mythtv at heathsworld
Dec 6, 2009, 9:45 AM
Post #6 of 6
(655 views)
Permalink
|
|
Re: Channel scanner duplicates with HDHomerun
[In reply to]
|
|
On Sun, Dec 6, 2009 at 11:00 AM, Myth User <mythtv [at] heathsworld> wrote:
> On Sat, Dec 5, 2009 at 9:14 PM, <stefan_jones [at] comcast> wrote:
>
>> On Wed, Nov 25, 2009 at 10:24 PM, Myth User <mythtv [at] heathsworld>wrote:
>> Ubuntu 9.10 with HDhomerun.
>>
>>> >
>>> > In mythtv-setup am trying to scan these 2 channels (to keep it simple):
>>> >
>>> > 114-1
>>> > 114-2
>>> > 114-3
>>> > 114-6
>>> >
>>> > 115-2
>>> > 115-4
>>> > 115-6
>>> >
>>> > When telling the channel scanner to scan, it will finish and tell me I
>>> have 3 new "non-conflicting" MPEG channels. I go ahead and add those (which
>>> I can't tell what they are but happen to be 115-2,4,6)... Next it says I
>>> have conflicting MPEG channels and have to add them manually. Why are the
>>> actual channel numbers 114 and 115 not being attached so the scanner knows
>>> the difference? Why can't it simply add the channels as listed above? The
>>> 114 and 115 part of the channel is omitted and the dash channels (sub
>>> channel numbers) are conflicting. I am looking at manually add all my
>>> channels if this is really broken in .22. Please let me know if I am
>>> missing something because it just worked in .21.
>>> >
>>> > Thanks in advance.
>>> > On Fri, Nov 27, 2009 at 8:44 PM, Mitch Gore <mitchell.gore [at] gmail>wrote:
>>
>>
>>> > yup, broken/issue. Thats what i think. Read the archives and you will
>>> find several issues like this, including me. Devs say that this is not an
>>> error/bug even tho it worked fine in .21
>>> >
>>> --
>>> > Mitchell
>>> >
>>>
>> ----- "Myth User" <mythtv [at] heathsworld> wrote:
>> > I see....... I did run across that thread but seemed silly that someone
>> would say it works but not tell the
>> >world how. That's really too bad because the truth is that it did work
>> right in .21. So if the dev's are so >certain that it really works (and
>> works better), why can't they provide documentation on how to click the
>> > right buttons? It's obviously not intuitive like in .21 because it just
>> worked. I've tried everything but a
>> > hammer and going back to .21....
>>
>>
>> My take on this:
>>
>> The .22 channel scanner may very well be far superior to that in .21.
>>
>> But the user interface isn't all there yet.
>>
>> The dialog boxes at the end are the GUI equivalent of a series of
>> questions at the end of a terminal-output application. In addition to being
>> a one-way trip, the choices are not well explained. Breaking up
>> problematical channels into SCTE, MPEG, etc. adds to the puzzlement.
>>
>> The current documentation does not help.
>>
>> Ideally, at the end of the scan the user would be presented with an
>> interactive list, sort of a Channel Pre-Editor, of problem channels. Each
>> entry would include the channel with which it conflicts. After deleting or
>> editing the entries, the survivors would be added to the list of valid
>> channels.
>>
>
> Agreed. However, as long as the scanner continues to omit the actual
> channel number and only add the sub channels there will always be
> irresolvable conflicts. There could be 100 channels that have "-1" or "-2"
> sub channels therefore 100 conflicting channels to resolve one at a time in
> a blind manner. The really crazy part is that if you edit a channel (after
> you manually assign them due to conflicts), the channel number has been
> detected and labeled inside each of the detected channels in another
> duplicate field. But, the scanner can't tell the difference between 114-1
> and 115-1 at this point because they both have a dash-one.
>
> In summary I believe the code should be changed to detect conflicts
> differently. Currently it matches like:
>
> If $subchannel = $subchannel, then conflict
>
> and should be:
>
> if ($channel)-($subchannel) = ($channel)-($subchannel), then conflict.
>
>
Those are not minus signs, but the dash in the sub channel.
if "($channel)-($subchannel)" = "($channel)-($subchannel)", then conflict.
|
