Test refactor, making them faster

[ahmadayubi]

Refactor and parallelism resulted in tests running in under 2 minutes (or less).
This requires the use of await page.waitForTimeout(200); when doing many keyboard interactions in a row to prevent missed keys.

• Certain tests are currently unstable, need fix

Closes #518 (indirect, as instead of launching a single browser, it's launching multiple browsers depending on the device)

[prushforth]

It's fantastic how fast they run. I ran them 4 times, but had a failure in one run, that disappeared in subsequent runs. Have you noticed this, and if so is it a setting that could be adjusted?

[ahmadayubi]

It's fantastic how fast they run. I ran them 4 times, but had a failure in one run, that disappeared in subsequent runs. Have you noticed this, and if so is it a setting that could be adjusted?

Do you remember the test? This is probably an issue related to actions being made too quickly. Since mapml-viewer is visual with animations, parts of the code need timeouts to allow the animations and loading to complete.

You can simply increase the overall sloMo value but this slows down parts that don't need slowing.

[prushforth]

No I didn't take note. I will play around to see if I encounter it again.

[Anshpreet8]

Do you remember the test? This is probably an issue related to actions being made too quickly. Since mapml-viewer is visual with animations, parts of the code need timeouts to allow the animations and loading to complete.

It appeared I had a failure for mapMLTemplatedFeatures and mapMLTemplatedImageLayer two seperate times, the failures seem to be a result of zooming:

[ahmadayubi]

It appeared I had a failure for mapMLTemplatedFeatures and mapMLTemplatedImageLayer two seperate times, the failures seem to be a result of zooming:

Try adding timeouts to parts, for example many keyboard actions at one point, see if that fixes it. Those files for me didn't cause issues but it's very hardware dependent (without timeouts).

[Anshpreet8]

Try adding timeouts to parts that seem, for example many keyboard actions at one point, see if that fixes it. Those files for me didn't cause issues but it's very hardware dependent (without timeouts).

yea, adding a larger timeout number seemed to fix it for me

[prushforth]

OK, I couldn't get the tests to fail again on that branch. It could have to do with caching resources, that happens at levels we might not be aware of. So, let's merge and deal with issues as they may come.