[mythtv] More scheduling scheduler

Paul Andreassen paulx at andreassen.com.au
Mon Apr 17 14:42:46 UTC 2006


On Mon, 17 Apr 2006 08:30 pm, Paul Andreassen wrote:
> > BUG #1: Setting or clearing a soft pad in "Setup" didn't trigger a
> > reschedule. That is, when I set my soft pad for the first time, nothing
> > happened until I changed one of my recordings.
> >
> > I think all that's needed to fix this is to add:
> >    ScheduledRecording::signalChange(0);
> > ... in mythfrontend/main.cpp after settings.exec().
>
> Sounds good.

Included in this patch.

> > BUG #2: The soft padding was sometimes mysteriously dropped when it
> > could have been preserved. I couldn't figure out why.
> >
> > Example 1:
> >
> > Title - Subtitle                    Chan Start  End   C I  T N   Pri
> > ABC News                               2 18:59-19:30  6 5  T 6   0/0
> > The Biggest Loser                      1 19:00-19:30  7 6  A 7   0/0
> > The 7.30 Report                        2 19:30-20:02  7 6  C 7   0/0
> > 20 to 1 - "Great Aussie Ideas, Icon    9 19:30-20:35  6 5  A 6   0/0
> >
> > Here THE BIGGEST LOSER has no 1-minute start-early pad, even though
> > tuner 7 is idle before that.
> >
> > When I added an override to prevent ABC NEWS from being recorded, THE
> > BIGGEST LOSER's start-early pad mysteriously re-appeared:
> >
> > Title - Subtitle                    Chan Start  End   C I  T N   Pri
> > The Biggest Loser                      1 18:59-19:30  7 6  A 7   0/0
> > ABC News                               2 19:00-19:30  0 0  O X   0/0
> > The 7.30 Report                        2 19:29-20:02  6 5  C 6   0/0
> > 20 to 1 - "Great Aussie Ideas, Icon    9 19:30-20:35  7 6  A 7   0/0
> >
> > Example 2:
> >
> > Title - Subtitle                    Chan Start  End   C I  T N   Pri
> > The Biggest Loser                      1 18:59-19:30  7 6  A 7   0/0
> > The 7.30 Report                        2 19:29-20:02  6 5  C 6   0/0
> > 20 to 1 - "Great Aussie Ideas, Icon    9 19:30-20:35  7 6  A 7   0/0
> >
> > Here 20 TO 1 has no soft padding. For some reason its end-late padding
> > has been dropped, even though there's nothing scheduled after it.
> >
> > When I changed the schedule so that 20 TO 1's start-early buffer could
> > be preserved (e.g. move its start time forward by a few minutes), its
> > end-late buffer mysteriously re-appeared also.
>
> I noticed this but before I put in some debugging statements and recompiled
> it went away.

This patch includes lots more PrintList(false) to help track this down.  Must 
be enable with 'mythbackend -v schedule'.

> > Example 3:
> >
> > Title - Subtitle                    Chan Start  End   C I  T N   Pri
> > Survivor: Panama - Exile Island - "    9 19:29-20:32  7 6  A 7   0/0
> > Thank God You're Here                  1 19:30-20:30  6 5  A 6   0/0
> > House - "Dnr"                          1 20:30-21:32  6 5  A 6   0/0
> > ER - "Two Ships"                       9 21:30-22:30  7 6  A 7   0/0
> >
> > "ER" has no soft padding, even though tuner 7 is idle before and
> > afterward.
> >
> > BUG #3: When a recording has a higher priority, its soft buffer turns
> > hard, causing conflicts with other shows.
> >
> > Title - Subtitle                    Chan Start  End   C I  T N   Pri
> > ABC News                               2 18:59-19:30  7 6  T 7   0/0
> > The Biggest Loser                      1 18:59-19:32  0 0  A C   0/0
> > Survivor: Panama - Exile Island - "    9 19:29-20:32  6 5  A 6   2/0
> > Thank God You're Here                  1 19:30-20:30  7 6  A 7   0/0
> >
> > Note that THE BIGGEST LOSER is conflicted, but it could be recorded if
> > its end-late padding and SURVIVOR's start-early padding was dropped.
> >
> > When I removed SURVIVOR's raised priority, this problem went away.
> > Similarly, if I elevated the priority of the other shows so they were
> > all +2, soft buffers were dropped in order to resolve conflicts.
>
> Have you turned on Utilities/Setup\Setup\Tv Settings\Recording
> Priorities\Set Recording Priorities\Reschedule Higher Priorities.  The
> default is off.

Changed the setHelpText to hopefully explain this.

> > SUGGESTIONS
> > -----------
> >
> > The following are mere ideas. Softpad will work perfectly well without
> > them. I hesitate to suggest them, because I would much rather have a
> > basic, working softpad in the trunk than a full-featured one stuck in a
> > branch. But I know other users want at least some of these, so I might
> > as well mention them as things to possibly consider in the future.
> >
> > SUGGESTION #1: In "Upcoming Recordings," recordings for which MythTV is
> > planning to drop the soft pad could be shown in a different color.
>
> I believe this is done by the themes.

Didn't add this.  Requires reusing a current colour or changing all themes.  
Not to hard.

> > Also, when a recording is highlighted, the lower information box could
> > show more detailed information. E.g. instead of just "Mon Apr 17, 8:29pm
> > - 9:32pm", it could display:
> > (a) Mon Apr 17, 8.29pm - 9:32pm (+1/+2)
> > ... or...
> > (b) Mon Apr 17, 8:30pm - 9:30pm (8:29pm - 9:32pm)
> > ... or something similar.
> >
> > These would help users identify recordings that will not have soft
> > buffers, and distinguish the guide data time from the padded time.
>
> This is easier because its just appending to a string.

Added the softend for display.  Not sure where else it'll be displayed.

> > SUGGESTION #2: When a program is conflicted, and the user views which
> > shows are conflicting it, MythTV could not list shows that only overlap
> > because of the soft padding. At present, it misleadingly suggests that
> > the conflict could be resolved by removing soft pads, which is not true.
>
> Haven't looked at this code.  Probably just lists all programs that are
> using the same time slot.

Its in Scheduler::getConflicting().

> > 
> > SUGGESTION #3: In "Setup," soft pads should be specified in minutes, not
> > seconds. Soft pads are "just in case" buffers, for which it's not
> > important to specify a time that is precise to the second.
>
> I believed some people would only require a small pad.  Some set top box
> take 10 seconds to change channels and remove their osd.
>
> > SUGGESTION #4: There are now two sets of buffer settings in "Setup",
> > distinguished only by the fact that one is called "soft" and the other
> > isn't called anything. However, it's backwards: pre-roll, being more
> > likely to be abandoned by the scheduler than this new system, is thus
> > actually softer.
> >
> > Perhaps pre-roll can be actually called "pre-roll" in the UI. Or I know
> > devs were considering limiting pre-roll/post-roll to less than a minute:
> > if this was done, and soft pad was specified in minutes as per
> > Suggestion #3, this would provide an even clearer distinction between
> > the two.
>
> Naming and describing this and the old padding is very difficult.

Did another attempt at this.

> > SUGGESTION #5: Users could cancel or modify the soft padding of
> > individual recordings/showings. Because fine-grained control is nice. :)
>
> I find it useful to enable it on a channel by channel basis.
>
> > Hope this helps! I'm happy to do more testing.
>
> Great.  Will do up another patch tonight.
>
> Paul

Will compile and install this tonight.

mythwebsoft3.patch.gz are the changes required for mythplugins.
sort3complete-3.patch.gz is all of above.
retrylist.patch should help when rescheduling.

Paul
-- 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: retrylist.patch
Type: text/x-diff
Size: 549 bytes
Desc: not available
Url : http://mythtv.org/pipermail/mythtv-dev/attachments/20060418/2ffe69b9/attachment.bin 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mythwebsoft3.patch.gz
Type: application/x-gzip
Size: 583 bytes
Desc: not available
Url : http://mythtv.org/pipermail/mythtv-dev/attachments/20060418/2ffe69b9/attachment-0001.bin 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: soft3complete-3.patch.gz
Type: application/x-gzip
Size: 2631 bytes
Desc: not available
Url : http://mythtv.org/pipermail/mythtv-dev/attachments/20060418/2ffe69b9/attachment-0002.bin 


More information about the mythtv-dev mailing list