"Jon Skeet [C# MVP]" <sk***@pobox.com> wrote in message
news:MPG.1e840b579aabf2ce98cf6a@msnews.microsoft.com...
> Bob <b**@nowhere.com> wrote:
> > Need to produce a Doc with no encoding info.
> > Is there anyway of doing this?
> > Thanks
> > Bob
> > i.e.
> > <?xml version=\"1.0\" ?>
>
> If you use XmlTextWriter and *explicitly* specify a null encoding, I
> believe that will work - but I haven't tried it.
>
> --
> Jon Skeet - <sk***@pobox.com>
> http://www.pobox.com/~skeet   Blog: http://www.msmvps.com/jon.skeet
> If replying to the group, please do not mail me too

"Jon Skeet [C# MVP]" <sk***@pobox.com> wrote in message
news:1142583262.683198.170010@z34g2000cwc.googlegroups.com...
> Bob wrote:
> > Thanks for your reply.
> > I have done a bit more investigation since posting and the problem is
that I
> > am writing into a stringbuilder.
> > It was reported previously as a bug by someone else but the MS response
was
> > that it was by design. If you are writing to a stringbuilder then you
will
> > end up with a string and strings are always UTF-16 Q.E.D.
> > I find this kind of logic dangerous.
>
> Yes - it's frankly ridiculous. They can't tell what encoding you'll end
> up using for converting the text data into a binary representation.
>
> > It  removes a level of control that I believe should remain with the
> > programmer.
>
> Agreed.
>
> > For what ever reason, the URI that I am posting to insists on no coding
> > attribute.
> >
> > So my logic of MakeXMLDoc -> string -> bytearray -> webrequest
poststream is
> > now
> > MakeXMLDOc ->string->modified string -> bytearray WebRequest.
> >
> > I am using Framework 2 and the recommendation is to use the XMLWriter.
> >
> > I tried using a memorystream and a XMLWriterSettings class with encoding
=
> > null but this didn't work.
>
> I think what you want is this:
>
> public class NullEncodingStringWriter : StringWriter
> {
>     public override Encoding Encoding
>     {
>         get { return null; }
>     }
> }
>
> If you create one of those, pass that to the XmlTextWriter, then call
> XmlDocument.Save, you'll find that it doesn't put on the encoding.
> Here's a sample:
>
> using System;
> using System.IO;
> using System.Text;
> using System.Xml;
>
> public class Test
> {
>     static void Main(string[] args)
>     {
>         XmlDocument doc = new XmlDocument();
>
>         doc.LoadXml ("<?xml version='1.0'
> encoding='UTF-8'?><element>text</element>");
>
>         StringWriter sw = new EncodingStringWriter();
>         XmlTextWriter xtw = new XmlTextWriter(sw);
>         doc.Save(xtw);
>         Console.WriteLine(sw.ToString());
>     }
> }
>
> public class NullEncodingStringWriter : StringWriter
> {
>     public override Encoding Encoding
>     {
>         get { return null; }
>     }
> }
>
> Alternatively, if you want to go to a MemoryStream anyway (to get the
> bytes out directly) you could use a StreamWriter which takes the
> MemoryStream as the stream to write to and uses null as the encoding.
> Unfortunately, StreamWriter prevents you from specifying a null
> encoding, so you need to create a derived type which overrides the
> Encoding property. Again, here's a sample:
>
> using System;
> using System.IO;
> using System.Text;
> using System.Xml;
>
> public class Test
> {
>     static void Main(string[] args)
>     {
>         XmlDocument doc = new XmlDocument();
>
>         doc.LoadXml ("<?xml version='1.0'
> encoding='UTF-8'?><element>text</element>");
>
>         MemoryStream stream = new MemoryStream();
>         StreamWriter writer = new NullEncodingStreamWriter (stream);
>         XmlTextWriter xtw = new XmlTextWriter(writer);
>         doc.Save(xtw);
>         Console.WriteLine(Encoding.UTF8.GetString(stream.ToArray()));
>     }
> }
>
> public class NullEncodingStreamWriter : StreamWriter
> {
>     public override Encoding Encoding
>     {
>         get
>         {
>             return null;
>         }
>     }
>
>     public NullEncodingStreamWriter (Stream stream) : base (stream)
>     {
>     }
> }
>
> Jon
>