def comment_analyse(bv):
url = 'https://www.bilibili.com/video/' + bv
comments = get_comment(url)
title = get_video_info(bv)['data']['title']
comment_words = get_words(comments)
comment_words_img_url, word_counts_top10 = word_cloud(comment_words, title)
return render_template('comment_analyse.html', url=url, title='分析',
comment_words_img_url=comment_words_img_url,
word_counts_top10=word_counts_top10, video_title=title,bv=bv)
def bullet_analyse(bv):
url = 'https://www.bilibili.com/video/' + bv
bullets = get_bullet_chat(url)
title = get_video_info(bv)['data']['title']
bullet_words = get_words(bullets)
bullet_words_img_url, word_counts_top10 = word_cloud(bullet_words, title)
bullets_img_url, bullet_counts_top10 = word_cloud(bullets, title)
return render_template('bullet_analyse.html', url=url, title='分析',
bullet_words_img_url=bullet_words_img_url,
word_counts_top10=word_counts_top10, bullet_counts_top10=bullet_counts_top10,
bullets_img_url=bullets_img_url, video_title=title,bv=bv)
def get_rhymes(word):
word = prepare_string(word).strip()
assert type(word) == str
assert len(word) >= 1
words = get_words()
rhymes = []
max_ending_size = min(4, len(word))
max_ending = word[-max_ending_size:]
smaller_ending_size = max(1, max_ending_size - 1)
smaller_ending = word[-smaller_ending_size:]
for candidate_word in words:
candidate_word_lower = prepare_string(candidate_word)
if candidate_word_lower == word:
# Discard, because we don't want to include the search word in the results
continue
if candidate_word_lower.endswith(smaller_ending):
rhyme = {"word": candidate_word, "score": 100}
if candidate_word_lower.endswith(max_ending):
rhyme["score"] *= 2
if candidate_word.lower().endswith(word):
rhyme["score"] *= 0.3
rhymes.append(rhyme)
rhymes.sort(key=lambda rhyme: rhyme["score"], reverse=True)
# Limit the number of words in the result
rhymes = rhymes[0:500]
for rhyme in rhymes:
rhyme["num_syllables"] = count_syllables(rhyme["word"])
if rhyme["word"][0] == rhyme["word"][0].upper():
# If the first letter is uppercase, it's probably an "egennamn". In that
# case, we link to Wikipedia
rhyme["url"] = "https://no.wikipedia.org/wiki/" + quote(
rhyme["word"])
else:
rhyme["url"] = ("https://ordbok.uib.no/perl/ordbok.cgi?OPP=" +
quote(rhyme["word"]) +
"&ant_nynorsk=5&nynorsk=+&ordbok=nynorsk")
return rhymes
def get_random_word():
words = get_words()
chosen_word = random.choice(words)
return chosen_word
def test_get_words(self):
words = get_words()
self.assertGreater(len(words), 400000)