Re: [Libcdio-devel] CD_MSF_FORMAT vs LBA on NetBSD

[Edd Barrett]

Hi Thomas,

On Sat, Dec 01, 2018 at 08:40:00PM +0100, Thomas Schmitt wrote:
>    if (track_num == CDIO_CDROM_LEADOUT_TRACK)
>            track_num = TOTAL_TRACKS + 1;

I think that's ok after all. If we look at the definition of
TOTAL_TRACKS, we see that it takes the starting track into account:

---8<---
#define TOTAL_TRACKS (_obj->tochdr.ending_track \
                      - _obj->tochdr.starting_track + 1)
--->8---

>   cdrskin -v dev=/dev/rcd0d cd_start_tno=7 trackA.wav trackB.wav trackC.wav

Thanks for this!

I made a CD that starts at track 7. It works perfectly in my Sony CDP-30
deck, and it correctly shows the first track number as track 7.

I was surprised to see that neither cd-info, nor cd-paranoia (the cdio
fork) could understand this disk however! cd-info suffers from the same
issue that the NetBSD driver had: assumes the start track is 1.

Here's the fix for that:
https://github.com/vext01/libcdio/commit/49b1910b19d9a4b41eb2200e61d77e20f1802f58

cd-paranoia looks more complicated. Can someone try a disk with an odd
starting track number on Linux? Does it work?

Note that cd-paranoia does work (on my fork and OpenBSD) for disks with
starting track 1.

Note also that the xiph version of cdparanoia also works with my odd
starting number disk.

-- 
Best Regards
Edd Barrett

http://www.theunixzoo.co.uk