"loop_end_position" does not update on eject

[gqzomer]

Bug Description

Discovered while writing a mapping for the reloop mixage series

I'm using the loop_start_position and loop_end_position controls to provide users feedback about the status of the loop for a given track. these leds are also updated when a new track is loaded or when a loaded track is ejected. The loop_end_position seems to be returning a incorrect value when a track is ejected.

expected behaviour
The loop_end_position return -1 just like the loop_start_position when a track has been ejected.

Observed behaviour

• the loop_end_position returns the position of the loop end point of the track that has just been ejected instead of -1)
• if a track new track is loaded (with no loop present or previously set) the loop_end_position returns the value of when it was last updated. This is usually the loop end point of a previously loaded track

steps to reproduce

• bind a simple logging function to the eject control, I used the script below

var test = {};

test.controlsToFunctions = {
    "eject": function(v) {
        if (v === 0) { 
            print("start " + engine.getValue("[Channel1]", "loop_start_position"));
            print("end " + engine.getValue("[Channel1]", "loop_end_position")); 
        }
    }
}

test.init = function(_id, _debugging) {
    script.bindConnections("[Channel1]", test.controlsToFunctions);
};

test.shutdown = function() {
    script.bindConnections("[Channel1]", test.controlsToFunctions, remove);
};

• load a either a new track or a track for which no loop points are present in the data base
  • eject the track, the logging function returns -1 for both loop_start_position and loop_end_position
• load a track with a loop set or set a new loop
  • eject the track, the function returns -1 for loop_start_position and the loop end point for loop_end_position
• load another track without any loop points present
  • eject the track, the function returns -1 for loop_start_position and the loop end point of the previous track for loop_end_position

I have tested it with both 2.3.6 and 2.4 and got the same result

Version

2.4-beta-199-g8c925617e3

OS

Fedora 38

[ronso0]

Thanks for reporting this.
FYI you can observe control values much easier with Options menu -> Developer Tools (need to start Mixxx with --developer argument).

[ronso0]

Fixed by #12224