开云体育

开云体育

I've finally figured out how to do this: The variable can be accessed as "self.application.lastBand" from inside the function "UpdateText(self)"

The problem was solved as follows.
#wFrame, h = frame.GetClientSize().Get()?# client window width
wFrame = conf.window_width - 15

Hi group,

I am trying to extend the following widget:

What I am trying to achieve is to automatically change the selected band on the Hardrock-50 via an API call. At the moment I have added a button for each band (not in the code linked above) and when I click on it a request is sent to the server component which then sends the serial command to the HR50. Now I want to automate this.
To do this, I need to continuously read the selected band (or frequency if band is not possible), but I can only do this at startup by adding

to the function Widgets_0x06(self, app, hardware, conf, frame, gbs, vertBox)

The function "UpdateText(self)" is called all the time so I tried to read out the value from there as a test but I failed since I cannot reach it from inside this function.

Can you please help me here?

Thanks and 73
Michael, DK1MI

The value of?wFrame in the second line is incorrect at first, and after the second reboot it becomes correct.

On Fri, Mar 1, 2024 at 01:28 AM, Rob Frohne (KL7NA) wrote:

Does anyone have any ideas on how to fix this problem?

Which compiler? With MSVC I assume since it does not have 'sys/time.h'.

Your student need to use MinGW. But I've no experience with "pip install" and MinGW.

Untested: "python setup.py build -c mingw32" could do it.

开云体育

The reason rigctld emulation is a bad idea is that it changes over time.? And the client may want vfo mode or not so most every command has to be checked for VFO.
A rig definition never changes -- but it can be extended.
Historically rigctld emulation was only necessary if you wanted a socket connection but that's not true anymore.
You can connect to any rig via a TCP socket and in 4.7 you'll be able to have a UDP/Multicast connection to control the rig.

So here's a rigctld emulation in code.

fgets(buf,sizeof(buf),fp);
char vfo[16];
long freqA,freqB;
int n = sscanf(buf,"f %s %ld", vfo, &freq);
if (n == 2) { // then there's no VFO
????int n = sscanf(buf,"f? %ld", &freq);
? ? vfo = current_vfo;? // you have to track which vfo is active -- PITA
}
if (vfo == 'VFOA') freqA = freq;
else if (vfo == 'VFOB') freqB = freq;
else printf("unknown vfo")
printf ("read freq %ld from vfo %s\n", freq, vfo);

And here's a TS-2000 emulator which is very straightforward and easy to understand.

fgets(buf,sizeof(buf),fp);
char vfo[16];
long freq;
int n = sscanf(buf,"FA%d",&freq);
printf ("read freq %ld from vfo %s\n", freq, vfo);
if (n == 1)
{
?freqA = freq;
}
n = sscanf(buf,"FB%d",&freq);
if (n == 1)
?freqB = freq;
}

Mike W9MDB

On Fri, Aug 18, 2023 at 01:50 PM, jimahlstrom wrote:

I can alter Quisk's Hamlib code if it makes it more convenient for the Hamlib community, but I need a description of what to change.

In?/g/n2adr-sdr/message/3043?we get a hint.

You probably should read the whole thread when you get a chance.

Personally I think emulating rigctld was a good idea, and I don't know why some think it was a bad idea.

I've read the code of an emulated TS-2000 and it reads like a hex dump.

Yet it seems the current "best practice" is to emulate a real radio closest to the SDR's capabilities, and use a radio-specific back end to tweak any back-end differences.

It seems to me this would lead to lots of copy/paste code since hamlib isn't object-oriented so their is no core object you can extend as far as I can tell.

It seems there are function pointers to call into other radio objects, but still it seems to take a few thousand lines of code to do a back end regardless.

As usual, these are just my opinions, and I'm willing to be corrected.

Regards,
RDP

What is the Rflna slider, what is its purpose?
What is the screen got when hitting the Bscope button?
How can I show the "slider 0 to 100% to change the transmit power", which is indicated in the HL2 - Bands - Tx level comment?
73 - Pierre - FK8IH

I'm using Quisk v.4.2.29 with my Hermes Lite 2. After one the last upgrades a "FDV button" appeared but it is in White color and is inoperative. I have already FreeDV v.1.9.8 installed on my computer. What did I miss to use it with Quisk?
73 - Pierre - FK8IH

?I think that here

? if conf.window_width > 0:

This is..

If you switch from -1 to any value (or from value to -1), then it's fine. But if you switch between different values, then there is a problem. For example, between 1366 and 1920 (forward and backward). The previous value is applied. If the operation is repeated, the new value is applied. You have to apply it twice (button "On").

Hi

Could sqelsh option be extended to cw?

It works with ssb/am and fm ..

It would be useful for monitoring a frequency preparatory to the start of a cw net or waiting for activity on a common calling frequency.
It can be fudged by using lsb (or usb) and offsetting the frequency .. but then has to be converted back to cw when activity occurs

Also .. in 4.2.26 there seems to be a default offset of 600Hz when initiating .. presumably to generate bfo freq for audio .. but there does not seem to be much
control over initial start up (like not? to have it when starting)

Thanks de zl1any stephen
\

I think I'm seeing the bigger picture now after reading Mike's comments and also reading the pihpsdr manual ( ref:??).

In that pihpsdr manual, search for?19090 to find where it discusses rig control (it's the port number pipsdr uses by default for rig control).

The manual says:

piHPSDR listens on port 19090 (by default, can be changed in the RigCtl menu). For standard digimode operation, choose the Kenwood TS-2000 radio model. If your digimode program can use the hamlib CAT connection library (for example, both Fldigi and WSJTX can), choose the ,,OpenHPSDR PiHPSDR” radio model.

And:

The CAT command set is explained in Appendix C. In most cases, using the Kenwood TS-2000 as the radio model would do it, if the digimode or laptop program uses hamlib to interface with radios, either choose TS-2000 or (preferably) the ,,OpenHPSDR PiHPSDR” radio model because this uses time- out values adapted to piHPSDR.

And:

Important here is to choose the correct rig model OpenHPSDR PiHPSDR and port :19090 (note the colon!).

So, if I understand this, pihpsdr is implementing TS-2000 commands passed over the network, and it also has its own back-end selected as "OpenHPSDR PiHPSDR" that still mostly implements TS-2000 commands but also extends some timeouts.

This agrees with my reading of pihpsdr's rigctl.c code that is implementing a bit-level protocol to emulate TS-2000, as opposed to quisk and others that implement a character-based protocol to emulate rigcrld.

And, if I can interpret what Mike is saying, quisk users (who have a new-enough version of Hamlib) presumably should select radio model 10 named "Quisk" to send rigctld CAT commands to Quisk, but only ones that Mike (and/or other Hamlib maintainers) think Quisk can/should handle correctly.

I'm typing all this in because reading the posts I linked above suggests there may be some disconnects on this topic.

I sure found it difficult to comprehend.

Thanks,
RDP

On Sun, Feb 25, 2024 at 05:30 PM, Mike Black wrote:

Because quick and sdrsharp are both trying to emulate rigctld which is a bad idea.

They'd be much better off emulating a rig like the TS-2000 or TS-480 or whatever matches their capability.

?

So had to disable some things that they cannot handle.

If I understand??correctly there was a time where the only way you could emulate a TS-2000 was if you had a (real or virtual) serial port, and many (most?)? SDRs do not have real serial ports, and most linuxes don't have virtual serial ports.? ?Maybe this is still the same now, maybe it has changed?

I also think the client programs forced you to provide serial port names (either COMx or /dev/ttyXX formats) and not TCP names (hostname:port?format) when using TS-2000 but there may be no real notion of connecting via TCP to a serial port so it's easy to understand why a client programmer would not code that.

This suggests that at the time there really was nothing workable except emulating rigctld.??

But this is just me guessing, so if you know more, please share.

PS: gqrx also emulates rigctld ( ref:??)...

Thanks,
RDP