Re: [Libcdio-devel] nrg audio not correct with dao

[Robert William Fuller]

R. Bernstein wrote:
> Robert William Fuller writes:
>  > Seems NRG didn't handle DAO in read_audio_sectors.  Ask me how much fun 
>  > I had with /usr/bin/cmp figuring this out.  Here's a clue:
>  > 
>  > 	cmp -l -i 0:352800 foo.bin /nfs/tmp/foo/foo.nrg
>  >   cmp: EOF on foo.bin
>  > 
>  > Note that 352800 is CDIO_PREGAP_SECTORS * 2352.
>
> Wow! And can you guess the number I'm thinking of right now?  

I'm afraid I can.  Is it the number 0x4b000?  (Decimal 307200.) 
Incidentally, the function with that number in it also seems to be 
missing the line "env->pos.lba = 0;" which appears in its peers in 
cdrdao.c and bincue.c.  I'm pretty sure its absence is a bug.  That will 
have to go on my list as well, but it will be pretty far down the list 
at this point.  Maybe someone else will pick this one up?

> Having to come play this sort of guessing game all too often is why I
> don't enjoy crossword puzzles and dread the all-too-common problem
> reports where crucial information from either the input or the output
> has been omitted (and replaced by the problem reporter's opinion of what's
> wrong which is correct less than half of the time.)

No doubt.

>  > The fix follows.  If it passes regression tests, I'm going to go ahead 
>  > and put it in.  (I'm going to keep checking things in until Rocky shoots 
>  > me in the head :-)
>
> Nah - libcdio needs some life breathed into it. Change a way! 

Sweet :-)

> But it would be real cool to create a DAO NRG image and add another
> regression test for that. Thanks.

Yeah that goes on the list too....  Sigh.  At some point I'm going to 
have to take a break from this and attend to some real life problems :-)