Broken GUI layout

[jflesch]

Hello,

I just tried installing Openhantek on GNU/Linux Debian. It seems there is a major problem with the layouts of the elements of the GUI. It makes a lot of features unusable.

GUI elements can be rearranged by the user, but when restarting OpenHantek, everything is a mess again.

Distribution : GNU/Linux Debian Jessie (8.2), 64bits
WM: Gnome 3 (dual screen)

Thanks in advance,

[geissdoerfer]

Same configuration and same issue for me. Apparently there's some buggy threading. If i try to change some settings (AC/DC):
QObject: Cannot create children for a parent that is in a different thread. (Parent is Hantek::Control(0x1c37960), parent's thread is QThread(0x1a2fdb0), current thread is Hantek::Control(0x1c37960)

[davidgraeff]

The current GUI is written in pure c++ and therefore not very maintenance friendly. I'm working on a successor written in a declarative style and with the current technology (Qt QML) in the other branch but still on holiday at the moment :)

[OliverHaag]

The threading issue and the GUI issue seem to be unrelated on first sight to me. I agree with David though - best wait for the QML rewrite and recheck both issues then.
Back in the Qt 4 times pure C++ was the best way to do it imo but nowadays I would also use QML since it's just much easier maintainable. Most probably some things have changed with Qt 5 (Guess Qt 5 was used, but this wasn't specified in the reports).

[geissdoerfer]

Ah okay i thought it was related, as for me the functionality is also affected and it kind of looked like the GUI elements could not communicate with control parts. Haven't had a closer look though tbh. Also i'm not sure about which Qt i was actually using, will check and try both. Anyway @davidgraeff, had a look into your branch and it looks very promising 👍 The killer feature for me would be a clean programmatic interface, that i can wrap for matlab/python

[cskau]

I ran into the same issue a couple days ago, and I went in an created a fix for at least part of the issue in this PR:
#11

I do not think this is related to threading.

[davidgraeff]

In my setup this problem is fixed, please open another issue if it is not for you.