complex things to try to accomplish this.

Thanks, we really need it.

[quaquel]

@quaquel this seems to already catch an error in WolfSheep related to the ABMSimulator?

in app.py, change the line to model = WolfSheep(simulator=simulator, grass=True). I don't understand how this has slipped through. My hunch is the split of #2470 into two separate PRs and the resulting rebase mess. I thought I tested everything afterwards, but I most somehow have overlooked this.

[quaquel]

There is another subtle bug still there. I have a meeting now. I'll fix those afterwards.

basically, reset will error with simulator.

fixed

[Corvince]

I did so way more complex things to try to accomplish this.

Same! Luckily, this is a simple solution

in app.py, change the line to model = WolfSheep(simulator=simulator, grass=True)

I thought I am crazy, because I wanted to do it on GitHub directly and it was already updated. But it was just you being faster. Thanks!