PlaybackPositionHelper.StorePlaybackPosition

[mvallevand]

Sub, I have a few questions on storing the resume position.

- if a file plays through to the end how is StorePlaybackPosition updated, ie. does it use the last PTS? This will general set a useless resume position one second or so early.

- if a seek manual or comskip is to end of file the PTS value, I am not seeing an update to StorePlaybackPosition, should this not store a value.

- what does NPVR do for resuming if the resume time and duration are the same?

Martin

[mvallevand]

Sub, I had time to do my own testing so could you confirm my find.

-it does seem to store the PTS

-if a seek past EOF is attempted, then it will try to seek 1-2 seconds early.

-there seems to be a threshold of around duration - ~5 seconds where resume will not be displayed

- Another observation on the duration == current position. In this case the Resume prompt on Recordings is shown, but the seek to location is odd (I see 115 seconds??)

Martin

[sub]

NPVR hasnt explicitly done anything to avoid showing the resume prompt next time if the user has watched to the end of a recording, but it should. I just dont typically watch the same recording twice, so I forgot to handle the situation.

The attached patch should remove the resume prompt in the Video Library or Recordings screen when the user has played a file to completion (or very near to it).

[mvallevand]

This still didn't do it for me on the NMT. In this case there is no StorePlaybackPostion of the last PTS when the file is stopped. The db still shows the same position that I started with.

The log has two examples skipping past EOF by ## Next and just letting it play.

Martin

[sub]

I see your logs dont seem to have a log message like this when it stops, whch is most likely a factor.

2011-03-07 08:11:41.760 [DEBUG][1] PlaybackPositionHelper.StorePlaybackPosition(D:\Downloads\TV\Caprica\caprica.s01e08.hdtv.xvid-fqm.avi, 184, 2598)

It does this correctly for me on the PC, but I havnt checked it on the NMT. I'll see if I can find why it's not happening on the NMT.

[mvallevand]

It works fine on the Stop from the remote the issue is at EOF.

Code:

2011-03-06 14:24:55.689    [DEBUG][3]    User pressed: S
2011-03-06 14:24:55.694    [DEBUG][3]    PlaybackPositionHelper.StorePlaybackPosition(C:\movies\Snowboarding\Snowboarding_20110305_16001700.ts, 118, 1045)

Martin

[sub]

It should happen on EOF too. It is here on the PC.

I took look at the NMT code, and cant see an obviously reason it's not triggering. I'll try running it up in the debugger a little later to see why not.

[mvallevand]

If you are running from the NMT, I suggest you use this mvpmcx2 http://forums.nextpvr.com/showthread.php...post405890 I send the duration send by NPVR as the last PTS on EOF seeks so you just seek past the end to test it quickly.

Martin

[sub]

I tried my patch with my NMT (using the mvpmcx2 I already had), and it seemed to work fine. If I played to the end, then it stored the playback position, and didnt prompt me for resume when I tried playing again.

Can you try again with this patch, then post your mvp.log? It adds a couple of logging messages to help me understand what exactly it's doing on your machine.

[mvallevand]

Here you go.

There is one thing in the logs that can't be right,

Code:

2011-03-06 15:49:48.484    [DEBUG][3]    User is trying to skip.
2011-03-06 15:49:48.484    [DEBUG][3]    Current position is: 6.07273333333333
2011-03-06 15:49:48.484    [DEBUG][3]    Asking for position: 966.072733333333
2011-03-06 15:49:48.484    [DEBUG][3]    Sending 'Play' Request
2011-03-06 15:49:48.484    [DEBUG][3]    Sending seek request to MVP/NMT. current position = 966, duration = 1045
2011-03-06 15:49:48.484    [DEBUG][3]    After skipping position reports as: 6.07273333333333
2011-03-06 15:49:48.484    [DEBUG][3]    OnKeyDown() complete

The "after skipping" is before the NMT even gets the request but this seems to be the value stored.

Martin