Zugg Software :: View topic

Apprentice Joined: 10 Oct 2000 Posts: 132 Location: USA

Well I wondered if someone could help me out with some pattern matching. I have it working with Regex's but there appears to be a pretty major memory leak in the Regex system in the beta so I want to find a solution using normal pattern matching until the leak gets fixed.

The patterns I need to match are:




Single summon patterns:


--> Poisonous Toad:+


--> Ghast:A


Double summon patterns:


--> Poisonous Toad:B, Ghast:A


Triple summon patterns:


--> Poisonous Toad:B, Giant Spider:A, Ghast:A


4 summon patterns:


--> Poisonous Toad:B, Giant Spider:A, Ghast:A, Werewolf:B

Now initially I tried the following patterns to match it:

Single:
--> (*):(*)$
Double:
--> (*):(*), (*):(*)$
Triple:
--> (*):(*), (*):(*), (*):(*)$
4:
--> (*):(*), (*):(*), (*):(*), (*):(*)$

Major problem with these is the Double will match 3 summons because it will assign the % variables as follows:

%1:Poisonous Toad:B, Giant Spider
%2:A
%3:Ghast
%4:A

Now if there is a way that I can get something like %w to match 2 words that would work nicely.

Jaerin

GURU Joined: 10 Oct 2000 Posts: 4379 Location: USA

First, I suggest using ? to match what comes after the colon since from your example it seems that this will always be just one character.

Second. try using this in the triggers:
#IF (!%pos(",", %1)) {commands for trigger}

What this does is to check if what is captured in %1 does not have a comma. If it does have one, then what we captured doesn't belong to that trigger. If it doesn't have one, then that trigger is correct for what was captured and we proceed to do the commands of the trigger.

Kjata

Apprentice Joined: 10 Oct 2000 Posts: 132 Location: USA

Well it seems that Zmud has had a bit of Regex code in it all along. I was able to put something together. This isn't going to look completely right because I know the forum doesn't like (plus) symbols for some reason but here it goes:




--> ([A-Z a-z 0-9 ~- %s]):([A-E ~~ ]), ([A-Z a-z 0-9 ~- %s]):([A-E ~~ ]), ([A-Z a-z 0-9 ~- %s]):([A-E ~~ ]), ([A-Z a-z 0-9 ~- %s]):([A-E ~~ +])$

Well you get the gist of it anyways. I tested it out and it works great. Probably even more efficiently than the way I was doing it before.

There actually is a possibility of two characters after the :. Either A-Z or(plus) or ~ and then A-Z or (plus)

Thanks for the help anyways

Jaerin