drj3122 drj3122 - 5 days ago 5
C# Question

DownloadString and Special Characters

I am trying to find the index of Mauricio in a string that is downloaded from a website using webclient and download string. However, on the website it contains a foreign character, Maurício. So I found elsewhere some code

string ToASCII(string s)
{
return String.Join("",
s.Normalize(NormalizationForm.FormD)
.Where(c => char.GetUnicodeCategory(c) != UnicodeCategory.NonSpacingMark));
}


that converts foreign characters. I have tested the code and it works. So the problem I have is that when I download the string, it downloads as MaurA-cio. I have tried both

wc.Encoding = System.Text.Encoding.UTF8;
wc.Headers.Add("Accept-Charset", "ISO-8859-1,utf-8;q=0.7,*;q=0.7");


Neither stop it from downloading as MaurA-cio.

(Also, I cannot change the search as I am getting the search term from a list).

What else can I try?
Thanks

Answer

DownloadString doesn't look at HTTP response headers. It uses the previously set WebClient.Encoding property. If you have to use it, get the headers first:

// call twice 
// (or to just do a HEAD, see http://stackoverflow.com/questions/3268926/head-with-webclient)
webClient.DownloadString("http://en.wikipedia.org/wiki/Maurício");
var contentType = webClient.ResponseHeaders["Content-Type"];
var charset = Regex.Match(contentType,"charset=([^;]+)").Groups[1].Value;

webClient.Encoding = Encoding.GetEncoding(charset);
var s = webClient.DownloadString("http://en.wikipedia.org/wiki/Maurício");

BTW--Unicode doesn't define "foreign" characters. From Maurício's perspective, "Mauricio" would be the foreign spelling of his name.