I think this would not have broken if MEM_reallocN was exactly like
the standard library realloc.  The slight differences probably mislead
whoever was making this change.

On Sat, Jan 26, 2013 at 6:42 AM, Ton Roosendaal <[email protected]> wrote:
> Hi all,
>
> Here's another nice illustration of code cleanup that caused bugs;
> http://projects.blender.org/scm/viewvc.php?view=rev&root=bf-blender&revision=54058
>
> The idea of "replace calloc + memcpy with recalloc" sounds nice, but it 
> breaks the case when memcpy is of length zero. (zero hairs, use 'add' brush 
> in particle mode).
>
> May I quote Joelonsoftware here?
>
> "Old code has been used. It has been tested. Lots of bugs have been found, 
> and they've been fixed. There's nothing wrong with it. It doesn't acquire 
> bugs just by sitting around on your hard drive. Au contraire, baby!"
>
> Great fun read:
> http://www.joelonsoftware.com/articles/fog0000000069.html
>
> Of course we should never be afraid to recode parts of Blender, but it 
> shouldn't be with the naive assumption that the old bad messy code wasn't 
> functional. New code should be totally and undeniably better, for users. Not 
> for coders. :)
>
> We can moan and complain about our current bad particle code a lot, but it 
> has offered a lot of good for Blender for many years. The challenge is not to 
> make a new system with clean design, but to make a new system that's far 
> superior.
>
> -Ton-
>
> ------------------------------------------------------------------------
> Ton Roosendaal  Blender Foundation   [email protected]    www.blender.org
> Blender Institute   Entrepotdok 57A  1018AD Amsterdam   The Netherlands
>
> _______________________________________________
> Bf-committers mailing list
> [email protected]
> http://lists.blender.org/mailman/listinfo/bf-committers
_______________________________________________
Bf-committers mailing list
[email protected]
http://lists.blender.org/mailman/listinfo/bf-committers

Reply via email to