rhymes.py

import pronouncing as P # pip install pronouncing
import random
import numpy as np
import sys
import re
import os

def words_to_line(words):
    print(" ".join(words)+",")
#     os.system("say {}".format(" ".join(words)))

class commonword_manager:
    def __init__(self,wordlistfile):
        commonwords = np.loadtxt(wordlistfile,dtype=str)

        common_phones = [P.phones_for_word(x) for x in commonwords]
        pronunciation_known = [len(x)>=1 for x in common_phones]
        self.commonwords = list(commonwords[pronunciation_known])
        self.nallwords = len(self.commonwords)

        common_phones = [P.phones_for_word(x)[0] for x in self.commonwords]
        common_nsyllables = [P.syllable_count(x) for x in common_phones]
        common_rhymes = [list(set(P.rhymes(x)) & set(commonwords)) for x in self.commonwords]
        self.common_rhyme_indices = [ [self.commonwords.index(x) for x in rhymes] for rhymes in common_rhymes ]
        self.common_stresses = [P.stresses_for_word(x)[0] for x in self.commonwords]

    def search_stress(self,stress_pattern):
        r = re.compile(stress_pattern)
        return [self.commonwords[i] for i in range(self.nallwords) if r.match(self.common_stresses[i])]
    
    def random_stress(self,stress_pattern):
        return random.choice(self.search_stress(stress_pattern))

    def search_stress_rhymes(self,word,stress_pattern):
        word_index = self.commonwords.index(word)
        word_rhymes_indices = self.common_rhyme_indices[word_index]
        word_rhymes = [self.commonwords[i] for i in word_rhymes_indices]
        r = re.compile(stress_pattern)
        return [self.commonwords[i] for i in word_rhymes_indices if r.match(self.common_stresses[i])]
    
    def random_stress_rhymes(self,word,stress_pattern):
        return random.choice(self.search_stress_rhymes(word,stress_pattern))
    
    def meter_line(self,stress_pattern_list,rhymeword=None):
        """Generates a line of verse based on the stress patten, with an option to force a rhyme for the last word
        
        This is the 'meat' of the code, and is incredibly inefficient and lazy. If it can't find a work that fits,
        it discards everything and starts over. If it finds a word is repeats in a line, it throws it out and starts again too
        """
        nsyllables = len(stress_pattern_list)
        while True:
            nsyllables_so_far = 0
            words_syllables = []
            while nsyllables_so_far<nsyllables:
                words_syllables.append(random.randint(1,3))
                nsyllables_so_far+=words_syllables[-1]
            if nsyllables_so_far>nsyllables:
                words_syllables[-1]-=nsyllables_so_far-nsyllables
            nwords = len(words_syllables)
            word_syllable_indices = np.cumsum(words_syllables)[:-1]
            word_syllable_indices=np.insert(word_syllable_indices,0,0)
            word_stresses = ["".join(stress_pattern_list[word_syllable_indices[i]:word_syllable_indices[i]+words_syllables[i]])+"$" for i in range(nwords)]
            ngenerate = nwords-1 if rhymeword else nwords
            words = [self.random_stress(word_stresses[i]) for i in range(ngenerate)]
            if rhymeword:
                words.append(self.random_stress_rhymes(rhymeword,word_stresses[-1]))
            if len(set(words))==len(words):
                return words

    def couplet_stanza(self,ncouplets):
        metre = ["0","[12]","0","[12]","0","[12]","0","[12]","0","[12]"] # iambic pentameter
        icouplet = 0
        while icouplet<ncouplets:
            try:
                first_line_words = self.meter_line(metre)
                first_line_lastword = first_line_words[-1]
                second_line_words = self.meter_line(metre,rhymeword = first_line_lastword)
                words_to_line(first_line_words)
                words_to_line(second_line_words)
                icouplet+=1
            except IndexError:
                pass # try again

    def generic_stanza(self,rhyme_lines_list,meters,meter_pattern):
        nlines = len(meter_pattern)
        stanza_complete = False
        while not stanza_complete:
            try:
                stanza = []
                rhyme_words = [None for x in rhyme_lines_list]
                for iline in range(nlines):
                    line_meter = meters[meter_pattern[iline]]
                    this_rhyme_index = None
                    for irhyme,rhyme_lines in enumerate(rhyme_lines_list):
                        if iline in rhyme_lines:
                            this_rhyme_index = irhyme
                    if this_rhyme_index is not None:
                        if rhyme_words[this_rhyme_index]:
                            line_words = self.meter_line(line_meter,rhymeword=rhyme_words[this_rhyme_index])
                        else:
                            line_words = self.meter_line(line_meter)
                            rhyme_words[this_rhyme_index] = line_words[-1]
                    else:
                        line_words = self.meter_line(line_meter)
                    stanza.append(line_words)
                stanza_complete = True
            except IndexError:
#                 print("failed attempt")
                pass # try again
        for line_words in stanza:
            words_to_line(line_words)
            


common3000 = commonword_manager("commonwords_3000.txt")

iambic_pentameter = ["0","[12]","0","[12]","0","[12]","0","[12]","0","[12]"]
iambic_tetrameter = ["0","[12]","0","[12]","0","[12]","0","[12]"]
iambic_trimeter = ["0","[12]","0","[12]","0","[12]"]
mixitup_meter = ["0","[12]","0","[12]","0","0","[12]"]
double_dactyl = ["[12]","0","0","[12]","0","0"]
dactyl_end = ["[12]","0","0","[12]"]

# kids story
print("STORYBOOK")
for istanza in range(6):
    for icouplet in range(2):
        common3000.generic_stanza([[0,1]],[iambic_tetrameter],[0,0])
    print()

# double dactyl
print("DOUCLEDACTYL")
for istanza in range(6):
    common3000.generic_stanza([[3,7]],[double_dactyl,dactyl_end],[0,0,0,1,0,0,0,1])
    print()

# ballad
print("BALLAD")
for istanza in range(6):
    common3000.generic_stanza([[0,2],[1,3]],[iambic_tetrameter,iambic_trimeter],[0,1,0,1])
    print()

# sonnet
print("SONNETS")
for istanza in range(4):
    common3000.generic_stanza([[0,2],[1,3],[4,6],[5,7],[8,10],[9,11],[12,13]],[iambic_pentameter],[0,0,0,0,0,0,0,0,0,0,0,0,0,0])
    print()

# heroic verse
print("HEROIC VERSE")
for icouplets in range(20):
    common3000.generic_stanza([[0,1]],[iambic_pentameter],[0,0])