def split_zh_en (zh_en_str):
tokenizer = Tokenizer()
mark = {"en":1, "zh":2}
zh_en_group = []
zh_set = []
en_set = []
status = ""
en = ""
zh = ""
for c in zh_en_str:
if tokenizer.is_zh(c):
if status == 'en':
zh_en_group.append ([mark["en"], ''.join(en_set)])
en += ''.join(en_set)
en_set = []
zh_set.append(c)
status = 'zh'
else:
if status == 'zh':
zh_en_group.append ([mark["zh"], ''.join(zh_set)])
zh += ''.join(zh_set)
zh_set = []
en_set.append(c)
status = 'en'
if en_set:
zh_en_group.append ([mark["en"], ''.join(en_set)])
en += ''.join(en_set)
elif zh_set:
zh_en_group.append ([mark["zh"], ''.join(zh_set)])
zh += ''.join(zh_set)
if en == "":
print 'error'
return zh_en_group, en, zh
def split_zh_en(zh_en_str):
tokenizer = Tokenizer()
mark = {"en": 1, "zh": 2}
zh_en_group = []
zh_set = []
en_set = []
status = ""
en = ""
zh = ""
for c in zh_en_str:
if tokenizer.is_zh(c):
if status == 'en':
zh_en_group.append([mark["en"], ''.join(en_set)])
en += ''.join(en_set)
en_set = []
zh_set.append(c)
status = 'zh'
else:
if status == 'zh':
zh_en_group.append([mark["zh"], ''.join(zh_set)])
zh += ''.join(zh_set)
zh_set = []
en_set.append(c)
status = 'en'
if en_set:
zh_en_group.append([mark["en"], ''.join(en_set)])
en += ''.join(en_set)
elif zh_set:
zh_en_group.append([mark["zh"], ''.join(zh_set)])
zh += ''.join(zh_set)
if en == "":
print 'error'
return zh_en_group, en, zh
import os
from my_class.Document import Document
from my_class.Tokenizer import Tokenizer
from doc_preprocessing import get_docs_list
from modules import json_io
from modules import csv_io
if __name__ == '__main__':
if len(sys.argv) > 1:
doc_input = sys.argv[1]
else:
doc_input = 'output/en_doc/'
document_list = get_docs_list(doc_input)
tokenizer = Tokenizer()
doc_id = 1
for doc in document_list:
doc_obj = Document(doc_id, doc, doc_input)
# tokenize
normalize_tokens = []
for line in doc_obj.get_lines():
tokens = tokenizer.to_tokens(line.decode('utf-8'))
for token in tokens:
if tokenizer.is_stop_word(token):
token = ""
elif token.isdigit():
normalize_tokens.append(token.encode('utf-8'))
else:
token = tokenizer.stemming(token)
normalize_tokens.append(token.encode('utf-8'))
import os
from my_class.Document import Document
from my_class.Tokenizer import Tokenizer
from doc_preprocessing import get_docs_list
from modules import json_io
from modules import csv_io
if __name__=='__main__':
if len(sys.argv) > 1:
doc_input = sys.argv[1]
else:
doc_input = 'output/en_doc/'
document_list = get_docs_list(doc_input)
tokenizer = Tokenizer()
doc_id = 1
for doc in document_list:
doc_obj = Document(doc_id, doc, doc_input)
# tokenize
normalize_tokens = []
for line in doc_obj.get_lines():
tokens = tokenizer.to_tokens(line.decode('utf-8'))
for token in tokens:
if tokenizer.is_stop_word(token):
token = ""
elif token.isdigit():
normalize_tokens.append(token.encode('utf-8'))
else:
token = tokenizer.stemming(token)
normalize_tokens.append(token.encode('utf-8'))