TCL(Tool Command Language) Scripting :: PROPOSAL: Tcl equivalent to .NET interop


	PROPOSAL: Tcl equivalent to .NET interop Unlike an earlier poster I can't see any great advantage to Tcl in .NET just to get the .NET GUI libraries. One thing that I do find useful when programming in C# is the .NET interop capabilities which allow you to call dll functions directly without having to get into the unpleasantness of C++ on Windows or writing your own wrapper stuff. In .NET the interop is based on a comprehensive byte code generation capability but nothing that elaborate is actually required for this purpose - just a specification of the input and output C types. Actually making the calls requires either some assembler :-( or perhaps pretending that we are calling something with signature f(int,int,int,int....? how many? ) :-( structure handling would be quite easily dealt with by mapping a structure description to a format for the binary command. If the specifications could be made so that there was a "simpleish" transform from the C# spec then we could leverage the existing knowledge in the pinvoke wiki at http://www.pinvoke.net to call windows stuff Hopefully somebody with Windows expertise will be inspired by this and make it happen - Personally I've always used C# or Tcl on Windows precisely because it is so difficult to program with anything else. I don't know enough about .NET to contribute meaningfully to this discussion, but does FFIDL provide (some of) the functionality you're requesting? http://www.elf.org/ffidl/ On May 28, 10:26 am, Aric Bills <aric.bi@gmail.com> wrote: > I don't know enough about .NET to contribute meaningfully to this > discussion, but does FFIDL provide (some of) the functionality you're > requesting? > http://www.elf.org/ffidl/ Yes - that seems to be going in the right direction although it seems that it is a work in progress especially for Windows which is where it is most needed e.g. library lookup is no good. (On Linux critcl is probably the easy way to go). Nick Hounsome wrote: > although it seems that it is a work in progress It's funny how everything is either a work in progress or abandoned and hasn't been updated in years. Anyone else notice that? :-) -- Darren New / San Diego, CA, USA (PST) His kernel fu is strong. He studied at the Shao Linux Temple. At Wed, 30 May 2007 08:06:57 -0700 Darren New <d@san.rr.com> wrote: > Nick Hounsome wrote: > > although it seems that it is a work in progress > It's funny how everything is either a work in progress or abandoned and > hasn't been updated in years. Anyone else notice that? :-) Most of the universe can be considered either 'work in progress' or 'abandoned and hasn't been updated in [billions of] years', depending on which of the various comosological theorys that are out there... -- Robert Heller -- 978-544-6933 Deepwoods Software -- Linux Installation and Administration http://www.deepsoft.com/ -- Web Hosting, with CGI and Database hel@deepsoft.com -- Contract Programming: C/C++, Tcl/Tk Darren New wrote: > Nick Hounsome wrote: >> although it seems that it is a work in progress > It's funny how everything is either a work in progress or abandoned and > hasn't been updated in years. Anyone else notice that? :-) Isn't any work either by definition a work in progress or abandoned? To release patches for something, work is in progress. If you refuse to work on it anymore ... it is abandoned. Jeff On May 30, 7:29 pm, Jeff Hobbs <j@activestate.com> wrote: > Darren New wrote: > > Nick Hounsome wrote: > >> although it seems that it is a work in progress > > It's funny how everything is either a work in progress or abandoned and > > hasn't been updated in years. Anyone else notice that? :-) > Isn't any work either by definition a work in progress or abandoned? To > release patches for something, work is in progress. If you refuse to > work on it anymore ... it is abandoned. I often hear people refer to abandoned work as "stable" work. To me, it seems that if a software package needs no improvements, fixes, documentation enhancements or corrections, then the software probably isn't very complex, or isn't being used very much. Larry W. Virden wrote: > I often hear people refer to abandoned work as "stable" work. To me, > it seems that if a software package needs no improvements, fixes, > documentation enhancements or corrections, then the software probably > isn't very complex, or isn't being used very much. There is virtue in not being very complex; it means that the software can actually be fixed to the point where it is correct. Donal.