def execute(self):
numberOfPages = 1
numberOfTweetsPerPage = 200
counter = 0
cloud = ""
txt = ""
if self.numberOfTweets > 200:
numberOfPages = ceil(self.numberOfTweets / 200)
else:
numberOfTweetsPerPage = self.numberOfTweets
for i in range(numberOfPages):
tweets = self.api.user_timeline(screen_name=self.username,
count=numberOfTweetsPerPage,
page=i)
for each in tweets:
cloud = each.text
cloud = re.sub(r'[A-Za-z@_]*', '', cloud)
counter += 1
txt = txt + ' ' + each.text
txt = re.sub(r'[A-Za-z@]*', '', txt)
twitter_mask = np.array(
Image.open(path.join(self.d, "templates/cloud/twitter-logo.jpg")))
stop = [
'می', 'من', 'که', 'به', 'رو', 'از', 'ولی', 'با', 'یه', 'این',
'نمی', 'هم', 'شد', 'ها', 'اما', 'تو', 'واقعا', 'در', 'نه', 'دارم',
'باید', 'آره', 'برای', 'تا', 'چه', 'کنم', 'بود', 'همه', 'دیگه',
'ای', 'اون', 'تی', 'حالا', 'بی', 'د', 'چرا', 'بابا', 'منم', 'کیه',
'توی', 'نیست', 'چی', 'باشه', 'که', 'بودم', 'می کنم', 'که', 'اینه',
'بهتر', 'داره', 'اینه', 'که', 'کردن', 'می', 'کن', 'بعد', 'دیگه',
'', '', '', ''
]
wc = WordCloudFa(
# font_path='IranNastaliq.ttf',
persian_normalize=True,
max_words=1000,
margin=0,
width=3000,
height=2500,
min_font_size=1,
max_font_size=1000,
background_color=self.backGround,
mask=twitter_mask,
include_numbers=False,
collocations=False)
wc.add_stop_words(stop)
wc.generate(txt)
directory = 'static/images/' + self.username + '.png'
directory = path.join(self.d, directory)
image = wc.to_image()
image.save(directory)
def get_image(UserID, chat_id, message, bot, update):
if len(UserID) < 2:
message(chat_id=chat_id.chat_id, text="آی دی نامعتبر است!")
else:
message(chat_id=chat_id.chat_id, text="در حال اتصال به اینستاگرام...")
allword = query.start(UserID, chat_id, message, bot, update)
allword_edited = removeWeirdChars(allword)
my_wordcloud = WordCloudFa(font_path="Sahel.ttf",
background_color="white",
width=720,
height=1280,
margin=2).generate(allword_edited)
image = my_wordcloud.to_image()
saved_dir = 'analysis/' + str(UserID) + '.jpg'
image.save(saved_dir)
message(chat_id=chat_id.chat_id, text="درحال ارسال عکس...")
return saved_dir
import numpy as np
d = path.dirname(__file__)
text = open(path.join(d, 'tweets/result.txt'), encoding='utf-8').read()
# Add another stopword
twitter_mask = np.array(Image.open(path.join(d, "input/southpark1.png")))
stopwords = set(STOPWORDS)
stopwords |= EN_STOPWORDS
# Generate a word cloud image
wordcloud = WordCloudFa(persian_normalize=True,
include_numbers=False,
max_words=200,
stopwords=stopwords,
margin=0,
width=3000,
height=3000,
min_font_size=10,
max_font_size=2300,
random_state=True,
background_color="black",
mask=twitter_mask).generate(text)
image = wordcloud.to_image()
image.show()
image.save('output/twitter_mask.png')
def make_wordcloud(font_path=None,
text_path='tweets/result.txt',
bw_img="input/southpark1.png",
img="input/southpark2.png",
add_stopwords=ADD_STOPWORDS,
bg_color='black',
include_numbers=True,
max_words=500,
random_color=False):
assert type(add_stopwords) == type(list())
# get data directory (using getcwd() is needed to support running example in generated IPython notebook)
d = path.dirname(__file__) if "__file__" in locals() else getcwd()
# load text
text = open(path.join(d, text_path), encoding='utf-8').read()
# load image. This has been modified in gimp to be brighter and have more saturation.
image = np.array(Image.open(path.join(d, bw_img)))
# subsample by factor of 3. Very lossy but for a wordcloud we don't really care.
mask_color = np.array(Image.open(path.join(d, img)))
# create mask white is "masked out"
twitter_mask = image.copy()
twitter_mask[twitter_mask.sum(axis=2) == 0] = 255
# some finesse: we enforce boundaries between colors so they get less washed out.
# For that we do some edge detection in the image
edges = np.mean([
gaussian_gradient_magnitude(mask_color[:, :, i] / 255., 2)
for i in range(3)
],
axis=0)
twitter_mask[edges > .01] = 255
# Add another stopword
stop_words = stopwords.words('english')
##stop_words_fa = stopwords.words('farsi')
for word in add_stopwords:
STOPWORDS.add(Normalizer().normalize(word))
stop_words.extend(STOPWORDS)
stop_words.extend(EN_STOPWORDS)
stop_words = set(stop_words)
# Getting rid of the stopwords
text_list = [word for word in text.split() if word not in stop_words]
# Converting the list to a text
text = ' '.join([str(elem) for elem in text_list])
text.replace('\u200c', '')
# Generate a word cloud image
wordcloud = WordCloudFa(font_path=font_path,
persian_normalize=True,
include_numbers=include_numbers,
max_words=max_words,
stopwords=stop_words,
margin=0,
width=3000,
height=3000,
min_font_size=1,
max_font_size=2300,
random_state=True,
background_color=bg_color,
mask=twitter_mask,
relative_scaling=0,
repeat=True).generate(text)
if not random_color:
image_colors = ImageColorGenerator(mask_color)
wordcloud.recolor(color_func=image_colors)
image = wordcloud.to_image()
image.show()
image.save('output/twitter_mask.png')
text = get_tweets_from_user(
username) # to get tweets of a specific user by its username
break
else:
print("you should enter f or u!")
text = get_tweets(text)
text = remove_bad_tweets(text)
text = "\n".join(text)
text = get_words(text)
print(len(text))
text = remove_bad_words(text)
print(len(text))
text1 = "\n".join(text)
text1 = removeWeirdChars(text1)
mask_array = np.array(Image.open(mask_path))
my_wc = WordCloudFa(width=1200,
height=1200,
background_color=background_color,
mask=mask_array,
persian_normalize=True,
repeat=False,
collocations=True)
my_wc.add_stop_words_from_file("../stop_words_kian.txt")
open("edited_tweets.txt", "w").write(text1)
my_wc.generate(text1)
image = my_wc.to_image()
image.show()
filename = datetime.now().strftime("%Y-%m-%d-%H-%M")
image.save('Images/{time}_photo.png'.format(time=filename))
counter += 1
txt = txt + ' ' + each.text
print(counter, cloud)
txt = re.sub(r'[A-Za-z@]*', '', txt)
twitter_mask = np.array(Image.open(path.join(d, "twitter-logo.jpg")))
stop = [
'می', 'من', 'که', 'به', 'رو', 'از', 'ولی', 'با', 'یه', 'این', 'نمی', 'هم',
'شد', 'ها', 'اما', 'تو', 'واقعا', 'در', 'نه', 'دارم', 'باید', 'آره',
'برای', 'تا', 'چه', 'کنم', 'بود', 'همه', 'دیگه', 'ای', 'اون', 'تی', 'حالا',
'بی', 'د', 'چرا', 'بابا', 'منم', 'کیه', 'توی', 'نیست', 'چی', 'باشه', 'که',
'بودم', 'می کنم', 'که', 'اینه', 'بهتر', 'داره', 'اینه', 'که'
]
wc = WordCloudFa(persian_normalize=True,
max_words=30000,
margin=0,
width=3000,
height=2500,
min_font_size=1,
max_font_size=3000,
background_color="white",
mask=twitter_mask,
include_numbers=False,
stopwords=stop).generate(txt)
image = wc.to_image()
image.show()
image.save('twitter.png')