Allow JavaScript accessors on API objects.

Allow JavaScript accessors on API objects.

(This is somewhat experimental; hence protected by #ifdef.)

R=epertoso@chromium.org, jochen@chromium.org
BUG=chromium:508898
LOG=N

Committed: https://crrev.com/ed110a1715e245a93fd8be60cc9edac06ba11887
Cr-Commit-Position: refs/heads/master@{#31002}

[vogelheim, 2015-09-24 11:29:47]

Please have a look, both of you. This doesn't do much, but I'd guess it's clean
enough to actually commit.

Two notes:

1, in api-accessors.cc, we have all the information we need to 'install' the
compatible receiver check. For Cpp accessors, CreateApiFunction called from
InstantiateFunction does that. So if we agree on *how* to do that, that would be
the right place.

2, Next, I'd then try to fiddle with the bindings to make use of this in some
reasonable form - also behind #ifdef - so Enrico can proceed with the
measurement.

[epertoso, 2015-09-24 12:09:06]

https://codereview.chromium.org/1367953002/diff/20001/src/api-natives.h
File src/api-natives.h (right):

https://codereview.chromium.org/1367953002/diff/20001/src/api-natives.h#newco...
src/api-natives.h:54: static void AddAccessorProperty(Isolate* isolate,
Handle<TemplateInfo> info,
These two overloads of AddAccessorProperty are internal only. Since down the
path we throw away the type-safety anyway, and their code is essentially the
same, how about just having one variant with type Handle<Object> for the getter
and setter arguments?

[vogelheim, 2015-09-24 16:35:49]

https://codereview.chromium.org/1367953002/diff/20001/src/api-natives.h
File src/api-natives.h (right):

https://codereview.chromium.org/1367953002/diff/20001/src/api-natives.h#newco...
src/api-natives.h:54: static void AddAccessorProperty(Isolate* isolate,
Handle<TemplateInfo> info,
On 2015/09/24 12:09:06, epertoso wrote:
> These two overloads of AddAccessorProperty are internal only. Since down the
> path we throw away the type-safety anyway, and their code is essentially the
> same, how about just having one variant with type Handle<Object> for the
getter
> and setter arguments?

Done.

[jochen (gone - plz use gerrit), 2015-09-25 09:20:51]

https://codereview.chromium.org/1367953002/diff/20001/src/api.cc
File src/api.cc (right):

https://codereview.chromium.org/1367953002/diff/20001/src/api.cc#newcode965
src/api.cc:965: DCHECK_EQ(v8::DEFAULT, access_control);
Why not just remove the parameter?

[jochen (gone - plz use gerrit), 2015-09-25 09:24:11]

https://codereview.chromium.org/1367953002/diff/40001/test/cctest/test-api-ac...
File test/cctest/test-api-accessors.cc (right):

https://codereview.chromium.org/1367953002/diff/40001/test/cctest/test-api-ac...
test/cctest/test-api-accessors.cc:26: // OF THIS SOFTWARE, EVEN IF ADVISED OF
THE POSSIBILITY OF SUCH DAMAGE.
plz use the shorter header

https://codereview.chromium.org/1367953002/diff/40001/test/cctest/test-api-ac...
test/cctest/test-api-accessors.cc:28: 
can you add

// TODO(jochen): Remove this after the setting is turned on globally.
#define V8_IMMINENT_DEPRECATION_WARNINGS

[vogelheim, 2015-09-25 12:59:24]

https://codereview.chromium.org/1367953002/diff/40001/test/cctest/test-api-ac...
File test/cctest/test-api-accessors.cc (right):

https://codereview.chromium.org/1367953002/diff/40001/test/cctest/test-api-ac...
test/cctest/test-api-accessors.cc:26: // OF THIS SOFTWARE, EVEN IF ADVISED OF
THE POSSIBILITY OF SUCH DAMAGE.
On 2015/09/25 09:24:11, jochen wrote:
> plz use the shorter header

Done.

https://codereview.chromium.org/1367953002/diff/40001/test/cctest/test-api-ac...
test/cctest/test-api-accessors.cc:28: 
On 2015/09/25 09:24:11, jochen wrote:
> can you add
> 
> // TODO(jochen): Remove this after the setting is turned on globally.
> #define V8_IMMINENT_DEPRECATION_WARNINGS

Done.

[vogelheim, 2015-09-25 12:59:45]

https://codereview.chromium.org/1367953002/diff/20001/src/api.cc
File src/api.cc (right):

https://codereview.chromium.org/1367953002/diff/20001/src/api.cc#newcode965
src/api.cc:965: DCHECK_EQ(v8::DEFAULT, access_control);
On 2015/09/25 09:20:51, jochen wrote:
> Why not just remove the parameter?

Done.

[epertoso, 2015-09-28 08:04:47]

lgtm

[jochen (gone - plz use gerrit), 2015-09-28 11:33:51]

lgtm

[vogelheim, 2015-09-29 10:51:38]

The CQ bit was checked by vogelheim@chromium.org

[commit-bot: I haz the power, 2015-09-29 10:51:51]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1367953002/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1367953002/60001

[commit-bot: I haz the power, 2015-09-29 11:16:06]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2015-09-29 11:16:28]

Patchset 4 (id:??) landed as
https://crrev.com/ed110a1715e245a93fd8be60cc9edac06ba11887
Cr-Commit-Position: refs/heads/master@{#31002}