We need to rev the IID for nsINavHistoryResultNode which was missed in bug 561450. If this will have no impact to binary add-ons or the web, please note that. +++ This bug was initially created as a clone of Bug #561450 +++ Both History.cpp and nsNavHistory.cpp use the synchronous GetNewSessionId(). If session ID counter hasn't been initialized a sync SQL query is done to fetch the biggest ID, which is incremented by 1.

I don't see a reason to rev the uuid there, touching constants doesn't require that, a binary component would just not compile, but there's no risk of miscalls (since constants don't change the vtable). Though, if you want to do that regardless for automated tools, I can fix it.

(In reply to Marco Bonardo [:mak] from comment #1) > I don't see a reason to rev the uuid there, touching constants doesn't > require that, a binary component would just not compile, but there's no risk > of miscalls (since constants don't change the vtable). > Though, if you want to do that regardless for automated tools, I can fix it. Sounds good - we can resolve/invalid if there's no add-on/web compatibility risk here.

I concur, this doesn't affect binary or JS compatibility.