c# - Having trouble with UTF-8 storing in NVarChar in SQL Server 2008

Question

I'm pulling data using System.Net.WebClient from a web site, and when the data comes back everything parses and looks good except letters with accents. For example, when it returns an é, SQL Server 2008 saves it as ??.

Just need to figure out how to convert these UTF-8 characters into something SQL Server can read. I'm storing it in an NVARCHAR(MAX) datatype.

I'm using Linq-to-SQL to insert into the database if you were curious.

Any thoughts on what I could do to convert it to the proper format?

Answer 1

Figured it out! When using the WebClient class, I was downloading the data as a string.

My Original Configuration...

System.Net.WebClient wc = new WebClient();
string htmlData = wc.DownloadString(myUri);

I tried to convert this data into a UTF-16...from it's current string, but since Microsoft operates in UTF-16, it had handled the conversion on its own.

Instead, I switched my approach to reading the actual byte[] array from the data like so...

System.Net.WebClient wc = new WebClient();
string htmlData = UTFConvert(wc.DownloadData(myUri));

private string UTFConvert(byte[] utfBytes)
{
    byte[] isoBytes = Encoding.Convert(Encoding.UTF8, Encoding.Unicode, utfBytes);
    return Encoding.Unicode.GetString(isoBytes);
}

This fixed the problem, and SQL correctly see's the accents in everything now. Yippee.

Cheers all, and thanks for your help!