Prevent ObjectBackedNativeHandler from calling itself once invalid.

Prevent ObjectBackedNativeHandler from calling itself once invalid.
Recommitting: added pesky HandleScopes.

BUG=181668
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=187995

[koz (OOO until 15th September), 2013-03-13 23:08:42]

This LGTM, but you might want to get someone more familiar with v8::Persistent
stuff to make sure you're freeing those objects correctly.

https://codereview.chromium.org/12693008/diff/2001/chrome/renderer/extensions...
File chrome/renderer/extensions/object_backed_native_handler.cc (right):

https://codereview.chromium.org/12693008/diff/2001/chrome/renderer/extensions...
chrome/renderer/extensions/object_backed_native_handler.cc:40: if
(!data->GetHiddenValue(v8::String::New(kIsValid))->BooleanValue()) {
Another option would be to add an is_valid field to RouterData itself, rather
than replacing it with a v8::Object...

Ah, but that would defeat the whole purpose here because RouterDatas are owned
by the ObjectBackedNativeHandler, which gets destroyed and then called into.
Right.

It might be good to add an explanatory comment to the class explaining the
motivations for this design.

[not at google - send to devlin, 2013-03-13 23:45:47]

Committed patchset #3 manually as r187975 (presubmit successful).

[not at google - send to devlin, 2013-03-13 23:49:25]

Also changed GetHiddenValue to just Get, silly me.

https://codereview.chromium.org/12693008/diff/2001/chrome/renderer/extensions...
File chrome/renderer/extensions/object_backed_native_handler.cc (right):

https://codereview.chromium.org/12693008/diff/2001/chrome/renderer/extensions...
chrome/renderer/extensions/object_backed_native_handler.cc:40: if
(!data->GetHiddenValue(v8::String::New(kIsValid))->BooleanValue()) {
On 2013/03/13 23:08:43, koz wrote:
> Another option would be to add an is_valid field to RouterData itself, rather
> than replacing it with a v8::Object...
> 
> Ah, but that would defeat the whole purpose here because RouterDatas are owned
> by the ObjectBackedNativeHandler, which gets destroyed and then called into.
> Right.
> 
> It might be good to add an explanatory comment to the class explaining the
> motivations for this design.

Done.

[not at google - send to devlin, 2013-03-14 01:11:39]

Committed patchset #6 manually as r187981.

[not at google - send to devlin, 2013-03-14 03:40:31]

Committed patchset #8 manually as r187995.