class SimpleSentimentAnalyzerTest(unittest.TestCase):
def setUp(self):
self.dc = SimpleSentimentAnalyzer()
self.pos_tweet = 'Congrats SolarCitys discount approach to going public pays off'
self.neg_tweet = 'Done with wack ass finals!!!'
def test_get_tweet_text_mood(self):
result_pos = self.dc.get_tweet_text_mood(self.pos_tweet)
self.assertEqual(2,result_pos)
result_neg = self.dc.get_tweet_text_mood(self.neg_tweet)
self.assertEqual(-4,result_neg)
def setUp(self):
self.dc = SimpleSentimentAnalyzer()
self.pos_tweet = 'Congrats SolarCitys discount approach to going public pays off'
self.neg_tweet = 'Done with wack ass finals!!!'
def index(request):
'''
Renders the view "index.html" - main webpage of web application.
Collects data to show in view. DataCollector is called to get the data
to show and analyze. SimpleSentimentAnalyzer is called to analyze the collected
tweets. The statistics are collected and passed to view.
'''
errors = []
tweets_to_show = [] # tweet' fields that are rendered on the page
texts_to_analyse = [] # texts of tweets that are analysed for sentiment
dc = DataCollector()
region = 'Anker Engelunds Vej 1, Denmark' # default region
radius = 15 # default radius
user_search = 0
if 'q' not in request.GET: # first start of website
context = {'first_map': region,
'errors':errors,
'user_search':user_search}
return render(request, 'twitter_region_sentiment/index.html', context)
elif 'q' in request.GET and request.GET['q'] != '': # normal behaviour
region = request.GET['q']
user_search = 1
elif 'q' in request.GET and request.GET['q'] == '': # empty textfield
errors.append('No region specified. Default region used.')
region = 'Anker Engelunds Vej 1, Denmark'
context = {'first_map': region,
'errors': errors,
'user_search': user_search}
return render(request, 'twitter_region_sentiment/index.html', context)
if 'r' in request.GET and request.GET['r'] != '':
radius = request.GET['r']
try:
tweets = dc.retrieve_tweets(region, radius)
except RegionError as r_error:
# exception is thrown when
errors.append(r_error.value)
region = 'Anker Engelunds Vej 1, Denmark'
context = {'first_map': region,
'errors': errors,
'user_search': user_search}
return render(request, 'twitter_region_sentiment/index.html', context)
# extract fields from tweets to show and texts to analyze as lists
for tweet in tweets:
tweets_to_show.append({
'text': tweet.text,
'user': tweet.user.AsDict()['screen_name'],
'datetime': tweet.AsDict()['created_at'],
'image': tweet.user.AsDict()['profile_image_url']
})
texts_to_analyse.append(tweet.text)
ssa = SimpleSentimentAnalyzer()
moods = ssa.get_region_moods(texts_to_analyse)
for idx, tweet in enumerate(tweets_to_show):
tweet['mood'] = moods[idx]
if moods[idx] == 0:
tweet['moodColor'] = 'blue'
elif moods[idx] < 0:
tweet['moodColor'] = 'red'
elif moods > 0:
tweet['moodColor'] = 'green'
else:
tweet['moodColor'] = 'black'
mood_stats = ssa.prepare_mood_stats(moods)
# 20 most frequent words are displayed
topics_temp = ssa.get_top_word_frequencies(texts_to_analyse)[0:21]
topics = []
for topic in topics_temp:
topics.append({'word':topic[0], 'freq':topic[1], 'mood':topic[2]})
# first_map = Map.objects.all()[0]
context = {
'first_map': region,
'errors': errors,
'tweets': tweets_to_show,
'user_search': user_search,
'moodList': moods,
'moods': mood_stats,
'topics': topics
}
return render(request, 'twitter_region_sentiment/index.html', context)
