跳至内容
菜单
此问题已终结
2 回复
10045 查看

 

I have the transient object:

class new_import(models.TransientModel):
     _name = 'new.import'
     data_file = fields.Binary('File')


I call a wizard to load the file in the binary field and then i want to parse it.

In the function to parse the file i do:

data_file = base64.b64decode(self.data_file)
lines = data_file.split('\n')
for line in lines:
    result = line[6:11]

but when i read the lines if there is a special character the length of the line isn't the same in all lines,

the file is utf8 in the origin

this must be an error when save the file in binary and then decode back to string

If i read the file directly with python with out store it in a binary field i can read it correctly:

codecs.open(file, mode='r', encoding='utf-8')
result = line[6:11].encode('utf-8')

How can i avoid storing the file in base64 or how can i after decode correctly to string?

Thanks in advance

形象
丢弃
编写者 最佳答案

Aswering Juan Vicente Pascual, that's not a solution.

The original txt file has 412 carachters per line. When i stored it in a binary field and then transform back in a string (

base64.b64decode(self.data_file)), lines has 411,412,413,414 or 415 in function of the special characters they have.

Some information is missed when the file is stored in a binary field.

Other option is that the user select the file in a form without using a binary field.

how can i do that?

then i could parse the file correctly: 

result = codecs.open(os.path.abspath(txt_file), mode='r', encoding='utf-8')

形象
丢弃
最佳答案

Try,

result = base64.encodestring(line[6:11].encode("iso-8859-1"))


Kind regards.


形象
丢弃
相关帖文 回复 查看 活动
1
5月 21
9097
1
11月 17
9242
2
2月 23
9421
2
5月 18
11608
0
7月 16
4564