Created
November 2, 2010 22:17
-
-
Save pingles/660402 to your computer and use it in GitHub Desktop.
This file contains hidden or 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
(ns tree | |
(:require [clojure.contrib.string :as str] | |
[clojure.contrib.io :as io])) | |
(def lines (rest (io/read-lines "/Users/paul/Work/pes500k.csv"))) | |
(def postcodes-from-file (map (fn [x] (let [parts (str/split #"," x) | |
postcode (str/replace-str " " | |
"" | |
(first parts)) | |
region-id (first (rest parts))] | |
{:postcode postcode | |
:region-id region-id})) | |
lines)) | |
(def postcodes '({:postcode "AB6" :region-id 9} | |
{:postcode "AL" :region-id 10} | |
{:postcode "AB" :region-id 12} | |
{:postcode "AL4" :region-id 10} | |
{:postcode "AL5" :region-id 9})) | |
(defn next-record | |
[record] | |
{:postcode (apply str (rest (:postcode record))) | |
:region-id (:region-id record)}) | |
(defn record-to-tree | |
([record] (record-to-tree record {})) | |
([record tree] | |
(let [postcode (:postcode record) | |
region-id (:region-id record)] | |
(if (nil? (seq postcode)) | |
tree | |
(assoc tree | |
(keyword (str (first postcode))) | |
(record-to-tree (next-record record) | |
{:region-id region-id})))))) | |
(defn merge-tree | |
[tree subtree] | |
(if (not (map? subtree)) | |
tree | |
(let [first-key (first (keys subtree))] | |
(if (contains? tree first-key) | |
(assoc tree first-key (merge-tree (first-key tree) | |
(first-key subtree))) | |
(assoc tree first-key (first-key subtree)))))) | |
;; tree=> (merge-tree {:a {:b 2}} {:a {:c 3}}) | |
;; {:a {:c 3, :b 2}} | |
;; tree=> (reduce merge-tree {} (map record-to-tree postcodes)) | |
;; {:A {:B {:region-id 12}, :L {:4 {:region-id 10}, :region-id 10}, | |
;; :region-id 10}} | |
(println "Loading results") | |
(def results (reduce merge-tree {} (map record-to-tree postcodes-from-file))) | |
(println "Results loaded") | |
(defn lookup-postcode | |
[results postcode] | |
(if (nil? (seq postcode)) | |
(:region-id results) | |
(recur ((keyword (str (first postcode))) results) | |
(rest postcode)))) | |
;; tree=> (lookup-postcode results "AB") | |
;; 12 | |
;; (println results) | |
(println (str "Region for B68 8QQ: " | |
(lookup-postcode results "B688QQ"))) | |
(println (str "Region for AL4 9HA: " | |
(lookup-postcode results "AL49HA"))) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment