def __init__(self, storedb, logfunc=None):
print 'Initializing ANEW module'
self.store_db = storedb
self.valence_db = ValenceDB()
self.stemmer = stemmer.PorterStemmer()
self.logfunc = logfunc
self.total_count = 0
# Connect to ANEW database
self.valence_db.connect()
def __init__(self, storedb, logfunc=None):
print 'Initializing ANEW module'
self.store_db = storedb
self.valence_db = ValenceDB()
self.stemmer = stemmer.PorterStemmer()
self.logfunc = logfunc
self.total_count = 0
# Connect to ANEW database
self.valence_db.connect()
class BasicAnalyzer(BaseAnalyzer):
def __init__(self, storedb, logfunc=None):
print 'Initializing ANEW module'
self.store_db = storedb
self.valence_db = ValenceDB()
self.stemmer = stemmer.PorterStemmer()
self.logfunc = logfunc
self.total_count = 0
# Connect to ANEW database
self.valence_db.connect()
def analyze_tweet(self, tweet):
try:
self.log('%s Received geolocated tweet (%d)' % (time.ctime(), self.total_count))
splits = [word.strip(string.punctuation) for word in tweet['text'].split()]
if len(splits) < 3:
self.log('\ttoo short')
return
self.log('\tanalyzing...')
count = 0
total = 0
f = 0.0
for split in splits:
tmp = self.valence_db.get_stat(split)
if tmp == -1:
# Word not found, so try stemming
tmp = self.valence_db.get_stat(self.stemmer.stem(split))
total +=1
if tmp != -1:
# Nothing found
count += 1
f += tmp
if count > 0:
avg = f/count
self.log('\t'+str(avg))
else:
avg = -1
self.log('\tCouldn\'t score')
# save tweet text, author, time, location, valence to DB
self.store_db.add_tweet(tweet, avg)
self.total_count += 1
except:
pass
def stop(self):
self.valence_db.close()
def log(self, msg):
if self.logfunc:
self.logfunc(msg)
else:
print msg
class BasicAnalyzer(BaseAnalyzer):
def __init__(self, storedb, logfunc=None):
print 'Initializing ANEW module'
self.store_db = storedb
self.valence_db = ValenceDB()
self.stemmer = stemmer.PorterStemmer()
self.logfunc = logfunc
self.total_count = 0
# Connect to ANEW database
self.valence_db.connect()
def analyze_tweet(self, tweet):
try:
self.log('%s Received geolocated tweet (%d)' %
(time.ctime(), self.total_count))
splits = [
word.strip(string.punctuation)
for word in tweet['text'].split()
]
if len(splits) < 3:
self.log('\ttoo short')
return
self.log('\tanalyzing...')
count = 0
total = 0
f = 0.0
for split in splits:
tmp = self.valence_db.get_stat(split)
if tmp == -1:
# Word not found, so try stemming
tmp = self.valence_db.get_stat(self.stemmer.stem(split))
total += 1
if tmp != -1:
# Nothing found
count += 1
f += tmp
if count > 0:
avg = f / count
self.log('\t' + str(avg))
else:
avg = -1
self.log('\tCouldn\'t score')
# save tweet text, author, time, location, valence to DB
self.store_db.add_tweet(tweet, avg)
self.total_count += 1
except:
pass
def stop(self):
self.valence_db.close()
def log(self, msg):
if self.logfunc:
self.logfunc(msg)
else:
print msg