headers
Michael Pruett | 2 Jan 21:23
Picon

Re: ATRAC3

On Tue, Dec 31, 2002 at 07:41:20PM +0000, Pete Ryland wrote:
> Hi Michael et al,
> 
> Thanks for libaudiofile.  It is very good software which I use daily and
> I hope you continue to support it for some time! :)
> 
> Anyway, I am interested in adding in support for ATRAC3-encoded files. 
> Whilst the format is not terribly well defined by open specifications, I
> understand that it's possible to use Sony's MS Windows codec to do the
> actual work and have a wrapper around this in libaudiofile.  How
> difficult is this, and can anyone provide a pointer to how to go about
> this?

Pete, adding support for ATRAC3 would be a useful feature.  While it
might be possible to use Sony's Windows codec to accomplish this task,
but this approach would have the disadvantage of requiring some sort of
Windows emulation layer which would likely only run on Intel x86
systems.  It would certainly be preferable to write an open source
codec which could be easily integrated into the Audio File Library.

Michael
Permalink | Reply | 1 comment |
headers
Guido Draheim | 10 Jan 12:49
Picon
Picon

libaudiofile not largefile sensitive

Through a discussion on alt.os.linux.mandrake, I had to notice
that every library has a risk of being broken when using the
system off_t type in its exported interfaces, including
structures and function synopsis.

The point comes from the fact that systems like linux and
solaris are "largefile sensitive" - a -D_LARGEFILE_SOURCE
will shift the "off_t" type from being a 32bit entity into
a 64bit entity. This will make for a nice and pretty
callframe mismatch.

Grepping a bit through header files installed on my system
it occurs to me that audiofile.h contains quite a lot of
these off_t arguments in its api calls - it can be only by
luck that no application had the misfortune to trap on one
of these, doesn't it.

A autoconf'ed lib 'centric treatment of the topic is given in:
http://article.gmane.org/gmane.comp.sysutils.autoconf.general/1893

Given the number of affected calls in this lib, it might be
the  best trick to just #error out as soon as a largefile
compile -Ddef will be seen. It's not the best thing, right.

cheers,
-- guido                                 http://ac-archive.sf.net
Permalink | Reply | 0 comments |
headers
Pete Ryland | 14 Jan 23:33
Favicon

Re: ATRAC3

On Thu, Jan 02, 2003 at 12:23:33PM -0800, Michael Pruett wrote:
> On Tue, Dec 31, 2002 at 07:41:20PM +0000, Pete Ryland wrote:
> > [adding in support for ATRAC3-encoded files]
> 
> Pete, adding support for ATRAC3 would be a useful feature.  While it
> might be possible to use Sony's Windows codec to accomplish this task,
> but this approach would have the disadvantage of requiring some sort of
> Windows emulation layer which would likely only run on Intel x86
> systems.  It would certainly be preferable to write an open source
> codec which could be easily integrated into the Audio File Library.

Yes, I would prefer not to be bound to a specific platform too, but I have
had a lot of trouble finding any specific documentation of the entire
algorithm - it seems to be fairly closed, unfortunately.  At
http://www.minidisc.org/ the applicable patents are conveniently listed, but
they don't really help that much either.  I have written to various Sony
divisions, but I have had little success there - they generally state that
the (NetMD device) manual is all you get.

I was under the impression that some other proprietary file formats are
supported by wrapping a Windows codec (at least for x86), which is why I
thought it might be the way to go until/if Sony releases more info.
Unfortunate as it is to go this way, a lot of people would still benefit
from it.

Pete
--

-- 
Pete Ryland
http://pdr.cx/
(Continue reading)

Permalink | Reply | 0 comments |

Gmane