Specification?

[KlfJoat]

Hi

I love the idea and promise of this project, especially since I was creating something like this for my own discs, a year or two ago, to help me out if I need to re-rip to better video codecs. Life got in the way, but I'm slowly ramping back up to finish my project.

Obviously, I'd like to help out and contribute my disc info from my automatic ripper & transcoder. That means not using your ImportBuddy software, but I'm fine with implementing your data file format. There's a lot of overlap.

To that end, is there some kind of specification for the file formats? The fields in those formats?

For example, I'm not a TypeScript developer, but my best understanding as I read the code is that you're only MD5 hashing file sizes sorted by filename, but not the filenames themselves.

Similarly, is there any kind of style guide?

I've run into some strange items as I browse around, like Looney Tunes shorts being mis-labeled as television episodes (they're really theatrical releases, thus Movies).

Thanks!

[lfoust]

The only file format that has any sort of spec is the summary files which identify the items on each disc:
https://github.com/TheDiscDb/data/wiki/Summary-File-Format

As far as a style guide - do you mean coding style or something related to the data? You can look at all the existing data to get a feel for the style.

As far as strange items (like the Looney Tunes) I just try to provide the best data I can to help people rip their own discs. Sometimes things don't fit well into one category so in the case of Looney Tunes I just tried to group it all together as much as possible. In the end, it is a database of what items to rip off of specific discs - and not a perfect catalog (like imdb or tmdb).

I welcome your help and contributions - let me know how I can help. If there are lessons you learned from trying a similar project - please do share

[KlfJoat]

I'm just in the idea phase right now, so I'm not going to be doing any of this tomorrow... Just brainstorming...

Since there isn't really a file format, what do you think about me creating and submitting a JSON-schema validator? And what about a data validator as well? I have such things for my data files, and I'm likely to create them anyway if I move to your file formats.

For a style guide, I meant data style... To use an example from your own Summary file format, like where you say to remove "(English)" from the Name field.

I'm still not clear on how the disc hash is being calculated, so if you could explain that I'd appreciate it.

One of the biggest lessons I learned is to allow for a separate IMDB/TMDb ID for each individual track/playlist. There are a number of reasons for this.

1. I have at least 1 DVD that has 2 separate IMDB/TMDb titles on it. https://www.amazon.com/dp/B000BMDD6U This has two documentaries, one on William Powell, one on Myrna Loy. This DVD pairs them together in part because of their amazing chemistry and longevity in the Thin Man series. In the current file format (as best I can tell), I'd have to choose one and then have to manually adapt for the second.

2. It provides greater fidelity for TV episode metadata. Providing an episode-specific IMDB/TMDb link for each on-disc title/playlist means that I can automatically grab episode images, descriptions, etc.

3. Sort of related to my number 1, occasionally a movie will be released with a track of some note that has its own IMDB/TMDb entry. I know I've got a couple of documentaries like that, like for The Matrix. Or a theatrical short of some note, like the 2000 Toy Story 2 DVD, which includes the first commercial release of Luxo, Jr., which was the proof of concept short of the Pixar style.

4. It solves the Looney Tunes problem (and other animation collections). Each LT/MM short was released as its own separate theatrical short subject, and each has an IMDB/TMDb page and MPAA certificate. Link each track to an IMDB/TMDb ID, and renaming that track becomes automatic.

[CuriousRipper]

I ran into a similar issue with RWBY. There are spinoff series with their own TVDB#s that are included as extras on sets for each volume.