[Closed] dotnet xml reading taking time

I’ve found that the bottleneck in situations like these is passing data from .NET to maxscript and vice versa. Especially complex datastructures that have to be wrapped can be very slow.
If you’re looking for something specific in the xml file, I’d suggest writing all xml parsing code in C#, or C++ or anything that compiles to a .net assembly. And then use a single function call to get exactly what you want with just one expensive ‘transfer’ of data.
But whether this approach will work nicely varies from case to case of course.

Found the same things here. The conversions are the slow part for sure. I did a test at one point of doing math with dotNet or Max script and it was way slower doing it with dotNet which really doesn’t make sense until you take into consideration the conversion from dotNet to Max script.

Ok thats interesting to hear about !
By the way, Paul, you will notice that the script is greatly inspiring from your dotnet tutorial that i found on your site. This was invaluable ressources for me! so thank you very much

I think i will give a try to what Pier Janssen says.
Creating a dotnet class in C and use a single function call that returns me an array of the data i need.
This specific problem of speed is not that important, but it gives me a good excuse to dive into this new world for me that is : “Creating a custom dotnet assembly and use it”

Does someone know good tutorials or documents to orient me in the good direction ?

• How to create a custom dotnet assemblie ?
• How to parse XML in C / C++ ?
• How to transfer an array of values from C to maxscript ?

Thank you

okay…
so to begin with all that stuff… as a total noob i decided to try making an assembly wich would return me a simple integer when i call it.

       So i opened VisualStudio, select Visual C++ and create a new "class library" project :
       this is the initial code  

// ParseXML.h
           
           #pragma once
           
           using namespace System;
           
           namespace ParseXML {
         
           	public ref class Class1
           	{
           		// TODO: Add your methods for this class here.
           	};
           }
           

       I compile this into a Dll wich i can load in max : 

 
           	dotnet.loadassembly "C:\ParseXML.dll"
           	dotnetclass "ParseXML.Class1"

       and this returns me in the listener :
       dotNetClass:ParseXML.Class1
       
       Nice!
       So next step is to add some methods i guess ^^
       
       I tried different things, by reading some c++ class documentation, and ended with this (no moquery please ^^)

#pragma once
  
  using namespace System;
  
  namespace ParseXML {
  
  	class Class1 {
  		public :
  			int getInteger() { return 5;};
  	};
  }

Then in max i load the new assembly, but when i do this :

showmethods (dotnetclass "ParseXML.Class1")

i get only this, and i have no getInteger method :s

  .[static]<System.Boolean>Equals objA objB
    .[static]<System.Boolean>ReferenceEquals objA objB

   I am a bit lost, What should i do to get an assembly that returns me an integer.
       
       For example in maxscript : 

C  = dotnetclass "ParseXML.Class1"
      C.GetInteger()
      5

       5 is the integer that the class should return to me ;)
 
 If i can get this, i get the basic !

The way you’re calling the getInteger() function is through the class, not through an object. You can only call functions that are declared static in this fashion. For member functions, you’ll need an object first:

(
   dotnet.loadassembly "C:\ParseXML.dll"
   local classInstance = dotnetObject "ParseXML.Class1"
   print classInstance.getInteger()
)

i get this error :s
– Runtime error: No constructor found which matched argument list: ParseXML.Class1
i guess i have something wrong in my C++ code…

Ah hm apparently a default constructor isn’t synthesized automatically. So add it to the Class1 class:

class Class1 {
  		public :
			Class1(); //constructor can be empty here.
  			int getInteger() { return 5;};
  	};

Well i keep getting the same error, with the constructor declared explicitly.
I tryed to change the c++ source code but i can’t find where the problem is.

Do you know a link to source code of a custom dotnet class in c++ or another language ?

Thank you:)

Hmm that’s odd… I’ve tried it on my pc, and it works just fine, without having to explicitly define a default constructor. Here’s the code I’m using:
C++

#pragma once

using namespace System;

namespace mxs_cpp_test {

	public ref class Class1
	{
	public:
		int getInteger() { return 5; }
		static int multiply(int i, int f) { return i * f; }
	};
}

mxs

(
	dotnet.loadAssembly @"C:\Documents and Settings\Administrator\My Documents\Visual Studio 2008\Projects\mxs_cpp_test\Release\mxs_cpp_test.dll"
	local class1 = dotnetclass "mxs_cpp_test.Class1"
	showmethods class1
	print (class1.multiply 2 4)
	local inst1 = dotnetobject "mxs_cpp_test.Class1"
	showmethods inst1
	print (inst1.getInteger())
)

result:

  .[static]<System.Boolean>Equals <System.Object>objA <System.Object>objB
  .[static]<System.Int32>multiply <System.Int32>i <System.Int32>f
  .[static]<System.Boolean>ReferenceEquals <System.Object>objA <System.Object>objB
8
  .<System.Boolean>Equals <System.Object>obj
  .[static]<System.Boolean>Equals <System.Object>objA <System.Object>objB
  .<System.Int32>GetHashCode()
  .<System.Int32>getInteger()
  .<System.Type>GetType()
  .[static]<System.Int32>multiply <System.Int32>i <System.Int32>f
  .[static]<System.Boolean>ReferenceEquals <System.Object>objA <System.Object>objB
  .<System.String>ToString()
5