Actually scratch (b) from my suggestions, I reckon everyone using this feature would expect (a) to happen.
See also bug 221445. This might get fixed there.
I'd rather see behavior (b) in this case since it might not be obvious to the user that his/her homepage(s) were loaded which can be confusing if a web page suddenly spawns several tabs.
(In reply to comment #3) > I'd rather see behavior (b) in this case since it might not be obvious to the > user that his/her homepage(s) were loaded which can be confusing if a web page > suddenly spawns several tabs. > In that case, would strtok solve this issue? I don't have a compiler so I can't make or test a patch myself...
As long bug 221445 doesn't change the separator from the pipe to some character which isn't valid in URLs (e.g. a space), splitting up the homepage is no option. The apparently mis-behaving code is also used by other applications (e.g. SeaMonkey) which allow pipes as regular characters. Breaking these just because some Firefox devs once made an unlucky decision isn't likely to be happening.
I'm going to guess that the reason this hasn't been fixed yet is because there hasn't been any agreement on what to make the home method do. I myself would like it to only open the first homepage in the tab, but if there could also be a preference to the behavior of the method, along with the home button. Only problem is that might confuse some people by making them think that if they set it to only open the first homepage, the browser will only open one page at start-up.
*** Bug 349414 has been marked as a duplicate of this bug. ***
*** Bug 347945 has been marked as a duplicate of this bug. ***
*** Bug 354678 has been marked as a duplicate of this bug. ***
(In reply to comment #11) > I guess I'm not the only > one who stumbles over this bug just because of the whatsnew-page right after > updating. Yes, you are not. I just upgraded to Firefox 184.108.40.206 and found the bug. window.home() is used here: http://www.mozilla.com/en-US/firefox/updated This is a PRIVACY ISSUE as it reveals to the server of my first home page the URLs of my others pages.
Marking blocking due to privacy issue. At the very least we could simply grab the first url and use that when window.home() is called
Created attachment 254589 [details] [diff] [review] Fix the privacy issue. This solves the privacy issue involved here, but doesn't actually fix the problem. I don't think we can, or even want to, fix the actual problem here before bug 221445 is fixed. This is based on Michael Ventnor's idea of using MOZ_PHOENIX to determine whether we need to do this or not.
Josh, is this something that we need to worry about for Camino as well?
Thanks Johnny, I had no idea NSStrings had those functions which is probably why I had so much trouble fixing it myself. :/ This looks like a nice, trivial bandaid for the 1.8 branch. Is it a good idea to request approval220.127.116.11?
Yes, given that this is a potential privacy problem, absolutely!
smorgan - is this an issue for Camino?
(In reply to comment #21) > smorgan - is this an issue for Camino? It shouldn't be, since Camino doesn't currently support multiple home pages, and isn't going to use Firefox's display method when we do. See also bug 189930 for how we're planning to do this.
Comment on attachment 254589 [details] [diff] [review] Fix the privacy issue. sr=dveditz
Privacy issue fixed on the trunk, leaving bug open to track the remaining problem that won't be fixed until bug 221445 is fixed.
Is there a reason why this is targetted for approval on 18.104.22.168 and not .0.3?
(In reply to comment #26) > Is there a reason why this is targetted for approval on 22.214.171.124 and not .0.3? The approval126.96.36.199 flag was renamed to approval188.8.131.52 in preparation for a quicker than usual 184.108.40.206 release for critical bug fixes only.
Comment on attachment 254589 [details] [diff] [review] Fix the privacy issue. approved for 220.127.116.11 and 18.104.22.168, a=dveditz for drivers
Unmarking the blocking1.9 flag here as the privacy issue here is now fixed. The reminder is not a blocker.