Last active
October 4, 2023 14:40
-
-
Save asherrecv/99d7f995a8676f4a60a82861e729cacf to your computer and use it in GitHub Desktop.
Clojure autocompletion with bigrams
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
(require '[clojure.set :as set]) | |
(defn bigrams-with-position [word] | |
(map (fn [a b i] [(str a b) i]) | |
(seq word) | |
(seq (drop 1 word)) | |
(range))) | |
(defn create-singleton-bigram-index [bigram position word] | |
{bigram {position #{word}}}) | |
(defn merge-position-index [lhs rhs] | |
(merge-with set/union lhs rhs)) | |
(defn merge-bigram-index [lhs rhs] | |
(merge-with merge-position-index lhs rhs)) | |
(defn create-bigram-index-of-word [word] | |
(reduce | |
merge-bigram-index | |
(map (fn [[bigram position]] | |
(create-singleton-bigram-index bigram position word)) | |
(bigrams-with-position word)))) | |
(defn create-bigram-index-of-dict [dict] | |
(reduce | |
merge-bigram-index | |
(map create-bigram-index-of-word dict))) | |
(defn query [bigram-index query-str] | |
(let [matches | |
(map (fn [[bigram position]] (get-in bigram-index [bigram position])) | |
(bigrams-with-position query-str))] | |
(if (empty? matches) | |
#{} | |
(reduce | |
set/intersection | |
matches)))) | |
(comment | |
(def dict ["go" "goroutine" "golang" "gogo" "gorage"]) | |
(bigrams-with-position "asd") | |
;; => (["as" 0] ["sd" 1]) | |
(create-bigram-index-of-word "gogog") | |
;; => {"go" {0 #{"gogog"}, 2 #{"gogog"}}, | |
;; "og" {1 #{"gogog"}, 3 #{"gogog"}}} | |
(create-bigram-index-of-dict ["go" "gol" "gogog"]) | |
;; => {"go" {0 #{"gol" "go" "gogog"}, 2 #{"gogog"}}, | |
;; "ol" {1 #{"gol"}}, | |
;; "og" {1 #{"gogog"}, 3 #{"gogog"}}} | |
) | |
(comment | |
(def dict ["go" "goroutine" "golang" "gogo" "gorage"]) | |
(def index (create-bigram-index-of-dict dict)) | |
(query index "gor") | |
;; => #{"gorage" "goroutine"} | |
(query index "gog") | |
;; => #{"gogo"} | |
(query index "gol") | |
;; => #{"golang"} | |
) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment