Vorbis file with packets on same page as Setup Header Options

[paultaylor]

I was having trouble with my jaudiotagger OggVorbis library code because it was failing to write changes to an OggVorbis file. The problem being that the original file contains additional (Audio?) packets on the same Ogg Page as the Setup Header, (I hadn't encountered this before , all my test files start a new page after the Setup header packet) - is it valid.

Below is a summary of the first three OggPages, as you can see the packets that follow immediately after the Setupheader are only one byte long, but the last packet is 255 to indicate it goes on to the next page, is this some attempt at padding within Ogg to allow the Vorbis Comment to expand, or a way to introduce a delay in the start of the music, real audio packet wouldnt be 1 byte long would it ?

Thanks anyone for any help you can give me with this,
Next Ogg Page Header:isValid:true:type:2:oggPageHeaderLength:28:length:30:seqNo:0:packetIncomplete:false:serNum:15
00
NextPkt(start:0:length:30), (CODEC INFO)

Next Ogg Page Header:isValid:true:type:0:oggPageHeaderLength:66:length:4325:seqNo:1:packetIncomplete:true:serNum:1
500
NextPkt(start:0:length:162), (VORBIS COMMENT)
NextPkt(start:162:length:3887), (SETUP HEADER)
NextPkt(start:4049:length:1),
NextPkt(start:4050:length:1),
NextPkt(start:4051:length:1),
NextPkt(start:4052:length:1),
NextPkt(start:4053:length:1),
NextPkt(start:4054:length:1),
NextPkt(start:4055:length:1),
NextPkt(start:4056:length:1),
NextPkt(start:4057:length:1),
NextPkt(start:4058:length:1),
NextPkt(start:4059:length:1),
NextPkt(start:4060:length:1),
NextPkt(start:4061:length:1),
NextPkt(start:4062:length:1),
NextPkt(start:4063:length:1),
NextPkt(start:4064:length:1),
NextPkt(start:4065:length:1),
NextPkt(start:4066:length:1),
NextPkt(start:4067:length:1),
NextPkt(start:4068:length:1),
NextPkt(start:4069:length:1),
NextPkt(start:4070:length:255),

Next Ogg Page Header:isValid:true:type:1:oggPageHeaderLength:51:length:4113:seqNo:2:packetIncomplete:false:serNum:
1500
NextPkt(start:0:length:393),
NextPkt(start:393:length:98),
NextPkt(start:491:length:149),
NextPkt(start:640:length:161),
NextPkt(start:801:length:153),
NextPkt(start:954:length:148),
NextPkt(start:1102:length:148),
NextPkt(start:1250:length:143),
NextPkt(start:1393:length:147),
NextPkt(start:1540:length:148),
NextPkt(start:1688:length:149),
NextPkt(start:1837:length:150),
NextPkt(start:1987:length:148),
NextPkt(start:2135:length:148),
NextPkt(start:2283:length:622),
NextPkt(start:2905:length:601),
NextPkt(start:3506:length:607),

[paultaylor]

Ran ogginfo on it

Appears it is valid, but with warnings

Processing file "test2.ogg"...

New logical stream (#1, serial: 000005dc): type vorbis
Warning: Vorbis stream 1 does not have headers correctly framed. Terminal header page contains additional packets or has non-zero granulepos
Vorbis headers parsed for stream 1, information follows...
Version: 0
Vendor: Xiph.Org libVorbis I 20020717 (1.0)
Channels: 2
Rate: 44100

Nominal bitrate: 260.486000 kb/s
Upper bitrate not set
Lower bitrate not set
User comments section follows...
ARTIST=Coolio
TITLE=Ganster Paradise
ALBUM=Black Groove
DESCRIPTION=CD'n'Go! Suite 2.00.945
GENRE=R&B
Vorbis stream 1:
Total data length: 7489653 bytes
Playback length: 4m:00.693s
Average bitrate: 248.935952 kb/s
Logical stream 1 ended

[Aoyumi]

Probably you make a mistake in the insertion point of the padding domain. The padding domain should be added after a comment header domain. And it is a part of the comment header domain.

This post has been edited by Aoyumi: Feb 19 2008, 15:29

[paultaylor]

Probably you make a mistake in the insertion point of the padding domain. The padding domain should be added after a comment header domain. And it is a part of the comment header domain.

Ok, I didnt actually create the file I was just trying to read it.

[kjoonlee]

I'm not sure if this helps, but there *are* some patched encoders that try to add padding, for future expansion of tag contents.

[paultaylor]

I'm not sure if this helps, but there *are* some patched encoders that try to add padding, for future expansion of tag contents.

Thanks, have you any links to a any ot these, do any run on Windows?