412 PurelyFunctional.tv Newsletter

00 Valid names.md

Valid names

This challenge looks like a fun experiment in building a simple rules-based validator.

Definitions:

1. A name is a sequence of terms separated by a space. It must have at least 2 terms. The last term must be a word.
2. A term is either an initial or a word.
3. An initial is a single capital letter followed by a period.
4. A word is a capital letter followed by one or more letters (upper or lower case).

Write a function that checks whether a string is a valid name.

Examples

Valid names:

• George R. R. Martin
• Abraham Lincoln
• J. R. Bob Dobbs
• H. G. Wells

Invalid names:

• J R Tolkien (no periods)
• J. F. K. (must end in word)
• Franklin (must have at least two terms)

Thanks to this site for the challenge idea where it is considered Expert in JavaScript. The problem has been modified from the original.

Please submit your solutions as comments on this gist.

@steffan-westcott you can do your regex using things like \p{Alpha} that the mean is closer to the description by one or more letters (upper or lower case)
\s is more like "space" that , there is \p{Upper} for "Capital", etc..

With thanks to @souenzzo, my answer becomes:

(defn name? [s]
  (re-matches #"(\p{IsUppercase}(\.|\p{IsAlphabetic}+) )+\p{IsUppercase}\p{IsAlphabetic}+" s))

I found that I needed to use the Unicode character categories to correctly handle some cases:

(some? (name? "Æthelred The Unready"))
=> true

I feel like this is 100% cheating, but instaparse is so much fun that it's illegal in 17 states:

; Start with clj -Sdeps '{:deps {instaparse/instaparse {:mvn/version "1.4.10"}}}' -r

(require '[instaparse.core :as i])

; Note: we could also use regexes here, especially for the basic tokens (initial, word, ws, etc.), but I feel like that's double cheating  😜
(def name-grammar "name    = term ws (term ws)* word
                   term    = initial | word
                   initial = upper '.'
                   word    = upper letter+
                   letter  = upper | lower
                   ws      = ' '
                   upper   = 'A' | 'B' | 'C' | 'D' | 'E' | 'F' | 'G' | 'H' | 'I' | 'J' | 'K' | 'L' | 'M' | 'N' | 'O' | 'P' | 'Q' | 'R' | 'S' | 'T' | 'U' | 'V' | 'W' | 'X' | 'Y' | 'Z'
                   lower   = 'a' | 'b' | 'c' | 'd' | 'e' | 'f' | 'g' | 'h' | 'i' | 'j' | 'k' | 'l' | 'm' | 'n' | 'o' | 'p' | 'q' | 'r' | 's' | 't' | 'u' | 'v' | 'w' | 'x' | 'y' | 'z'")

(def name-parser (i/parser name-grammar))

(defn name?
  [s]
  (not (i/failure? (name-parser s))))

; Testing

; Valid names
(name? "George R. R. Martin")
(name? "Abraham Lincoln")
(name? "J. R. Bob Dobbs")
(name? "H. G. Wells")

; Invalid names
(name? "J R Tolkien")   ; (no periods)
(name? "J. F. K.")      ; (must end in word)
(name? "Franklin")      ; (must have at least two terms)

(defn ->terms [name]
  (clojure.string/split name #"\s"))

(def initial?
  (partial re-matches #"[A-Z]\."))

(def word?
  (partial re-matches #"[A-Z][a-zA-Z]+"))

(defn valid-name? [name]
  (let [valid-terms (map #(or (initial? %) (word? %)) (->terms name))]
    (and (every? seq valid-terms)
         (word? (last valid-terms))
         (>= (count valid-terms) 2))))

(defn initial? [s]
  (re-matches #"[A-Z]\." s))

(defn word? [s]
  (re-matches #"[A-Z][a-z|A-Z]+" s))

(defn term? [s]
  (or (initial? s) (word? s)))

(defn name? [s]
  (let [terms (map term? (clojure.string/split s #"\s"))]
    (boolean
     (and (every? identity terms)
          (>= (count terms) 2)
          (word? (last terms))))))

;; (name? "George R. R. Martin") ;; => true
;; (name? "Abraham Lincoln") ;; => true
;; (name? "J. R. Bob Dobbs") ;; => true
;; (name? "H. G. Wells") ;; => true

;; (name? "J R Tolkien") ;; => false
;; (name? "J. F. K.") ;; => false
;; (name? "Franklin") ;; => false

(defn word? [x]
  (boolean (re-matches #"^[A-Z]\w+$" x)))

(defn initial? [x]
  (boolean (re-matches #"^[A-Z]\.$" x)))

(defn term? [x]
  (or (word? x)
      (initial? x)))

(defn name? [x]
  (let [terms (clojure.string/split x #"\s")]
    (and (<= 2 (count terms))
         (every? true? (map term? terms))
         (word? (last terms)))))

(ns eric-normand-newsletter-challenges.valid-names
  (:require [clojure.spec.alpha :as s]
            [clojure.string :as string]))

(defn words [s]
  (->> (string/split s #"\s")
       (filter (complement string/blank?))))

(s/def ::name #(re-matches #"[A-Z][a-zA-Z]+" %))
(s/def ::initial #(re-matches #"[A-Z]\." %))

(s/def ::term (s/or :initial ::initial
                    :name ::name))
(s/def ::full-name (s/cat :names-or-initials (s/+ ::term)
                          :last-name ::name))

(defn name? [s]
  (s/valid? ::full-name (words s)))

@pmonks solution is better but I use the same cheatcodes:

(require '[instaparse.core :as insta])
(def name? (complement (comp insta/failure? (insta/parser "valid-name = <whitespace>? term ( <whitespace> term )* ( <whitespace> word ) <whitespace>?
<term> = initial | word
initial = #'[A-Z]\\.'
word = #'[A-Z][A-Za-z]+'
<whitespace> = #'[\\s]+'"))))

(name? "P. W. A. Breed") ;; => true
(name? "Pieter W. A.") ;; => false
(name? "Pieter W. A. Breed") ;; => true

I like the regex solutions, and I expect that they're the fastest. Just for fun, here's a different approach that uses a state machine hidden in a transduce reducing function. Note: the state is a transient. [Updated to fix a bug with something like "A.Bad Dot".]

;; state :cap? = first letter of term is capital, :cnt = count of characters in term,
;; :initial? = term is an initial, :toks = number of tokens
(defn valid-name? [s]
  (let [step (fn step
               ([] (transient {:cap? false :cnt 0 :initial? false :toks 0}))
               ([state]
                (and state
                     (:cap? state) 
                     (>= (:cnt state) 2)
                     (not (:initial? state))
                     (>= (:toks state) 1)))
               ([state c]
                (let [cnt (inc (:cnt state))]
                  (case  c
                    (\A \B \C \D \E \F \G \H \I \J \K \L \M \N \O \P \Q \R \S \T \U \V \W \X \Y \Z)
                    (cond (not (:initial? state)) (assoc! state :cap? true :cnt cnt)
                          (= cnt 1) (assoc! state :cap? true :initial? false :cnt cnt)
                          :else (reduced false))

                    (\a \b \c \d \e \f \g \h \i \j \k \l \m \n \o \p \q \r \s \t \u \v \w \x \y \z) 
                    (if (and (:cap? state) (not (:initial? state)))
                      (assoc! state :cnt cnt)
                      (reduced false))

                    \space 
                    (if (and (:cap? state) (> cnt 2))
                      (assoc! state :cap? false :cnt 0 :toks (inc (:toks state)))
                      (reduced false))

                    \. 
                    (if (and (:cap? state) (= cnt 2))
                      (assoc! state :initial? true :cnt cnt)
                      (reduced false))

                    (reduced false)))))]
    (transduce identity step s)))

@sztamas Nice to see another usage of spec! This was my first time using the library, really interesting stuff 👍

(ns spec-names
  (:require [clojure.spec.alpha :as s]
            [clojure.string :as str]))

; input values
(def validnames ["Abraham Lincoln"
                 "George R. R. Martin"
                 "J. R. Bob Dobbs"
                 "H. G. Wells"])

(def invalidnames ["J R Tolkien"
                   "J. F. K."
                   "Franklin"])
; predicates
(defn isCapitalized? [string] (= string (str/capitalize string)))
(defn lenGt2? [string] (> (count string) 2))
(defn endsWithPeriod? [string] (= "." (str (last string))))

; specs
(s/def ::valid-name (s/and isCapitalized? lenGt2?))
(s/def ::valid-initial (s/and isCapitalized? endsWithPeriod?))
(s/def ::nameorinitial (s/or :name ::valid-name :initial ::valid-initial))

; validator
(defn name-validator [name]
  (let [splitname (str/split name #" ")
        lastname (last splitname)
        init-terms (drop-last splitname)]

    (every? true?
            [(s/valid? (s/+ ::nameorinitial) init-terms)
             (s/valid? ::valid-name lastname)])))

; run tests!
(mapv name-validator invalidnames)
(mapv name-validator validnames)

@andyfry01 Yes, I thought spec can be used nicely for this kind of problem!

@pieterbreed you're too kind; in the real world I would absolutely use a solution closer to yours than the one I posted. Regexes are awesome! 😉

Spec really is shining in this one wow

(ns pftv.challenges.412
  (:require [clojure.spec.alpha :as spec]
            [clojure.string :as str]))

(spec/def ::initial (partial re-matches #"[A-Z]\."))

(spec/def ::word (partial re-matches #"[A-Z][A-Za-z]+"))

(spec/def ::term (spec/or :initial ::initial :word ::word))

(spec/def ::name (spec/cat
                  :terms (spec/+ ::term)
                  :ends-in-word ::word))

(defn valid-name? [name]
  (let [name-vec (str/split name #" ")]
    (spec/valid? ::name name-vec)))

(def valid-names ["George R. R. Martin"
                  "Abraham Lincoln"
                   "J. R. Bob Dobbs"
                  "H. G. Wells"])

(def invalid-names ["J R Tolkien"
                      "J. F. K."
                      "Franklin"])
(every? true?
        (map valid-name? valid-names))
;; => true

(every? false?
        (map valid-name? invalid-names))
;; => true