def processReplyRetweetHashtag(self, topicDict=None):
if not topicDict:
print "processReplyRetweetHashtag: topic dict is none. all topics are included."
with open(self.infile, "r") as f:
linenum = 0
while 1:
oneline = f.readline()
if not oneline:
break
linenum += 1
if len(oneline) is 0 or oneline[0] is not 'T':
continue
tline = oneline
uline = f.readline()
wline = f.readline()
lp = lineParser(tline, uline, wline)
self.G.add_node(lp.user)
self.G.add_nodes_from(lp.refuser)
for refusr in lp.refuser:
if not self.G.has_edge(lp.user, refusr):
self.G.add_edge(lp.user, refusr, weight=1)
else:
self.G[lp.user][refusr]['weight'] += 1
for ht in lp.hashtag:
if topicDict and not topicDict.has_key(ht):
continue
nx.set_node_attributes(self.G, ht, {lp.user:True})
def todayTopTenHashtag(self):
with open(self.infile, "r") as f:
linenum = 0
hashtags = collections.Counter()
while 1:
oneline = f.readline()
if not oneline:
break
linenum += 1
if len(oneline) is 0 or oneline[0] is not 'T':
continue
tline = oneline
uline = f.readline()
wline = f.readline()
lp = lineParser(tline, uline, wline)
if not len(lp.hashtag):
continue
for ht in lp.hashtag:
hashtags[ht] += 1
self.todayTopic = dict(hashtags.most_common(30))
