andyjk15/tweet-collector
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[14.10.20] Testing

Browse Source
This commit is contained in:
andrewso 2020-10-14 09:48:40 +01:00
parent ab669a1ed7
commit 8e15f8c9f8
1 changed files with 36 additions and 44 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

80
src/tweets/collector.py
View File

@ -23,6 +23,7 @@ from urllib3.exceptions import ProtocolError
hourStack = [] hourStack = []
dumpStack = [] dumpStack = []
processStack = []
class keys(): class keys():
@ -67,7 +68,7 @@ class Streamer():
while True: while True:
try: try:
stream = Stream(auth=api.auth, listener=listener, tweet_mode='extended') stream = Stream(auth, listener=listener, tweet_mode='extended')
stream.filter(languages=["en"], track=hashtag) stream.filter(languages=["en"], track=hashtag)
except IncompleteRead: except IncompleteRead:
log("Incomplete Read Error", 'ERR') log("Incomplete Read Error", 'ERR')
@ -75,9 +76,6 @@ class Streamer():
except ProtocolError: except ProtocolError:
log("Protocol Error", 'ERR') log("Protocol Error", 'ERR')
continue continue
except:
log("Some other Error", 'ERR')
continue
class Listener(StreamListener): class Listener(StreamListener):
@ -91,73 +89,63 @@ class Listener(StreamListener):
if (time() - self.start_time) < self.limit: if (time() - self.start_time) < self.limit:
data = json.loads(data) data = json.loads(data)
log(len(dumpStack), 'INFO')
# Check if tweet is a retweet # Check if tweet is a retweet
if 'retweeted_status' in data: if 'retweeted_status' in data:
if 'extended_tweet' in data['retweeted_status']: if 'extended_tweet' in data['retweeted_status']:
#if tweet is over the 140 word limit #if tweet is over the 140 word limit
text = data['retweeted_status']['extended_tweet']['full_text'] text = data['retweeted_status']['extended_tweet']['full_text']
processTweet(text, self.hashtag) dumpStack.append({'type': self.hashtag, 'tweet': text})
# dumpStack.append({'type': self.hashtag, 'tweet': text})
else: else:
text = data['retweeted_status']['text'] text = data['retweeted_status']['text']
processTweet(text, self.hashtag) dumpStack.append({'type': self.hashtag, 'tweet': text})
# dumpStack.append({'type': self.hashtag, 'tweet': text})
else: else:
# Else if a normal Tweeet # Else if a normal Tweeet
if 'extended_tweet' in data: if 'extended_tweet' in data:
# If tweet is over 140 word limit # If tweet is over 140 word limit
text = data['extended_tweet']['full_text'] text = data['extended_tweet']['full_text']
processTweet(text, self.hashtag) dumpStack.append({'type': self.hashtag, 'tweet': text})
# dumpStack.append({'type': self.hashtag, 'tweet': text})
else:
processTweet(data["text"], self.hashtag)
def processTweet(text, type): def processTweet():
log("hourStack size :: [{}]".format(len(hourStack)), "INFO") processStack = dumpStack.copy()
dumpStack.clear()
log("Processing [{}] Tweet...".format(text), 'INFO') # log("Processing [{}] Tweet...".format(text), 'INFO')
removedLines = fixLines(text) if len(processStack) != 0:
log("fixLines", 'DEBUG') for tweet in processStack:
removedSpecialChars = cleanTweet(removedLines) removedLines = fixLines(str(tweet))
log("cleanTweet", 'DEBUG')
removedSpacing = removeSpacing(removedSpecialChars[0])
log("removeSpacing", 'DEBUG')
tweetLength = checkLength(removedSpacing) removedSpecialChars = cleanTweet(removedLines)
log("checkLength", 'DEBUG') removedSpacing = removeSpacing(removedSpecialChars[0])
if tweetLength == True:
checkIfEnglish = detectLaguage(removedSpecialChars[0]) tweetLength = checkLength(removedSpacing)
log("Boop2", 'DEBUG') if tweetLength == True:
if checkIfEnglish == True: checkIfEnglish = detectLaguage(removedSpecialChars[0])
tweetText = remove_non_ascii(removedSpacing) if checkIfEnglish == True:
log("Boop3", 'DEBUG')
# log("Cleaned Tweet: {}".format(tweetText), 'INFO') tweetText = remove_non_ascii(removedSpacing)
cleanedTweet = tweetText + ' ' + removedSpecialChars[1] # log("Cleaned Tweet: {}".format(tweetText), 'INFO')
if callSpamFilter(cleanedTweet) != 'spam': cleanedTweet = tweetText + ' ' + removedSpecialChars[1]
log("After spam call", 'DEBUG')
pos, neu, neg, compound = callSentimentAnalyser(cleanedTweet) if callSpamFilter(cleanedTweet) != 'spam':
log("After Sentiment Call", 'DEBUG') pos, neu, neg, compound = callSentimentAnalyser(cleanedTweet)
if compound != 0.0 | neu <= 0.8: if compound != 0.0 or neu <= 0.8:
log("if not 0.0 or above 0.8", 'DEBUG') hourTweet = {'pos': 1, 'neu': 1, 'neg': 1, 'compound': 1, 'type': ""}
hourTweet = {'pos': pos, 'neu': neu, 'neg': neg, 'compound': compound, 'type': type} hourStack.append(hourTweet)
processStack.clear()
hourStack.append(hourTweet) else:
# processStack.clear() log("Dump Stack was Empty", 'WARN')
# else:
# log("Dump Stack was Empty", 'WARN')
def collector(hashtag): def collector(hashtag):
@ -183,11 +171,15 @@ def createHourJob():
log("Creating hour job...", 'INFO') log("Creating hour job...", 'INFO')
schedule.clear("sendToArtemis") schedule.clear("sendToArtemis")
ovPos, ovNeu, ovNeg, ovCompound = 0, 0, 0, 0 ovPos, ovNeu, ovNeg, ovCompound = 0, 0, 0, 0
type = ""
global timeF global timeF
timeF = timeFunction() timeF = timeFunction()
processTweet()
processStack = hourStack.copy() processStack = hourStack.copy()
hourStack.clear()
log("Extracting sentiment scores...", 'INFO') log("Extracting sentiment scores...", 'INFO')
@ -208,7 +200,7 @@ def createHourJob():
if type == "bitcoin": if type == "bitcoin":
type = 'btc_usd' type = 'btc_usd'
hourStack.clear() processStack.clear()
sendToArtemis(pos, neu, neg, compound, type) sendToArtemis(pos, neu, neg, compound, type)
else: else: