-
Notifications
You must be signed in to change notification settings - Fork 26
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Multiple independent bandmap windows #593
Conversation
This looks great @kyleboyle |
I have tried several concepts on how to implement this. I started that the new bands will be in the same widget but it was not user-friendly, then I tried to create new widgets inside or outside the dock as a new docked widget. But it seemed quite unclear to me. But I definitely like your approach and I welcome this PR. I didn't look into the source code, why it can't be compiled to QT5, but hopefully it will work. Thank you. |
I will test it as soon as possible. in the meantime. @aa5sh @kyleboyle Do you have any idea why MacOS CI Github Action is failing with this?
It seems that something changes in QT between 6.8.1 and 6.8.2. In Windows this module has also been moved to a separate external module. But I don't know what needs to be installed in Homebrew to make it work. |
@foldynl I'm using the qt version from homebrew which is old 6.7.3
Ill try updating my formula to the newest and see what happens. |
I upgraded to 6.8.2 and it compiles and runs in QT Creator 15. I started from a fresh project by deleting all the user generated files in the project directory. However this feature is broken for me now - clicking on the plus button doesn't launch a new bandmap widget. I will continue to investigate. All i did was a |
Ok the codebase is functional now, and my dev env seems to be producing much more stable builds so thanks for forcing me to update qt :D Could you try using macos-14 instead. There are older posts about this problem and the os version seems to play a part. |
I just used the MacOS15 Image in testing_0.42 GitHub CI and everything is working. |
It means that it is possible to compile the testing_0.42 branch under MacOS...Now I am going to test this PR |
Sorry, I was working at my church trying to rewire some outside speakers.
Glad yall got it sorted out. I will update my github action to use 15 as
well. I have my flow basically just where it will build as well as
codesign/notarize it.
…On Sat, Feb 8, 2025 at 12:46 PM Ladislav ***@***.***> wrote:
everything is working.
It means that it is possible to compile the testing_0.42 branch under
MacOS...Now I am going to test this PR
—
Reply to this email directly, view it on GitHub
<#593 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AUEEMXSDI7OSHKQQLLGC2DT2OZGHZAVCNFSM6AAAAABWU4KMHCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMNBVHA4TSOBSHA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
I also had 6.7.2 install and just updated to 6.8.2 in homebrew. two things I noticed quickly:
But I do like the ease of opening a new bandmap, the context menu to change bands, can click one of the new windows and all seems well. Michael |
I tested the change. I like its simplicity. I only found these issues:
Point 2 is mostly cosmetic, but point 1 needs to be fixed. I’ll also take a look to see how this could be efficiently handled in the Activity Manager. |
I have these types of issues as well but with all floating windows on macos - think perhaps there are dock window / tools window mac os bugs in general, at least while running with qt creator builds. I don't seem to have the same crashes when running official builds from you.
I've had this happen when crashes occur as the geometry / layout gets out of sync with the saved non vfo bandmap state |
Lada this shows my unfamiliarity with all of the features :) I thought that the activity profiles were only for setting the qso fields and associated equipment. You mention that you will investigate how to integrate with activity profiles - I'll leave it with you for now and if you want me to look into it please let me know.
This is also something that I thought about as well. My initial thought was that if the user doesn't want a bandmap to follow the VFO they may close it and rely on multiple non vfo bandmaps set at the bands they are switching between. Though I personally wouldn't want to introduce the complexity in the first version - in the future we could add an option to only sync the vfo bandmap if there is not another bandmap open which is on the new band. |
I thought about that as well with the crashes when running it from qt creator after I posted it.
… On Feb 9, 2025, at 9:08 AM, Kyle Boyle ***@***.***> wrote:
* If I undock a bandmap then create a new one I was moving it and docked it in the same window as original and it acted weird. Like he header wasn't there to undock it. I clicked around and it appeared and I was able to undock it. I tried again and crashed qlog.
I have these types of issues as well but with all floating windows on macos - think perhaps there are dock window / tools window mac os bugs in general, at least while running with qt creator builds. I don't seem to have the same crashes when running official builds from you.
* If I don't have a bandmap open by default it's like I send a phantom bandmap window appear then close as qlog as opening.
I'll spend some more time with it but two quick comments.
I've had this happen when crashes occur as the geometry / layout gets out of sync with the saved non vfo bandmap state
—
Reply to this email directly, view it on GitHub <#593 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AUEEMXWOLBY5GGHZHP6DR632O5VQNAVCNFSM6AAAAABWU4KMHCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMNBWGM2DQOJQGM>.
You are receiving this because you were mentioned.
|
That's not a problem. I'll take a look at it. Unfortunately, it's not as simple as it was in the beginning. The main issue is that the Qt bugs this and this significantly complicate things on the Linux platform.
Yes, I agree with that. It needs to be predictable and simple. And I think it is. It's the only thing that "surprised" me as a user. Thank you for this PR. |
I tried to reorganize it so that it would be easy to integrate it into Activity Manager. Can @kyleboyle please test it and see if it works on MacOS? I have created a new branch BandmapMultiDockWidget2 for it. |
Works as expected, great job. I can switch activities and the arrangements follow. |
Could we consider changing the readme and wiki so it says MacOS is
supported now and not just for developers? I feel like MacOS version in
general is pretty stable now?
…On Thu, Feb 27, 2025 at 8:01 PM Kyle Boyle ***@***.***> wrote:
Works as expected, great job. I can switch activities and the arrangements
follow.
—
Reply to this email directly, view it on GitHub
<#593 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AUEEMXWK5MZUURKXX5GKBC32R67O3AVCNFSM6AAAAABWU4KMHCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMOBZGUZDQNRTGA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
[image: kyleboyle]*kyleboyle* left a comment (foldynl/QLog#593)
<#593 (comment)>
Works as expected, great job. I can switch activities and the arrangements
follow.
—
Reply to this email directly, view it on GitHub
<#593 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AUEEMXWK5MZUURKXX5GKBC32R67O3AVCNFSM6AAAAABWU4KMHCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMOBZGUZDQNRTGA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
Thanks for the test. There's still work to do but your test confirmed that it's going in the right direction. |
I agree. If you say it's OK, I can change it. I also put it there because I wasn't able to give people advice. If I have both of you as @aa5sh @kyleboyle, if you agree, I’d like to add you to the Authors list in Help->About. I think it would be good to add you to a place where regular users can see it. If you agree, just send me the string (name/nickname/callsign) I should add. I'm talking about this dialog. Thanks for your work! |
Thank you I do plan to be around for a while and help where I can. I really like QLog it is something I was wanting for a while being cross platform and just good powerful application. I’ve also enjoyed learning for you guys. For me just like you with my name and callsign - Michael Morgan AA5SH
Thanks
… On Feb 28, 2025, at 1:26 AM, Ladislav ***@***.***> wrote:
foldynl
left a comment
(foldynl/QLog#593)
Could we consider changing the readme and wiki so it says MacOS is
supported now and not just for developers? I feel like MacOS version in
general is pretty stable now?
I agree. If you say it's OK, I can change it. I also put it there because I wasn't able to give people advice.
If I have both of you as @aa5sh @kyleboyle, if you agree, I’d like to add you to the Authors list in Help->About. I think it would be good to add you to a place where regular users can see it. If you agree, just send me the string (name/nickname/callsign) I should add. I'm talking about this dialog.
image.png (view on web)
Thanks for your work!
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you were mentioned.
<https://github.com/aa5sh> <https://github.com/kyleboyle> <https://github.com/user-attachments/assets/dae1149a-be3a-4820-b51f-a55160b5db49> <#593 (comment)> <https://github.com/notifications/unsubscribe-auth/AUEEMXSM7PNWOQQ5B5HLBG32SAFSJAVCNFSM6AAAAABWU4KMHCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMOBZHEZDKOBVGY>
foldynl
left a comment
(foldynl/QLog#593)
<#593 (comment)>
Could we consider changing the readme and wiki so it says MacOS is
supported now and not just for developers? I feel like MacOS version in
general is pretty stable now?
I agree. If you say it's OK, I can change it. I also put it there because I wasn't able to give people advice.
If I have both of you as @aa5sh <https://github.com/aa5sh> @kyleboyle <https://github.com/kyleboyle>, if you agree, I’d like to add you to the Authors list in Help->About. I think it would be good to add you to a place where regular users can see it. If you agree, just send me the string (name/nickname/callsign) I should add. I'm talking about this dialog.
image.png (view on web) <https://github.com/user-attachments/assets/dae1149a-be3a-4820-b51f-a55160b5db49>
Thanks for your work!
—
Reply to this email directly, view it on GitHub <#593 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AUEEMXSM7PNWOQQ5B5HLBG32SAFSJAVCNFSM6AAAAABWU4KMHCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMOBZHEZDKOBVGY>.
You are receiving this because you were mentioned.
|
Ladislav, I see progress in the 43 branch, great stuff! Please notify if/when you'd like me to do more testing. Sorry I missed your previous comment asking for call sign - VE9KZ.. thanks so much for including me! |
The testing branch is dedicated for this pre-release testing. I'm quite worried about this change because some GUI combinations can lead to crashes—for example, when you have detached windows and merge them into one outside the main application window. Then, if you detach one of them again, the application crashes. It is known QT issue, but without fix and without any work-around. |
Are the qt crashes a result of the bandmap windows or do the crashes happen when reconfiguring any of the other widgets as well? Perhaps it is more worrisome with this change because users are more likely to be playing around with the window layouts with all of the bandmap windows? |
It's better to show it. The crash occurs when I try to detach a window. This is running under Qt 6.8.2 on Linux. It's not an issue with the new bandmaps; it's an issue of Qt —I found a bug report about it somewhere. Under Qt 5.x, the behavior is slightly different—the application exits, but the detached window remains. I probably won’t be able to fix this. I'm still testing the bandmaps because I've already found many issues related to profile switching and the allocation/deallocation of bandmaps. |
So is this a Linux issue or have you seen it on WIndows? I just did the
same action 4-5 times without a crash. I can play with it more this
evening.
I know I have seen oddities with crashes with windows, even resizing, when
I have launched QLog from the QT Creator, but didn't have them when I built
it separately and ran it. But just know I did run it from QT Creator with
no crashes.
…On Tue, Mar 25, 2025 at 8:39 AM Ladislav ***@***.***> wrote:
It's better to show it. The crash occurs when I try to detach a window.
This is running under Qt 6.8.2 on Linux. It's not an issue with the new
bandmaps; it's an issue of Qt —I found a bug report about it somewhere.
Under Qt 5.x, the behavior is slightly different—the application exits, but
the detached window remains. I probably won’t be able to fix this.
I'm still testing the bandmaps because I've already found many issues
related to profile switching and the allocation/deallocation of bandmaps.
crash.webm
<https://github.com/user-attachments/assets/467674f6-2881-4fe3-9899-164f61ebd85d>
—
Reply to this email directly, view it on GitHub
<#593 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AUEEMXVBFIX242KBBZVQVRT2WFFATAVCNFSM6AAAAABWU4KMHCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDONJRGI4TQMZZGU>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
[image: foldynl]*foldynl* left a comment (foldynl/QLog#593)
<#593 (comment)>
It's better to show it. The crash occurs when I try to detach a window.
This is running under Qt 6.8.2 on Linux. It's not an issue with the new
bandmaps; it's an issue of Qt —I found a bug report about it somewhere.
Under Qt 5.x, the behavior is slightly different—the application exits, but
the detached window remains. I probably won’t be able to fix this.
I'm still testing the bandmaps because I've already found many issues
related to profile switching and the allocation/deallocation of bandmaps.
crash.webm
<https://github.com/user-attachments/assets/467674f6-2881-4fe3-9899-164f61ebd85d>
—
Reply to this email directly, view it on GitHub
<#593 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AUEEMXVBFIX242KBBZVQVRT2WFFATAVCNFSM6AAAAABWU4KMHCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDONJRGI4TQMZZGU>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
On Windows, it's more unpredictable. Sometimes it behaves the same as on Linux—crashing. Other times, it correctly performs the detach, but the moved window becomes unresponsive. But as I mentioned, it is a QT bug, so there's probably nothing we can do about it in QLog. I'm mainly testing in general to make sure the multiple bands windows behave as expected for users. And what does "expected" means - I don’t have an exact definition for it. I must not be surprised by what it does. |
I think Michael and I have griped about this before. I support his claim that running from qt creator seems to increase the likelihood of crashing (for me in macos). I rarely get crashes when running official builds from his repository for macos, and never on windows (I haven't been purposefully testing much on windows, but I use windows when on the radio). Ladislav: I also have not been able to get deterministic crashing in macos - sometimes the floating window layout changes work, sometimes they crash. |
I'm aiming for a release on April 4th. It's not good to release QLog this week right before the CQ WW WPX SSB – even though I don't think anyone would use it during the contest, but I'll be able to test it with high QSO traffic from DXC. |
Talking about high QSO traffic on DXC did you notice how something was
flooding the networks on Sunday for an hour or so off and on? I was
thinking it was going odd on my end/QLog but when I went to several web
clusters thaeyhad similar results. LIke spots were replicating on
different bands and all for some reason. I remember one of the sites
wouldn't even load. Somebody had some bad code somewhere was sending bogus
spots. At one point I had over 500 alerts in QLog when the most I
typically see is 5-10 unless it is a contest weekend then the most I have
ever seen in my case is like 40-50.
…On Tue, Mar 25, 2025 at 10:30 AM Ladislav ***@***.***> wrote:
I'm aiming for a release on April 4th. It's not good to release QLog this
week right before the CQ WW WPX SSB – even though I don't think anyone
would use it during the contest, but I'll be able to test it with high QSO
traffic from DXC.
—
Reply to this email directly, view it on GitHub
<#593 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AUEEMXVB4JLKRSZYHY3PJPL2WFSANAVCNFSM6AAAAABWU4KMHCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDONJRGY3DINRRHE>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
[image: foldynl]*foldynl* left a comment (foldynl/QLog#593)
<#593 (comment)>
I'm aiming for a release on April 4th. It's not good to release QLog this
week right before the CQ WW WPX SSB – even though I don't think anyone
would use it during the contest, but I'll be able to test it with high QSO
traffic from DXC.
—
Reply to this email directly, view it on GitHub
<#593 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AUEEMXVB4JLKRSZYHY3PJPL2WFSANAVCNFSM6AAAAABWU4KMHCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDONJRGY3DINRRHE>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
The last couple big contests have seen some bad actors flooding the clusters with garbage spots. If you go to dxheat and look at the chat usually people there will be complaining about it if you want to verify if the flooding is happening or not. |
I know this doesn't belong in this thread, but could this be prevented somehow in QLog? We have QSO deduplication. Wasn't that effective? |
Yeah I'm sorry just made me think of it when you mentioned the spots
rates.
I don't think what I saw was just a lot of spots. It was a lot of bad
spots. I should have made a screenshot or something. It would be one DX
call sign but I would get 50 or so spots all different frequencies, modes
all at the same time.
…On Tue, Mar 25, 2025 at 11:47 AM Ladislav ***@***.***> wrote:
I know this doesn't belong in this thread, but could this be prevented
somehow in QLog? We have QSO deduplication. Wasn't that effective?
—
Reply to this email directly, view it on GitHub
<#593 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AUEEMXRNU32KWEICEBOVBZ32WF27HAVCNFSM6AAAAABWU4KMHCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDONJRHA4TQMZRGY>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
[image: foldynl]*foldynl* left a comment (foldynl/QLog#593)
<#593 (comment)>
I know this doesn't belong in this thread, but could this be prevented
somehow in QLog? We have QSO deduplication. Wasn't that effective?
—
Reply to this email directly, view it on GitHub
<#593 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AUEEMXRNU32KWEICEBOVBZ32WF27HAVCNFSM6AAAAABWU4KMHCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDONJRHA4TQMZRGY>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
From what I remember the spots were technically valid but they were fake, using callsigns that dont exist or weren't active |
added to release 0.43 |
closes #339
Attempt at adding multiple bandmap windows, using the existing right click menu to change which band is displayed. The VFO linked bandmap serves as the "master" which manages the spot data model shared with the non-vfo bandmaps. The user may close the vfo bandmap but it continues to operate in the background as it always has. This master vfo bandmap is also used to launch new non-vfo bandmaps and continues to provide the existing
clear all
and spot expiry features. I've tried to reuse as much of the existing bandmap as possible for the non-vfo bandmaps. Basically they are stripped of rig link and data management.This change introduces a new concept around dockwidgets, which is the actual deletion and removal of dockwidgets when they are closed. This goal so far has been unexpectedly buggy, which is why you see some weird management code in mainwindow, even though this feature is meant to be built into qt but it doesn't work in many different scenarios. (WA_DeleteOnClose)
There are some remaining items to clean up which I will continue with if @foldynl gives the go ahead for this feature.

Kapture.2025-02-06.at.21.41.34.webm
(I gave up working on split mode support, too many design choices to do on my own)