Zugg Software :: View topic

Newbie Joined: 22 Nov 2004 Posts: 9

Hello all -

Trying make a trigger that will capture certain elements of some mud output. Here is a sample of that output:

Frontier - Boone: Boondock(T) Carson(I) Fremont(B)
Garr - Garioso: Ashten(R)
Gephoria - Gwarrior: Gforce(A)
Gliese - Deklax: Crash(I) Valis(A)
Gough - Dunavant: Gough Prime(R) New Gough(T)
Granite - Freya: Pebble(T)
Happy - Priscilla: Smile(R)
Havefun - Bourbon: Bourbonballs(A) Whiskey Map(T)

The elements I want to capture, are the name that appear after the colon, in each line. Here is the pattern I started with:

(%w) - (%w)~: (%w)~({A|R|T|I|B|L|N}~)

In this example, I want to be able to add %3 to a string list (note that the letter in the parens will always be one of the seven in the trigger - and I do NOT want that part for the string list). So in the output above, in the case of the 'Garr' line, I want to just get 'Ashten', for the list.

I know I might have to make three seperate triggers to account for the fact that each line of output might have 1-3 elements I need to get in it (as in the 'Frontier' line, above).

Here's where I run into trouble...some lines will have two word names, instead of one (as in the 'Gough' line above). I don't know how to catch the whole name (which I need), without using (%*)...however, if I use (%*) the trigger barfs on other lines, because it grabs 2 or even 3 elements where I only wanted one.

Basically, given the example above, I would want to nab:
Boondock|Carson|Fremont|Ashten|Gforce|Crash|Valis|Gough Prime|New Gough|Pebble|Smile|Bourbonballs|Whiskey Map

Can anyone give me some insight? TIA

Enchanter Joined: 05 Mar 2003 Posts: 593 Location: Canada

This is all untested:

Posted: Wed May 14, 2008 5:28 pm

This pattern has perhaps two too many captures. You may find that

Enchanter Joined: 05 Mar 2003 Posts: 593 Location: Canada

If words per items vary the gambit [^\(] might be a good way to go.
If however the number of items can go beyond 3 I think manually extending the pattern with additional (?: (\a+(?: \a+)?)\([ARTIBLN]\))?
Is probably the right move and then just test for %1,%2,...,%n
if instead you do something like (?: (\a+(?: \a+)?)\([ARTIBLN]\))*
then you'll need to use something like %regex or %subregex to break the string down once more into its component parts.

Newbie Joined: 22 Nov 2004 Posts: 9

Thanks for the help guys! I haven't used regex scripting before..now I see I might have to start learning it, as it is obviously more flexible...

Still running into a snag, however...The script works perfectly, in that it will capture all the elements that I want, in a line-by-line basis (tested all possible combinations, and they work). However, in setting up the trigger, I am having it add the captured elements to a string list, with this:

(Note: the elements I want in the list will be %3, %4, and %5 - using #additem, instead of %additem to avoid creating duplicates, in the string list)

#additem all_planets %3
#if (%4) {#additem all_planets %4}
#if (%5) {#additem all_planets %5}

The problem is - now - that when I run the script, it seems to be adding corrupted data to the string list - for example, using the sample output from my first post, instead of adding "Whiskey Map" to the string list....it might just add "Whiskey"..or "Map"...or even just a blank line!

Since, in testing, everything works fine, I am suspecting that the problem lies in the fact that there is too much data coming from the mud all at once, for zmud to handle efficiently. When I issue the command to get the output that I need, to operate this script on, the mud dumps out 145 lines of output, all at once - and I am asking zmud to process each line individually...

As it happens, I own a copy of cmud. After figuring out how to add the trigger in cmud, I tested it on the same thing, and everything works perfectly, and the proper string list is generated, which further inclines me to believe that zmud's problem is too much data, too fast. (And before you suggest it, I would really prefer to stay with zmud - I know cmud is supposed to be bigger, better, faster, etc...but it is just too confusing for me - too much of an old dog, I guess)

So - does anyone have any suggestions as to how I can get zmud to process all the data more quickly and efficiently, so I can test my theory?

Sorry for the long-ish post, and TIA