Re: [E-devel] Ebits design idea

[Chris b. R. <ch...@da...>]

On Wed, Feb 27, 2002 at 08:39:43PM -0600, Nathan Ingersoll muttered...
: On Thu, Feb 28, 2002 at 01:16:34AM +0100, Christopher 'smugg' Rosendahl is quoted as saying:
: . . .
: > Implementing this wouldnt make the existing bits.db break or anything,
: > it would just add more functionality to Ebits.
: > And since the three states normal, hilited and clicked will be builtin (ebits
: > would allways know of theese), clicking & hiliting would work as it currently
: > does, but it would search the list for properties for that specific state.
: > Instead of just grabbing ebits_object_bit_state->hilited.blabla.
: > All the state names will be saved lower-case, but when triggering states
: > it will be case-insensitive.
: > 
: > 
: > I hope i make myself clear here, if i don't just bark really LOAD!
: > 
: > ......I'm glad that you read this far, thanks for listening ;P
: 
: from my perspective, this would make ebits more flexible with little extra
: overhead to do so.  This would definitely help in some corner cases we've
: encountered in ewl. With relatively short lists of states and reasonable name
: lengths for the states, the searching overhead should be fairly minimal.

You could still have the optimised case of hacing normal, hilted and clicked
in the structure like now, but also have a list of states that aren't those,
that way when looking for a custom state you need not check the standard ones.

Regards,

Chris
-- 
+------------------------------------------------------------------+
| Chris Ross   |      ch...@da... | ct...@fe...        |
|              | http://www.darkrock.co.uk | http://www.ferite.org |
+------------------------------------------------------------------+
"SYSOP \sis' op\ n: the person laughing as you type."