Bug Stomp

Also - sorry for the second post - but I realized I may not have given enough info about the problem I was having with the seek I mentioned before, so here it is from my language center, if that helps.

you do not have (any|a|the|) (noun) [40,0] <?PF rem (key2) as only "last_sob"; ?>
    
    You're right! Could I borrow yours?
        + my what, what [0]
        Your (mem-last _sob), (mem-nickname), aren't you paying attention?

This keyphrase worked fine until today, catching: Julie Tinkerbell!
^julie tinkerbell ! (re) [127,0] <?PF raw ?>

but no longer seems to work as of tonight.
I also noticed the word "dude" is BLAB. Is that intentional? I've attempted at least half a dozen workarounds, from ^dude$ (re), to breaking it down letter by letter. Many thanks - and Happy New Year.

"quite fine" goes to xcommand, even though neither word is a verb. I thought perhaps "quite" was spellcorrected to "quiet" or "quit" in preprocessing, but that was not the case in debug.

I have the keyphrase: do those people (verb), I have tried it ranked anywhere between 0,0 - 100,0
that is somehow never matched to, for example,
"Do those people dance"
but found instead is the keyphrase: (verb) 0,0
Now that is just odd. Also odd is, in debug, the processor calls "people" the sv in the sentence, the v is "do people" and "dance" is the sob, considered part of the ob "those dance".

Irina Khalidar (not Irina Khalidar 2) by Irina, an M-rated bot with development = 11,768, can't be chatted with (the message "Incorrect bot information" appears). So something is screwed up - hopefully this can be fixed! I think she is one of the more interesting and innovative bots at the Forge, and it's a shame no one can see her.

I'm not sure if this is a bug or not, but this no longer works:

    goto game start
game start [10,0] <?PF raw; ?>
    yada yada
        + whatever [0]
        goto game second <?PF rem "rock" as only "music"; ?>
game second [10,0] <?PF raw; ?>
    yada yada
        + whatever [0]
        goto game last
game last [10,0] <?PF raw; ?>
    yada yada

This used to work (before last year's changes to the engine), but currently it gets to "game second" but never gets to "game last." Also, the memory assigned after that first seek doesn't seem to hold for whatever reason.

The above is incorrect (sorry, I should have narrowed the problem down better before posting - I'm a little rusty!). Actually, "game last" wasn't reached because getting there was contingent on the the memory "music" having the assigned value.

So, the following does not result in the memory getting assigned. Replacing the goto with normal text results in the memory assignment working correctly. Presumably the presence of the goto is interfering with the memory assignment

        goto game second <?PF rem "rock" as only "music"; ?>

I also tried the following. The goto worked, but the memory assignment does not.
        <?PF rem "rock" as only "music"; ?> goto game second

I believe this is a bug.

Dr Ben, I believe it is that you can't script a goto response. However, I also believe that if there is memory script in the keyphrase the goto goes to, and/or in the response triggered by it, that will make the memory hold.

Thanks for responding, ezzer. You can add script to a goto response (hopefully we are talking about the same thing here). My bot relies very very heavily on this and it works very well. There only seems to be a problem when the goto is a response to a seek. This used to work a couple of years ago - that's why I think its failure to work now is a bug (perhaps the Professor broke something when he was tinkering this past year?).

This problem isn't a big deal for me since I found an acceptable work-around for the affected code, but I figured I'd report the bug since it might affect other bots.

Yes. Come to think of it, I have to correct myself. I use conditional script on some goto responses - it is indeed the ones in seeks that do not seem to work. I remember now having addtional issues at one point with goto seeks being broken and not working at all...I need to try and jog my memory for how that ended up being resolved,and also review some forum posts I read a while back - in particular, an exchange between the Professor and another botmaster in either Personality or AI Engine regarding the placement of script and proper memory storage/gotos...Sorry I wasn't helpful! Hopefully we'll figure it out.

The download for large transcripts doesn't seem to work correctly. The entire 93 Mb downloaded, but my text editor cannot detect the character encoding. So something seems to be messed up. Perhaps packaging it in a zip file might help (and make the download smaller!)? This isn't a problem for me since I don't use that feature, but I figured I'd report the problem.

I usually download a day at a time by pressing ctrl-A on a transcript page to select all, copying, then pasting into a text editor. Then I delete the web page stuff at the top and bottom of the file. This works well, but it does remove the emoticons (which isn't a big deal). The only problem is that for one of my machines, the clipboard seems to max out at about 13,000 lines. But this machine was able to copy all 23,000-some lines of yesterday's transcript without problem. I'm not sure what is causing the difference between the two machines. I then run a little TCL script that sorts the transcript into individual conversations. Just thought I'd share some ideas for anyone that's interested.

Hi, I'm trying to make a bot that's a good conversationalist but can also be a sex bot if it likes you enough. I'm in the early stages, not into anything yet really. But my bot doesn't recognize "hi" or "hello" or any other greeting really. Is anyone else having an issue with xHello? If not, what might be wrong and how could I fix it?