[mythtv] ATSC scan, default channel number

Klaas de Waal klaas.de.waal at gmail.com
Thu Jul 15 21:44:52 UTC 2021


On Wed, 14 Jul 2021 at 22:24, Klaas de Waal <klaas.de.waal at gmail.com> wrote:

>
>
> On Tue, 13 Jul 2021 at 19:49, John P Poet <jppoet at gmail.com> wrote:
>
>> On Tue, Jul 13, 2021 at 11:36 AM Klaas de Waal <klaas.de.waal at gmail.com>
>> wrote:
>>
>>>
>>>
>>> On Sat, 10 Jul 2021 at 17:44, John P Poet <jppoet at gmail.com> wrote:
>>>
>>>> On Sat, Jul 10, 2021 at 7:30 AM Klaas de Waal <klaas.de.waal at gmail.com>
>>>> wrote:
>>>>
>>>>>
>>>>>
>>>>> On Sat, 10 Jul 2021 at 10:32, Klaas de Waal <klaas.de.waal at gmail.com>
>>>>> wrote:
>>>>>
>>>>>>
>>>>>>
>>>>>> On Fri, 9 Jul 2021 at 22:57, John P Poet <jppoet at gmail.com> wrote:
>>>>>>
>>>>>>> Hi Klaas,
>>>>>>>
>>>>>>> I am setting up a new development machine to do some work. In the
>>>>>>> process I noticed that when scanning ATSC channels, no default channel
>>>>>>> number is given and the user must manually choose one for every single
>>>>>>> channel. This is when the user chooses to manually pick which channels get
>>>>>>> inserted into the DB.
>>>>>>>
>>>>>>> For example, the message I am getting is:
>>>>>>>
>>>>>>>> You chose to manually insert channel 7_1 (KOAT-DT). Please enter a
>>>>>>>> unique channel channel number. Default value is
>>>>>>>>
>>>>>>>
>>>>>>> If I just click on OK, it just redraws the same dialog box. Instead
>>>>>>> I have to select Edit and type in 7_1. I could have sworn this was more
>>>>>>> automatic the last time I did this. Any chance you can have it pick a good
>>>>>>> default?
>>>>>>>
>>>>>>> Hi John,
>>>>>>
>>>>>> This definitely should be automatic so there is something really
>>>>>> wrong.
>>>>>> The correct behavior is to use the received channel number without
>>>>>> asking and when there is a conflict the suggestion channel number is with a
>>>>>> letter added.
>>>>>> There should also be the option in the dialog for "OK for all"  so
>>>>>> that it is never asked again.
>>>>>> It could be a coincidence but there are also threads in the forum
>>>>>> about ATSC cable scanning not working at all or crashing once the scan is
>>>>>> complete. And that is with v31. I assume you are using master.
>>>>>> At least you still get the correct channel numbers and names, which
>>>>>> the ATSC cable scans do not even do anymore.
>>>>>> What makes this a puzzle is that I have not changed anything
>>>>>> significant in the scanning code for a long time now.
>>>>>> On the other hand, maybe people do not do channel scans that often....
>>>>>> The scanning does work correctly with DVB-T2/C/S2 as you might have
>>>>>> guessed.
>>>>>> I will have a look at the code this weekend and possibly try to do a
>>>>>> simple ATSC scan with a captured ATSC transport stream. Have done that in
>>>>>> the past so I should be able to do that again....
>>>>>> In the meantime, if you have an idea what is wrong be welcome to
>>>>>> explore and fix it!
>>>>>>
>>>>>> Hi John,
>>>>>
>>>>> Can you please send me a complete backup of your database after you
>>>>> have done one or more scans?
>>>>> Rationale:
>>>>> In master the "Scan type"  "Import Existing Scan" allows playing back
>>>>> again a scan previously done. I expect the bug to be in the channel
>>>>> importer where the options such as "Complete scan data required" etc are
>>>>> processed and this can be replayed from an existing scan. Which then
>>>>> hopefully enables me to debug the channel numbering code with your ATSC
>>>>> scan.
>>>>>
>>>>> Thanks,
>>>>> Klaas.
>>>>>
>>>>
>>>> Yes, I can do that. I will replicate what I did when I saw the problem,
>>>> which was to do a scan on an initial setup, i.e. brand new database.
>>>>
>>>>
>>>> Hi John,
>>>
>>> I've resurrected the capability to scan from a single transport stream,
>>> according to your description in
>>>
>>> https://github.com/MythTV/mythtv/commit/8828cc5e4698f72e25fcfc31782c42fda5079717#diff-09ba904030e3795f57e6332cba578167
>>> "Add the ability to scan for the programs in a MPTS stream from an IPTV
>>> source."
>>> I needed a small fix in channelscan_sm.cpp to make this work but that is
>>> another story.
>>> The good news is that I should now be able to reproduce and debug the
>>> channel importer issue.
>>>
>>> Question: can you send me a full MPTS that does not scan correctly?
>>>
>>> Thanks,
>>> Klaas.
>>>
>>
>> Hi Klaas,
>>
>> Yes, I can do that. It may take me a couple of days, but I will get it to
>> you by the end of this week.
>>
>>
>> Hi John,
>
> It looks like I can reproduce the problem. I found a transport stream from
> the beginning of 2020 which I think I actually got from you.
> When I scan this with the IPTV UDP method I do get wrong results.
> It has e.g.  KWBQ-TV which then gets channel number 5001-3, the 5001
> being the UDP port number.
> However, in the scan log it looks like KWBQ-TV should be 19-1 and the
> other channels 19-2, 19-3 and 19-4.
> This is the VCT:
> (HandleVCT) - ChannelScanSM[7]: Got a Virtual Channel Table for 0
> (LogLines) - Terrestrial Virtual Channel Section
> (LogLines) - PSIP tableID(0xc8) length(215) extension(0x817)
> (LogLines) - version(0) current(1) section(0) last_section(0)
> (LogLines) - atsc_protocol_version(0)
> (LogLines) - channel_count(4) tsid(0x0817)
> (LogLines) - Channel #0 name(KWBQ-TV) 19-1 mod(ATSC 8-VSB) cTSID(0x0817)
> (LogLines) - pnum(3) ETM_loc(1) access_ctrl(0) hidden(0) hide_guide(0)
> service_type(ATSC TV)
> (LogLines) - source_id(1)
> (LogLines) - descriptors length(23) count:1
> (LogLines) - Service Location Descriptor tag(0xa1) length(21) Dumping
> (LogLines) - 000 e0 31 03 02 e0 31 00 00 00 81 e0 34 65 6e 67 81 |.1...1
> .....4eng.|
> (LogLines) - 010 e0 35 73 70 61 |.5spa|
> (LogLines) - Channel #1 name(Grit) 19-2 mod(ATSC 8-VSB) cTSID(0x0817)
> (LogLines) - pnum(4) ETM_loc(1) access_ctrl(0) hidden(0) hide_guide(0)
> service_type(ATSC TV)
> (LogLines) - source_id(2)
> (LogLines) - descriptors length(17) count:1
> (LogLines) - Service Location Descriptor tag(0xa1) length(15) Dumping
> (LogLines) - 000 e0 41 02 02 e0 41 00 00 00 81 e0 44 65 6e 67
> |.A...A.....Deng|
> (LogLines) - Channel #2 name(Laff) 19-3 mod(ATSC 8-VSB) cTSID(0x0817)
> (LogLines) - pnum(5) ETM_loc(1) access_ctrl(0) hidden(0) hide_guide(0)
> service_type(ATSC TV)
> (LogLines) - source_id(3)
> (LogLines) - descriptors length(17) count:1
> (LogLines) - Service Location Descriptor tag(0xa1) length(15) Dumping
> (LogLines) - 000 e0 51 02 02 e0 51 00 00 00 81 e0 54 65 6e 67
> |.Q...Q.....Teng|
> (LogLines) - Channel #3 name(Ion) 19-4 mod(ATSC 8-VSB) cTSID(0x0817)
> (LogLines) - pnum(6) ETM_loc(0) access_ctrl(0) hidden(0) hide_guide(0)
> service_type(ATSC TV)
> (LogLines) - source_id(4)
> (LogLines) - descriptors length(17) count:1
> (LogLines) - Service Location Descriptor tag(0xa1) length(15) Dumping
> (LogLines) - 000 e0 61 02 02 e0 61 00 00 00 81 e0 64 65 6e 67
> |.a...a.....deng|
>
> I think that this is the problem so I will start looking into this.
>
> The channel numbers like 5001-3 are generated because they are scanned as
IPTV channels.
This is done in channelscan_sm.cpp lines 1712-1714.
When I select a manual install then the channel numbers lik 5001-3 are
suggested as default. Pressing OK accepts this. Which is correct.
It would be possible to use ATSC major-minor as default channel number also
for IPTV channels but that is then new functionality, not a bug fix.
I have not yet been able to find the path in the code that generates an
empty default channel number when using the manual install.
It would be nice to have an "OK all" option when manual install is
selected, to accept the default for all following channels.
Question: do you use IPTV or HDHomeRun or /dev/dvb tuners?

To be continued!
Klaas.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mythtv.org/pipermail/mythtv-dev/attachments/20210715/71befbfe/attachment.htm>


More information about the mythtv-dev mailing list