Re: [VCDImager Bugs/Devel] Further to my prev email re Chinese VCD

[Herbert Valerio Riedel]

ok, I've identified the problem... seems as if the authoring tool copied
the root IDR to the PVD, which has only a very limited space reserved for
the root idr, but failed to adjust the record length field... thus causing
a kind of harmless buffer overflow...

well, I'll replace that assertion by a warning and a return...

On Sat, 23 Jun 2001, Nick Andrew wrote:

> G'day,
>
> Here's a hex dump of the idr structure at the place where it detects
> the missing XA letters ...
>
> 0000 30 00 14 00 00 00 00 00 00 14 00 08 00 00 00 00  |0...............|
> 0010 08 00 63 05 11 16 12 14 00 02 00 00 01 00 00 01  |..c.............|
> 0020 01 00 20 20 20 20 20 20 20 20 20 20 20 20 20 20  |..              |
> --DEBUG: 48 1 14,
>
> I changed the program to return instead of suicide and tried --norip.
> It ran my debugging code 16 times, and printed the following:
>
> --DEBUG: VCD 2.0 detected
> ++ WARN: fyi, entry point and auto pause locations cannot be determined 
> without mpeg extraction -- hope that's ok
>    INFO: writing xml description to `videocd.xml'...
>
> I have attached the videocd.xml file it wrote.
>
> Going to a real read now, it looks clean except for:
>
> --DEBUG: reading 00:08:41
> --DEBUG: reading 00:08:42
> --DEBUG: cn +edge @492
> --DEBUG: reading 00:08:43
> --DEBUG: reading 00:08:44
>
> What does "cn +edge @492" mean?
>
> and so on until:
>
> --DEBUG: encountered subheader EOF @279709
> --DEBUG: truncating file to 279173 packets
>    INFO: writing xml description to `videocd.xml'...
>    INFO: done
>
> I get this truncation message on all rips (I don't know if that means
> the mpeg is broken, but in the case of The Matrix it certainly truncated
> the video) ...
>
> Nick.
>

-- 
Herbert Valerio Riedel       /    Phone: (EUROPE) +43-1-58801-18840
Email: address@hidden       /    Finger address@hidden for GnuPG Public Key
GnuPG Key Fingerprint: 7BB9 2D6C D485 CE64 4748  5F65 4981 E064 883F 4142