The test I added in bug 642338 relies on a Firefox-specific pref, that you can't hide the location bar. Since this is a core test, it shouldn't rely on this pref being set.
Wrong bug ID.
Bug 642338 comment 37:
Serge Gautherie (:sgautherie) 2011-05-31 17:07:04 PDT
> If not, I'll review a patch to modify the test so it checks the pref before
> the is() call.
To be explicit, is() check is fine in both cases, but its expected value should simply depend on "dom.disable_window_open_feature.location"(!?) preference value.
More details are in bug 653996.
So just to be clear, I *can* rely on a pref in all.js?
> pref("dom.disable_window_open_feature.status", true);
Created attachment 537580 [details] [diff] [review]
> So just to be clear, I *can* rely on a pref in all.js?
Actually, nevermind. That's not necessary.
Comment on attachment 537580 [details] [diff] [review]
[Mozilla/5.0 (Windows NT 5.0; rv:7.0a1) Gecko/20110605 Firefox/7.0a1 SeaMonkey/2.2a1pre] (nightly, SM 2.4a1pre !)
I like this code. And it fixes bug 653996 :-)
From my bug 653996 comment 3:
Nit: you might want to add "from HTML" (or "content" or whatever applies) to the comment, "ftr".
Created attachment 537657 [details] [diff] [review]
Comment on attachment 537657 [details] [diff] [review]
I meant to ask for Smaug's review on this patch, but I apparently missed that.