Hello,
Here's what I don't understand.
Maybe someone can help me here.
How can someone build a compatible API without getting information from the origional source code?
No I'll try to explain this more practically below.
And if you build api's that act exactly like the origional enviornment, then you would have to write it the same way, although you would extend it further.
And the available documentation that you speak of also would have to be source code.
Correct?
So no matter how you say it, it always comes back to the origional source code.
If you're going to emulate the Amiga OS and extend it, you're gonna have to do it with the origional source code.
Correct?
What other documentation would there be that would even start to help that is not source code?
Doesn't make sense to me.
Well what do you have in an API documentation generally:
You have the API prototypes, example for a function:
int toto (char* titi, int ta);
Than you have explaination so the programmer understand how to use the function and what it does exactly:
Ex: The toto function is a function to pass a message to titi and tell if this is the final message (ta=1 if yes, 0 if not).
It return 1 if the message have been received by titi, 0 if not.
Ok this is a very simple example. Then you the only thing you have to take care when you want to recreate this function is that your recreated function takes the same arguments, do the same job (send a message to titi and tell him it's the last one, as well as check if he received it) and return the same value (1 if titi received the message, 0 if not).
In a nutshell you have to do the same work that have been done by the original engineer of the function when he created it. That is to say, you have to reinvent it.
That's what have done the AROS Team and that what have done the MOS Team (with the help of the already done work of the AROS Team).
I hope this help you to understand better (even if the example is very simple
)
Regards
