
from helpers_new import *
import_all()
import matplotlib.pyplot as plt
import tweepy as tw
plt.style.use('pinks.mplstyle')
%matplotlib inline


%%html
<style>
a:link {color: #f6b48f !important; font-weight: 600 !important;}
a:visited {color: #f6b48f !important; font-weight: 600 !important;}
</style>


client = tw.Client('')


query = '#fed -is:retweet lang: en'
tws = client.search_recent_tweets(query=query,
								  tweet_fields = ['context_annotations', 'created_at'],
								  max_results = 10)


for tweet in tws[0]:
	pretty(tweet.text, f'Tweet id: {tweet.id}')


dow_30_df = get_tweets('#dow30', 10)


API_KEY = ''
API_SECRET = ''
BEARER = ''
TOKEN = ''
TOKEN_SECRET = ''


auth = tw.OAuth1UserHandler(API_KEY, API_SECRET, TOKEN, TOKEN_SECRET)


api = tw.API(auth)


# tweet = api.update_status('Whaddup, World! Ignore everything I have ever said on Twitter before...unless you LIKE what I have said. This is the new DATA SCIENCE me. (Tweeted via Python and Tweepy. Tweep Tweep!)')


# tws = api.home_timeline(count = 5)

# for tweet in tws[0]:
# 	pretty(tweet.text, f'Tweet id: {tweet.id}')


# tws = api.user_timeline(count = 2)

# for tweet in tws[0]:
# 	pretty(tweet.text, f'Tweet id: {tweet.id}')


query = '#NYSE -is:retweet lang: en'
tws = client.search_recent_tweets(query=query,
								  tweet_fields = ['context_annotations', 
												  'created_at'],
								  max_results = 10)


for tweet in tws[0]:
	print(tweet.created_at)

2023-02-02 16:10:11+00:00
2023-02-02 16:10:07+00:00
2023-02-02 16:10:03+00:00
2023-02-02 16:09:53+00:00
2023-02-02 16:09:44+00:00
2023-02-02 16:09:35+00:00
2023-02-02 16:09:30+00:00
2023-02-02 16:09:27+00:00
2023-02-02 16:09:18+00:00
2023-02-02 16:09:02+00:00


for tweet in tws[0]:
	print(tweet.id)

1621179182184030208
1621179162739474433
1621179147446886401
1621179106179022854
1621179069936066566
1621179028437696512
1621179008225349633
1621178994866397191
1621178958292127744
1621178890344448003


for tweet in tws[0]:
	print(tweet.text)

$AMZN  new alert at https://t.co/A7qrDarJHY  #stocks #daytrading  #NYSE #NASDAQ #market 883
Spiking 🚀 Today's Acquisition
➡02 February 2023 : 
First American acquires 1031 Solutions 
#First #American #Solutions #NYSE #FAF 
To learn more, visit :https://t.co/NdENbVpvlo
$NEWT  new alert at https://t.co/A7qrDarJHY  #stocks #daytrading  #NYSE #NASDAQ #market 881
$AAPL  new alert at https://t.co/A7qrDarJHY  #stocks #daytrading  #NYSE #NASDAQ #market 879
$AAPL  new alert at https://t.co/A7qrDarJHY  #stocks #daytrading  #NYSE #NASDAQ #market 877
$MCO  new alert at https://t.co/A7qrDarJHY  #stocks #daytrading  #NYSE #NASDAQ #market 875
$BLNK Stock Alert: @BlinkCharging plan to battle naked short sellers! #ShortSqueeze 
https://t.co/Wb7nz6Yg72

#stockmarket #stocks #trading #investing #markets #EVcharging #EV #electricvehicles #EVchargers #ESG #SwingTrading #fintwit #nyse #dowjones #sp500 #NASDAQ #wallstreetbets https://t.co/CxKUcfuLAr
$XOM  new alert at https://t.co/A7qrDarJHY  #stocks #daytrading  #NYSE #NASDAQ #market 873
$BDN  new alert at https://t.co/A7qrDarJHY  #stocks #daytrading  #NYSE #NASDAQ #market 871
$GOOS  new alert at https://t.co/A7qrDarJHY  #stocks #daytrading  #NYSE #NASDAQ #market 867


# tweeting the tweet that will be removed
# bad_tweet = api.update_status('bad_tweet')

# removing the tweet
# api.destroy_status(bad_tweet.id)


# api.create_friendship(screen_name = '#YouTube')


# api.destroy_friendship(screen_name = '#YouTube')


client = tw.Client('AAAAAAAAAAAAAAAAAAAAAH4tlgEAAAAAJz3BHC1CYpWAtTrmvqMEc6ZHlqQ%3D1vpTrgfbZHqdQJlljVcso99m9rb4abITpUYbrAoyWiaISNGOLr')


query = '#stockmarket -is:retweet lang: en'
tws = client.search_recent_tweets(query=query,
								  tweet_fields = ['context_annotations', 
												  'created_at'],
								  max_results = 10)


for tweet in tws[0]:
	pretty(tweet.text, f'Tweet id: {tweet.id}')


import re
from nltk.sentiment.vader import SentimentIntensityAnalyzer
from textblob import TextBlob
import nltk
nltk.download('vader_lexicon')
from PIL import Image
from wordcloud import WordCloud, STOPWORDS

[nltk_data] Downloading package vader_lexicon to
[nltk_data]     /Users/evancarr/nltk_data...
[nltk_data]   Package vader_lexicon is already up-to-date!


query = '#apple -is:retweet lang:en'
paginator = tw.Paginator(client.search_recent_tweets, 
						 query = query,
						 max_results = 100,
						 limit = 10)


tweets = []
for tweet in paginator.flatten():
	tweets.append(tweet)


tweets_df = pd.DataFrame(tweets)


head_tail_vert(tweets_df, 3, '100 Imported Tweets')


def preprocess_tweet(sentence):
	sentence = sentence.lower()
	
	# remove RT
	sentence = re.sub('RT @\w+: ', " ", sentence)
	# remove punctuation and numbers
	sentence = re.sub("(@[A-Za-z0-9]+)|([^0-9A-Za-z \t])|(\w+:\/\/\s+)", ' ', sentence)
	# single character removal
	sentence = re.sub(r'\s+[a-zA-Z]\s+', ' ', sentence)
	#remove multiple spaces
	sentence = re.sub(r'\s+', ' ', sentence)
	sentence = sentence.replace('https', '')
	
	return sentence


cleaned_tweets = []
for tweet in tweets_df['text']:
	cleaned_tweet = preprocess_tweet(tweet)
	cleaned_tweets.append(cleaned_tweet)


tweets_df['cleaned'] = pd.DataFrame(cleaned_tweets)
tweets_df.drop(columns = ['edit_history_tweet_ids'], inplace = True)


head_tail_vert(tweets_df, 2, 'New Cleaned Column')


cleaned = list(tweets_df.cleaned.items())


cleaned[0][1]

' aapl apple earnings are today iphonesales will be key for the stock  co awyawakzpi via '


pretty(TextBlob(cleaned[0][1]).sentiment, 'TextBlob(cleaned[0][1]).sentiment')
polarity, subjectivity = TextBlob(cleaned[0][1]).sentiment
pretty(polarity, 'polarity')
pretty(subjectivity, 'subjectivity')


pretty(SentimentIntensityAnalyzer().polarity_scores(cleaned[0][1]),
	   'SentimentIntensityAnalyzer().polarity_scores(cleaned[0][1])')


def add_sentiment_columns(df, text_column):
	from nltk.sentiment.vader import SentimentIntensityAnalyzer
	from textblob import TextBlob
	
	df[['polarity', 
		   'subjectivity']] = df['cleaned']\
							  .apply(lambda Text: pd.Series(TextBlob(Text).sentiment))

	for idx, row in df[text_column].items():
		score = SentimentIntensityAnalyzer().polarity_scores(row)
		neg = score['neg']; neu = score['neu']
		pos = score['pos']; comp = score['compound']
		if comp <= -0.05:
			df.loc[idx, 'sentiment'] = 'negative'
		elif comp >= 0.05:
			df.loc[idx, 'sentiment'] = 'positive'
		else:
			df.loc[idx, 'sentiment'] = 'neutral'
			
		df.loc[idx, 'neg'] = neg
		df.loc[idx, 'neu'] = neu
		df.loc[idx, 'pos'] = pos
		df.loc[idx, 'compound'] = comp

	return df


tweets_df = add_sentiment_columns(tweets_df, 'cleaned')


head_tail_vert(tweets_df[['cleaned', 'polarity', 'subjectivity', 
						  'sentiment', 'neg', 'neu', 'pos', 'compound']], 
			   2, 
			   'DF with Sentiment Columns')


negative_tweets = tweets_df[tweets_df.sentiment == 'negative']
positive_tweets = tweets_df[tweets_df.sentiment == 'positive']
neutral_tweets = tweets_df[tweets_df.sentiment == 'neutral']


sentiments_df = pd.DataFrame(tweets_df.sentiment.value_counts())
sentiments_df['percentage'] = sentiments_df.sentiment / len(tweets_df) * 100


see(sentiments_df, 'Sentiments DataFrame with Percentages')


labels = sentiments_df.index
percentages = sentiments_df['percentage']

donut = plt.Circle((0,0), 0.7, color = '#35193e')
plt.pie(percentages, 
		labels = labels,
	    colors = ['C0', 'C1', 'C2'])
plt.title('Sentiment Distribution')
plotster = plt.gcf()
plotster.gca().add_artist(donut)

<matplotlib.patches.Circle at 0x7faf69cbeef0>


def word_cloud_it(text, title):
	mask = np.array(Image.open('./twitter_materials_skillcate/a1_inputs/cloud.png'))
	stopwords = set(STOPWORDS)
	word_cloud = WordCloud(background_color = 'white',
						  mask = mask,
						  max_words = 100,
						  stopwords = stopwords,
						  repeat = True)
	word_cloud.generate(str(text))
	word_cloud.to_file('my_word_cloud.png')
	pretty('Your WordCloud has been successfully saved.')
	header_text(title)
	display(Image.open('my_word_cloud.png'))


word_cloud_it(tweets_df.cleaned.values, 'WordCloud: All Tweets')


word_cloud_it(positive_tweets.cleaned.values, 'WordCloud: Postive Tweets')


word_cloud_it(negative_tweets.cleaned.values, 'WordCloud: Negative Tweets')


tweets_df['characters'] = tweets_df['cleaned'].astype(str).apply(len)
tweets_df['num_words'] = tweets_df['cleaned'].apply(lambda x: len(str(x).split()))


head_tail_vert(tweets_df[['cleaned', 'characters', 'num_words']], 3, 'Added Length Columns')


see(pd.DataFrame(tweets_df.groupby('sentiment').characters.mean()),
	'Average Number of Characters by Sentiment')


see(pd.DataFrame(tweets_df.groupby('sentiment').num_words.mean()),
	'Average Number of Words by Sentiment')


tweets_df.to_csv('tweets_df.csv')

tweet_id	tweet_text
1621163221871910912	#DOW30 (Indices Future) : Live example on T.F. 15m, demark 9, supertrend, LRegr.Forec., Interc., Slope, RSquar., Supp and Resist. Lorenz attractor area, Pivot and Target price, and new Open Interest Long (AreaGreen) Short (AreaRed). https://t.co/4jpETboyt9
1621055322902372352	At December’s meeting, bolstering optimism that #inflation is cooling, but still giving no indication of an upcoming pause in #RateHikes. The #SP500 increased 42.61pts (+1.05%) to 4119, the #Nasdaq100 jumped 261.17pts(+2.16%) to 12363 & the #Dow30 rose 6.92pts (+0.02%) to 34092. https://t.co/PwSsAItmnW
1620859477389107200	#DOW30 (Indices Future) : Live example on T.F. 15m, demark 9, supertrend, LRegr.Forec., Interc., Slope, RSquar., Supp and Resist. Lorenz attractor area, Pivot and Target price, and new Open Interest Long (AreaGreen) Short (AreaRed). https://t.co/VeEUDbNPiK
1620782852874051584	#DOW30 (Indices Future) : Live example on T.F. 15m, demark 9, supertrend, LRegr.Forec., Interc., Slope, RSquar., Supp and Resist. Lorenz attractor area, Pivot and Target price, and new Open Interest Long (AreaGreen) Short (AreaRed). https://t.co/AktFlN4eyt
1620710422495236096	February 01/2023, #StockMarketNews: On Tuesday, #USIndices closed higher as the #SP500 rose 58.83pts (+1.46%) to 4076, the #Nasdaq100 increased 189.54pts (+1.59%) to 12101 and the #Dow30 gained 368.95pts (+1.09%) to 34086.#StocksToTrade #stocksinnews #TradingOpportunity #trade https://t.co/LQn7nq06aG
1620491934023442432	#DOW30 (Indices Future) : Live example on T.F. 15m, demark 9, supertrend, LRegr.Forec., Interc., Slope, RSquar., Supp and Resist. Lorenz attractor area, Pivot and Target price, and new Open Interest Long (AreaGreen) Short (AreaRed). https://t.co/6mI2v5e1mx
1620453054708596736	#DOW30 (Indices Future) : Live example on T.F. 15m, demark 9, supertrend, LRegr.Forec., Interc., Slope, RSquar., Supp and Resist. Lorenz attractor area, Pivot and Target price, and new Open Interest Long (AreaGreen) Short (AreaRed). https://t.co/copNjMdcFs
1620135138456928256	#DOW30 (Indices Future) : Live example on T.F. 15m, demark 9, supertrend, LRegr.Forec., Interc., Slope, RSquar., Supp and Resist. Lorenz attractor area, Pivot and Target price, and new Open Interest Long (AreaGreen) Short (AreaRed). https://t.co/D0UCvYel3R
1620088967730462720	Drugmaker Amgen $AMGN said on #Monday that it let go ~300 U.S. employees or about 1.2% of its total workforce, citing recent organizational changes to its commercial team. $DIA $SPY $QQQ #Dow30 #stocks #SwingTrading #Mondayvibes $IBB $XBI $ATOS $INM https://t.co/5zFYSARkES
1620061074296045568	#DOW30 (Indices Future) : Live example on T.F. 15m, demark 9, supertrend, LRegr.Forec., Interc., Slope, RSquar., Supp and Resist. Lorenz attractor area, Pivot and Target price, and new Open Interest Long (AreaGreen) Short (AreaRed). https://t.co/3U75IjUh85

	edit_history_tweet_ids	id	text
0	[1621178996493885440]	1621178996493885440	$AAPL #Apple’s Earnings Are Today. #IPhoneSales Will Be Key for the Stock. https://t.co/aWyAWAkZpi via @BarronsOnline
1	[1621178787592536064]	1621178787592536064	#Apple should be required to let #iPhone users sideload apps, #Biden admin says. That's probably one of the few things, the @GOP somewhat agrees on.😉 https://t.co/CdBS3fa1fR
2	[1621178669317201922]	1621178669317201922	Investors Cheer Meta’s Slimmed-Down Ambitions\n\n#Activision #Adani #Adani #Alphabet #Amazon.com #Apple #Bank #Berkshire #Blackstone #European #Strive #Meta https://t.co/Eitl5wiQYf

	edit_history_tweet_ids	id	text
996	[1620955902705639424]	1620955902705639424	Apple's ane-transformers - experiences? #nlp #apple https://t.co/OpJavNehzt
997	[1620955803825094657]	1620955803825094657	Is there any Hacking attempts on your device? DM now for assistance and help. #tech #technology #iphone #technews #gadgets #apple #innovation #android #business #smartphone #samsung #pro #techno #instatech #programming #engineering #electronics #coding #gadget #computer #gaming
998	[1620955719456391168]	1620955719456391168	" Everybody loves $rndr " \n\n#ai #metaverse #gpu #top10crypto #apple https://t.co/rohD5jBLwz

	id	text	cleaned
0	1621178996493885440	$AAPL #Apple’s Earnings Are Today. #IPhoneSales Will Be Key for the Stock. https://t.co/aWyAWAkZpi via @BarronsOnline	aapl apple earnings are today iphonesales will be key for the stock co awyawakzpi via
1	1621178787592536064	#Apple should be required to let #iPhone users sideload apps, #Biden admin says. That's probably one of the few things, the @GOP somewhat agrees on.😉 https://t.co/CdBS3fa1fR	apple should be required to let iphone users sideload apps biden admin says that probably one of the few things the somewhat agrees on co cdbs3fa1fr

	id	text	cleaned
997	1620955803825094657	Is there any Hacking attempts on your device? DM now for assistance and help. #tech #technology #iphone #technews #gadgets #apple #innovation #android #business #smartphone #samsung #pro #techno #instatech #programming #engineering #electronics #coding #gadget #computer #gaming	is there any hacking attempts on your device dm now for assistance and help tech technology iphone technews gadgets apple innovation android business smartphone samsung pro techno instatech programming engineering electronics coding gadget computer gaming
998	1620955719456391168	" Everybody loves $rndr " \n\n#ai #metaverse #gpu #top10crypto #apple https://t.co/rohD5jBLwz	everybody loves rndr ai metaverse gpu top10crypto apple co rohd5jblwz

	cleaned	polarity	subjectivity	sentiment	neg	neu	pos	compound
0	aapl apple earnings are today iphonesales will be key for the stock co awyawakzpi via	0.00	1.00	neutral	0.00	1.00	0.00	0.00
1	apple should be required to let iphone users sideload apps biden admin says that probably one of the few things the somewhat agrees on co cdbs3fa1fr	-0.20	0.10	positive	0.00	0.94	0.06	0.13

	cleaned	characters	num_words
0	aapl apple earnings are today iphonesales will be key for the stock co awyawakzpi via	88	15
1	apple should be required to let iphone users sideload apps biden admin says that probably one of the few things the somewhat agrees on co cdbs3fa1fr	150	26
2	investors cheer meta slimmed down ambitions activision adani adani alphabet amazon com apple bank berkshire blackstone european strive meta co eitl5wiqyf	154	21

	cleaned	characters	num_words
996	apple ane transformers experiences nlp apple co opjavnehzt	59	8
997	is there any hacking attempts on your device dm now for assistance and help tech technology iphone technews gadgets apple innovation android business smartphone samsung pro techno instatech programming engineering electronics coding gadget computer gaming	255	35
998	everybody loves rndr ai metaverse gpu top10crypto apple co rohd5jblwz	71	10

Tweet id: 1621179092468142080
👨‍⚕️👷‍♀️NFP PREVIEW VIDEO 👷‍♀️👨‍⚕️ See the key figures you need to watch ahead of the #NFP report, including what it might mean for the #Fed. Our full preview report has potential trade setups in the #FX market: https://t.co/jN2VpXn0zB This video is not intended for the US audience. https://t.co/mrFXUjvlyD

Tweet id: 1621178305490866178
📊 The #Fed increased interest rates by 25 bp & indicated a need to tighten monetary policy further in its effort to combat #inflation. The #FOMC raised its rate to a range of 4.5% to 4.75%, the highest level since Oct 07. #LeverageShares Capital at risk. https://t.co/I0SYnhswwe

Tweet id: 1621178136758042629
The #FederalReserve and the #EuropeanCentralBank have raised their interest rates. 📊 The #FED raised rates a quarter point, from 4.5% to 4.75%, and the #ECB hiked interest rates by 0.5%. https://t.co/g7pPyZVlgD

Tweet id: 1621178059675009024
"US \| Fed slows to 25bp hike and recognizes that “the…" - BBVA #fed https://t.co/MfMlm6HWjw