1122758 - Get rid of XPCJSObjectHolder::newHolder()

Status: RESOLVED FIXED

(Core :: XPConnect, defect)

Description

[Andrew McCreight [:mccr8]]

This is a weird function.  I think it just makes sense to inline it and make the ctor public.

Comment 1

[Andrew McCreight [:mccr8]]

Attachment: Get rid of XPCJSObjectHolder::newHolder().

I'm not entirely sure this is a good idea, but it seems like an improvement to me.

try run: https://treeherder.mozilla.org/#/jobs?repo=try&revision=c1f3b868c1f6

Comment 2

[Gabor Krizsanits (INACTIVE)]

Comment on attachment 8554621 [details] [diff] [review]
Get rid of XPCJSObjectHolder::newHolder().

Review of attachment 8554621 [details] [diff] [review]:
-----------------------------------------------------------------

I think it's definitely an improvement, and sorry about the delay I was having a holiday.

::: js/xpconnect/src/nsXPConnect.cpp
@@ +823,5 @@
>  
>      if (NS_SUCCEEDED(rv) && !rval.isPrimitive()) {
> +        JSObject *obj = rval.toObjectOrNull();
> +        if (!obj)
> +            return NS_ERROR_FAILURE;

I think the !rval.isPrimitive() ensures that rval is a valid object, so why do we need this check here?

Comment 3

[Andrew McCreight [:mccr8]]

(In reply to Gabor Krizsanits [:krizsa :gabor] from comment #2)
> I think the !rval.isPrimitive() ensures that rval is a valid object, so why
> do we need this check here?

Good point.

try run with that check removed: https://treeherder.mozilla.org/#/jobs?repo=try&revision=f78fec40d632

Comment 4

[Andrew McCreight [:mccr8]]

I removed the redundant check.

https://hg.mozilla.org/integration/mozilla-inbound/rev/40aeb4e4292c

Comment 5

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/40aeb4e4292c