Right now, things I have planned to add:
- non-blocking sleep()'s
- "native-looking" "dollar variables" (IE, maybe $target(id) could be accessed this way:
ie, as a tied Perl hash. Behind the scenes, the Perl module would do a lookup against the actual "$target(id)" value, and return it as a scalar here. The only problem with this method is that I don't believe that this construct:Code: Select all
$target{id};
is valid Perl. I'll have to poke around -- I'm open to suggestions.Code: Select all
$target{hp,cur}; - along the same lines, lots of tied scalars. IE values that look like variables, but return useful information without needing to be explicitly set. Two that would be useful are $stand_state (sitting, standing, ducking, etc.) and $run_state (standing still, walking, running, etc.). Of course, these will require that some new mem locs be found, but I'm confident in the abilities of our 'leet cracker teams :P
- a cleaner way to "unload" and "reload" modules. /perl stop|/perl start WORKS, but it's clunky, and you may not want to zap the ENTIRE Perl state just to load a new version of a module.
- MUCH MORE ERROR CHECKING! Right now, MQPerl::KillTimer(35000) will almost certainly kill MQPerl, MQ, and EQ. The array index isn't verified. This is a simple fix that I'll implement tonight, but there are lots of cases like this.
- more "standard" code in utils.pl that is distributed. I've already got an "autofollow" script in there, what other types of tasks are common enough that they should have scripts built into the core MQPerl distro?
