for i in range(how_many):
sentences = []
names = []
genres = []
mechanics = []
for line, source in corpus.assemble("0.l"):
if no_punctuation_at_end.search(line):
line += "."
sentences.append(line)
names.append(source['name'])
genres.append([genre for id, genre in source.get('boardgamecategory', [])])
mechanics.append([mechanic for id, mechanic in source.get('boardgamemechanic', [])])
# Make assemblers for single- and multi-word names.
single_word_assembler = WordAssembler()
multi_word_assembler = Assembler()
# Create a composite assembler that will choose single- and
# multi-word names in appropriate proportion.
name_assembler = CompositeAssembler([single_word_assembler, multi_word_assembler])
for name in names:
words = whitespace.split(name)
if len(words) == 1:
single_word_assembler.add(name)
else:
multi_word_assembler.add(words)
assembler, choice = name_assembler.assemble()
if assembler == single_word_assembler:
separator = ''
else:
import json
from queneau import WordAssembler
from corpus import Corpus
import textwrap
assembler = WordAssembler(Corpus.load("dinosaurs"))
dinos = []
for i in range(2):
dino = assembler.assemble_word()
if dino[0] in 'AEIO':
dino = "an " + dino
else:
dino = "a " + dino
dinos.append(dino)
print "Look! Behind that ridge! It's %s fighting %s!" % tuple(dinos)
import json
from queneau import WordAssembler
from corpus import Corpus
import textwrap
c = Corpus.load("toxic_chemicals")
assembler = WordAssembler(c['chemicals'])
for i in range(50):
words = []
for j in range(2):
b = assembler.assemble_word()
words.append(b)
print("%s %s"%(tuple(words)))
import json
from queneau import WordAssembler
from data import load_json
import textwrap
assembler = WordAssembler(load_json("dinosaurs.json"))
dinos = []
for i in range(2):
dino = assembler.assemble_word()
if dino[0] in 'AEIO':
dino = "an " + dino
else:
dino = "a " + dino
dinos.append(dino)
print "Look! Behind that ridge! It's %s fighting %s!" % tuple(dinos)
from queneau import WordAssembler
import nltk
from nltk.tokenize import WordPunctTokenizer
from textblob.tokenizers import word_tokenize
poem = 'data/poems/book01.txt'
with open(poem,'rb') as f:
raw = f.read()
### tokenizer = WordPunctTokenizer()
### tokens = tokenizer.tokenize(raw)
### tokens = nltk.word_tokenize(raw)
### text = nltk.Text(tokens)
tokens = list(word_tokenize(raw))
words = [w.lower() for w in tokens]
vocab = sorted(set(words))
vocab = vocab[15:] # cut out the punctuation
corpus = WordAssembler(vocab)
for i in range(1000):
print corpus.assemble_word(min_length=5)
import json
from queneau import WordAssembler
import textwrap
common_corpus = WordAssembler(json.load(
open("data/english_words.common.json")))
full_corpus = WordAssembler(json.load(open("data/english_words.json")))
print 'You know "%s", "%s", and "%s".' % tuple(common_corpus.assemble_word()
for i in range(3))
print 'But have you heard of "%s", "%s", or "%s"?' % tuple(
full_corpus.assemble_word() for i in range(3))
import json
from queneau import WordAssembler
from corpus import Corpus
import textwrap
common_corpus = WordAssembler(Corpus.load("english_words"))
full_corpus = WordAssembler(Corpus.load("english_words"))
print 'You know "%s", "%s", and "%s".' % tuple(common_corpus.assemble_word()
for i in range(3))
print 'But have you heard of "%s", "%s", or "%s"?' % tuple(
full_corpus.assemble_word() for i in range(3))
from queneau import Assembler, WordAssembler
import textwrap
corpus = Assembler.load(open("data/minor_planets.min.json"),
tokens_in='citation')
how_many = 100
for i in range(how_many):
sentences = []
names = []
for sentence, source in corpus.assemble("f.l", min_length=3):
sentences.append(sentence)
names.append(source['name'])
# Make a new assembler from the names of the asteroids that were chosen.
name_assembler = WordAssembler(names)
name = name_assembler.assemble_word()
print name
for s in textwrap.wrap(" ".join(sentences)):
print s
if i < how_many - 1:
print
import random
import json
from queneau import WordAssembler
from corpus import Corpus
import textwrap
c = Corpus.load("cannabis")
assembler = WordAssembler(c)
verbs = ["Puff on that",
"Puff the",
"Smoke that",
"Smokin on the",
"Smoke on that",
"Smoke on this",
"Hit the",
"Take a toke of the",
"Pass the",
"Hit the",
"Hit this",
"Blaze up that",
"Blaze on that",
"Fire up that",
"Light up the",
"Blaze some"]
for _ in range(25):
print("%s %s"%(verbs[random.randint(0,len(verbs)-1)], assembler.assemble_word()))
from queneau import Assembler, WordAssembler
import textwrap
from corpus import Corpus
corpus = Assembler.loadlist(Corpus.load("minor_planets"),
tokens_in='citation')
how_many = 100
for i in range(how_many):
sentences = []
names = []
for sentence, source in corpus.assemble("f.l", min_length=3):
sentences.append(sentence)
names.append(source['name'])
# Make a new assembler from the names of the asteroids that were chosen.
name_assembler = WordAssembler(names)
name = name_assembler.assemble_word()
print(name)
for s in textwrap.wrap(" ".join(sentences)):
print(s)
if i < how_many-1:
print()
import json
from queneau import WordAssembler
import textwrap
common_corpus = WordAssembler(json.load(open("data/english_words.common.json")))
full_corpus = WordAssembler(json.load(open("data/english_words.json")))
print 'You know "%s", "%s", and "%s".' % tuple(common_corpus.assemble_word() for i in range(3))
print 'But have you heard of "%s", "%s", or "%s"?' % tuple(full_corpus.assemble_word() for i in range(3))
import json
from queneau import WordAssembler
from corpus import Corpus
import textwrap
c = Corpus.load("nsa_projects")
assembler = WordAssembler(c['codenames'])
for i in range(50):
words = []
for j in range(2):
b = assembler.assemble_word(pattern="011.", length=5)
words.append(b)
print("%s has been replaced by %s"%(tuple(words)))
for i in range(how_many):
sentences = []
names = []
genres = []
mechanics = []
for line, source in corpus.assemble("0.l"):
if no_punctuation_at_end.search(line):
line += "."
sentences.append(line)
names.append(source['name'])
genres.append([genre for id, genre in source.get('boardgamecategory', [])])
mechanics.append([mechanic for id, mechanic in source.get('boardgamemechanic', [])])
# Make assemblers for single- and multi-word names.
single_word_assembler = WordAssembler()
multi_word_assembler = Assembler()
# Create a composite assembler that will choose single- and
# multi-word names in appropriate proportion.
name_assembler = CompositeAssembler([single_word_assembler, multi_word_assembler])
for name in names:
words = whitespace.split(name)
if len(words) == 1:
single_word_assembler.add(name)
else:
multi_word_assembler.add(words)
assembler, choice = name_assembler.assemble()
if assembler == single_word_assembler:
separator = ''
else: