Update most LayoutTests to be agnostic to scrollTopLeftInterop mode

Update most LayoutTests to be agnostic to scrollTopLeftInterop mode

Replace most uses of body.scrollTop/Left with scrollingElement so that
we can experiment with turning the ScrollTopLeftInterop feature on and
off without too much pain.

BUG=157855
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=195624

[Rick Byers, 2015-05-16 05:48:48]

rbyers@chromium.org changed reviewers:
+ tdresser@chromium.org

[Rick Byers, 2015-05-16 05:48:49]

Tim, PTAL

These changes were generated by doing an interactive search and replace for
'\.body\.scroll' -> '.scrollingElement.scroll', running the tests with
scrollTopLeftInterop enabled and disabled and fixing issues.  I also went
through the test changes jochen@ made when he took ScrollTopLeftInterop out of
the experimental set and made sure I had all of those.  There are still a few
failures when we turn on scrollTopLeftInterop for tests that are explicitly
verifying that we're not yet shipping that mode (so we still have test coverage
of the difference - it's just localized).

Thanks,
   Rick

[tdresser, 2015-05-19 13:53:11]

LGTM, with nits.

https://codereview.chromium.org/1133693002/diff/40001/LayoutTests/fast/dom/wi...
File LayoutTests/fast/dom/window-scroll-scaling.html (right):

https://codereview.chromium.org/1133693002/diff/40001/LayoutTests/fast/dom/wi...
LayoutTests/fast/dom/window-scroll-scaling.html:20: body.style["height"] =
window.innerHeight + 100 + "px";
Is this correct?
Should we be setting the size of the scrollingElement, or the body?

https://codereview.chromium.org/1133693002/diff/40001/LayoutTests/fast/dom/wi...
LayoutTests/fast/dom/window-scroll-scaling.html:31: //
scrollingElement.scrollWidth (scrollHeight) will still be measuring the document
nit: scrollingElement->document.scrollingElement

https://codereview.chromium.org/1133693002/diff/40001/LayoutTests/fast/events...
File LayoutTests/fast/events/scroll-in-scaled-page-with-overflow-hidden.html
(right):

https://codereview.chromium.org/1133693002/diff/40001/LayoutTests/fast/events...
LayoutTests/fast/events/scroll-in-scaled-page-with-overflow-hidden.html:31: //
box can still be scrolled progamatically).
progamatically -> programmatically 
body -> document.body

[Rick Byers, 2015-05-20 15:10:24]

https://codereview.chromium.org/1133693002/diff/40001/LayoutTests/fast/dom/wi...
File LayoutTests/fast/dom/window-scroll-scaling.html (right):

https://codereview.chromium.org/1133693002/diff/40001/LayoutTests/fast/dom/wi...
LayoutTests/fast/dom/window-scroll-scaling.html:20: body.style["height"] =
window.innerHeight + 100 + "px";
On 2015/05/19 13:53:11, tdresser wrote:
> Is this correct?
> Should we be setting the size of the scrollingElement, or the body?

Either would work, but since this test is explicitly testing quirks mode
behavior I'll revert this back to just 'body'.  Technically the test is equally
valid in non-quirks mode if it used scrollingElement but I'm not aiming to
increase any test coverage with this CL so let's just keep it as-is.

There may be some other cases where I unnecessarily updated quirks-mode tests to
use scrollingElement (since mostly this was a bulk-refactor), but I don't think
there's much downside (they're still correct) so I'm not planning on looking
explicitly for such cases.

https://codereview.chromium.org/1133693002/diff/40001/LayoutTests/fast/events...
File LayoutTests/fast/events/scroll-in-scaled-page-with-overflow-hidden.html
(right):

https://codereview.chromium.org/1133693002/diff/40001/LayoutTests/fast/events...
LayoutTests/fast/events/scroll-in-scaled-page-with-overflow-hidden.html:31: //
box can still be scrolled progamatically).
On 2015/05/19 13:53:11, tdresser wrote:
> progamatically -> programmatically 
> body -> document.body

Done.

[Rick Byers, 2015-05-20 15:11:03]

The CQ bit was checked by rbyers@chromium.org

[Rick Byers, 2015-05-20 15:11:04]

The patchset sent to the CQ was uploaded after l-g-t-m from
tdresser@chromium.org
Link to the patchset: https://codereview.chromium.org/1133693002/#ps60001
(title: "tdresser cr feedback")

[commit-bot: I haz the power, 2015-05-20 15:11:10]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1133693002/60001

[commit-bot: I haz the power, 2015-05-20 17:08:15]

Committed patchset #4 (id:60001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=195624