Skip to content
Fetching contributors…
Cannot retrieve contributors at this time
executable file 87 lines (73 sloc) 3.21 KB
#! /bin/sh
#| Hey Emacs, this is -*-scheme-*- code!
exec racket -l errortrace --require "$0" --main -- ${1+"$@"}
;; Some code to reply in an alarmingly-human-like way. Idea, but not
;; the code, utterly stolen from Peter Danenberg (aka "klutometis"),
;; to whom all credit is due.
;; The basic idea: someone says something to the bot. The bot doesn't
;; recognize that input as one of its built-in commands, so this code
;; runs. This code breaks the input into words, and ranks each word
;; based on how frequently it appears in the "corpus" (just a log of
;; all the input the bot has seen). It picks the "most interesting"
;; word -- i.e., the one that appears the least -- and then finds all
;; the utterances in the corpus that contain it. It then returns an
;; utterance chosen at random from that set, favoring the longer
;; (presumably more-interesting) ones.
#lang racket
(only-in "vars.rkt" *incubot-logger*))
(include "incubot-tests.rkt")
(provide log)
(define (log fmt . args)
(apply (or (*incubot-logger*)
(lambda (fmt . args)
(apply fprintf (current-error-port) fmt args)
(newline (current-error-port))))
(string-append "incubot-server:" fmt) args))
(define (omit-needless words)
;; Finding the popularity of a word takes time proportional to that
;; popularity. For example, to find the popularity of the word
;; "it", we must scan through the log_word_map index, counting the
;; number of occurrences. For a common word like "it", that's slow!
;; Since we're ultimately only interested in _unpopular_ words, we
;; can simply filter out a bunch of the more popular words from our
;; input set ``words'', and rank only the remaining words. An
;; example I tried showed that without such filtering, it took about
;; 8 seconds to return a witticism; with filtering, it was .4
;; seconds. Seems worthwhile!
;; I got this list by querying a reasonbly-current copy of the
;; corpus like so:
;; select distinct(word), count(word) c from log_word_map group
;; by word having c> 20000 order by c desc limit 20;
(define noise (set "a" "and" "be" "but" "emacs" "for" "have" "i" "if" "in" "is" "it" "not" "of" "on" "that" "the" "to" "with" "you"))
(set-subtract words noise))
(provide incubot-sentence)
(define incubot-sentence
[(list (? list? s) (? corpus? c))
(incubot-sentence (wordlist->wordset s) c)]
;; only tests use this
[(list (? string? s) (? corpus? c))
(incubot-sentence (string->lowercased-words s) c)]
[(list (? set? ws) (? corpus? c))
(let ([rare (rarest (omit-needless ws) c)])
(and rare
(random-choose-string-containing-word rare c)))]
(log "incubot-sentence invoked with bogus arglist: ~s" bogon)
(define/contract (rarest ws c)
(-> set? corpus? (or/c string? #f))
(let ([ranked (corpus-rank-by-popularity c ws)])
(log "~a" ranked)
(and (not (null? ranked) )
(let ([chosen-pair (car ranked)])
(log "incubot chose ~s, which appears ~a times"
(vector-ref chosen-pair 0)
(vector-ref chosen-pair 1))
(vector-ref chosen-pair 0)))))
Jump to Line
Something went wrong with that request. Please try again.