[mythtv] Channel scanning/validation in trunk (feedback post r14355)

[Jarod Wilson]

On Aug 31, 2007, at 10:10 AM, Daniel Kristjansson wrote:

> On Fri, 2007-08-31 at 08:58 -0500, Willy Boyd wrote:
>> On 8/30/07, Daniel Kristjansson <danielk at cuymedia.net> wrote:
>>> Anyway, I'm not looking for feedback on the channel scanner anymore
>>> than usual. At the moment I'm more interested in whether the channel
>>> validation prevents channels from being added to the a Video Source
>>> when they should be added, or adds channels when it shouldn't be
>>> adding them. If someone is using esoteric recorders like DBOX2 or
>>> IPTV, or using a diverse collection of recorder types, feedback is
>>> especially appreciated.
>
>> Would this improvement also apply to a HDHomeRun and QAM channels?
>
> Again I'm not asking for any special testing of the scanner. In trunk
> the HDHomeRun scanner is much smarter about detecting encrypted  
> channels
> than it used to be, but there is a bug where the wrong tuning
> information is sometimes used, usually from the transport immediately
> before or after the correct one in the scan. I think Jarod might have
> backported the fix that let MythTV detect transports which looked
> encrypted on first glance, but upon further inspection revealed that
> the A/V streams were not actually encrypted.

Correct, changesets 13238 and 13243 were the backported bits to deal  
with that issue.

> I remember feeling really
> nervous when I noticed commits along that line in 0.20-fixes,

I tested the living heck out of 'em before committing, and since  
channel scanning was completely busted at the time anyhow, I figured  
it'd be just fine to bring 'em along. :) (hindsight being 20/20,  
looks like a good call!)

> but as I said before the 0.20.2 & 0.20-fixes channel scanner is the  
> best yet
> in any MythTV release.

Both 0.20-fixes and trunk seem to work equally well for me, but then  
I don't have anything particularly interesting with my setup -- just  
run of the mill QAM cable.

--jarod