Fixes conversion from NSImage to ImageSkia when NSImageRep has size which isn't it's pixel size

NSImageRep can have a size which is not it's pixel size.
Fix conversion between NSImage to ImageSkia for when this is the case.

Also fix the unittest so that it passes on HiDPI mac and changed unittest constants to be kWidth1x, kHeight1x etc

Test=ImageMacTest.NSImageWithResizedNSImageRepToImageSkia
Bug=None

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=141616

[Nico, 2012-06-07 21:25:15]

LGTM

When does this happen?

[pkotwicz, 2012-06-07 21:39:07]

I am unsure as to how this can happen other than the way that I triggered it in
the unittest. It seems that in most cases NSImageRep::size gives the same value
as pixelsWide and pixelsHigh
However, given the NSImageRep API this situation can happen.

[pkotwicz, 2012-06-07 21:39:46]

sky for OWNERS

[sky, 2012-06-07 22:18:29]

Rubber stamp LGTM

[commit-bot: I haz the power, 2012-06-08 13:05:23]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/pkotwicz@chromium.org/10545069/2001

[commit-bot: I haz the power, 2012-06-08 14:55:39]

Change committed as 141206

[pkotwicz, 2012-06-11 01:42:18]

Nico, can you check the style for CreateBitmapImageRep in image_mac_unittest.
Not sure about Objective C style.

[Nico, 2012-06-11 05:05:14]

http://codereview.chromium.org/10545069/diff/10002/ui/gfx/image/image_mac_uni...
File ui/gfx/image/image_mac_unittest.mm (right):

http://codereview.chromium.org/10545069/diff/10002/ui/gfx/image/image_mac_uni...
ui/gfx/image/image_mac_unittest.mm:35: NSRect request_rect = NSMakeRect(0, 0,
static_cast<float>(width) / x_scale,
Remove static_cast<>, promotion rules do that implicitly

http://codereview.chromium.org/10545069/diff/10002/ui/gfx/image/image_mac_uni...
ui/gfx/image/image_mac_unittest.mm:40: [image unlockFocus];
…hm, this is pretty convoluted. Maybe it's simpler to create a NSBitmapImageRep
directly (
https://developer.apple.com/library/mac/#documentation/Cocoa/Reference/Applic...
) and then draw to it using graphicsContextWithBitmapImageRep: (
https://developer.apple.com/library/mac/#documentation/Cocoa/Reference/Applic...:
)?

[pkotwicz, 2012-06-11 15:03:05]

Create NSBitmapImageRep directly as recommended. Don't draw anything as it is
not necessary for the NSBitmapImageRep to actually have content for the purpose
of the tests

[Nico, 2012-06-11 15:23:07]

lgtm with comments addressed

http://codereview.chromium.org/10545069/diff/1003/ui/gfx/image/image_mac_unit...
File ui/gfx/image/image_mac_unittest.mm (right):

http://codereview.chromium.org/10545069/diff/1003/ui/gfx/image/image_mac_unit...
ui/gfx/image/image_mac_unittest.mm:21: *image_rep = [[NSBitmapImageRep alloc]
initWithBitmapDataPlanes:NULL
nit: break before initWith…, and align subsequent lines on the :

http://codereview.chromium.org/10545069/diff/1003/ui/gfx/image/image_mac_unit...
ui/gfx/image/image_mac_unittest.mm:28: colorSpaceName:@"NSDeviceRGBColorSpace"
I think NSDeviceRGBColorSpace is a named constant and shouldn't be in @"".
Remove @" and ".

[commit-bot: I haz the power, 2012-06-11 16:50:21]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/pkotwicz@chromium.org/10545069/6006

[commit-bot: I haz the power, 2012-06-11 18:46:39]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/pkotwicz@chromium.org/10545069/6006

[commit-bot: I haz the power, 2012-06-12 03:27:57]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/pkotwicz@chromium.org/10545069/6006

[commit-bot: I haz the power, 2012-06-12 04:59:42]

Change committed as 141616