> Currently with AVI, if the subs stream is muxed into the file ( bad bad > bad :( ), it has to be done this way, so the parser has to detect what > type of subtitle stream it is ( text, SSA, ASS, etc. ). ??? The avi parser in dshow knows nothing about the format of a subtitle file muxed into, it just says it's text and dvobsub reads it in a format it can figure out. > For this very > reason VobSub, the most used subtitles editor on Windows, will write a > tag line into the subs stream on the very beginning as a kind of identifier. If you mean submux (and also avimux recently), it puts a predefined "GAB2" zero terminated string at the beginning of the generic text sample and dumps the real subtitle file after that. > This is making us a lot of problems for our DirectShow parser filter, as > stupid DShow only knows one type of subtitle streams also. We decided > now to 'extend' Dshow a bit here ( ;) ) and agreed with Gabest, the > author of VobSub who is also the developer of DVobSub, the most used > DShow based subtitlte display filter, to add new 'media subtypes' to > DirectShow to make a 'cleaner' solution here in future. What are the new subtypes? I meant I'd be good to have 1 new subtype which differs from MEDIASUBTYPE_Text and has some kind of format block to store the needed global data like style declarations. http://usf.corecodec.org