On Thu, 28 Mar 2002, Jim Schram wrote: > >> So, the moral of the story is: Don't call System-Use-Only functions. Ever. > >> (Thanks! :o) > > > > with that type of attitude you limit yourself a lot :) > > Well, with your type of attitude, we can never innovate and keep existing > software running. (Touch�! ;o)
good programmers check that they can do stuff before they start messing around :) unfortunately, not everyone is like that - so, most of us suffer, or resort to poking around ourselves (which, causes other problems) :P > But seriously... we fully support "hacking" as a viable method of creating > new products and opening new markets! What we do not support (actually > cannot support) is hacking which does not restrict itself to specific OS > versions against which the hack has been fully tested and qualified. > > In other words... customers assume your software will continue to work in > all future Palm OS versions, so if you do things that aren't kosher > (without at least informing the user your software wasn't designed for > this "future" OS version and therefore it might not work correctly) the > customer gets upset and blames Palm. From their perspective, anything > your software is able to do must be supported by the OS, right? Wrong. > This leads to an incredible number of support calls as you can imagine. thats why it is good to get access to the tools for the new platforms asap.. most of the hacks i had to write for lemmings caused it NOT to work on palmos 5.0 - but, lucky for me, 5.0 gave me more heap space to not have to use my little hacks :) i congratulate palm for giving us access to their "new" stuff way ahead of time :0 allows most of us to fix things.. but, in regards to what we had to do to get the product out to begin with, not all users have flashy 5.0 units :) so, we have to make best with what we have :) system-only routines can make that happen.. :) but, it is nice to see the Simualtor barf bigtime when you call something like MemSemaphore* :P [that reminds, me, time to update some code] :P > > question remains, what if you dont have a SysHandleEvent loop? > > or, simply calling SysHandleEvent without enqueuing it should > > be sufficient to make it happen ASAP :) > > If there are no events in the queue when you call SysHandleEvent, the > device will behave as if the call to SysHandleEvent occurred from your > main event loop. However, if there is even one other event in the queue, > your SysHandleEvent will process that event, leaving the actual power-off > event to be either handled in your main event loop or possibly ignored > altogether depending on how the code is structured. It's basically a > nested event loop problem. heh, maybe you could write your own event loop? :) *g* or, check if there is an event on the queue before "posting" the power off key :P // az "more than one way to skin a cat, just dont eat it" :P [EMAIL PROTECTED] http://www.ardiri.com/ -- For information on using the Palm Developer Forums, or to unsubscribe, please see http://www.palmos.com/dev/support/forums/
