Bug Stomp

A keyphrase for "thank you" shouldn't take much rank at all. For 'lol' make a keyphrase for "haha". It catches several varieties of typed laughter.

I seem to recall the number rank of +37 as what overcomes most automated responses...

Ok, there may be a user error here, but I seem to have found Julie's NVR problem. I loaded all my xnone Responses with a conditional:

<?PF if (mem-nickname) is not "Pinquey"; ?>

Yesterday I added a new one, and did NOT include that conditional. Now when Julie gets into the Xnones, she only chooses that Response. And I'm pretty sure these folks are not nicknamed "Pinquey."

She had a pretty funny conversation yesterday in which she made him crazy by responding to every proposition by asking him what sort of music he liked. But it was a revealing accident, not intentional.

So my suspicions of the Is/Is not thingie are firming up a bit. As I said, this could be some sort of user error, but if it is, I can't find it.

I've never used "not" in a conditional statement. My conditional statements without "not" work just fine. *shrugs*

My positive conditionals work too. But the Book says negatvies (is not and does not exist) work too. Is anyone using the negatives successfully?

M

Yes - I have many NOT statements in BJ's toast-related keywords (eg: {?PF if (mem-nickname) is not "Talkie Toaster"; ?},) and they all seem to work fine.

How frustrating. Mine are obviously not working, and the positives are. Back to the drawing board.

M

I think a found the problem. If (mem-something) does not exist, then the is not comparison fails just like the is comparison does. So "is not" actually means "is something other than". Hmm. More drawing board.

Ah, I hadn't noticed that. But then I default (name) to (mem-nickname), and I don't think I have any other NOTs.

Just default whatever you need to a random value, and all should be well.

>> Just default whatever you need to a random value,
>> and all should be well.
Got to be careful of that. If I default to a random value, then I am carrying that value AND whatever was already there. If I default to ONLY a random value, I lose whatever I already tagged the visitor with previously.

Here's the puzzle. I assign a triggering nickname to an annoying visitor. Then when he comes back I check for that nickname, and hang up if it's there.

I guess I need some clarification on the memory lists. Does mem-nickname contain "MickMcA, Mick, Doughboy" or "MickMcA|Mick|Doughboy"? In other words, if I say
if (mem-nickname) is "Mick";
am I asking if "Mick" is in the list, or if the one pulled from the list is "Mick"? Is that clear? I don't want the guy to get past the trap because I've got more than one nickname stored for him.

mems are comma-delimited, and (I'm pretty sure, though I don't recall trying this myself,) it checks if it is in the list, rather than picking one at random.

I'm running a test bot in the debugger, so the questions are coming hard and heavy.

KP: I have (artadjnoun) [50,0]
does not match "I have a red gem," "I have a gem," or "I have gems." Why?

KP: I have (art|) (adj|) (noun|) [50,0]
does match, but (key1), (key2), and (key3) are empty even if the statement being parsed is "I have a green stone." However, (obmod) does contain "a green stone."

And a tip for the unwary. If a literal (key1) turns up in memories or transcripts, that means the KP didn't contain a (key1). Not exactly what I thought was supposed to happen.

M