Features are function names if they have a parent feature.

Features are function names if they have a parent feature.

There's no need to check whether the parent is accessible to the current context.

BUG=305000
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=242154

[Jeffrey Yasskin, 2013-12-18 02:05:14]

Actually review patch set 1. It's based on
https://codereview.chromium.org/109683007, so I couldn't 'try' it directly.

[not at google - send to devlin, 2013-12-18 02:15:21]

https://codereview.chromium.org/117843003/diff/1/chrome/renderer/extensions/d...
File chrome/renderer/extensions/dispatcher.cc (right):

https://codereview.chromium.org/117843003/diff/1/chrome/renderer/extensions/d...
chrome/renderer/extensions/dispatcher.cc:754: // then this must be a function or
event, so we should not register.
yak shave:

a check which would make me feel more confident about this change is that we
don't load any bindings until after the context has been initiated. this is
actually what prompted this noparent stuff: we had lazy bindings for "app" and
"app.window", and when setting up the "app.window" binding it would
inadvertently load the "app" binding, which broke things.

I would even go so far as to say that no lazily initialised code should ever be
run during DidCreateScriptContext, and you could implement that similar to
ModuleSystem::NativesAllowedScope (or whatever it's called).

[not at google - send to devlin, 2013-12-18 02:15:45]

> initiated.

initialized.

[not at google - send to devlin, 2013-12-18 02:40:16]

lgtm

[commit-bot: I haz the power, 2013-12-20 17:29:51]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jyasskin@chromium.org/117843003/40001

[commit-bot: I haz the power, 2013-12-20 20:00:53]

Change committed as 242154