[CSS Shapes] inset corner radii are not flipped for vertical writing modes

[CSS Shapes] inset corner radii are not flipped for vertical writing modes

When creating an inset shape, convert the radii to logical coordinates as well as the box.

BUG=352785
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=169288

[Bem Jones-Bey (adobe), 2014-03-12 19:06:17]

https://codereview.chromium.org/196723010/diff/1/LayoutTests/fast/shapes/shap...
File
LayoutTests/fast/shapes/shape-outside-floats/shape-outside-floats-inset-rounded-different-writing-modes-left.html
(right):

https://codereview.chromium.org/196723010/diff/1/LayoutTests/fast/shapes/shap...
LayoutTests/fast/shapes/shape-outside-floats/shape-outside-floats-inset-rounded-different-writing-modes-left.html:61:
assert_approx_equals(rect.top, 91, 0.1);
I'd just make the epsilon in the tests 0.5 instead of 0.1. When writing the test
in WebKit, I didn't see any fractions even though AFAIK they've turned on
subpixel layout. 0.5 should nicely account for rounding (or the lack thereof)

[leviw_travelin_and_unemployed, 2014-03-12 23:43:15]

On 2014/03/12 19:06:17, Bem Jones-Bey (adobe) wrote:
>
https://codereview.chromium.org/196723010/diff/1/LayoutTests/fast/shapes/shap...
> File
>
LayoutTests/fast/shapes/shape-outside-floats/shape-outside-floats-inset-rounded-different-writing-modes-left.html
> (right):
> 
>
https://codereview.chromium.org/196723010/diff/1/LayoutTests/fast/shapes/shap...
>
LayoutTests/fast/shapes/shape-outside-floats/shape-outside-floats-inset-rounded-different-writing-modes-left.html:61:
> assert_approx_equals(rect.top, 91, 0.1);
> I'd just make the epsilon in the tests 0.5 instead of 0.1. When writing the
test
> in WebKit, I didn't see any fractions even though AFAIK they've turned on
> subpixel layout. 0.5 should nicely account for rounding (or the lack thereof)

They've definitely enabled sub-pixel layout (woohoo!)

[rwlbuis, 2014-03-13 01:53:00]

On 2014/03/12 19:06:17, Bem Jones-Bey (adobe) wrote:
>
https://codereview.chromium.org/196723010/diff/1/LayoutTests/fast/shapes/shap...
> File
>
LayoutTests/fast/shapes/shape-outside-floats/shape-outside-floats-inset-rounded-different-writing-modes-left.html
> (right):
> 
>
https://codereview.chromium.org/196723010/diff/1/LayoutTests/fast/shapes/shap...
>
LayoutTests/fast/shapes/shape-outside-floats/shape-outside-floats-inset-rounded-different-writing-modes-left.html:61:
> assert_approx_equals(rect.top, 91, 0.1);
> I'd just make the epsilon in the tests 0.5 instead of 0.1. When writing the
test
> in WebKit, I didn't see any fractions even though AFAIK they've turned on
> subpixel layout. 0.5 should nicely account for rounding (or the lack thereof)

Done in patch set 2, works on my mac, not sure yet about linux....

[Bem Jones-Bey (adobe), 2014-03-14 21:40:50]

lgtm, assuming it works on linux.

[rwlbuis, 2014-03-14 22:04:27]

Levi, Bem likes the patch as you can see. It is not tested on linux yet since
the trybots are slow to pick it up, but I think we could even try CQ, we'd get
notified in case it fails on linux.

[leviw_travelin_and_unemployed, 2014-03-14 22:44:48]

Yup. LGTM.

[rwlbuis, 2014-03-14 23:25:32]

The CQ bit was checked by rob.buis@samsung.com

[commit-bot: I haz the power, 2014-03-14 23:25:36]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rob.buis@samsung.com/196723010/70001

[commit-bot: I haz the power, 2014-03-14 23:33:04]

Change committed as 169288