On 7/20/06, Denis Oliver Kropp <[EMAIL PROTECTED]> wrote:
> Mike Emmel wrote:
> > Here is a first pass at property support.
> >
> > Its basically just a shared meme hash based off of glibs hash and a
> > ugrade of the one in direct to use the same algorithm but use malloced
> > mem. They could easily be combined.
>
> Have only looked at the hash table so far. Seems to have some smaller
> advantages.
>
> I'm not sure why it's not commited, but once I made a modification
> to FusionVector that allows passing a NULL shared memory pool during
> creation, which made the implementation use D_CALLOC() instead of
> SHCALLOC() for that vector.
>
> Apart from indention I only see that you're using bool sometimes
> where DirectResult would be more conform, i.e. use bool only if
> "is" or "has" is in the function name. Returning int directly
> from functions like _size() is also ok. But nobody could know that
> fusion_hash_remove() returns the inverse of fusion_hash_insert()
> in case it succeeds.
>
Okay :)

> Will have a look at your property changes, now :)
>
I thoght about passing in null for pool to signal a local allocation but this
hides a bug. I think its better to add a func to explitly create a local hash.
The same can be done for vector.  The one in direct can then be deleted ?
I can make the other clean ups you suggest also.




> --
> Best regards,
>    Denis Oliver Kropp
>
> .------------------------------------------.
> | DirectFB - Hardware accelerated graphics |
> | http://www.directfb.org/                 |
> "------------------------------------------"
>

_______________________________________________
directfb-dev mailing list
[email protected]
http://mail.directfb.org/cgi-bin/mailman/listinfo/directfb-dev

Reply via email to