This function tends to be a total footgun, and people generally want JS_GetGlobalForScopeChain. There may still end up being a few consumers of this, but we should rename the resulting function to JS_GetDefaultGlobal or somesuch.

Yes, please. IIUC, this takes us closer to bug 604813 and the bright and shiny future of JSContext-per-JSRuntime (which would then allow us to merge the two into one thing), right?

(In reply to Luke Wagner [:luke] from comment #1) > Yes, please. IIUC, this takes us closer to bug 604813 and the bright and > shiny future of JSContext-per-JSRuntime (which would then allow us to merge > the two into one thing), right? Yes, though as discussed on IRC, there will still be some consumers of that stuff for a while.

Here's the first batch of work on this: https://tbpl.mozilla.org/?tree=Try&rev=b0d3e16ed3f3

This one is easy to infer, because we subsequently call JS_CallFunctionValue, which asserts that cx, obj, and fval are all same-compartment. So assuming this code doesn't compartment mismatch right now, this should be equivalent.

The primary consumer of this is the whole inner/outer DOM window setup, which uses the default global to track the current inner. But there are few other random ones as well. We use this as an opportunity to convert a bunch of consumers from the two-step GetNativeContext() -> JS_GetGlobalObject() into just |GetNativeGlobal()|. This will make things much easier to convert when we start tracking the current inner explicitly.

\o/

remote: https://hg.mozilla.org/integration/mozilla-inbound/rev/ccddc9279bae remote: https://hg.mozilla.org/integration/mozilla-inbound/rev/bc3d298479a5 remote: https://hg.mozilla.org/integration/mozilla-inbound/rev/4fc957c28503

https://hg.mozilla.org/mozilla-central/rev/ccddc9279bae https://hg.mozilla.org/mozilla-central/rev/bc3d298479a5 https://hg.mozilla.org/mozilla-central/rev/4fc957c28503

(In reply to Bobby Holley (:bholley) (PTO thurs/fri) from comment #13) > remote: https://hg.mozilla.org/integration/mozilla-inbound/rev/bc3d298479a5 There's a bunch of cx -> mContext changes that weren't in the reviewed patch.

(In reply to :Ms2ger from comment #15) > (In reply to Bobby Holley (:bholley) (PTO thurs/fri) from comment #13) > > remote: https://hg.mozilla.org/integration/mozilla-inbound/rev/bc3d298479a5 > > There's a bunch of cx -> mContext changes that weren't in the reviewed patch. Hm, not sure how that happened - probably happened while dealing with merge conflicts with bug 868130. Anyway, those should all be ok - good catch though!