• Open Menu Close Menu
  • Apple
  • Shopping Bag
  • Apple
  • Mac
  • iPad
  • iPhone
  • Watch
  • TV
  • Music
  • Support
  • Search apple.com
  • Shopping Bag

Lists

Open Menu Close Menu
  • Terms and Conditions
  • Lists hosted on this site
  • Email the Postmaster
  • Tips for posting to public mailing lists
Re: From Panther to Tiger
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: From Panther to Tiger


  • Subject: Re: From Panther to Tiger
  • From: Roy Lovejoy <email@hidden>
  • Date: Mon, 16 May 2005 13:01:50 -0700


On May 16, 2005, at 12:14 PM, Matt Neuburg wrote:

My anecdotal experience (i.e. guesswork) suggests a surprising
amount of Tiger-caused breakage to existing apps. This is a developer list
and developers make apps, so one expects that not a few people on this list
had their apps break due to Tiger, and I think that a lot of us would
benefit from hearing how. That's what I took the OP to be asking for and
it's something I'd be interested in as well. m.




I can add one really odd thing..

I am maintaining a Carbon application whose history goes back to pascal code circa 1985..
They were using third party source that handled floating palette z-ordering (legacy code from before Mac Window Manager handled it)..


Introduced in Tiger, and _never_ before occurring, if you call GetWindowList(), the first window in the list will actually be the menu bar.

The code base was relying on a third party library with an erroneous 'IsWindowActive()' call that basically compared a WindowPtr in question to the first non-floating palette in the GetWindowList() chain..

In Tiger's case, it was this new 'Menubar window' (my guess it's an HIView)..

So for 21 years, this code was kosher, albeit crufty. In Tiger, this new behavior caused *ALL* of the mouse clicks in *ALL* the windows to be ignored, because the app was in constant state of "the user clicked in a non-foreground window, Select it, then the next click will be valid"...

Now this isn't a bug on Tiger's part by any means.. There are many new (within last 10 year) API's that do the job better and would have avoided this bug, but it is an instance where there was a radical shift in api definition that caused the breaking of an App that was by most accounts 'stable'..

(and yes, I pointed out to the company that Tiger didn't 'break' the app, the company's lack of keeping up, at least in the right decade of API usage broke the app.)

but it was unexpected none-the-less.

_______________________________________________
Do not post admin requests to the list. They will be ignored.
Cocoa-dev mailing list      (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden


  • Follow-Ups:
    • Re: From Panther to Tiger
      • From: John Stiles <email@hidden>
References: 
 >Re: From Panther to Tiger (From: Matt Neuburg <email@hidden>)

  • Prev by Date: Re: From Panther to Tiger
  • Next by Date: Re: FSCopyObjectAsync: useless and crippled
  • Previous by thread: Re: From Panther to Tiger
  • Next by thread: Re: From Panther to Tiger
  • Index(es):
    • Date
    • Thread