#!/usr/bin/python
# -*- encoding: UTF-8 -*-

import sys
import codecs

window = [('_B','E')]

def process_window():
    global window

    tag = window[1][1]
    c0 = window[1][0]

    if c0 != 'E_':
        tp = window[0][1]
        cp = window[0][0]
        cn = window[2][0]

        U00 = 'U00-%s' % cp
        U01 = 'U01-%s' % c0
        U02 = 'U02-%s' % cn
        U03 = 'U03-%s/%s' % (cp, c0)
        U04 = 'U04-%s/%s' % (c0, cn)
        U05 = 'U05-%s/%s' % (cp, cn)

        event=(tag, U00, U01, U02, U03, U04, U05, tp)
        print ' '.join(event).encode('UTF-8')

    del window[0]

def process_file(file):
    global window

    for line in file:
        if line[0] == '\n':
            word, tag = 'E_', 'E'
        else:
            word, tag = line.strip().split()

        if len(window) == 3:
            process_window()

        window.append ((word, tag))

if __name__ == "__main__":
    try:
        file = codecs.open(sys.argv[1], "r", "utf-8")
    except:
        print "corpus file is not specified, or open failed!"
        sys.exit()
    
    process_file(file)
    file.close()