Support named query generation

Support named query generation

This patch support generation of named query
for IDLs with named enumerator.

Using the feature, following custom implementation are removed.
- V8DOMStringMap::namedPropertyQuery
- V8Storage::namedPropertyQuery

R=haraken@chromium.org
BUG=229740
TEST=TestEventTarget.idl
     TestInterface.idl
     TestObject.idl

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=152534

[kojih, 2013-06-17 07:27:34]

r?

Diff of generated code:

+void V8DOMStringMap::namedPropertyQuery(v8::Local<v8::String> name, const
v8::PropertyCallbackInfo<v8::Integer>& info)
+{
+    DOMStringMap* collection = toNative(info.Holder());
+    AtomicString propertyName = toWebCoreAtomicString(name);
+    ExceptionCode ec = 0;
+    int returnValue = 0;
+    bool result = collection->namedPropertyQuery(propertyName, returnValue,
ec);
+    if (ec) {
+        setDOMException(ec, info.GetIsolate());
+        return;
+    }
+    if (!result)
+        return;
+    v8SetReturnValueInt(info, 0);
+}
+
 
+void V8Storage::namedPropertyQuery(v8::Local<v8::String> name, const
v8::PropertyCallbackInfo<v8::Integer>& info)
+{
+    Storage* collection = toNative(info.Holder());
+    AtomicString propertyName = toWebCoreAtomicString(name);
+    ExceptionCode ec = 0;
+    int returnValue = 0;
+    bool result = collection->namedPropertyQuery(propertyName, returnValue,
ec);
+    if (ec) {
+        setDOMException(ec, info.GetIsolate());
+        return;
+    }
+    if (!result)
+        return;
+    v8SetReturnValueInt(info, 0);
+}
+

[haraken, 2013-06-17 07:41:02]

LGTM

https://codereview.chromium.org/17225003/diff/1/Source/bindings/scripts/CodeG...
File Source/bindings/scripts/CodeGeneratorV8.pm (right):

https://codereview.chromium.org/17225003/diff/1/Source/bindings/scripts/CodeG...
Source/bindings/scripts/CodeGeneratorV8.pm:3667: v8SetReturnValueInt(info, 0);

0 => returnValue.

However, I might want to remove 'returnValue' from the namedPropertyQuery()
parameters and hard-code 'v8::None' here for the following reasons:

- Per the comment in v8.h, we should use v8::None instead of 0.
- However, we might not want to write v8::None in core/ (since whether we should
expose v8:: to core/ is controvertial).
- At this point, we cannot imagine any case where we want to return other than
0.

https://codereview.chromium.org/17225003/diff/1/Source/core/storage/Storage.cpp
File Source/core/storage/Storage.cpp (right):

https://codereview.chromium.org/17225003/diff/1/Source/core/storage/Storage.c...
Source/core/storage/Storage.cpp:120: if (name == "length")

It's OK to leave this check in this CL, but this check should be replaced with
GetRealNamedPropertyInPrototypeChain() or something like that, shouldn't it? It
looks strange to hard-code "length" and treat it specially.

[kojih, 2013-06-17 07:52:57]

https://codereview.chromium.org/17225003/diff/1/Source/bindings/scripts/CodeG...
File Source/bindings/scripts/CodeGeneratorV8.pm (right):

https://codereview.chromium.org/17225003/diff/1/Source/bindings/scripts/CodeG...
Source/bindings/scripts/CodeGeneratorV8.pm:3667: v8SetReturnValueInt(info, 0);
OK.

https://codereview.chromium.org/17225003/diff/1/Source/core/storage/Storage.cpp
File Source/core/storage/Storage.cpp (right):

https://codereview.chromium.org/17225003/diff/1/Source/core/storage/Storage.c...
Source/core/storage/Storage.cpp:120: if (name == "length")
I agree.

[commit-bot: I haz the power, 2013-06-17 08:34:21]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/kojih@chromium.org/17225003/2002

[commit-bot: I haz the power, 2013-06-17 09:48:05]

Change committed as 152534