Race condition in `end-file` event?

[alterNERDtive]

mpv Information

mpv 0.39.0 Copyright © 2000-2024 mpv/MPlayer/mplayer2 projects
libplacebo version: v7.349.0
FFmpeg version: 7.0.2
FFmpeg library versions:
   libavcodec      61.3.100
   libavdevice     61.1.100
   libavfilter     10.1.100
   libavformat     61.1.100
   libavutil       59.8.100
   libswresample   5.1.100
   libswscale      8.1.100

Other Information

- Linux version: Fedora 41
- Kernel Version: 6.11.7-300.fc41.x86_64
- GPU Model: Navi 32 [Radeon RX 7700 XT / 7800 XT]
- Mesa/GPU Driver Version: Mesa 24.2.6 (LLVM 19.1.0)
- Window Manager and Version: Sway 1.10
- Source of mpv: Fedora Repo
- Latest known working version: 0.37 (Fedora 40)
- Issue started after the following happened: upgrade Fedora 40 → 41, mpv 0.37 → 0.39

Reproduction Steps

I am using the following script to exit fullscreen after the last item in a playlist has ended (with --idle=yes):

mp.register_event("end-file", function()
    if mp.get_property_native("playlist-pos") == -1 then
        mp.set_property_native("fullscreen", false)
    end
end)

After updating to 0.39, this stopped working. Investigation has turned out that there seems to be a race condition now (or always has been and I had never run into it) that leads to playlist-pos not resetting to -1 at the end of the play list fast enough for this event handler to work.

If I add a couple of

    mp.msg.info(mp.get_property_native("playlist-pos"))

to the start of the handler, I get an output of e.g.

[exit_fullscreen] 0
[exit_fullscreen] 0
[exit_fullscreen] -1

(and then the script works, since it’s -1 when the condition is hit).

Expected Behavior

playlist-pos is -1 in the end-file event of the last entry in the playlist.

Actual Behavior

playlist-pos still holds the value of the last entry in the playlist for a little while before assuming the value -1.

Log File

output.txt

Sample Files

No response

I carefully read all instruction and confirm that I did the following:

• [X] I tested with the latest mpv version to validate that the issue is not already fixed.
• [X] I provided all required information including system and mpv version.
• [X] I produced the log file with the exact same set of files, parameters, and conditions used in "Reproduction Steps", with the addition of --log-file=output.txt.
• [X] I produced the log file while the behaviors described in "Actual Behavior" were actively observed.
• [X] I attached the full, untruncated log file.
• [X] I attached the backtrace in the case of a crash.

[CounterPillow]

Pretty sure this was always racey, I think you'll want playlist-playing-pos:

Index of the "playing" item on playlist. A playlist item is "playing" if it's being loaded, actually playing, or being unloaded. This property is set during the MPV_EVENT_START_FILE (start-file) and the MPV_EVENT_START_END (end-file) events. Outside of that, it returns -1. If the playlist entry was somehow removed during playback, but playback hasn't stopped yet, or is in progress of being stopped, it also returns -1. (This can happen at least during state transitions.)

Though I don't know whether "or is in progress of being stopped" refers to reaching the end of playlist, worth looking into. Otherwise you might just want to observe the idle-active property.

[kasper93]

During end-file event, the playlist id corresponding to the item that has been just unloaded is provided, see docs.

Processing events, doesn't stop the player from progressing, so any get_property_native inside event handler are "racey", scripts are run asynchronous.

Note that valid value for playlist-pos on end-file is 0 in your case. By repeatedly calling the get_property_native you basically forcing script to wait until this value is cleared and mpv switches to idle.

has ended (with --idle=yes)

Observe idle-active instead.

[alterNERDtive]

I think you'll want playlist-playing-pos

I thought I had double checked that, but … apparently not. Seems to work as expected, thanks!

Otherwise you might just want to observe the idle-active property.

That one exhibits the same behaviour as reported, just with false → true.

[kasper93]

That one exhibits the same behaviour as reported, just with false → true.

What do you mean by false → true?

It reports idle after finishing playback correctly for me.

mp.observe_property("idle-active", "native", function(_, val)
    if val then
        mp.set_property_native("fullscreen", false)
    end
end)

[alterNERDtive]

Oh. Yes. My bad.