steve.lhomme at free.fr
Sun Sep 4 12:05:16 CEST 2005
>> On Friday 02 September 2005 18:36, Yang wrote:
>>> I asume so, becouse after 2 muxings with same setings, the only
>>> diference, what i can find in MatroskaPropv2.8 in advanced view, is a
>>> diferent UID.
>> Muxing date, muxing application, writing application, segment UID, track
>> UIDs, attachment UIDs, chapter UIDs, tag UIDs. All those would have to
>> be forced. The exactly same version of mkvmerge would have to be used.
>> Sorry, too much work. I won't do it anytime soon.
> Chapter UID is already predefined, same as edition UID. Same version of
> mkvmerge is quite obvious. So there is only needed track, tag and
> attachment UID's to be forced. And only problem then would be a date of
> muxing. But hell with it, it could be also predefined ^_^ (but as
> default it woul be readed from system clock). I dont know how exacly
> matroska is built, but if you already done half of the job, then i think
> it woudnt be so hard to do. mkvmerge already creates those UID by
> random, so i asume that there is only need for comand, that changes that
> random number to predefined one.
No, the numbers are random precisely to identify different files
globally (like on P2P networks). What you are asking for, if I
understand correctly, is not possible. If a file has different content
(whatever it is) the Segment UID has to be different. And the checksum
of the file will *never* be the same if the content is different.
More information about the Matroska-users