[mythtv] [DVB] [PATCH] TS timing issue fix(s) - which is best?

[William Uther]

On Tue, 29 Mar 2005 17:01:50 +0800, Tj NG <htpc at treblid.dyndns.org> 
wrote:

> Nathan Manzi wrote:
>
>> William Uther wrote:
>>
>>> Patch:
>>>
>>> This patch implements the first fix above.  Note: This patch is
>>> untested.  I have tested commenting out the line that sets
>>> _scanning_pes_header_for_gop to false, but this patch removes that
>>> variable entirely.
>>>
>>>
>>>
>> Using CVS and the patch returns during make:
>>
>> dtvrecorder.cpp: In member function `virtual void 
>> DTVRecorder::Reset()':
>> dtvrecorder.cpp:86: error: `_scanning_pes_header_for_gop' undeclared
>> (first use
>>   this function)
>> dtvrecorder.cpp:86: error: (Each undeclared identifier is reported 
>> only once
>>   for each function it appears in.)
>> make[2]: *** [dtvrecorder.o] Error 1
>> make[2]: Leaving directory `/cvs/cvs/mythtv/libs/libmythtv'
>> make[1]: *** [sub-libmythtv] Error 2
>> make[1]: Leaving directory `/cvs/cvs/mythtv/libs'
>> make: *** [sub-libs] Error 2
>>
>>
>> I don't think you've removed all the instances where it's called. I'm
>> gonna play around with it for a bit, I'll post the results.

Sigh.  I am running CVS from a week ago on my main box - I didn't want 
to get distracted by the configure changes while working on the TS bug. 
  I made that patch in an up-to-date working copy on a different 
machine.

I guess I shouldn't have tried to cut corners.  As I said, commenting 
out that one line is the core of the fix - those who want to test right 
away can make the change easily enough.

I'll post a working patch soon. (tomorrow morning some time?)

> Anything else I'm missing? I applied the patch and removed other
> occurances of _scanning_pes_header_for_gop. I still can't ff/rewind a 
> TS
> recording (the time is wrong).. but when I hit 'E', the time is 
> correct.

Once a recording is recorded with the broken code, the timestamps will 
be all wrong and you'll have the problem.  You should be able to 
rebuild the keyframe information using ?commflag? and that should fix 
the problem.  Rebuilding the keyframe info didn't fix things for me 
before this change, and I don't think this change affects the 
rebuilding code, which suggests that there may be another bug in that 
part of the code... but I haven't tested that yet.

Be well,

Will         :-}

--
Dr William Uther                           National ICT Australia
Phone: +61 2 9385 6357               Computer Science and Engineering
Email: willu at cse.unsw.edu.au          University of New South Wales
Web: http://www.cse.unsw.edu.au/~willu/     Sydney, Australia