Attack The Music custom domain for Bandcamp no longer works (+style proposal)

MusicBrainz
, ,
1

The label Attack The Music used to have a custom domain set up for their Bandcamp page at https://shop.attackthemusic.com, and many releases such as EmoCosine’s Shake The Future! EP have been created with URL relationships pointing to this domain as if it were Bandcamp (since, by all means, it was).

Unfortunately, the label seems to have messed with their DNS/webserver configuration, and https://shop.attackthemusic.com/<path> now redirects to https://attackthemusic.com/shop/<path>, which is not Bandcamp.

If anyone has a bot set up, or you personally have a lot of time on your hands, please consider editing all former Bandcamp custom domain Attack The Music URL relationships to their official Bandcamp equivalents, which can be found at https://attackthemusic.bandcamp.com/<path> — yes, all you have to do is change the domain.

You can find all relationships with the API:

I have personally been adding both the custom domain and the official Bandcamp domain URLs for releases of which that applies in fear of something like this happening. (as in, I’ve done it like, twice.)

Thus, I have a style proposal. I believe it would be a good idea to require the official Bandcamp domain to be used in relationship URLs, optionally in place of any potential custom ones. According to the Bandcamp help article linked above, the official domain (<artist/label>.bandcamp.com) should automatically redirect to the custom one, but this is “currently” broken (has been for the last two years apparently), and custom ones redirect to the official one (if the DNS records aren’t misconfigured).

Style proposal ticket: https://tickets.metabrainz.org/browse/STYLE-2798

6 Likes
2

I wonder if there is something that could be added to the server that makes a snapshot of a site to the Wayback Machine when a link is added and if the link breaks, can revert to the backup.

Obviously this doesn’t solve the current issue, but just something to consider to avoid this in the future.

Edit: I’ll try to make some time to work on editing the links later today.

We may also want to check in with the label and make sure they’re not switching to their own storefront or something like that. We don’t want to have to do this twice.

1 Like
3

Taking a snapshot was a GSoC project, see the blog post here, which resulted in the auto-archiver ‘melba’: GitHub - metabrainz/melba: Musicbrainz’s External Links wayBack machine Archiver · GitHub

I assume it’s running/working?

However I don’t think anything changed in the MB UI/display regarding this. There are some other tickets for that stuff.

2 Likes
4

Makes sense, so I’m not surprised someone has already had the idea and implemented the code. I’m curious how it was implemented here though like you said. Seems it may not have been. I’ll have to go check it out.

One thing that has become apparent in the last few days is that everyone who is working on the backend for MB are very busy. Which I suppose is inevitable in an open-source project of this size. I’m genuinely impressed by how few bugs there are here, it’s inevitable some things will be missed.

1 Like