Making an autoroller that can capture and store rolls

I've seen a lot of autorollers on the forums. I currently have one that works. However, I've never seen one that has done autorolling for non-numerical rolling systems AND saves a current "best" roll as sort of like a tally system to see if you need to lower your roller settings to something that is more realistic.

I'd like to get some advice on how I would actually go about doing something like this without needing a third-party application.

The MUD in question has a rolling system that creates rolls looking like this:

EDIT: Autoformat messes up what the rolls look like.

There is a single space in front of Str and 3 spaces in between each attribute which follows. There is no space in front of the first stat and 2 spaces in between each statistic value.



Where the stats can be (from low to high) - Awful, Low, Poor, Aver, Fair, Good

My current autoroller looks something like this:

To Accept Rolls:

^(Good|Fair).. (Good|Fair).. (Good|Fair).. (Good|Fair)..(Good).. (Good).. (Good|Fair|Aver).. (Good).. (Good|Fair).. (Good|Fair).. (Good|Fair)

And sends back "y" if the roll meets the set specifications.

With: Enabled, Keep Evaluating, Regular Expression, Repeat On Same Line checked and a sequence value of 99

And the second part of the roller is:

To Reject Rolls:

Use this body

and sends back "n", basically for any roll that does not meet the set specifications.

All options the same but with a sequence value of 100.

Edited by Nick to show the layout better

You probably need to make a different regular expression that matches all possibilities, and then tests them in a short script. eg.


^
(?P<str>Awful|Low|Poor|Aver|Fair|Good)\s+
(?P<int>Awful|Low|Poor|Aver|Fair|Good)\s+
(?P<wis>Awful|Low|Poor|Aver|Fair|Good)\s+
(?P<dex>Awful|Low|Poor|Aver|Fair|Good)\s+
(?P<agi>Awful|Low|Poor|Aver|Fair|Good)\s+
(?P<con>Awful|Low|Poor|Aver|Fair|Good)\s+
(?P<cha>Awful|Low|Poor|Aver|Fair|Good)\s+
(?P<wil>Awful|Low|Poor|Aver|Fair|Good)\s+
(?P<voi>Awful|Low|Poor|Aver|Fair|Good)\s+
(?P<per>Awful|Low|Poor|Aver|Fair|Good)\s+
(?P<app>Awful|Low|Poor|Aver|Fair|Good)\s*
$


I have shown the trigger on multiple lines for clarity, you would just join them together when entering. The \s+ means "one or more spaces" which allows for the variable number of spaces between each word.

I have used named wildcards to give each matching thing a "wildcard name" (eg. "str" for the Str item).

You do not need to "repeat on same line" for this one - it will only match once.

Now set the trigger to "send to script" and do something like this:



What this will do is use wildcard substitution to put the actual matching words in instead of %<str> or whatever, so you can test the words. If you want to do more, like noting your best score so far, or whatever, you could.

One approach would be to convert each one to a number and then add the numbers up, so you can see how close you are getting.

eg.



You can then use this function to get each one into a number, like this:

A regular expression with the same effect can be done like this:


^(?x) (?P<str>Awful|Low|Poor|Aver|Fair|Good)\s+ (?P<int>(?1))\s+ (?P<wis>(?1))\s+ (?P<dex>(?1))\s+ (?P<agi>(?1))\s+ (?P<con>(?1))\s+ (?P<cha>(?1))\s+ (?P<wil>(?1))\s+ (?P<voi>(?1))\s+ (?P<per>(?1))\s+ (?P<app>(?1))\s*$


This saves repeating the "Awful|Low|Poor|Aver|Fair|Good" bit (and therefore makes it easier to change if you need to change what the options are).

This uses (?1) to mean "whatever pattern #1 was".