Remove double constructors that assume degrees from CSSRotation

Remove double constructors that assume degrees from CSSRotation

Instead, use only constructors that take CSSAngleValues, so that the
unit is explicit, then actually store this information.

This is to reflect the spec change that removed these constructors in
https://github.com/w3c/css-houdini-drafts/commit/535591271f88d71fd7ca60dc4132db15f93b980a

Also includes:
- Adding support for making a CSSAngleValue out of a
    CSSPrimitiveValue, to continue supporting getting
    CSSTransform(CSSRotation) from inline StylePropertyMaps.
- Moves some of the inlineStyle CSSRotation tests to use the
    property-suite for more thorough testing. I was not able
    to convert all the tests, as I still have to figure out
    how to test the behaviour when the value is set using the
    old CSSOM to something like "rotateX", "rotateY" and
    "rotateZ".

BUG=545318
Committed: https://crrev.com/0eebfb8ffb2b8730c2679d41c011f2871b28cdf1
Cr-Commit-Position: refs/heads/master@{#439404}

[meade_UTC10, 2016-12-05 04:56:35]

Description was changed from

==========
[WIP] Half baked look at removing double constructors from CSSRotation

BUG=
==========

to

==========
Remove double constructors that assume degrees from CSSRotation

Instead, use only constructors that take CSSAngleValues, so that the
unit is explicit.

This is to reflect the spec change that removed these constructors in
https://github.com/w3c/css-houdini-drafts/commit/535591271f88d71fd7ca60dc4132...

Also includes:
- Adding support for making a CSSAngleValue out of a 
    CSSPrimitiveValue, to continue supporting getting 
    CSSTransform(CSSRotation) from inline StylePropertyMaps.
- Moves some of the inlineStyle CSSRotation tests to use the 
    property-suite for more thorough testing.

BUG=545318
==========

[meade_UTC10, 2016-12-05 04:58:14]

Description was changed from

==========
Remove double constructors that assume degrees from CSSRotation

Instead, use only constructors that take CSSAngleValues, so that the
unit is explicit.

This is to reflect the spec change that removed these constructors in
https://github.com/w3c/css-houdini-drafts/commit/535591271f88d71fd7ca60dc4132...

Also includes:
- Adding support for making a CSSAngleValue out of a 
    CSSPrimitiveValue, to continue supporting getting 
    CSSTransform(CSSRotation) from inline StylePropertyMaps.
- Moves some of the inlineStyle CSSRotation tests to use the 
    property-suite for more thorough testing.

BUG=545318
==========

to

==========
Remove double constructors that assume degrees from CSSRotation

Instead, use only constructors that take CSSAngleValues, so that the
unit is explicit.

This is to reflect the spec change that removed these constructors in
https://github.com/w3c/css-houdini-drafts/commit/535591271f88d71fd7ca60dc4132...

Also includes:
- Adding support for making a CSSAngleValue out of a 
    CSSPrimitiveValue, to continue supporting getting 
    CSSTransform(CSSRotation) from inline StylePropertyMaps.
- Moves some of the inlineStyle CSSRotation tests to use the 
    property-suite for more thorough testing. I was not able
    to convert all the tests, as I still have to figure out
    how to test the behaviour when the value is set using the
    old CSSOM to something like "rotateX", "rotateY" and
    "rotateZ".

BUG=545318
==========

[meade_UTC10, 2016-12-05 05:03:46]

Description was changed from

==========
Remove double constructors that assume degrees from CSSRotation

Instead, use only constructors that take CSSAngleValues, so that the
unit is explicit.

This is to reflect the spec change that removed these constructors in
https://github.com/w3c/css-houdini-drafts/commit/535591271f88d71fd7ca60dc4132...

Also includes:
- Adding support for making a CSSAngleValue out of a 
    CSSPrimitiveValue, to continue supporting getting 
    CSSTransform(CSSRotation) from inline StylePropertyMaps.
- Moves some of the inlineStyle CSSRotation tests to use the 
    property-suite for more thorough testing. I was not able
    to convert all the tests, as I still have to figure out
    how to test the behaviour when the value is set using the
    old CSSOM to something like "rotateX", "rotateY" and
    "rotateZ".

BUG=545318
==========

to

==========
Remove double constructors that assume degrees from CSSRotation

Instead, use only constructors that take CSSAngleValues, so that the
unit is explicit, then actually store this information.

This is to reflect the spec change that removed these constructors in
https://github.com/w3c/css-houdini-drafts/commit/535591271f88d71fd7ca60dc4132...

Also includes:
- Adding support for making a CSSAngleValue out of a 
    CSSPrimitiveValue, to continue supporting getting 
    CSSTransform(CSSRotation) from inline StylePropertyMaps.
- Moves some of the inlineStyle CSSRotation tests to use the 
    property-suite for more thorough testing. I was not able
    to convert all the tests, as I still have to figure out
    how to test the behaviour when the value is set using the
    old CSSOM to something like "rotateX", "rotateY" and
    "rotateZ".

BUG=545318
==========

[meade_UTC10, 2016-12-05 05:20:08]

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

[commit-bot: I haz the power, 2016-12-05 05:20:19]

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

[meade_UTC10, 2016-12-05 06:18:17]

meade@chromium.org changed reviewers:
+ rjwright@chromium.org, timloh@chromium.org

[meade_UTC10, 2016-12-05 06:18:18]

There's a lot going on in this CL, let me know if you think I should split it
up!

[meade_UTC10, 2016-12-05 06:18:18]

meade@chromium.org changed reviewers:
+ rjwright@chromium.org, timloh@chromium.org

[meade_UTC10, 2016-12-05 06:18:19]

There's a lot going on in this CL, let me know if you think I should split it
up!

[commit-bot: I haz the power, 2016-12-05 06:47:52]

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

[commit-bot: I haz the power, 2016-12-05 06:47:53]

Dry run: Try jobs failed on following builders:
  mac_chromium_rel_ng on master.tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[Timothy Loh, 2016-12-06 06:31:40]

https://codereview.chromium.org/2550063002/diff/40001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/css/cssom/CSSAngleValue.cpp (right):

https://codereview.chromium.org/2550063002/diff/40001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/css/cssom/CSSAngleValue.cpp:22:
DCHECK(!value.isCalculated());
It's not obvious to me that this is correct -- are there tests for this?

[Timothy Loh, 2016-12-06 06:34:39]

BTW typedcssom/inlinestyle/transform.html needs to be updated too

[meade_UTC10, 2016-12-06 06:35:58]

https://codereview.chromium.org/2550063002/diff/40001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/css/cssom/CSSAngleValue.cpp (right):

https://codereview.chromium.org/2550063002/diff/40001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/css/cssom/CSSAngleValue.cpp:22:
DCHECK(!value.isCalculated());
On 2016/12/06 06:31:39, Timothy Loh wrote:
> It's not obvious to me that this is correct -- are there tests for this?

Yes, it's tested in
LayoutTests/typedcssom/inlilnestyle/properties/transform.html.

If this didn't work, "Getting transform with a CSSTransformValue whose value is
rotate(30deg)" etc would fail (see transform-expected.txt)

[Timothy Loh, 2016-12-06 06:44:17]

On 2016/12/06 06:35:58, Eddy wrote:
>
https://codereview.chromium.org/2550063002/diff/40001/third_party/WebKit/Sour...
> File third_party/WebKit/Source/core/css/cssom/CSSAngleValue.cpp (right):
> 
>
https://codereview.chromium.org/2550063002/diff/40001/third_party/WebKit/Sour...
> third_party/WebKit/Source/core/css/cssom/CSSAngleValue.cpp:22:
> DCHECK(!value.isCalculated());
> On 2016/12/06 06:31:39, Timothy Loh wrote:
> > It's not obvious to me that this is correct -- are there tests for this?
> 
> Yes, it's tested in
> LayoutTests/typedcssom/inlilnestyle/properties/transform.html.
> 
> If this didn't work, "Getting transform with a CSSTransformValue whose value
is
> rotate(30deg)" etc would fail (see transform-expected.txt)

I mean for calcs in particular, e.g. "rotate(calc(30deg + 100grad))". It doesn't
look like we are explicitly rejecting these values anywhere and I don't think we
simplify these values into non-calcs (I think we simplify down to
"rotate(calc(120deg))")

[meade_UTC10, 2016-12-09 04:03:13]

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

[commit-bot: I haz the power, 2016-12-09 04:03:42]

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

[meade_UTC10, 2016-12-09 04:07:16]

On 2016/12/06 06:44:17, Timothy Loh wrote:
> On 2016/12/06 06:35:58, Eddy wrote:
> >
>
https://codereview.chromium.org/2550063002/diff/40001/third_party/WebKit/Sour...
> > File third_party/WebKit/Source/core/css/cssom/CSSAngleValue.cpp (right):
> > 
> >
>
https://codereview.chromium.org/2550063002/diff/40001/third_party/WebKit/Sour...
> > third_party/WebKit/Source/core/css/cssom/CSSAngleValue.cpp:22:
> > DCHECK(!value.isCalculated());
> > On 2016/12/06 06:31:39, Timothy Loh wrote:
> > > It's not obvious to me that this is correct -- are there tests for this?
> > 
> > Yes, it's tested in
> > LayoutTests/typedcssom/inlilnestyle/properties/transform.html.
> > 
> > If this didn't work, "Getting transform with a CSSTransformValue whose value
> is
> > rotate(30deg)" etc would fail (see transform-expected.txt)
> 
> I mean for calcs in particular, e.g. "rotate(calc(30deg + 100grad))". It
doesn't
> look like we are explicitly rejecting these values anywhere and I don't think
we
> simplify these values into non-calcs (I think we simplify down to
> "rotate(calc(120deg))")

I adjusted the code to do something more sensible, and added some tests.

[meade_UTC10, 2016-12-09 04:15:00]

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

[commit-bot: I haz the power, 2016-12-09 04:15:19]

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, 2016-12-09 05:26:55]

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

[commit-bot: I haz the power, 2016-12-09 05:26:55]

Dry run: Try jobs failed on following builders:
  linux_chromium_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[rjwright, 2016-12-16 04:33:03]

https://codereview.chromium.org/2550063002/diff/80001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/css/cssom/CSSAngleValue.cpp (right):

https://codereview.chromium.org/2550063002/diff/80001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/css/cssom/CSSAngleValue.cpp:23: return nullptr;
We won't be supporting calc angles ever?

[rjwright, 2016-12-16 04:33:34]

LGTM

[meade_UTC10, 2016-12-19 01:32:47]

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

[commit-bot: I haz the power, 2016-12-19 01:32:56]

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

[meade_UTC10, 2016-12-19 02:25:47]

meade@chromium.org changed reviewers:
+ sashab@chromium.org

[meade_UTC10, 2016-12-19 02:25:48]

+sasha since Tim's OOO now.

PTAL? Thanks!

[sashab, 2016-12-19 03:30:18]

LGTM :)

[meade_UTC10, 2016-12-19 04:00:01]

Thanks Sasha!

[meade_UTC10, 2016-12-19 04:00:08]

The CQ bit was unchecked by meade@chromium.org

[meade_UTC10, 2016-12-19 04:00:13]

The CQ bit was checked by meade@chromium.org

[meade_UTC10, 2016-12-19 04:00:14]

The patchset sent to the CQ was uploaded after l-g-t-m from
rjwright@chromium.org
Link to the patchset: https://codereview.chromium.org/2550063002/#ps100001
(title: "Fix test")

[commit-bot: I haz the power, 2016-12-19 04:00:30]

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

[meade_UTC10, 2016-12-19 04:01:21]

https://codereview.chromium.org/2550063002/diff/80001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/css/cssom/CSSAngleValue.cpp (right):

https://codereview.chromium.org/2550063002/diff/80001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/css/cssom/CSSAngleValue.cpp:23: return nullptr;
On 2016/12/16 04:33:03, rjwright wrote:
> We won't be supporting calc angles ever?

We might, but it's not specced yet...
I'm planning on leaving it like this until we have defined behaviour for what
it's supposed to do.

[commit-bot: I haz the power, 2016-12-19 04:05:18]

CQ is committing da patch.

Bot data: {"patchset_id": 100001, "attempt_start_ts": 1482120013645510,
"parent_rev": "6ff26824c5c70458f7ba168c108b2e6869302aff", "commit_rev":
"48ca9e40f500a9f5bb46c879a1fda3952e45f515"}

[commit-bot: I haz the power, 2016-12-19 04:10:11]

CQ is committing da patch.

Bot data: {"patchset_id": 100001, "attempt_start_ts": 1482120013645510,
"parent_rev": "8a60fa535c4ce7ef797c73038631c1c486292c0a", "commit_rev":
"324a4ed62e5e6f4b1cd47bfe248d798c35ccbd2f"}

[commit-bot: I haz the power, 2016-12-19 04:10:38]

Description was changed from

==========
Remove double constructors that assume degrees from CSSRotation

Instead, use only constructors that take CSSAngleValues, so that the
unit is explicit, then actually store this information.

This is to reflect the spec change that removed these constructors in
https://github.com/w3c/css-houdini-drafts/commit/535591271f88d71fd7ca60dc4132...

Also includes:
- Adding support for making a CSSAngleValue out of a 
    CSSPrimitiveValue, to continue supporting getting 
    CSSTransform(CSSRotation) from inline StylePropertyMaps.
- Moves some of the inlineStyle CSSRotation tests to use the 
    property-suite for more thorough testing. I was not able
    to convert all the tests, as I still have to figure out
    how to test the behaviour when the value is set using the
    old CSSOM to something like "rotateX", "rotateY" and
    "rotateZ".

BUG=545318
==========

to

==========
Remove double constructors that assume degrees from CSSRotation

Instead, use only constructors that take CSSAngleValues, so that the
unit is explicit, then actually store this information.

This is to reflect the spec change that removed these constructors in
https://github.com/w3c/css-houdini-drafts/commit/535591271f88d71fd7ca60dc4132...

Also includes:
- Adding support for making a CSSAngleValue out of a 
    CSSPrimitiveValue, to continue supporting getting 
    CSSTransform(CSSRotation) from inline StylePropertyMaps.
- Moves some of the inlineStyle CSSRotation tests to use the 
    property-suite for more thorough testing. I was not able
    to convert all the tests, as I still have to figure out
    how to test the behaviour when the value is set using the
    old CSSOM to something like "rotateX", "rotateY" and
    "rotateZ".

BUG=545318

Review-Url: https://codereview.chromium.org/2550063002
==========

[commit-bot: I haz the power, 2016-12-19 04:10:39]

Committed patchset #6 (id:100001)

[commit-bot: I haz the power, 2016-12-19 04:13:07]

Description was changed from

==========
Remove double constructors that assume degrees from CSSRotation

Instead, use only constructors that take CSSAngleValues, so that the
unit is explicit, then actually store this information.

This is to reflect the spec change that removed these constructors in
https://github.com/w3c/css-houdini-drafts/commit/535591271f88d71fd7ca60dc4132...

Also includes:
- Adding support for making a CSSAngleValue out of a 
    CSSPrimitiveValue, to continue supporting getting 
    CSSTransform(CSSRotation) from inline StylePropertyMaps.
- Moves some of the inlineStyle CSSRotation tests to use the 
    property-suite for more thorough testing. I was not able
    to convert all the tests, as I still have to figure out
    how to test the behaviour when the value is set using the
    old CSSOM to something like "rotateX", "rotateY" and
    "rotateZ".

BUG=545318

Review-Url: https://codereview.chromium.org/2550063002
==========

to

==========
Remove double constructors that assume degrees from CSSRotation

Instead, use only constructors that take CSSAngleValues, so that the
unit is explicit, then actually store this information.

This is to reflect the spec change that removed these constructors in
https://github.com/w3c/css-houdini-drafts/commit/535591271f88d71fd7ca60dc4132...

Also includes:
- Adding support for making a CSSAngleValue out of a
    CSSPrimitiveValue, to continue supporting getting
    CSSTransform(CSSRotation) from inline StylePropertyMaps.
- Moves some of the inlineStyle CSSRotation tests to use the
    property-suite for more thorough testing. I was not able
    to convert all the tests, as I still have to figure out
    how to test the behaviour when the value is set using the
    old CSSOM to something like "rotateX", "rotateY" and
    "rotateZ".

BUG=545318

Committed: https://crrev.com/0eebfb8ffb2b8730c2679d41c011f2871b28cdf1
Cr-Commit-Position: refs/heads/master@{#439404}
==========

[commit-bot: I haz the power, 2016-12-19 04:13:08]

Patchset 6 (id:??) landed as
https://crrev.com/0eebfb8ffb2b8730c2679d41c011f2871b28cdf1
Cr-Commit-Position: refs/heads/master@{#439404}