Do you want me to test under Arch Linux ?

I compile Jamulus on CentOS 6 with Qt version 5.6.1.
That means that Qt6 should remain compatiible with
older Qt. I will let you know when Jamulus git master
has problems (unless I get a heads-up before).

OK this is not going to work as the IPv6 code uses a class that no longer exists QRegExp in Qt6.
It needs to be rewritten with QRegularExpression :(

OK this is not going to work as the IPv6 code uses a class that no longer exists QRegExp.
It needs to be rewritten with QRegularExpression :(

Ok, well we can fix that after 3.8.1 - we aren't trying to squeeze Qt6 work into this upcoming release.

OK this is not going to work as the IPv6 code uses a class that no longer exists QRegExp in Qt6.
It needs to be rewritten with QRegularExpression :(

Thanks to your instructions @dcorson-ticino-com I've rewritten the QRegExp code as follows:

    QRegularExpression rx1 ( "^\\[([^]]*)\\](?::(\\d+))?$" ); // [addr4or6] or [addr4or6]:port
    QRegularExpression rx2 ( "^([^:]*)(?::(\\d+))?$" );       // addr4 or addr4:port or host or host:port

    QString strPort;
    
    QRegularExpressionMatch rx1match = rx1.match ( strAddress );
    QRegularExpressionMatch rx2match = rx2.match ( strAddress );

    // parse input address with rx1 and rx2 in turn, capturing address/host and port
    if ( rx1match.capturedStart () == 0 )
    {
        // literal address within []
        strAddress   = rx1match.captured ( 1 );
        strPort      = rx1match.captured ( 2 );
        bLiteralAddr = true; // don't allow hostname within []
    }
    else if ( rx2match.capturedStart () == 0 )
    {
        // hostname or IPv4 address
        strAddress = rx2match.captured ( 1 );
        strPort    = rx2match.captured ( 2 );
    }

Would you be able to verify if this works for you using the new Qt6.2.0? I'm actually curious to see how Qt6 affects mobile devices.

Would you be able to verify if this works for you using the new Qt6.2.0? I'm actually curious to see how Qt6 affects mobile devices.

On vacation without my dev environment this week. I will look at this around Wed next week.
D

I hope that Qt5 will remain an option for a good while by use of a compile option (or Qt6 option).

@ngocdh The code you provided compiles OK under 6.1.

Unfortunately there are other problems now in socket.cpp
socket.cpp:287: error: 'inet_pton' was not declared in this scope
I have the impression that this is a Win10 problem, but have not found a solution.

Where did you get Qt 6.2 from? The Qt utility does not offer me more than 6.1

Unfortunately there are other problems now in socket.cpp socket.cpp:287: error: 'inet_pton' was not declared in this scope I have the impression that this is a Win10 problem, but have not found a solution.

This line compiles fine in the Windows build environment on Github, so I guess there might be something missing in your local build environment?

Where did you get Qt 6.2 from? The Qt utility does not offer me more than 6.1

Maybe update the utility first? I'm not sure though, just saying

I merged my iOS code and yours, compiled ok, but the app couldn't launch. I got the following output:

2021-10-03 23:08:43.928993+0200 Jamulus[15122:6998706] - allocated port number: 22134
2021-10-03 23:08:43.941280+0200 Jamulus[15122:6998706] Error: You are creating QApplication before calling UIApplicationMain.
If you are writing a native iOS application, and only want to use Qt for
parts of the application, a good place to create QApplication is from within
'applicationDidFinishLaunching' inside your UIApplication delegate.
dyld4 config: DYLD_LIBRARY_PATH=/usr/lib/system/introspection DYLD_IMAGE_SUFFIX=_debug DYLD_INSERT_LIBRARIES=/Developer/usr/lib/libBacktraceRecording.dylib:/Developer/Library/PrivateFrameworks/DTDDISupport.framework/libViewDebuggerSupport.dylib
(lldb) 

Someone knows how to fix this please? @emlynmac maybe?
Here's my branch: https://github.com/ngocdh/jamulus/tree/Qt6

I can have a look; seems like startup is not happening in the right order.
Ok, what are the steps here to build for iOS and how do I get QT6 installed in the right place?
I have QT5 installed with home-brew.

Compiling instructions can be found here: https://github.com/jamulussoftware/jamulus/blob/master/COMPILING.md

To install Qt6.2.0 LT, I simply used the maintenance tool.

Any updates here? Having Qt6 support would be great.

I could help out with git wrangling if you need it. But it looks like github doesn't think there are any conflicts to be resolved?

@npostavs you can ofc. take over the PR and get the it to work by forking it.

But it looks like github doesn't think there are any conflicts to be resolved?

That's because the code is still on my PC. Git claims merging problems that I am unable to resolve with any tool I have tried.

That's because the code is still on my PC. Git claims merging problems that I am unable to resolve with any tool I have tried.

Could you perhaps just push your changes to a new branch without resolving anything so we can see them?

Could you perhaps just push your changes to a new branch without resolving anything so we can see them?

Will do next week, leaving now for a long weekend :)

@dcorson-ticino-com can you maybe just upload the code as zip file if everything fails with git?

To where can I upload it ?

I did some work on this branch a few weeks ago to get it up to date with master. I resolved the conflicts and successfully built with Qt5, but didn't yet get it to build with Qt6 for Windows on the github CI. See https://github.com/softins/jamulus/actions/runs/1491465763 and my branch Qt6-merged

Thanks Tony.
I hope that with qt6 the MAC problems can be resolved.
I must say that the never-ending problems with git have severely braked my interest in working on the Jamulus SW, unfortunately.

It heads into the bushes at about line 464 of the list not finding files.
I remember I had to change some of the Windows Autobuild files .ps1 to get it to compile, locally at least, but unfortunately couldn't tell you now what that was.

@softins I think the problem is that the initialise build environment function needs the QT variables. Probably we need to modify this function?

@dcorson-ticino-com going to close this in favor of #1795 (comment)