John Marshall wrote in message <4227@palm-dev-forum>...
>
...
>Right. Most applications want to use the newest SDK and then query for
>available features at runtime (*). But you want to solve something that's
>basically the *opposite* problem.
For example, continuing to use the OLD version of an API
function when the new one (using the same name it had in
the old API) has changed it's signature.
btw ... has Palm decided that this is a Bad Thing(tm) yet?
Instead of renaming the old traps XxxVy() and adding new
traps with the old Xxx() name, they really ought to let the
old API keep using Xxx() and require the new, enhanced
method to use a new name, XxxEnh() or something...
>Richard Hartman <[EMAIL PROTECTED]> wrote:
>> It might be nice if an SDK_VER #define were added to future
>> versions of the SDK...
...
>it seems unnecessary.
Not to everybody....
>Besides, it's a little too late to add it to the old SDKs!
Well, that excuse can be used forever, can't it? It's not
to late to keep the problem from getting worse for -future-
SDKs!
For the old SDKs we could always add docs that say
something like
when compiling under SDK 1.0 add
"#define SDK_VER 100" to your
project prefix header.
--
-Richard M. Hartman
[EMAIL PROTECTED]
186,000 mi/sec: not just a good idea, it's the LAW!
--
For information on using the Palm Developer Forums, or to unsubscribe, please see
http://www.palm.com/devzone/mailinglists.html
