from twitter.twitter import *
import cld2full

tt=xtwc.sents('20100128.txt')
corp=[]

for t in tt:
 at=[w.lower() for w in t if w.isalpha()]
 if len(at) >= 5:
  corp.append((' '.join(at)).encode('utf8'))
len(corp)

rcorp=[r for r in ((cld2full.detect(t),t) for t in corp) if r[0].is_reliable]
len(rcorp)

mecorp=[r for r in rcorp if r[0].details[0].language_code=='en']
len(mecorp)

eecorp=[r for r in mecorp if r[0].details[1].language_code=='un']
len(eecorp)

necorp=[r for r in rcorp if (r[0].details[0].language_code!='en') and (r[0].details[1].language_code!='en') and (r[0].details[2].language_code!='en')]
len(necorp)