My hunch about multiple like-named albums was correct. My request for an artist’s albums was grabbing all regions instead of just the US. I’m comfortable using only US releases for my project’s data.
I don’t know how that happened because every other function (none of which needed it) included the “US” argument but not the one that needed it.
While digging through the JSON for clues, I noticed at the bottom that there were a total of 60 albums my request could have grabbed if I hadn’t set the limit to 50. Sixty! And that’s with the album type limited to album (instead of including compilations, singles, etc.).
I do want to include US compilations, however, and am working on an algorithm that takes the same track from different albums into account.
If you care, Alice Cooper has a total of:
- 67 albums, including compilations, in all regions
- 37 albums, not including compilations, in the US
About that second number … it includes two versions of Billion Dollar Babies–one “regular” and one “deluxe”–which I expected. What I didn’t expect was two albums named Nobody Likes Me. A copy and paste of the album ID for one of them into my browser’s address bar for the Spotify Web Player gave me this:
That “1 more release” button jumped out at me even before I could paste in the second album ID. As it turns out, clicking it did that for me.
Different year and what I assume is the label in the copyright.
This poor album has been released with so many different titles and covers over the years. I bought it on vinyl (because I had to, not because I was a hipster) as Live at the Toronto Rock-n-Roll Revival (no “volume 4” when I bought it) in 1982.
The songs that are actually by Alice Cooper have had incorrect titles on every version and, apparently, two of them aren’t even them. For a band and manager who have always made such a big deal about their publishing, I’m surprised this has been out there so long.
As long as they’re both available in the US, I’ll consider those tracks as needing to be combined in my world-changing algorithm.
So, between refining those two request parameters and using arrays to reduce the number of requests, this project continues to speed along.