def _messageCB(self, con, msg):
"""Called when a message is recieved"""
if not msg.getBody(): # Dont process blank messages
return
# Jabber IDs contain a slash between the server name and the
# resource name. This presents a problem, since Alice uses
# the session name as the filename for the log/session files.
# So the session name for Jabber sessions is just
# "username@server".
source = str(msg.getFrom()).split("/")[0]
# In some case, message body is Unicode, and some bot doesn't like
# Unicode. Convert it to ASCII with str() before submitting.
#response = self.submit(str(msg.getBody()), source+"@JABBER")
try:
# for Thai, it need word segmentation
cmd = wordseg( msg.getBody() )
# get the answer
response = self._bot.reply(cmd).encode("utf-8")
# learn something new
self._bot.learn(cmd)
except UnicodeEncodeError:
response = "Unicode Encoding Error!"
time.sleep( random.random() * self._maxdelay )
self.display(response, source)
def test_wordseg():
for hashtag in hdata:
wordseg(hashtag)
def test_wordseg():
for hashtag in hdata:
wordseg(hashtag)
def test_wordsegment():
for hashtag in hdata:
segment(hashtag)
hdata = open('hashtags.txt').read().split()
n = 100
print("==wordseg==\t\t==wordsegment==")
for hashtag in hdata:
print("[" + " ".join(wordseg(hashtag)[0]) + "]\t\t[" +
" ".join(segment(hashtag)) + "]")
print("==wordseg==\t\t==wordsegment==")
print(
str(
timeit.timeit("test_wordseg()",
number=n,
setup="from __main__ import test_wordseg")) + "\t\t" +
str(
timeit.timeit("test_wordsegment()",
number=n,
setup="from __main__ import test_wordsegment")))
#!/opt/local/bin/python2.7
# -*- coding: utf-8 -*-
import sys
sys.path+=['.']
import wordseg
s=wordseg.wordseg()
s.load_dict("data/dict.txt")
for l in open("data/p.txt"):
l=l.strip()
if len(l)>0:
b=s.segment(l,1,0)
print l, " => ",
for i in b:
print i,
print
