Re: [linux-minidisc] First patch for write support on HiMD available
>Betreff : Re: [linux-minidisc] First patch for write support on HiMD available
>Gesendet: 11.03.2011 02:56:17
>An : "Adrian Glaubitz <glaubitz@physik.fu-berlin.de>"
>Von: "Adrian Glaubitz <glaubitz@physik.fu-berlin.de>"
>----- Originale Nachricht -----
>On Mar 11, 2011, at 2:00 AM, Adrian Glaubitz wrote:
>>
>> On Mar 10, 2011, at 10:51 PM, manner.moe@gmx.de wrote:
>>>
>>>> where the track you just wrote to the HiMD should appear as the last one, in this case
>>>> it's a track from the F-Zero SNES game original soundtrack =).
>>>
>>> I think there is a bug in writing the trkidxXX.hma file. qhimdtransfer
>>> doesn´t show the newly uploaded track, SonicStage cannot read the disc
>>> ("not initialized by SonicStage") and my RH1 doesn´t show the track, too.
>>> There ist a key and EKB number stored in the track info table but these
>>> entries should be zero on mp3 files.
>>> Also the content ID shouldn´t be zero.
>>
>> Just for the record. Thomas and I have independently looked into that bug and Thomas
>> was a bit faster to discover the problem: it's the number of tracks which is not properly
>> updated in the track index file (track count located at 0x100 in the track index file),
>> see [1].
>I added a
>>setbeword16(himd->tifdata + 0x100, himd_track_count(himd) + 1);
>as a dirty hack at the end of "himd_add_track_info" which forces an increment
>of the track count in the track index file. QHiMDTransfer shows the proper
>number of tracks now, what makes me wonder, however, is that the track info for
>the added track is not set at all, e.g. the track info data is empty for that
>track in QHiMDTransfer while himddump shows the complete and proper
>track information.
You have to add an entry in the play order table, too (see trackslot section
in libhimd) at 0x102 + 2 x index.
QHiMDTransfer uses trackslots, so we have to create one.
Thomas
