# coding=utf-8

import xlrd

'''读取评论新闻数据，并按照评分将其分成正向和负向两个文本'''

def read_data(file_path, sheet_name):

work_book = xlrd.open_workbook(file_path)

work_sheet = work_book.sheet_by_name(sheet_name)

data = []

pos_res = []

neg_res = []

# 遍历sheet1中所有行row

num_rows = work_sheet.nrows

for curr_row in range(num_rows):

if curr_row == 0:

continue

row = work_sheet.row_values(curr_row)

# 星级大于30的为正向情感评论

if row[1] >= '30':

pos_res.append(row[3].replace(' ', '').replace('\n', '').replace('\t', '').replace('\r', ''))

# 星级小于30的为负评论

if row[1] < '30':

neg_res.append(row[3].replace(' ', '').replace('\n', '').replace('\t', '').replace('\r', ''))

# data.append(pos_res, neg_res)

# print('row%s is %s' % (curr_row, row))

return pos_res, neg_res

# # 遍历sheet1中所有列col

# num_cols = work_sheet.ncols

# for curr_col in range(num_cols):

# col = work_sheet.col_values(curr_col)

# print('col%s is %s' % (curr_col, col))

#

# # 遍历sheet1中所有单元格cell

# for rown in range(num_rows):

# for coln in range(num_cols):

# cell = work_sheet.cell_value(rown, coln)

# print cell

def read_data_test():

file_path = '/Users/li/workshop/MyRepository/DeepNaturalLanguageProcessing/DeepNLP/spider/豆瓣影评数据.xls'

tmp = read_data(file_path, u'豆瓣影评数据')

# pos_res = []

# neg_res = []

# for i in tmp:

# if i[1] >= '30':

# pos_res.append(i[3])

#

# for i in tmp:

# if i[1] < '30':

# neg_res.append(i[3])

#

# print len(pos_res), len(neg_res), len(tmp)

for i in tmp[0]:

print(i)

print('\n')

for i in tmp[0]:

print(i)

if __name__ == "__main__":

read_data_test()