Remove the ENABLE_IFRAME_SEAMLESS compile-time flag.

Remove the ENABLE_IFRAME_SEAMLESS compile-time flag.

This patch drops the ~10 instances of '#IFDEF ENABLE(IFRAME_SEAMLESS)' from the
codebase. It shouldn't effect any web-visible behavior, as the interesting bits
are all hidden safely behind a runtime flag, whose behavior is untouched by
this patch.

The only non-mechanical change in this patch is the addition of an annotation
to the iframe@seamless attribute, ensuring that it's only exposed if the
runtime flag is set (that was an oversight in the original code).

R=eseidel@chromium.org

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=147704

[Mike West, 2013-04-05 11:59:56]

Hi Eric! Would you mind taking a look at this IFDEF-killing patch?

Thanks!

-mike

[eseidel, 2013-04-05 19:39:45]

LGTM.  We may need to patch M27 to not expose seamless if the property is
leaking as it appears to be.

https://codereview.chromium.org/13646006/diff/1/Source/WebCore/html/HTMLIFram...
File Source/WebCore/html/HTMLIFrameElement.idl (left):

https://codereview.chromium.org/13646006/diff/1/Source/WebCore/html/HTMLIFram...
Source/WebCore/html/HTMLIFrameElement.idl:30: [Reflect,
Conditional=IFRAME_SEAMLESS] attribute boolean seamless;
It appears we were doing this wrong before. :)  We were exposing iframe.seamless
in M27??

[Mike West, 2013-04-05 20:12:35]

On 2013/04/05 19:39:45, Eric Seidel (Google) wrote:
> LGTM.  We may need to patch M27 to not expose seamless if the property is
> leaking as it appears to be.

Indeed. Are we still using webkit.org for those release branches?

[Mike West, 2013-04-05 23:32:53]

Committed patchset #2 manually as r147704 (presubmit successful).