[mythtv] MPEG2 fixup tool

[Cory Papenfuss]

> Yes it needs to compile against either SVN mythtv or the latest CVS
> ffmpeg code.  They changed how timestamps were handled sometime
> recently.
> Not sure about the problem on line 127 though.  I didn't think that
> the the way codecs were attached had been changed for ages.
>
> Also note, it can provide useful mpeg analysis by commenting out the
> #define WRITE and setting debug=1 or 2.
>
> If you get it to compile, and it doesn't fix your broken mpegs,
> senging me the logs of the above could help me get a handle on the
> problem.
> .Geoff

 	Well, I fried up the SVN and got your code to compile.  I've tried 
it on a couple of videos.  It appears to work, although I didn't see any 
sort of indication of frame insertion on one, and on another, I got a 
number of these:

[mpeg2video @ 0xb7ebda24]concealing 352 DC, 352 AC, 352 MV errors
[mpeg2video @ 0xb7ebda24]00 motion_type at 9 22
[mpeg2video @ 0xb7ebda24]00 motion_type at 17 23
[mpeg2video @ 0xb7ebda24]00 motion_type at 8 24
[mpeg2video @ 0xb7ebda24]00 motion_type at 4 25
[mpeg2video @ 0xb7ebda24]00 motion_type at 20 26
[mpeg2video @ 0xb7ebda24]00 motion_type at 9 27
[mpeg2video @ 0xb7ebda24]ac-tex damaged at 1 28
[mpeg2video @ 0xb7ebda24]00 motion_type at 4 29
[mpeg2video @ 0xb7ebda24]concealing 352 DC, 352 AC, 352 MV errors
[mpeg2video @ 0xb7ebda24]00 motion_type at 8 24
[mpeg2video @ 0xb7ebda24]00 motion_type at 4 25
[mpeg2video @ 0xb7ebda24]00 motion_type at 20 26
[mpeg2video @ 0xb7ebda24]00 motion_type at 9 27
[mpeg2video @ 0xb7ebda24]ac-tex damaged at 1 28
[mpeg2video @ 0xb7ebda24]00 motion_type at 4 29

 	I cranked up debug on another stream and got these:
Frame type: 2 pos: 80da490
Frame type: 3 pos: 805aa80
Frame type: 3 pos: 80fee08
Frame type: 1 pos: 81016f8
Rate: 3003
VID: I-Frame #:0 pts: 16200 exp: 16200 pos: 806e1aa
VID: P-Frame #:3 pts: -9223372036854775630 exp: 25209 pos: 805ea30
VID: B-Frame #:1 pts: 19203 exp: 19203 pos: 8073b80
VID: B-Frame #:2 pts: 22206 exp: 22206 pos: 80d3dd0
VID: P-Frame #:6 pts: -9223372036854775630 exp: 34218 pos: 8076708
VID: B-Frame #:4 pts: 28212 exp: 28212 pos: 80d71b0
VID: B-Frame #:5 pts: 31215 exp: 31215 pos: 80de730
VID: P-Frame #:9 pts: -9223372036854775630 exp: 43227 pos: 807a1d0
VID: B-Frame #:7 pts: 37221 exp: 37221 pos: 80e1930
VID: B-Frame #:8 pts: 40224 exp: 40224 pos: 8058de0
VID: P-Frame #:12 pts: -9223372036854775630 exp: 52236 pos: 80da490
VID: B-Frame #:10 pts: 46230 exp: 46230 pos: 805aa80
VID: B-Frame #:11 pts: 49233 exp: 49233 pos: 80fee08
AUD: pts: 31320
AUD: pts: 33480
AUD: pts: 35640

 	Overflow in the PTS-parsing?

-- 

*************************************************************************
* Cory Papenfuss                                                        *
* Electrical Engineering candidate Ph.D. graduate student               *
* Virginia Polytechnic Institute and State University                   *
*************************************************************************