FLAC cuesheet support in a FUSE filesystem. Designed to integrate with Plex.
When running, this filesystem will automatically expand .cue files into separate tracks. When accessing the tracks, it will use ffmpeg to extract the appropriate segment of the files for playback. This filesystem also allows accessing portions of any music file using a specific filename structure: "/flaccue/{file name}.flaccuesplit.{start_time}.{end_time}.{ext}". The "/flaccue" portion is the path where the FUSE filesystem is mounted. {filename} is the full path to the base file on the original fileystem, including the file extension. {ext} should match whatever extension option is chosen for the FUSE filesystem (default wav). The default settings also cache the parsed cue files, using up a noticeable amount of memory for increased speed (although, with my library, still less memory than the extracted tracks use while actively listening to music). Actively extracted tracks will persist in memory from when the track is opened until one minute past the last file handle accessing the track is closed--while generally this will only involve a couple of tracks in memory at any given time, this can balloon somewhat during scanning and similar.
This structure will directly work with the native Plex scanners, allowing use of the full Plex Music Scanner unlike previous versions of this package. The filesystem overhead does matter (e.g., encoding the tracks into FLAC files causes failures in Plex on my system when track lengths exceeded ~15 minutes, causing hangups in playback) but the default settings work reliably on my Synology NAS (with output files in wav format vs flac being the biggest and most necessary speed increase I found). To use in this method, you would simply tell the Plex Scanner to scan the /flaccue/{path_to_music} folder.
The Plex Media Server directory includes an updated Scanner for parsing cue sheets for the FLAC files, bypassing the direct FUSE cue parsing. It will create Plex albums for the cue sheets with the album available as disc 9999, track [Disc #] (comment out line 90 in flaccuelib.py to disable the full album) and also the split up tracks. The tracks give Plex filenames with "/flaccue/{file name}.flaccuesplit.{start_time}.{end_time}.{ext}". This causes each track to have a unique filename (seemingly necessary) and also provides the track timing information. This bypasses the need for the FUSE filesystem to parse cue files to access specific tracks, somewhat increasing access speed at the cost of needing a custom Scanner. That said, in getting the FUSE filesystem to work directly I found many places to improve reliability. Sadly, you will need to rescan any libraries using this scanner as I needed to slightly change the filename format for accessing tracks through the flaccue filesystem.
My FLAC files also include "Disc 1", "Disc 2", and similar as the final part of the title for multi-disk sets. Something like: "Artist - Album Title Disc 3.flac" The scanner is designed to pull disc information from this and to group albums together, while the direct FUSE filesystem will number tracks including the disc number (e.g., 203 for track 3 on disc 2).
The file "flaccue.py" is the FUSE filesystem. This is a python script that requires the mirrored directory ('/') and the mount location ('/flaccue') as parameters. You can access any portion of a FLAC file through this filesystem--I've also used it to extract specific song subportions out of the FLAC file for burning onto CDs. You just provide the appropriate start and end times in the filename. Note that FLAC times are specified in cue sheets as MM:SS:FF where MM is minutes, SS is seconds, and FF is frame information (1/75th of a second) and this is the format FLACCue understands.
You do not want to run this script as root as this will give anyone read access to any file. Use something like this instead: sudo --user=flaccue nohup flaccue.py / /flaccue/ &
As my Plex server runs on a Synology webserver, I've also created a Synology package to run the FLACCue script automatically. The source for creating this package can be found in the synology_package directory. You can compile this with pkgscripts from Synology: https://github.com/SynologyOpenSource/pkgscripts-ng
After installing the toolkit and checking out branch DSM7.2 (or similar), you put the FLACCue folder in the "toolkit/source" directory and run this command from the "toolkit/" directory (you may need to change the processor family or DSM version appropriately): sudo pkgscripts-ng/EnvDeploy -v 7.2 -p apollolake sudo pkgscripts-ng/PkgCreate.py -v 7.2 -p apollolake -c FLACCue
Note that this does not appear to work directly on Synology boxes and so you may need to have a separate Linux system (VM is fine) to build this.
Note that you may also need to set permissions on INFO.sh to be executable. Also note that the python script is called "FLACCue" in this package instead of flaccue.py. The files are the same other than the name.
I've included an unsigned spk file you can install directly at your own risk in the synology_package directory. FFMPEG will need to be installed (and the spk should warn you of this if it's not already there).
Synology no longer allows root access for a package that isn't signed by Synology. As such, for installing the spk, you'll also need to log in as root and run:
sudo -i
curl -k https://bootstrap.pypa.io/get-pip.py | python3
pip install ffmpeg-python mutagen numpy
mkdir /flaccue
chown flaccue:flaccue /flaccueYou'll need to do this after the package is installed and before it will successfully load. You'll also need to provide permissions for the flaccue user to access any media files you want it to be able to. Read-only access is likely best.
Note that on more recent versions of the Synology OS, you may need to perform the above steps after upgrading the OS. You will then either need to manually start the FLACCue package or restart the server.
You may also need to modify the pip command to instead be:
python -m pip install --upgrade --no-cache-dir --use-deprecated=legacy-resolver ffmpeg-python mutagen numpy