How do I decode a base64 encoded string containing an XML document that contains characters with accents (á,é,í,ó,ú) in C#?

Question

How do I decode a base64 encoded string containing an XML document that contains latin letters (á,é,í,ó,ú)?

I am aware of this question How do I encode and decode a base64 string? But the solutions provided do not work well with letters that have accents.

So far I've tried:

xmlBase64 = System.Text.Encoding.ASCII.GetString(System.Convert.FromBase64String(XmlDoc));
xmlBase64 = System.Text.Encoding.Unicode.GetString(System.Convert.FromBase64String(XmlDoc));
xmlBase64 = System.Text.Encoding.UTF8.GetString(System.Convert.FromBase64String(XmlDoc));
xmlBase64 = System.Text.Encoding.UTF32.GetString(System.Convert.FromBase64String(XmlDoc));

But in all cases the latin letters (spanish characters) are replaced with ? or similars.

EDIT:

This is the base64 encoded string

This is the Decoded string

what encoding was used to encode data? note aside - google latin alphabet and what letters are in it — Iłya Bursov, Jul 27 '17 at 19:41
@litelite, no I've check the document and the letters are there. — Alejandro Montilla, Jul 27 '17 at 19:42
@Adam, I think the same way, but for some reason is not working. — Alejandro Montilla, Jul 27 '17 at 19:43
Please post sample data of your Xml, both as plain text and base64 string. — Filburt, Jul 27 '17 at 19:45
@Ilya Bursov, `UFT-8`, in fact if I check the string in an onine decoder, the result is fine, but not in my program. — Alejandro Montilla, Jul 27 '17 at 19:46
First test would be finding how it is encoded in the first place. If UTF8 is viewed as ASCII some of the extended characters will be converted to invalid characters. Once that byte array is encoded into base64 it doesnt matter how you decode it... it will still have the invalid characters. — Matthew Whited, Jul 27 '17 at 20:27
@MatthewWhited The encoding process was sucessfull, in fact if you decode the string with an online converted, all the caracter are fine. I've use https://www.base64decode.org/ to check the string. — Alejandro Montilla, Jul 27 '17 at 20:31

score 5 · Accepted Answer · answered Jul 27 '17 at 21:11

It's helpful to see the bytes produced by System.Convert.FromBase64String(XmlDoc).

I've done that and took a look at the word "metálicas" in your original string (this was just the first word I found with an accent). This portion of the string is converted to the byte array 6D 65 74 E1 6C 69 63 61 73.

From that byte array it's easy to see two things:

This is a single byte encoding
It is not UTF-8: In UTF-8, bytes greater than 7F never occur on their own, always in groups of 2-4.

From there I guessed it would be some form of extended ASCII, Windows-1252 seems to work. Try the following:

xmlBase64 = System.Text.Encoding.GetEncoding(1252).GetString(System.Convert.FromBase64String(XmlDoc));

Great!!! That solve the problem. Not really sure how the encoding was made with Windows-1252. Thank you for your time. — Alejandro Montilla, Jul 27 '17 at 21:19

How do I decode a base64 encoded string containing an XML document that contains characters with accents (á,é,í,ó,ú) in C#?

1 Answers1