Re: Odd error logged when coming back from 'Browse All Versions'
Re: Odd error logged when coming back from 'Browse All Versions'
- Subject: Re: Odd error logged when coming back from 'Browse All Versions'
- From: Graham Cox <email@hidden>
- Date: Thu, 25 Aug 2011 12:24:53 +1000
On 25/08/2011, at 12:03 PM, Quincey Morris wrote:
> As to why this isn't automatic, the answer is hinted at, earlier in the video. The versions browser doesn't know what parts of your user interface are necessary for interacting with the document window during version browsing. It could just as easily turn out that you *do* want a palette shown during browsing. It all depends on the function of the palette. Currently, it appears, palette windows are hidden by default on entry to the mode, and not restored on exit, and there's a hint that this behavior is not set in stone.
While I can't imagine how a palette would come into version browsing, I will accept that there's a use case beyond my imagination. Fair enough.
What I do find hard to comprehend is how the use case of leaving them all hidden all by default is reasonable. The system takes care of hiding all and restoring all on a suspend/resume, on the basis that the app's state hasn't really changed. Surely that's true for version browsing as well? Sure, the document itself may have changed its internal state, but the app state has not - you've reverted to an earlier version of the document or not, but what you're doing with that document is the same.
> Adding two 1-line delegate methods in order to make your intentions explicit doesn't seem unnecessarily onerous. I *think* what you're actually discovering is that *after* taking a run at implementing various new 10.7 features, it's worthwhile to watch the videos again. At that point, you're going to be primed to absorb how to deal with the wiggly bits around the edges, which are to a great extent what the WWDC sessions are about.
It's not that onerous, just surprising. It means that to adopt the new document features, which <appears> to be a case of returning a simple YES/NO switch, your app suddenly behaves in a way that seems broken without additional work.
The additional work would be eased if the app delegate received some sort of global notification about entering and leaving the version browsing mode, but it doesn't - only the document window does, meaning you have to punt the notification up to a global app level within your app to act upon it (in my case, not untypical, my inspectors are global objects which are sensitive to any document context). So it's not just a 2-line override, it amounts to quite a bit more than that. Another alternative would be that the palettes themselves received a notification about entering/leaving this mode. It just feels that it hasn't been fully thought through.
I will take a look at the sessions.
--Graham
_______________________________________________
Cocoa-dev mailing list (email@hidden)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden