Clean up the initializer mess in Document.

Clean up the initializer mess in Document.

Before this, some members were initialized in the initializer list and
others were initialized in the body of the constructor. As far as I can
tell, it's OK for everything to just be in the initializer list.

As a bonus, I found two completely unused members in Document (and a
local static) and removed them as well.

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=151593

R=abarth@chromium.org

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=151695

[abarth-chromium, 2013-05-31 22:47:58]

lgtm

https://codereview.chromium.org/16276002/diff/1/Source/core/dom/Document.cpp
File Source/core/dom/Document.cpp (right):

https://codereview.chromium.org/16276002/diff/1/Source/core/dom/Document.cpp#...
Source/core/dom/Document.cpp:477: , m_textAutosizer(TextAutosizer::create(this))
I'd leave these in the body of the constructor because they might run a
non-trivial amount of code, and that code might want to see the document in a
more constructed state.

[dcheng, 2013-05-31 22:54:53]

https://codereview.chromium.org/16276002/diff/1/Source/core/dom/Document.cpp
File Source/core/dom/Document.cpp (right):

https://codereview.chromium.org/16276002/diff/1/Source/core/dom/Document.cpp#...
Source/core/dom/Document.cpp:477: , m_textAutosizer(TextAutosizer::create(this))
On 2013/05/31 22:47:58, abarth wrote:
> I'd leave these in the body of the constructor because they might run a
> non-trivial amount of code, and that code might want to see the document in a
> more constructed state.

Before I made this change, I inspected and manually verified that they don't
depend on Document being in a fully constructed state. Is there a concern that
this will change in the future? If it does change in the future, wouldn't that
be something we want to know about by explicitly breaking so the dependency can
be more clearly documented?

[abarth-chromium, 2013-05-31 23:00:35]

Ok.  If you checked it, then it's probably ok.

[commit-bot: I haz the power, 2013-05-31 23:17:15]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/dcheng@chromium.org/16276002/1

[commit-bot: I haz the power, 2013-06-01 03:05:06]

Change committed as 151593

[dcheng, 2013-06-03 21:03:35]

PTAL. This patch was reverted due to a debug assert; it turns out I made a silly
logic error when calling setURL around ~Document.cpp:500 in the previous
version; the behavior differs if |m_frame| is null but |url| is not empty. This
patch should now match previous behavior exactly.

[abarth-chromium, 2013-06-03 21:14:36]

lgtm

[dcheng, 2013-06-03 23:36:02]

Committed patchset #2 manually as r151695 (presubmit successful).