get_haiku.py

import os
import re
import random
import requests
from nltk import word_tokenize
from nltk import pos_tag
from nltk.tokenize import SyllableTokenizer
from random_word import RandomWords

rw = RandomWords()
SSP = SyllableTokenizer()
DIR = os.path.dirname(os.path.abspath(__file__))


def get_word():
    url = "https://raw.githubusercontent.com/first20hours/google-10000-english/master/google-10000-english-usa-no-swears.txt"
    r = requests.get(url)
    words = r.text.splitlines()
    word = random.choice(words)
    while not pos_tag(word_tokenize(word))[0][1].startswith("NN") or syllables_in_text(word) > 3:
        word = random.choice(words)
    return word


def random_word():
    word = rw.get_random_word()
    while not pos_tag(word_tokenize(word))[0][1].startswith("NN") or syllables_in_text(word) > 3:
        word = rw.get_random_word()
    return word


def syllables_in_word(word):
    return len(SSP.tokenize(word))


def syllables_in_text(text):
    count = 0
    for word in word_tokenize(text):
        count += syllables_in_word(word)
    return count


def get_haiku(word):
    url = "https://dataviz.ei.columbia.edu/api/haiku"
    headers = {'Content-type': 'application/json'}
    data = {'subject': word}
    counts = []
    while counts != [5, 7, 5]:
        r = requests.post(url, headers=headers, json=data)
        lines = re.sub(r'[^\w\s]', '', r.text).splitlines()
        lines = [l for l in lines if len(l) > 1]
        counts = [syllables_in_text(l) for l in lines]
        print(r.text)
        print(counts)
        print([[SSP.tokenize(token)
                for token in word_tokenize(line)] for line in lines])
    return r.text


def main():
    word = input("Enter a word or type 'r' to randomly choose a word: ")
    if word.strip().lower() == 'r':
        word = random_word()  # from library
        # word = get_word()  # from word list
    print(f"The word is {word}.")
    haiku = get_haiku(word)
    print(haiku)


if __name__ == "__main__":
    main()