[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Collider and Mover code
From: |
David Philippi |
Subject: |
Re: Collider and Mover code |
Date: |
Sat, 25 Jan 2003 11:26:25 +0100 |
User-agent: |
KMail/1.4.1 |
On Saturday 25 January 2003 04:33, Gervase Lam wrote:
> needs another code review.
Let's start with the suggestions/corrections:
- templates parameters should be named by using typename instead of class,
there are cases where this is much easier to read
mover.hxx:
- the template requires an #include vector.hxx since you actually use one
collider.hxx:
- this one doesn't requires an #include vector.hxx, it may be moved to
collider.cxx as well as the world.hxx
upright_pingu_collider.cxx
- either use bool collide or return true/false - right now the bool is
useless. Using a compiler with named return value optimization collided=true;
break; and a return collided at the end or using direct returns should create
identical code
- the static_cast<float>(pingu_height) is redundant - pos.y is a float so the
result type of pos.y - pingu_height is already float. You only need to cast
if there's no float involved otherwise
linear_mover.hxx
- why are the Collider parameters passed by value instead of const reference?
- why does the for loop in update use a float as counter?
Bye David
- Collider and Mover code, Gervase Lam, 2003/01/12
- Re: Collider and Mover code, Gervase Lam, 2003/01/16
- Re: Collider and Mover code, Gervase Lam, 2003/01/20
- Re: Collider and Mover code, Gervase Lam, 2003/01/24
- Re: Collider and Mover code,
David Philippi <=
- Re: Collider and Mover code, Gervase Lam, 2003/01/25
- Re: Collider and Mover code, Gervase Lam, 2003/01/27
- Re: Collider and Mover code, Gervase Lam, 2003/01/28