Used to work fine, but recently (not sure when) existing files are not being detected properly. Worse, when a recording is aborted and the partial file deleted the previously fine file is now gone.
I have always used pianobarfly remotely, writing to a gvfs-smb mounted share. Tonight I ran pianobarfly locally and it did correctly identify existing files and behaved as I expected it would. Since gvfs has seen a number of updates lately and the mount point has changed. I have to wonder if too much of the path is being used to determine if the file exists or not. Seems to me that anything before "mp3" should be ignored. I'll take a look at the code but I'm hoping someone who is already familiar with it will recognize the issue (and fix it :-)
Oops, clicked the wrong button.