656810 - Implement js::NumberObject

Status: RESOLVED FIXED

(Core :: JavaScript Engine, defect)

Description

[Jeff Walden [:Waldo]]

Attachment: Patch

As object subclasses go it's on the low end of the importance scale, but we should have it to get away from the getPrimitiveThis/setPrimitiveThis punning.

Comment 1

[Nicholas Nethercote [inactive]]

Comment on attachment 532079 [details] [diff] [review]
Patch

Review of attachment 532079 [details] [diff] [review]:
-----------------------------------------------------------------

r=me

::: js/src/vm/NumberObject.h
@@ +69,5 @@
> +    }
> +
> +  private:
> +    inline void init(jsdouble d) {
> +        setSlot(PRIMITIVE_THIS_SLOT, NumberValue(d));

Is PRIMITIVE_THIS_SLOT the best name now?  Would something simpler like VALUE_SLOT be better?  And if you make that change, would setValue() be a better name than init()?

@@ +78,5 @@
> +    ::js_InitNumberClass(JSContext *cx, JSObject *global);
> +
> +  private:
> +    NumberObject();
> +    NumberObject &operator=(const NumberObject &so);

Add a comment here, please, explaining these are declared private so they can't be used.  (And maybe why you don't want them to be used.)

Comment 2

[Luke Wagner [:luke]]

5 out 5 JS engineers just surveyed agree that private cctor/op= is a basic C++ idiom and not deserving of comment.

Comment 3

[Jeff Walden [:Waldo]]

(In reply to comment #1)
> Is PRIMITIVE_THIS_SLOT the best name now?  Would something simpler
> like VALUE_SLOT be better?  And if you make that change, would
> setValue() be a better name than init()?

I had thought I was mimicking the name of the spec internal property, but that was actually [[PrimitiveValue]], so I renamed to PRIMITIVE_VALUE_SLOT and setPrimitiveValue.

> @@ +78,5 @@
> > +    ::js_InitNumberClass(JSContext *cx, JSObject *global);
> > +
> > +  private:
> > +    NumberObject();
> > +    NumberObject &operator=(const NumberObject &so);
> 
> Add a comment here, please, explaining these are declared private so
> they can't be used.  (And maybe why you don't want them to be used.)

A poll of the office says private constructor/operator= are idiomatic for "don't do that", so I'm leaving this as-is.  It doesn't seem important to clarify why, but anyone who's truly curious shouldn't have to think much past "this is GC-managed" to explain.

This is atop a bunch of other patches, so it'll be awhile before it lands, I think (but for 6, I believe).

Comment 4

[Jeff Walden [:Waldo]]

...and four and a half months later, those other patches have made their way through reviews and such into the tree -- landed:

http://hg.mozilla.org/integration/mozilla-inbound/rev/66736ca9c8c3

Comment 5

[Ed Morley [:emorley]]

https://hg.mozilla.org/mozilla-central/rev/66736ca9c8c3