ViteFalcon

This is awesome!

Considering that curiosity is not willing to share it, my take is to not try to push him to it. It's his hard work and it's decision to make what he would like to do with his code. To me, the best things that can happen with this project are these (in no particular order): Learn more about the client itself: Even if it means that the code is not out there, just learning a lot about the client and how things work in the client is a great piece of knowledge. If curiosity is willing to share some of that, it would be awesome! A client that is truly moddable/customizable: At the end of the day, what we all want is a client that is open to modding and allows custom artwork and actions to show up in the client. While having a closed-source might limit a lot of things, it is still possible to have a client that's vastly better than trying to hex-hack the official client to do have custom things built, IMO. I could be dead wrong here, but I see the potential. Motivate other developers to build a client: Take this knowledge and learn to build another client. Who knows that other person might just release the source to the wild .

Too much good information to be hidden away in a form post. I took the liberty to copy this over the wiki: A brief look at RagExe. Thanks, curiosity!

What do you plan to use for graphics (+ GUI), audio and input?

What do you mean by emulator in C++ and client in C#? Just want to know what you mean by emulator really. AFAIK, they're actually building the whole client in C++ after having some issues with porting their C# client to Linux. I believe it's mainly because client's performance in Linux was bad. And it's currently on-hold as the main developer Henko is finishing his masters, which I believe should be done by now. Edit: Ignore this post. I thought you were talking about RagnaXNA.

Actually there are 9 'goto' statements in grf.cpp alone. I haven't taken the time to hunt for other goto statements. I had trouble compiling it in VS2010 due to the variable template argument functions. Rewriting those functions to use va_list, va_start to get variable arguments to send it to your custom printf and aprintf functions works for me. I should mention though that it's really fast to open and list contents in a GRF file. I'm trying to find out how fast it would be to find and read contents of a file. I see that you're using boost. Good choice! I'm still studying the code. Will let you know of my comments soon. As an example of what you should've done instead of goto statement, you could've created a function called: mark_dirty(); The code that you jump to should be the body of that function. Does that make sense?

Thanks for sharing your code. I do however have one concern, you're using 'goto' in your code. You should ideally not use goto statements. In your case, you use goto to exit a conditional block, which is bad practice. You should ideally be throwing an exception, catching it and doing whatever is required if a bad_grf exception happens. That's just one of the problems I find with your code. I'll have a better look later and report if I find more. Don't take this critisim in a bad way. Take it as points to improve your programming skills