[Lcdproc] MOU-AL202C-XR

[Ethan Dicks]

On Thu, May 15, 2008 at 9:19 PM, Ray Molenkamp
<rmolenkamp@matrixorbital.ca> wrote:
> Ethan,
>
>  Please review attached patch.
>
>  The K was not caused by the cursor command
>  but by the GPO output commands. On units with
>  only 1 GPO you send 254 'V' or 'W' to turn it
>  on or off. On units with more GPOs  it takes a
>  parameter specifying which GPO to turn on/off.
>
>  What was happening is on a unit that had multiple
>  GPO's the codepath for a single GPO was chosen
>
>  254 'V/W' was send, with the unit waiting for the parameter
>
>  Next command in the pipeline was a cursor command
>  254 'K' , the GPO command ate the 254 and the K
>  was displayed on the display.

Your explanation makes perfect sense.  I see exactly how that could be
happening.

>  Since there is no way to tell how many
>  GPO's a unit has and even if we kept list
>  there's always custom units that deviate from
>  the standard I moved the setting to the config file.

That probably makes the most sense.  I didn't write the current
driver, but I have watched it evolve from supporting one Matrix
Orbital display to supporting many.  I think the choices were made on
the basis of what models existed and what sorts of generalities there
were amongst all the models.  There are now so many models that it's
obviously hard to make some of these generalities.  I didn't know
there were any models that had a number of GPOs other than 1 or 6
until this topic came up.

>  Seemed like the most logic thing to do.
>
>  --Ray

I'll have a look at the patch and see how it works with my VKD204, but
from casual inspection, it looks fine.  The only issue I can see is if
the mixed case of "GPOs" conflicts stylistically, or if the name
should be more explicit like "NUM_GPOS" or something similar.  It
might also make sense to still default to old behavior (1 for LCD/VFD,
6 for LKD/VKD) if someone forgets to set the number, since the vast
majority of models _do_ fit that scheme.

-ethan