If you're reading non Unicode data you need to encode it for Unique.

When you read from the file stream you need to use the proper encoding for the file which from what you describe most likely is a native ANSI codepage.

When you read the file apply the appropriate encoding (ie. Encoding.Default or a specific codepage) to get the proper translation into Unicode strings.

+++ Rick ---

Look at Encoding.Default.ToBytes() and Encoding.Default.ToString() which in combination can encode data from raw

>I try with this but no, my error is read flat file.
>
>>You should make sure your requestEncoding and responseEncoding in the globalization tag is set to UTF-8. This will ensure that pages get generated with UTF-8 encoding and input is accepted and passed back in UTF-8 (which is basically Unicode encoded into an ASCII character set).
>>
>>This will work if you retrieve browser input, but there may be issues with encoding if clients send data directly to your via HTTP and are not properly encoding their content as UTF-8.
>>
>>+++ Rick ---
>>
>>>Hi
>>>
>>>I've a problem , when I try read a text with especial characters for example ö, ñ , ó my app change to squares , how i do to resolve this.
>>>
>>>Thank's