[mythtv] MARK_COMM_START, MARK_COMM_END

[Robert Tsai]

On Sun, May 21, 2006 at 07:01:01PM +0200, Lucas Meijer wrote:
> > 	- Which frame should be marked as MARK_COMM_END? The last
> > 	frame of the commercial break, or the first frame of the
> > 	content? If the latter, that means it is technically
> > 	impossible to represent a single-frame commercial-break; is
> > 	that intended? (I acknowledge that a single-frame break is
> > 	more likely a falsely-indicated break, so that this scenario
> > 	is not real-worldly expected to ever happen.)
>
> I don't think nobody really defined that yet.
>
> > 	- What am I doing wrong? :)
>
> Not much help here, other than having looked at your code snipplet,
> and seeing that that is also what I would expect to work. The
> ClassicCommDetector returns a map, with framenumbers as key, and
> either 0, MARK_COMM_START or MARK_COMM_END as values. maybe your
> problem is indeed not related to your commflagger. Do classicly
> flagged recordings work fine?

I think 0 (a.k.a. MARK_CUT_END) is only used for when a cutlist is
loaded; I don't think it's directly "returned" by the commflagger for
commercial-detection purposes.

> Does your database contain 4's and 5's after you ran your flagger?
> 
> do you run breathe() from time to time (shouldn't matter, but who
> knows)
> 
> do you inform mythcommflag that you have new commflag info to share
> with it? (gotNewCommercialBreakList() signal).
> 
> I'm pretty sure you'd do all those things already, but other than
> that, I have no idea what's going on. the interfacing with
> mythcommflag regarding your results look fine to me.

Yeah, everything else is as you suggest, and I think everything is
fine now (as per earlier e-mail from Steve).

--Rob