phetsims/qa

Sanity check testing for ES6 modules

Closed this issue · 21 comments

Under https://phet-dev.colorado.edu/modules-testing/, I've manually "deployed" three simulations: build-a-molecule, john-travoltage and molarity.

Under /phet/ should be the phet-brand links, and under /phet-io/ should be the password-protected parts for phet-io testing. It seems like phet-io is operating normally.

If things seem very broken, feel free to contact @jonathanolson or @samreid before continuing.

Notes from phetsims/chipper#867 about what we want tested:

Stage 1: Basic test on all platforms, all 3 sims:

  1. Does the sim run and behave normally?
  2. Fuzz the sim for 30 seconds.

Stage 2: Just 2 sims on a few platforms:

  1. Do the strings display naturally?
  2. Does English work
  3. Does Spanish work?
  4. Do simplified Chinese characters (zh_cn) work?
  5. Check phet-brand, phet-io-brand, and adapted-from-phet including credits and licensing.
  6. Does a11y view run and seem OK?
  7. Do the sims work in an iframe?
  8. Do the sims work in XHTML?
  9. Do the PhET-iO wrappers work?
  10. Do these sims look good in the debug version of the app?
  11. Make sure the sound works.
  12. Test some query parameters, e.g. ?screens=2,1.

For 1-4, use the "all" file and locale specific files.

@jbphet has created a side issue for making sure Rosetta works on a built ES6 sim. See phetsims/chipper#883.

@KatieWoe next top priroity once #480 is finished off.

If you have questions, feel free to ping @jonathanolson

  • iPad 10.12 Safari
    • Basic Test
    • Locale Tests
  • iPad 10.13 Safari
    • Basic Test
    • String Tests
    • a11y View
    • App Test
  • Mac 10.10 Safari
    • Basic Test
    • Brand Tests
  • Mac 10.11 Safari
    • Basic Test
    • String Tests
  • Mac 10.11 Chrome
    • Basic Test
    • iframe Test
  • Mac 10.12 Safari
    • Basic Test
    • Locale Tests
  • Mac 10.12 Chrome
    • Basic Test
    • Sound Test
  • Mac 10.13 Safari
    • Basic Test
    • Sound Test
  • Mac 10.13 Chrome
    • Basic Test
    • iframe Test
  • Mac 10.14 Safari
    • Basic Test
    • XHTML Test
  • Mac 10.14 Chrome
    • Basic Test
    • Query Parameter Test
  • Mac 10.15 Safari
    • Basic Test
    • Brand Tests
    • Wrapper Tests
  • Mac 10.15 Chrome
    • Basic Test
    • a11y View
    • Query Parameter Test
  • Win 10 IE (Fail)
    • Basic Test
    • Locale Tests (X)
    • iframe Test
    • XHTML Test
    • Sound Test
  • Win 10 Edge
    • Basic Test
    • String Tests (X)
    • XHTML Test
  • Win 10 Firefox
    • Basic Test
    • Locale Tests
    • Wrapper Tests
    • Sound Test
  • Win 10 Chrome
    • Basic Test
    • a11y View
    • Wrapper Tests
    • Query Parameter Test
  • ChromeOS
    • Basic Test
    • Brand Tests
    • XHTML Test
    • Query Parameter Test

@jonathanolson @samreid Let me know if there any changes you want to the above test checklist.

Looks good to me!

Also, I think my biggest worry would be how it works on Win10 IE, is it possible to test that next?

I can do that. What aspects? @jonathanolson

Per @samreid doing phet-io testing on Molarity

Noted elsewhere, but recording here. String test query parameters do not seem to be working. Non-locale query parameters should still work.

IE11 and stringTest fixes are included in https://phet-dev.colorado.edu/modules-testing2. I'll be working on other items, but can we get those tests done under that?

Dragging sliders in Molarity in IE is a bit more laggy than in published, but not really bad enough to be issue worthy.

Blue floor and blue wall? Is this a thing that we know about?

1

I saw this on macOS 10.11.6 + Safari 11.1.2. @KatieWoe?

Floor and wall change to normal color if you move John.

If you resize the screen, the blue color goes away.

Were dev tools open in Safari or something?

@jonathanolson, no devtools open, but I may have had them open earlier.

Pending @KatieWoe's approval, QA is done.

I think #482 (comment) should probably have an issue. I'll make one and then we'll finish up.

QA is done

Assigning @ariel-phet since this is important for Thursday.

@samreid @jonathanolson since QA is complete, would be good for both of you to review the found issues. Nothing here looks like show stoppers for the migration to me.

@samreid, @chrisklus, and @zepumph looked over all open issues this morning, and non seemed related to or blocking es6 module migration. We feel good to close, but wanted to ping @jonathanolson to see if there is anything else for this issue.

@jonathanolson @ariel-phet can this be closed?

Yes, thank you!