[mythtv] Seek (back and forth) broken in MythTV trunk r23186

R. G. Newbury newbury at mandamus.org
Mon Feb 1 23:00:11 UTC 2010

On 02/01/2010 02:34 PM, Michael T. Dean wrote:

>> So am I correct in reading the changeset, that 'StickyKeys' continues
>> to exist but as an internal code? C++ is NOT my bag!
> Nope. :)  (It's still in your DB, but the only place the setting exists
> in code is in the DB updates that fix key bindings for users.)
> The StickyKeys setting was a setting that just changed the meaning of
> key bindings.  The proper solution to getting the StickyKeys behavior is
> to have the appropriate actions available and allow users to customize
> their key bindings as desired.
> Basically, we already had something for configuring how keys work (key
> bindings), so using a setting to configure how keys work just obscured
> the behavior and made configuring Myth more confusing.  Really, how many
> people would guess, "To adjust my keys, rather than go into Edit Keys, I
> need to go into TV Playback settings and ..."?
> So, I took what was a setting for changing key bindings and made it into
> what it should have been--key bindings.
> Granted, at the time the setting was introduced, we didn't have
> MythControls (an easy GUI approach for editing key bindings), and we
> didn't have the absolute seeks, and ...  So, it's one of those things
> that made sense at the time, but is OBE (and, IMO, needed updating for
> the current design of Myth).
> Now, if you mean, "Does the functionality provided by StickyKeys exist?"
> the answer is, "Yes."  Had I recognized the change to the
> relative/absolute seeks when I did the original change, I would have
> handled it at that point.  Unfortunately, I just forgot the overloaded
> functionality for those keys.  So thank you (and thanks to Albert) for
> letting me know about the additional effects of the change.
> My goal is to take a lot of the settings we've acquired over the years
> that are now buried in the settings area of MythTV and move them to
> places where they make more sense.***  The shear number of times on the
> list I quote settings and their help text and have to explain where to
> find them tells me we need to simplify things a bit.  :)  (Not to
> mention all the "if you have this set, then this happens, otherwise, if
> you have this set, then this other things happens, unless you also have
> this set, then ..." type responses I send.)
> Mike
> ***And, once we change the action processing as David Engel mentioned
> earlier in the thread, this particular change will make configuration
> even more intuitive as we'll allow users to associate FFWDSTICKY and
> or ... (or whatever the new "unbundled" actions turn out to be) to the
> same key.

Thanks for the explanation. Of course, it makes sense that things just 
grew like topsy.
Regarding your goal and the problems of explanations, I have long 
thought that the docs need a re-write to cover some of this. Keys.txt 
needs to be better laid out and explained, and include a reverse lookup 
(what does 'E' do?). And the error messages need to be referenced 
somehow I think.

Even Linux Format magazine trashes mythtv for the trials and 
tribulations of installation, when theproblems are not due to myth at 
all. Thankfully we have this list.


         Please let me know if anything I say offends you.
          I may wish to offend you again in the future.

          Tux says: "Be regular. Eat cron flakes."

More information about the mythtv-dev mailing list