[mythtv-users] Schedules Direct TBA Listings

[Michael T. Dean]

On 07/17/2014 11:51 AM, Gary Buhrmaster wrote:
> On Thu, Jul 17, 2014 at 10:37 AM, Michael T. Dean wrote:
>> On 07/16/2014 06:44 PM, Thomas Mashos wrote:
> ....
>>> Actually, I believe the --dd-grab-all is the best fix, as it refreshes
>>>
>>> all of your data and is less intensive for SD.
>>
>> Yes.  Definitely.
> Perhaps, finally, on 0.28, '--dd-grab-all' should be the default.
> WIth an option such as '--dd-grab-old-slow-and-incomplete'
> added (ok, probably '--no-dd-grab-all', but I like my option
> better as being more descriptive).  As I recall, there is some
> minor complexity due to option conflict checks, but certainly
> doable.  The release notes (which no one reads, but...) would
> have to warn those running substantially under-resourced
> systems to override the new default, but still it would seem
> that the needs of the many outweigh the needs of the few....

The biggest problem with making it the default is that it causes hard 
failures on systems that can't handle it (never-started/"lost" 
recordings and can possibly even cause complete scheduler failures which 
may result in the backend running for days without doing any recordings 
and/or crashes and/or system failures due to unexpected "pulling the rug 
out" data loss).  With the current default, there are no hard 
failures--only "less efficient" scheduling due to not knowing about 
updates of current day and +2 through +12 listings.

I definitely won't be able to lead the change/handle issues for a while, 
but my preference would be to make the whole thing robust enough to 
prevent those failures (and efficient enough that even underpowered 
systems could handle it) before we change it.  That said, I wouldn't 
object to someone else's changing it if they take the lead on handling 
all the issues (and complaints :) that will undoubtedly surface.

However, the time spent on making --dd-grab-all work (or at least making 
it not cause major failures) for underpowered systems may be better 
spent on figuring out how to integrate the JSON-based delta-only 
grabbing into MythTV (and/or XMLTV).

Mike