I have to admit that when I first saw this solution, I thought "hmmm -
clever but feels like it is exploiting a loophole". Depending on the
engne to quietly ignore error-containing stacks just "feels wrong".
I have NOT tried it, but a conceptually cleaner (to me) solution would
be to have a "splash-stack" approach, where the splash stack would
determine the major version in use, then load any behaviour stacks (i.e.
dependent on the version number if appropriate) and then finally load
the app stack, so that the version-dependent behaviours would come into
play. (I haven't been paying enough attention to know whether it would
need to "re-bind" the behaviours or not.)
-- Alex.
On 11/04/2016 23:04, Monte Goulding wrote:
On 12 Apr 2016, at 6:18 AM, Richard Gaskin <ambassa...@fourthworld.com> wrote:
I would be concerned if the engine ignores errors that prevent it from using a
script.
You only get script compilation errors when you set the script not every time a
message passes through an object that has an uncompiled script. If we had the
latter it might make it very hard to fix some issues. Additionally the engine
will load a stack file that has uncompilable scripts silently. At some point it
might be helpful to get a list of issues popping up at which point my solution
won’t be a good idea so I would switch to dynamically switching behaviors
depending on the version.
Cheers
Monte
_______________________________________________
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
_______________________________________________
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
