CSS Animations: Fix flaky test transform-origin-vs-functions.html

CSS Animations: Fix flaky test transform-origin-vs-functions.html

The test used setTimeout, and relied on precise timing.

We now use testharness.js, and no longer rely on callback timing.

BUG=248938, 664857, 674715
Committed: https://crrev.com/53db6d2749de2904b040c2d86936c9a4d1e7dde3
Cr-Commit-Position: refs/heads/master@{#439391}

[Eric Willigers, 2016-12-16 02:49:22]

Description was changed from

==========
CSS Animations: Fix flaky test transform-origin-vs-functions.html

The test used setTimeout, and relied on precise timing.

We now use testharness.js, and no longer rely on callback timing.

BUG=664857
==========

to

==========
CSS Animations: Fix flaky test transform-origin-vs-functions.html

The test used setTimeout, and relied on precise timing.

We now use testharness.js, and no longer rely on callback timing.

BUG=248938,664857,674715
==========

[Eric Willigers, 2016-12-16 02:49:22]

ericwilligers@chromium.org changed reviewers:
+ suzyh@chromium.org

[Eric Willigers, 2016-12-16 02:49:27]

The CQ bit was checked by ericwilligers@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-12-16 02:50:28]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[suzyh_UTC10 (ex-contributor), 2016-12-16 03:07:09]

https://codereview.chromium.org/2582713002/diff/20001/third_party/WebKit/Layo...
File
third_party/WebKit/LayoutTests/animations/3d/transform-origin-vs-functions.html
(right):

https://codereview.chromium.org/2582713002/diff/20001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/animations/3d/transform-origin-vs-functions.html:66:
function epsilonToZero(value) {
Hmmm... This approach makes me a little uncomfortable. I would rather have
something parse "matrix3d(...)" into a list of numbers that we can
assert_approx_equals on. What options do we have for libraries or existing
functions that would do that parsing for us?

https://codereview.chromium.org/2582713002/diff/20001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/animations/3d/transform-origin-vs-functions.html:80:
}, "transform-origin vs. Transform Functions");
While you're editing this file, can you elaborate this test name into a more
self-explanatory statement?

[alancutter (OOO until 2018), 2016-12-16 03:19:33]

alancutter@chromium.org changed reviewers:
+ alancutter@chromium.org

[alancutter (OOO until 2018), 2016-12-16 03:19:33]

https://codereview.chromium.org/2582713002/diff/20001/third_party/WebKit/Layo...
File
third_party/WebKit/LayoutTests/animations/3d/transform-origin-vs-functions.html
(right):

https://codereview.chromium.org/2582713002/diff/20001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/animations/3d/transform-origin-vs-functions.html:66:
function epsilonToZero(value) {
On 2016/12/16 at 03:07:09, suzyh wrote:
> Hmmm... This approach makes me a little uncomfortable. I would rather have
something parse "matrix3d(...)" into a list of numbers that we can
assert_approx_equals on. What options do we have for libraries or existing
functions that would do that parsing for us?

I wrote a function like that for one of my tests:
https://cs.chromium.org/chromium/src/third_party/WebKit/LayoutTests/animation...

[suzyh_UTC10 (ex-contributor), 2016-12-16 03:26:06]

On 2016/12/16 at 03:19:33, alancutter wrote:
>
https://codereview.chromium.org/2582713002/diff/20001/third_party/WebKit/Layo...
> File
third_party/WebKit/LayoutTests/animations/3d/transform-origin-vs-functions.html
(right):
> 
>
https://codereview.chromium.org/2582713002/diff/20001/third_party/WebKit/Layo...
>
third_party/WebKit/LayoutTests/animations/3d/transform-origin-vs-functions.html:66:
function epsilonToZero(value) {
> On 2016/12/16 at 03:07:09, suzyh wrote:
> > Hmmm... This approach makes me a little uncomfortable. I would rather have
something parse "matrix3d(...)" into a list of numbers that we can
assert_approx_equals on. What options do we have for libraries or existing
functions that would do that parsing for us?
> 
> I wrote a function like that for one of my tests:
>
https://cs.chromium.org/chromium/src/third_party/WebKit/LayoutTests/animation...

Then let's pull that out into a shared resources/something.js file and reuse it
here.

[Eric Willigers, 2016-12-16 05:30:15]

The CQ bit was checked by ericwilligers@chromium.org to run a CQ dry run

[Eric Willigers, 2016-12-16 05:30:25]

https://codereview.chromium.org/2582713002/diff/20001/third_party/WebKit/Layo...
File
third_party/WebKit/LayoutTests/animations/3d/transform-origin-vs-functions.html
(right):

https://codereview.chromium.org/2582713002/diff/20001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/animations/3d/transform-origin-vs-functions.html:66:
function epsilonToZero(value) {
On 2016/12/16 03:19:33, alancutter wrote:
> On 2016/12/16 at 03:07:09, suzyh wrote:
> > Hmmm... This approach makes me a little uncomfortable. I would rather have
> something parse "matrix3d(...)" into a list of numbers that we can
> assert_approx_equals on. What options do we have for libraries or existing
> functions that would do that parsing for us?
> 
> I wrote a function like that for one of my tests:
>
https://cs.chromium.org/chromium/src/third_party/WebKit/LayoutTests/animation...

That function splits 6.12323e-17 into 6.12323 and -17.

I've modified the function and added in to the new file helpers.js  (I created a
new helpers.js instead of adding to animation-test-helpers.js, as
animation-test-helpers.js is designed for non-testharness tests.)

I haven't changed the existing transform-post-multiplication.html test to use
the new function - it would start failing as it currently accidentally considers
numbers like '1.2344e+17' and '1.2346e+17' to be within 0.01 of each other in
absolute value.

https://codereview.chromium.org/2582713002/diff/20001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/animations/3d/transform-origin-vs-functions.html:80:
}, "transform-origin vs. Transform Functions");
On 2016/12/16 03:07:09, suzyh wrote:
> While you're editing this file, can you elaborate this test name into a more
> self-explanatory statement?

Changed the wording.

[commit-bot: I haz the power, 2016-12-16 05:30:41]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-12-16 07:08:42]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-12-16 07:08:43]

Dry run: This issue passed the CQ dry run.

[suzyh_UTC10 (ex-contributor), 2016-12-18 23:35:23]

lgtm, thanks!

[Eric Willigers, 2016-12-18 23:42:50]

The CQ bit was checked by ericwilligers@chromium.org

[commit-bot: I haz the power, 2016-12-18 23:43:05]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-12-19 01:09:49]

CQ is committing da patch.

Bot data: {"patchset_id": 40001, "attempt_start_ts": 1482104570532580,
"parent_rev": "a420dfe484a4ae507599bed1d62e94d04db9fb9a", "commit_rev":
"a71cf1bc351234989fcf093042eb3b1dfa04047f"}

[commit-bot: I haz the power, 2016-12-19 01:10:52]

Description was changed from

==========
CSS Animations: Fix flaky test transform-origin-vs-functions.html

The test used setTimeout, and relied on precise timing.

We now use testharness.js, and no longer rely on callback timing.

BUG=248938,664857,674715
==========

to

==========
CSS Animations: Fix flaky test transform-origin-vs-functions.html

The test used setTimeout, and relied on precise timing.

We now use testharness.js, and no longer rely on callback timing.

BUG=248938,664857,674715

Review-Url: https://codereview.chromium.org/2582713002
==========

[commit-bot: I haz the power, 2016-12-19 01:10:53]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2016-12-19 01:14:08]

Description was changed from

==========
CSS Animations: Fix flaky test transform-origin-vs-functions.html

The test used setTimeout, and relied on precise timing.

We now use testharness.js, and no longer rely on callback timing.

BUG=248938,664857,674715

Review-Url: https://codereview.chromium.org/2582713002
==========

to

==========
CSS Animations: Fix flaky test transform-origin-vs-functions.html

The test used setTimeout, and relied on precise timing.

We now use testharness.js, and no longer rely on callback timing.

BUG=248938,664857,674715

Committed: https://crrev.com/53db6d2749de2904b040c2d86936c9a4d1e7dde3
Cr-Commit-Position: refs/heads/master@{#439391}
==========

[commit-bot: I haz the power, 2016-12-19 01:14:09]

Patchset 3 (id:??) landed as
https://crrev.com/53db6d2749de2904b040c2d86936c9a4d1e7dde3
Cr-Commit-Position: refs/heads/master@{#439391}