Make sure api interceptors don't change the store target w/o storing

Make sure api interceptors don't change the store target w/o storing

BUG=chromium:619166
R=verwaest@chromium.org

Committed: https://crrev.com/2c8ca9ad09281d4138ae363566051e45afd0838c
Cr-Commit-Position: refs/heads/master@{#37152}

[Toon Verwaest, 2016-06-20 14:48:33]

lgtm with a question:

Shouldn't we verify access on the store target? If this happens for
window.location, the store target is window.location itself, rather than a
"hidden prototype". Shouldn't we verify access again in that case?

[jochen (gone - plz use gerrit), 2016-06-20 15:08:59]

On 2016/06/20 at 14:48:33, verwaest wrote:
> lgtm with a question:
> 
> Shouldn't we verify access on the store target? If this happens for
window.location, the store target is window.location itself, rather than a
"hidden prototype". Shouldn't we verify access again in that case?

I could just do the API check for both cases of the branch, wdyt?

[jochen (gone - plz use gerrit), 2016-06-20 15:17:36]

like this

[Toon Verwaest, 2016-06-20 18:34:42]

You'd have to cache verify the map of the store-target instead in that case. The
actual object stays the same I presume.

[jochen (gone - plz use gerrit), 2016-06-21 08:29:53]

ptal

[Toon Verwaest, 2016-06-21 13:09:24]

https://codereview.chromium.org/2082633002/diff/40001/src/objects.cc
File src/objects.cc (right):

https://codereview.chromium.org/2082633002/diff/40001/src/objects.cc#newcode4334
src/objects.cc:4334: Utils::ApiCheck(*store_target == *it->GetStoreTarget(),
Don't you want to do exactly the same check in both cases?

[jochen (gone - plz use gerrit), 2016-06-21 14:44:26]

ptal

[Toon Verwaest, 2016-06-21 14:48:15]

Sorry, yet another comment :)

https://codereview.chromium.org/2082633002/diff/60001/src/objects.cc
File src/objects.cc (right):

https://codereview.chromium.org/2082633002/diff/60001/src/objects.cc#newcode4327
src/objects.cc:4327: if (maybe_attributes.FromJust() == ABSENT) break;
You break out here already if it's absent. The check should be above that break?

[jochen (gone - plz use gerrit), 2016-06-21 17:01:31]

On 2016/06/21 at 14:48:15, verwaest wrote:
> Sorry, yet another comment :)
> 
> https://codereview.chromium.org/2082633002/diff/60001/src/objects.cc
> File src/objects.cc (right):
> 
>
https://codereview.chromium.org/2082633002/diff/60001/src/objects.cc#newcode4327
> src/objects.cc:4327: if (maybe_attributes.FromJust() == ABSENT) break;
> You break out here already if it's absent. The check should be above that
break?

updated

[Toon Verwaest, 2016-06-21 18:01:07]

lgtm ;)

[jochen (gone - plz use gerrit), 2016-06-21 18:05:51]

The CQ bit was checked by jochen@chromium.org

[commit-bot: I haz the power, 2016-06-21 18:05:59]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2082633002/80001

[commit-bot: I haz the power, 2016-06-21 18:08:05]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2016-06-21 18:09:59]

Description was changed from

==========
Make sure api interceptors don't change the store target w/o storing

BUG=chromium:619166
R=verwaest@chromium.org
==========

to

==========
Make sure api interceptors don't change the store target w/o storing

BUG=chromium:619166
R=verwaest@chromium.org

Committed: https://crrev.com/2c8ca9ad09281d4138ae363566051e45afd0838c
Cr-Commit-Position: refs/heads/master@{#37152}
==========

[commit-bot: I haz the power, 2016-06-21 18:10:00]

Patchset 5 (id:??) landed as
https://crrev.com/2c8ca9ad09281d4138ae363566051e45afd0838c
Cr-Commit-Position: refs/heads/master@{#37152}