See https://bugzilla.mozilla.org/show_bug.cgi?id=824717#c4 We also need this for the MMI implementation.

This blocks a leo+ bug so leo+.

I still want to test it with a real use case like bug 883178 before asking for r? but I could use some feedback in the meantime :)

Comment on attachment 766606 [details] [diff] [review] v1 Review of attachment 766606 [details] [diff] [review]: ----------------------------------------------------------------- ::: dom/base/test/test_domrequest.html @@ +66,5 @@ > +var _error = new DOMError("detailedError"); > +ok(true, _error); > +ok(true, reqserv); > +ok(true, reqserv.fireError); > +ok(true, reqserv.fireDetailedError); Ignore lines 67-70, please. They were there just for debugging purposes.

Comment on attachment 766606 [details] [diff] [review] v1 Clearing the feedback flag as this actually doesn't work in a real scenario. I get: JavaScript Error: "Error: Permission denied to access property 'valueOf'" when trying to access DOMError properties from content.

Naive question: can't you simply add a .fireError() function that takes a nsIDOMDOMError?

AFAIK there is no nsIDOMDOMError in m-c anymore

(In reply to Fernando Jiménez Moreno [:ferjm] (needinfo instead of CC, please) from comment #6) > AFAIK there is no nsIDOMDOMError in m-c anymore That sounds like a problem to me but that was not the point of my comment. You could add .fireError() that takes a nsISupports* then. The point being that there is no need to have a new function name, we are just going to add a new signature.

(In reply to Mounir Lamouri (:mounir) from comment #8) > (In reply to Fernando Jiménez Moreno [:ferjm] (needinfo instead of CC, > please) from comment #6) > > AFAIK there is no nsIDOMDOMError in m-c anymore > > That sounds like a problem to me but that was not the point of my comment. > You could add .fireError() that takes a nsISupports* then. The point being > that there is no need to have a new function name, we are just going to add > a new signature. Oh, I see. I was just following Jonas' suggestion at https://bugzilla.mozilla.org/show_bug.cgi?id=824717#c4 I can overload .fireError I guess

I'm fine either way. I'm not a huge fan of overloads though, they work poorly with the typelessness of JavaScript.

Though, most importantly, this is a class that's going to go away as we move to using Promises instead of DOMRequests, so we should do whatever is easiest.

Comment on attachment 767134 [details] [diff] [review] v1 (m-c) We agreed via IRC to keep it simple and take the fireDetailedError solution.

I'll have a b2g18 patch tomorrow (CST)

Thanks Jonas! http://hg.mozilla.org/projects/birch/rev/105681e909e5

Backed out from b2g18 due to chrome tests failure. https://hg.mozilla.org/releases/mozilla-b2g18/rev/2d9072f92b52 I won't be able to fix it until tomorrow as I am taking a 20 hours flight.

Re-pushed to b2g18: https://hg.mozilla.org/releases/mozilla-b2g18/rev/b5e2fec86fc4