Obviously there are two different issues to be considered:
-if a scripted command doesnt work as its supposed to
-or is the script not compatible with changes in a new version
Agreed. And I didn't mean to say anything else. But the fact it
did work is at least a positive hint
.
When someone reports a bug that could be script related, the obvious place to start is with a clean copy of mIRC to see if the problem exists without scripts.
I (we) did and, of course, reinstalled the script just after. Same crashes.
But you should agree also that a script can do
legitimate actions or sequences of actions which are not being done in interactive standard use... Especially when using the remote event control... obviously made for this kind of use.
Either this is just a 'closed' software and yes, obviously, a mere 'blank' test is sufficient, or this includes scripting capabilities and thus we can also test scripts with it. Just common sense.
The more information ppl have, the better able they are to track down a problem in any situation
100% agreed... and it is almost the conclusion of my previous post
mIRC is a great product becoming more and more complex and I know this kind of bug can't be avoided. And scripting just makes it worse. Because any undocumented or non-standard command use can work in one version and become impossible in the next one. No doubt about this, I have many examples in other languages/kits. That's the reason why the "incompatible change" chapter is soooo important in every sdkit.
But, once again, there are other threads here focusing on some close -c $nick inconsistencies. And this seems to be related also to this random crash since the logs I get always point to a chat close as the bug trigger. I'm really trying to trace the problem and if I manage to have more details I will give them to whoever can use them...