def move(self, tokenizedfile):
sourcefile = tokenizedfile
tokenfile = tokenizedfile + '.tok'
utils.move(os.path.join(self.input_loc, sourcefile), os.path.join(config.backup_area, sourcefile))
utils.move(os.path.join(self.input_loc, tokenfile), os.path.join(config.processing_area, tokenfile))
print '%s - Moved files: %s and %s.' % (unicode(datetime.datetime.now()), sourcefile, tokenfile)
self.status = 0
def tag(self, tokenfile):
'''
Tags text files.
:param tokenfile: Name of a tokenized file.
:return: No return value, instead creates a file that contains all detected entitites.
'''
# print '%s - Now tagging: %s' % (unicode(datetime.datetime.now()), tokenfile)
open_tokenfile = open(os.path.join(self.processing_loc, tokenfile), 'r')
token_list = open_tokenfile.readlines()
tagged_list = self.tagger.tag(token_list)
tag_resultfile = codecs.open(os.path.join(self.processing_loc, '%s.tag' % tokenfile), mode='w+', encoding='utf-8')
position = 1
for tagged_token in tagged_list:
# Increase position counter only for words, not for punctuation marks and quotations.
if (tagged_token[0] not in '!?&"#,;:.-\/()[]{}=') and (tagged_token[0] != '``') and (tagged_token[0] != '\'\''):
position += 1
# Write token + tag + position to file, if token is not tagged as 'O' (other).
if tagged_token[1] != 'O':
# tag_resultfile.write('%s %s %s \n') % (tagged_token[1], unicode(position), tagged_token[0])
tag_resultfile.write(tagged_token[1] + ' ' + unicode(position) + ' ' + tagged_token[0] + '\n')
print '%s - Successfully tagged the file: %s' % (unicode(datetime.datetime.now()), tokenfile)
tag_resultfile.close()
utils.move(os.path.join(self.processing_loc, '%s.tag' % tokenfile), os.path.join(self.output_loc, '%s.tag' % tokenfile))
self.status = 1
