From: Christiaan H. <cmh...@gm...> - 2008-01-06 22:06:17
|
On 6 Jan 2008, at 10:39 PM, Adam R. Maxwell wrote: > > On Jan 6, 2008, at 1:33 PM, Christiaan Hofman wrote: > >> >> On 6 Jan 2008, at 10:30 PM, Adam R. Maxwell wrote: >> >>> >>> On Jan 6, 2008, at 7:10 AM, Adam R. Maxwell wrote: >>> >>>> >>>> On Jan 6, 2008, at 5:06 AM, Christiaan Hofman wrote: >>>> >>>>> Adam, can you confirm the last remark in bug # 1830798, e.g. by >>>>> commenting the window autosavename and/or the side pane sizing in >>>>> [SKMainWindowController windowDidLoad]? Perhaps the order of >>>>> setting >>>>> PDF settings / changing window size / loading PDF is important? >>>> >>>> OK, I'll check that later today. >>> >>> Seems to work okay on the MacBook Pro with that commented, but I >>> need >>> to try on the G5 at work since that system fails pretty reliably. >>> I'm >>> pretty sure I've tried changing most of that stuff and reordering >>> it. >> >> And of course it should not matter, but apparently it does. Also it's >> not an option to remove it. > > I was wondering if it might be a problem with a nil message with a > float or struct return value? Maybe setting a view size to a negative > value or something. I wish I'd reported it to Apple during Leopard > seeding, but there was a more serious bug that kept it from even > opening (which wasn't fixed until 10.5.0). You mean in our code or internal? I cannot imagine this to happen in our code. I don't think a view cab ever get negative size. E.g. the window frame in the prefs should always be consistent with the side pane widths. Christiaan |