cdparanoia changes

[Jason Voegele]

Monty has made a few changes to Xiph's cdparanoia recently.  I am wondering if
you have any plans of porting ongoing changes to cdparanoia to the CdIo
paranoia library, or if the sources have diverged too much to make this
possible?

--
Jason Voegele
I am examining you on your fool ideas that no intelligent Christian
on earth believes.
                -- Clarence Darrow, to William Jennings Bryan

[Peter Creath]

Thank you for pointing this out.  I had recently sent Monty the merged
patches/comments I made to libcdio-paranoia long ago.  I didn't
realize he had merged them into the main trunk.

Not only did he merge them, but he appears to have answered all of my
outstanding questions.  Hurrah!

Have you noticed any functional changes he made?

[Jason Voegele]

On Friday 16 May 2008 10:55:58 am Peter Creath wrote:
> Thank you for pointing this out.  I had recently sent Monty the merged
> patches/comments I made to libcdio-paranoia long ago.  I didn't
> realize he had merged them into the main trunk.
>
> Not only did he merge them, but he appears to have answered all of my
> outstanding questions.  Hurrah!
>
> Have you noticed any functional changes he made?

Not yet.  I just updated my copy of the source from Subversion and noticed
several recent commits.  I've built a local copy and I'm going to experiment
with it.

My overarching concern is whether to use cdparanoia from Xiph or
cdio-cdparanoia from libcdio as the engine for my CD ripper.  If the intent
is that libcdio will incorporate changes from Xiph then that would make my
decision easy. :)

--
Jason Voegele
"Well," Brahma said, "even after ten thousand explanations, a fool is no
wiser, but an intelligent man requires only two thousand five hundred."
                -- The Mahabharata.

[R. Bernstein]

Jason Voegele writes:
 > My overarching concern is whether to use cdparanoia from Xiph or
 > cdio-cdparanoia from libcdio as the engine for my CD ripper.  If the intent
 > is that libcdio will incorporate changes from Xiph then that would make my
 > decision easy. :)

It looks as though in this case the intent was Xiph was incorporating
the changes from libcdio :)

[Jason Voegele]

On Friday 16 May 2008 11:22:14 am R. Bernstein wrote:
> Jason Voegele writes:
>  > My overarching concern is whether to use cdparanoia from Xiph or
>  > cdio-cdparanoia from libcdio as the engine for my CD ripper.  If the
>  > intent is that libcdio will incorporate changes from Xiph then that
>  > would make my decision easy. :)
>
> It looks as though in this case the intent was Xiph was incorporating
> the changes from libcdio :)

Ah, that is even better. :)


--
Jason Voegele
Even if you aren't in doubt, consider the mental welfare of the person who
has to maintain the code after you, and who will probably put parens in
the wrong place.  -- Larry Wall in the perl man page

[Peter Creath]

I quickly skimmed Monty's changes to paranoia/, and almost all of them
look like the changes backported from libcdio (mostly comments and a
few bug fixes that originated in libcdio).

He has made a handful of other code changes that I need to examine (to
understand and document), and then I plan to port those over to
libcdio so that it's in sync with Xiph.

Note that there will continue to be a little bit of divergence with
Xiph, mostly due to the underlying infrastructure (libcdio vs.
paranoia's direct SCSI access), but also because Monty chose not to
merge in the TRACE_PARANOIA debug code.

Interestingly, he also updated cdparanoia to LGPLv3 from GPLv2.

[R. Bernstein]

Jason Voegele writes:
 > On Friday 16 May 2008 11:22:14 am R. Bernstein wrote:
..
 > > It looks as though in this case the intent was Xiph was incorporating
 > > the changes from libcdio :)
 >
 > Ah, that is even better. :)

But that said, lest I give the wrong impression, it's really Peter
Creath seems to have did the work either way -- so thanks Peter.

[R. Bernstein]

Peter Creath writes:
 > Note that there will continue to be a little bit of divergence with
 > Xiph, mostly due to the underlying infrastructure (libcdio vs.
 > paranoia's direct SCSI access), but also because Monty chose not to
 > merge in the TRACE_PARANOIA debug code.

I wonder why not. I think this code in fact started from paranoia. And
testing is kind of a neat thing to do, no?

 >
 > Interestingly, he also updated cdparanoia to LGPLv3 from GPLv2.

As I said, maybe the intent was to follow libcio :-)

[R. Bernstein]

Peter Creath writes:
 >
 > No, TRACE_PARANOIA was something I added to libcdio's version of
 > paranoia while I was documenting it.  He has other debugging messages
 > floating around, but TRACE_PARANOIA provides more verbose logging of
 > stage 1 and 2.  I presume Monty doesn't need it, but it turns out to
 > be useful in libcdio's regression tests (when they fail).

Ah - okay. Thanks for the clarification and info.
 >
 > > As I said, maybe the intent was to follow libcio :-)
 >
 > I knew libcdio moved to v3, but did it also move to lesser GPL?

Ditto - I didn't notice that. No GPL (v3), not LGPL. Again clearly
others are paying more close attention than I am. So thanks for
filling me in.

[R. Bernstein]

Peter Creath writes:
 > Not only did he merge them, but he appears to have answered all of my
 > outstanding questions.  Hurrah!

What were the questions and what are the answers?

(I vaguely recall something like the issue of silence versus silence
using 0-valued amplitude; it might be good to have them written down
somewhere other than old email.)

Thanks.

[Peter Creath]

On Fri, May 16, 2008 at 8:30 PM, R. Bernstein <rocky@...> wrote:
> What were the questions and what are the answers?

They're all in the comments.  I'll be porting them over when I get a chance.

[Robert William Fuller]

Peter Creath wrote:
> On Fri, May 16, 2008 at 8:30 PM, R. Bernstein <rocky@...> wrote:
>> What were the questions and what are the answers?
>
> They're all in the comments.  I'll be porting them over when I get a chance.

An aside....  What does it mean when paranoia indicates a "skip" error?
    How would this relate to reading from the disc pre-gap?  Note that
the sector number would be in the range (-150, 0].