Newcomers

In that case, I'd suggest having the non-apology seeks and KPs require emotion of a certain level, and only the demanding an apology available below that. if emotion > -4 on every other KP, if emotion = -5 on apology demand, and setting emotion based on apology or whatever they need to apologize for.

Seeks are the best way to do that properly.

Keyword: I don't like trains
Reply: Apologize!
Seek ID:123: sorry
reply: Okay, we're good now
Seek ID 456: xnomatch
reply: No, I insist you apologize
__Seek: sorry
__goto: 123
__Seek: xnomatch
__goto 456

Something like that is recursive, and will hold the user in the loop until they say "sorry"

Thank you both. I think I prefer Zeig Wolf's answer because it seems easier than adding emotion range for every key phrase. This one, however probably won't work if user resets their browser.

Resetting browsers will get you out of a loop like that - handy when you mess up your gotos If you give a KP or seek for insults 24 or more rank it should outrank the xinsult default.

Thanks.

@Zeig, this sort of setup should work. I would recommend setting the ranks very high so that your user won't accidentally be able to trigger some xcommand or something like that. You should be able to retain this situation even after the person leaves using memories:

Punch: (ID:1)
Response: Ouch! That's not nice. Apologize!
rem "punch" as only "tmp"
Seeks:
--sorry (Rank 125) ID:2
"It's okay"
rem "nothing as only "tmp"

--xnomatch (Rank 99) ID:3
goto 1

Then in your xinitiate sequence, include a conditional like this:

xinitiate
"Hello again" if (mem-tmp) is not "punch"
goto 1 if (mem-tmp) is "punch"

Personally, I don't like relying on the emotion system for anything as it is very easy to manipulate it.

Thaks. One more thing: "rank" is script right? Should I put it under keywords or under response? Or any will do?

Never mind. I figured it out. I just didn't see the box at first.

Okay, dumb question time from me: when uploading, what order do things get checked for errors in? I have an error being thrown on the line + xnomatch [0,0], which is ridiculously vague (indent error, so probably hanging tab on previous line). Am I right in thinking that the engine parses down the file from the start, find an error, and throws there? That would let me tab every so often and cause an intentional breakage on a non-xnomatch, and use that to narrow down where the actual error is?

I do wish that the errors told the line number, though.

I believe it goes line by line, top to bottom, for error checking, that or I just have weirdly consistent errors. It does stop at the first error it finds. It does seem to give the exact issue and not a line beforehand issue though.

Having a line number would be great...

i'm having trouble with maryguise's roleplay bot. the instructions on the chat start are too vague and i don't know how to start any of the storys