Given the following MGrammar:

module Demo.Employees{
language EmployeeDSL
{
syntax Main = e:Employee*
=> Employees { valuesof(e) };

syntax Employee = employeeName n:text endEmployee
=> id(n) {Name {n}};

@{Classification["Keyword"]}
token employeeName = "Name: ";

token text = ( 'A'..'Z' | 'a'..'z')+;

token endEmployee = "*";

interleave WhiteSpacing = " " | "\r" | "\n";
}
}


And this input:

Name: Bob
*
Name: Alice
*

...Both Intellipad and mgx.exe procude the following output (Intellipad does not include the module name of course):

module Demo.Employees {
Employees {
Bob {
Name => "Bob"
},
Alice {
Name => "Alice"
}
}
}

When I do thesame programmatically with DynamicParser like this:

    class Program
    {
        static void Main(string[] args)
        {
            DynamicParser parser = null;

            string fileToParse  = "EmployeeData.ed";
            string moduleName   = "Demo.Employees";
            string languageName = "EmployeeDSL";

            System.Reflection.Assembly assembly = System.Reflection.Assembly.GetExecutingAssembly();
            using (var img = MImage.LoadFromResource(assembly, "EmployeeParser.mx"))
            {
                var factory = img.ParserFactories[moduleName+ "." + languageName];
                parser = factory.Create();
                parser.GraphBuilder = new NodeGraphBuilder();
                Node root = (Node)parser.Parse(fileToParse, new ParserErrorReporter());

                // Save the graph as an m file
                string destinationFile = Path.ChangeExtension(fileToParse, "m");
                StringBuilder fullMFile = new StringBuilder();
                fullMFile.AppendFormat("module {0} ", moduleName);
                fullMFile.AppendLine("{");
                fullMFile.Append(root.WriteToString());
                fullMFile.AppendLine("}");
                File.WriteAllText(destinationFile, fullMFile.ToString());
            }
        }
    }

...the generated graph looks like the following:

module Demo.Employees {
Employees {
Bob {
Name {
'Bob'
}
},
Alice {
Name {
'Alice'
}
}
}

Again, for comparison, the mgx.exe version:

module Demo.Employees {
Employees {
Bob {
Name => "Bob"
},
Alice {
Name => "Alice"
}
}
}


Whatam I missing here?

module Demo.Employees{
language EmployeeDSL
{
syntax Main = e:Employee*
=> Employees { valuesof(e) };

syntax Employee = employeeName n:text endEmployee
=> id(n) {Name => n};

@{Classification["Keyword"]}
token employeeName = "Name: ";

token text = ( 'A'..'Z' | 'a'..'z')+;

token endEmployee = "*";

interleave WhiteSpacing = " " | "\r" | "\n";
}
}

Let me know if that helps.

For more details on the new graph model, check out the paper by Clemens Szyperski (http://msdn.microsoft.com/en-us/library/dd878360.aspx).

David

module Demo.Employees{
language EmployeeDSL
{
syntax Main = e:Employee*
=> Employees { valuesof(e) };

syntax Employee = employeeName n:text endEmployee
=> id(n) {Name => n};

@{Classification["Keyword"]}
token employeeName = "Name: ";

token text = ( 'A'..'Z' | 'a'..'z')+;

token endEmployee = "*";

interleave WhiteSpacing = " " | "\r" | "\n";
}
}

Let me know if that helps.

For more details on the new graph model, check out the paper by Clemens Szyperski (http://msdn.microsoft.com/en-us/library/dd878360.aspx).

David