Re: [Tlf-devel] Few modifications in Tlf core

[Thomas Beierlein]

Am Sun, 5 Jan 2014 14:09:21 +0100
schrieb Ervin Hegedüs - HA2OS <address@hidden>:

> Hello Thomas, hello Nate,
> 
> a few days ago I sent a patch to Nate (through this list) for
> hamlib, to fix this bug in case of Kenwood rigs. May be he will
> merge...
> 
> But I think the original problem will not solved with this,
> namely I would like to keep untouched my filter state, when I
> switch band/VFO. You _must_ to pass an explicit value to
> rig_set_mode() as 4th argument, which is the filter width value.
> That mean if you call that function, that set the state of
> filter - except if you pass that value what actually set up.
> 
Right, thats what I am trying to say with my last mail. The main
problem is to retrieve the actual bandwidth setting. 

You can call rig_get_mode() but that gives you only the bandwidth for
the actual mode. If that is SSB and not CW you have to switch to CW.
But to do that you must set a new bandwidth. 

The only possible way is to handle it similar to the frequency
handling. That would mean, that tlf has to remember the last bandwidth
setting before mode or band changed and then recall that value. For
first startup we still have to rely on some default value.

73, de Tom DL1JBE
> 
> 73,
> 
> 
> Ervin
> HA2OS
> 
> 
> On Sun, Jan 05, 2014 at 01:54:33PM +0100, Thomas Beierlein wrote:
> > Hi Ervin, hi Nate,
> > 
> > I think you both have your points....
> > 
> > 
> > RIG_PASSBAND_NORMAL is defined as a special coded width value. Every
> > rig should interpret that value as 'use the defined _normal_
> > bandwidth for that mode. As it seems that there are backends which
> > misses these behavior I think the problem could be solved at a
> > higher layer, namely in src/rig.c. 
> > 
> > In the code for the rig_set_mode function the width parameter
> > should be checked against 'RIG_PASSBAND_NORMAL' and then replaced
> > by the value from rig_passband_normal() before calling the back end
> > code. What do you think?
> > 
> > 
> > 
> > Am Fri, 3 Jan 2014 11:06:54 +0100
> > schrieb Ervin Hegedüs - HA2OS <address@hidden>:
> > 
> > > 
> > > Btw: I think all of these meditation will NOT solve my
> > > problem, namely I don't want the Tlf modify the state of filters
> > > if I change band and/or VFO. :)
> > > 
> > Right you are. By design hamlib is not able to simply switch mode
> > without requesting a bandwidth setting. We could not even ask the
> > actual bandwidth with a rig_get_mode() before as the rig may be in
> > an different mode (say SSB) when asked and then we want to switch
> > to CW but not width the SSB bandwidth. 
> > 
> > 
> > Btw Nate, from our problems here I would suggest for the new API v3
> > to have function calls to get/set the mode and the bandwidth
> > separately. 
> > 
> > 73, de Tom DL1JBE
> > 
> > 
> > -- 
> > "Do what is needful!"
> > Ursula LeGuin: Earthsea
> > --
> > 
> 



-- 
"Do what is needful!"
Ursula LeGuin: Earthsea
--