Use plus/minus 0.5 instead of 1 when calculating zoomed length.

Use plus/minus 0.5 instead of 1 when calculating zoomed length.

Plus/minus 0.5 to the truncated value is more reasonable than
directly adding 1, because 0.5 is the average of the fraction
value.

This patch fixes Acid3 test 99/100 when zoom is 125%, and also
improves overall image zooming.

This patch requires a content shell change to set zoom factor
in script (CL 253493006).

BUG=360713
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=172977

[wjywbs, 2014-04-27 18:22:30]

Please take a look. Thanks.

[eae, 2014-04-28 14:44:47]

https://codereview.chromium.org/251063002/diff/1/Source/core/rendering/style/...
File Source/core/rendering/style/RenderStyle.h (right):

https://codereview.chromium.org/251063002/diff/1/Source/core/rendering/style/...
Source/core/rendering/style/RenderStyle.h:1791: inline int
adjustForAbsoluteZoom(int value, float zoomFactor)
We really shouldn't be using ints for layout values at all and the long term fix
here would likely be to use the float or LayoutUnit versions instead,

As a short term intermix fix, however, I suppose this does make more sense.

Could you please add a FIXME to reduce/remove our dependency on zoom adjusted
int values?

[wjywbs, 2014-04-29 02:13:03]

On 2014/04/28 14:44:47, eae wrote:
>
https://codereview.chromium.org/251063002/diff/1/Source/core/rendering/style/...
> File Source/core/rendering/style/RenderStyle.h (right):
> 
>
https://codereview.chromium.org/251063002/diff/1/Source/core/rendering/style/...
> Source/core/rendering/style/RenderStyle.h:1791: inline int
> adjustForAbsoluteZoom(int value, float zoomFactor)
> We really shouldn't be using ints for layout values at all and the long term
fix
> here would likely be to use the float or LayoutUnit versions instead,
> 
> As a short term intermix fix, however, I suppose this does make more sense.
> 
> Could you please add a FIXME to reduce/remove our dependency on zoom adjusted
> int values?

I added the fixme comment. I don't know why someone closed this CL hours ago.
PTAL, thanks.

[leviw_travelin_and_unemployed, 2014-04-29 02:36:04]

LGTM. eae may have a bug to point that fixme at. It'd be nice to have something
more actionable in the comment.

[eae, 2014-04-29 15:43:47]

LGTM

[eae, 2014-04-29 15:43:57]

The CQ bit was checked by eae@chromium.org

[commit-bot: I haz the power, 2014-04-29 15:44:17]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/wjywbs@gmail.com/251063002/20001

[wjywbs, 2014-04-29 15:47:55]

The CQ bit was unchecked by wjywbs@gmail.com

[wjywbs, 2014-04-29 15:50:04]

On 2014/04/29 15:43:57, eae wrote:
> The CQ bit was checked by eae@chromium.org

Thanks, but the content shell change should be committed first in chromium side
(cl 253493006).

[wjywbs, 2014-04-30 11:58:51]

The CQ bit was checked by wjywbs@gmail.com

[commit-bot: I haz the power, 2014-04-30 11:58:59]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/wjywbs@gmail.com/251063002/20001

[commit-bot: I haz the power, 2014-04-30 14:46:59]

Change committed as 172977