Re: [boost] [gsoc15] Proposal
The METALISCPP project was only a starter. My idea was to implement a very minimal syntactical abstraction over the metaprogramming constructs and build a prolog like library over it. As you can see, the METALISCPP simply uses macros to hide all those long template-based codes and provide an easier and more convenient syntax. The main motivation behind it was to have a library that would handle polices, inheritance, complex dependencies and such conveniently at compile time with a logic oriented syntax. Then I started thinking about extending the idea for a complete First Order Logic Programming Library and got stuck there. If you are willing to mentor it this year, we can try to improve the performance and syntactically modify the metalog library, maybe we'll end up with something as convenient as Boost::Phoenix. Best Wishes Ganesh PS I've forked metalog and trying to figure out what useful things can be done with it in 3 months time.
Oh I'm sorry if I gave the wrong idea, but I certainly do not qualify for
mentoring (yet hopefully).
I was just amazed by the fact you were actually doing something so similar
to metalog and decided to share it with you. I thought I was alone there
and that it was, well, just a bad idea, but your goal seems to be pretty
similar to mine: handle complex inheritance patterns and other type
relationships in a convenient and powerful manner.
Anyways I wish you luck on your project.
On Feb 12, 2015 4:07 PM, "Ganesh Prasad"
The METALISCPP project was only a starter. My idea was to implement a very minimal syntactical abstraction over the metaprogramming constructs and build a prolog like library over it. As you can see, the METALISCPP simply uses macros to hide all those long template-based codes and provide an easier and more convenient syntax.
The main motivation behind it was to have a library that would handle polices, inheritance, complex dependencies and such conveniently at compile time with a logic oriented syntax. Then I started thinking about extending the idea for a complete First Order Logic Programming Library and got stuck there.
If you are willing to mentor it this year, we can try to improve the performance and syntactically modify the metalog library, maybe we'll end up with something as convenient as Boost::Phoenix.
Best Wishes Ganesh
PS I've forked metalog and trying to figure out what useful things can be done with it in 3 months time.
participants (2)
-
Bruno Dutra
-
Ganesh Prasad