From: Adam R. M. <ama...@ma...> - 2005-01-29 16:34:12
|
On Jan 29, 2005, at 02:22, Christiaan Hofman wrote: > I am not very happy that it is still possible to have an invalid > cite-key format, which could give rise to errors in the parser. I do > not see a good way to avoid it. But AFAIKS, the only way to set an > invalid format is by the following steps: > 1) set the invalid format by hand (command line or pref file) > 2) start BD and at the warning choose Go To Preferences > 3) ignore the pref panel (which should have the format shown in red > and the triangle) and close it > > So that is almost deliberate. I am guessing that anyone who can set it > by hand must know enough not to do this. So should we worry more about > this possibility? > There is only so much that you can do. Someone who wants to screw things up will find a way, but I think there's enough there to protect people who are paying attention. Thinking about it, though, colorblindness may limit the effectiveness of the red text/yellow icon. What about showing a warning sheet when they try to leave the pref pane with an invalid format, either by switching panes or closing the window? Even a warning sheet when you commit editing on the text field wouldn't be that bad, IMO, since you're not going to be changing it frequently. BTW, thanks for catching my dumb commit on commenting out the assertions; I found they were a problem by messing with my plist by hand, then opening a file. I should have done BOOL yn = [scanner...], then NSAssert( yn,...). -- Adam |