I thought I’d take the opportunity to share a Bash script I made to automate ripping music off CUE/BIN files. It splits BINs into separate files, so it’s 1 file per track, and strips pregap data, encodes audio tracks to FLAC or Ogg Vorbis, and it also generates new CUE sheets.
Link to the script:
https://github.com/linux4ever07/scripts/blob/main/cuebin_extract.sh
The idea came to me some years ago when I noticed that GOG packaged some of their games in a dumb way. It was specifically DOS games (bundled with DOSBox) that had CD audio. They would include the original BIN file, but with a modified CUE sheet that would not let you access the high quality CD audio. On top of that, they included Ogg Vorbis tracks, wasting HDD space for no reason by effectively storing the music twice, but only letting you access the lower quality Vorbis tracks. So, I thought, why not just split the BIN, encode the audio tracks to FLAC, and that way you both get better audio quality and also use less HDD space. DOSBox supports CUE sheets that list FLAC, Ogg Vorbis and even Opus tracks.
I took inspiration from ‘bchunk’, which is a program that does something similar. However, bchunk converts data tracks to ISO files, which is not what I wanted. I wanted to keep the original tracks completely untouched, so my script will copy data tracks as normal BIN files. bchunk also can’t encode audio tracks to FLAC or Ogg Vorbis, but it will produce uncompressed WAV files. And bchunk doesn’t produce new CUE sheets for the created files.
I also use the script to extract the OST from random games. I put together a playlist with some of my favorite retro video game music that I extracted using the script:
https://www.youtube.com/watch?v=nHI7ghR6XX4
There’s other uses for the script, such as splitting BINs in general. For example, many BeOS / ZETA disc images contain boot floppies as the first track, and once you split the BIN you can access the boot floppy image. And you can create frankenstein disc images by exchanging tracks between different disc images.
The script is made for Linux, but should work on macOS / FreeBSD as well if you have a recent version of Bash installed, as well as ffmpeg, flac and oggenc.
I just checked the macOS / FreeBSD man page for ‘stat’, and noticed the syntax differs from the version in GNU coreutils (which is what’s used in Linux). That’s probably the only thing that would need to be changed to make the script work on those other systems. It’s on line 526.
Well pretty much no one is using BSD
Fair enough. But macOS has more users than Linux, and is partly based on FreeBSD. The shell and the userland tools are from FreeBSD. I prefer Linux of course and haven’t used a Mac in years, but I still think it’s nice making scripts compatible with all *nix systems.
Fair enough, I just don’t pester the developer if you can avoid it. (Foss devs often receive demands)
For sure. If you get something for free then it is what it is. Some of my scripts probably won’t work outside of Linux but I still make an effort to not use external commands if there’s no need to. I try to use the internal features of Bash as much as possible, mostly cause it’s just faster that way. A consequence of that is that the scripts are at least more likely to work on other systems (that have Bash).