ByoungSooPark/modeling-app
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Fix: stable E2E tests on ubuntu localhost (#5269)

Browse Source 
* fix: Needed to increase timeout for this test. waitForExecutionDone does not work since there are errors in the KCL code

* fix: again another wait for execution does not work

* fix: bug that desyncs codeManager, executeCode, lspPlugin

* fix: fixing react race condition on parsing numeric literals in command bar on open

* fix: adding comment to clarify the gotcha

* fix: saving off debugging...

* fix: added wait for execution done

* fix: removing testing code

* fix: adding wait for execution done

* fix: adding execution done wait

* fix: only fixes the chamfer point and click delete

* fix: add 1250ms before every progresscmdbar, removed model loaded scene state that is flaky, added toast if someone presses the command bar too quickly

* fix: adding a wait for execution

* fix: updating wait for execution

* fix: wait for execution done

* fix: wait for execution done

* fix: not waiting for scene, not waiting for command bar

* fix: restoring name

* fix: auto fixes

* A snapshot a day keeps the bugs away! 📷🐛 (OS: namespace-profile-ubuntu-8-cores)

* A snapshot a day keeps the bugs away! 📷🐛 (OS: namespace-profile-ubuntu-8-cores)

* A snapshot a day keeps the bugs away! 📷🐛 (OS: namespace-profile-ubuntu-8-cores)

* A snapshot a day keeps the bugs away! 📷🐛 (OS: namespace-profile-ubuntu-8-cores)

* A snapshot a day keeps the bugs away! 📷🐛 (OS: namespace-profile-ubuntu-8-cores)

* A snapshot a day keeps the bugs away! 📷🐛 (OS: namespace-profile-ubuntu-8-cores)

* fix: bad prompt fix

* fix: Fixed testing selections with wait

* fix: last wait fix

* fix: trying to resolve more flakes when running with more workers

* chore: adding a skipLocalEngine tag

* fix: fixing test when using local engine

* fix: codespell

* fix: big if true

* chore: skipping one more local engine tests that fails

* fix: auto fix:

* fix: restoring this testing code

---------

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
This commit is contained in:
Kevin Nadro
2025-02-11 11:13:25 -06:00
committed by GitHub
parent 162856064b
commit 28311d160a
20 changed files with 411 additions and 320 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
e2e/playwright/fixtures/cmdBarFixture.ts
View File

@ -112,6 +112,9 @@ export class CmdBarFixture {
* and assumes we are past the `pickCommand` step.
*/
progressCmdBar = async (shouldFuzzProgressMethod = true) => {
// FIXME: Progressing the command bar is a race condition. We have an async useEffect that reports the final state via useCalculateKclExpression. If this does not run quickly enough, it will not "fail" the continue because you can press continue if the state is not ready. E2E tests do not know this.
// Wait 1250ms to assume the await executeAst of the KCL input field is finished
await this.page.waitForTimeout(1250)
if (shouldFuzzProgressMethod || Math.random() > 0.5) {
const arrowButton = this.page.getByRole('button', {
name: 'arrow right Continue',
@ -128,6 +131,23 @@ export class CmdBarFixture {
}
}
// Added data-testid to the command bar buttons
// command-bar-continue are the buttons to go to the next step
// does not include the submit which is the final button press
// aka the right arrow button
continue = async () => {
const continueButton = this.page.getByTestId('command-bar-continue')
await continueButton.click()
}
// Added data-testid to the command bar buttons
// command-bar-submit is the button for the final step to submit
// the command bar flow aka the checkmark button.
submit = async () => {
const submitButton = this.page.getByTestId('command-bar-submit')
await submitButton.click()
}
openCmdBar = async (selectCmd?: 'promptToEdit') => {
// TODO why does this button not work in electron tests?
// await this.cmdBarOpenBtn.click()