RE: [AD] SAMPLE loading problem

["Vincent Penecherc'h" <Vincent.Penecherch@xxxxxxxxxx>]

> I don't think so.  The way pack_igetw is implemented, a 16-bit -1 will
> be returned as 0x0000FFFF rather than 0xFFFFFFFF, hence the 
> test against
> EOF should be safe.  See the code a bit further down from your patched
> lines which casts the value to a signed short.
> 
> Of course, if you have a counter-example...

Ah, it returns int, and not short.
Makes me wonder why it barfed on one of my samples, then.
Maybe the sample was corrupted somehow. I'll see if I can
have the code fail again with the patch removed, though
since this was over a year ago, chances are I don't have
the same sample set anymore. Especially as I'm using Vorbis
now (for size). We'll see...



______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
______________________________________________________________________