pick_random_album.py + build_weighted_album_csv.py + add_weights_to_csv.py

I got an idea for randomizing my music into a playlist but weighted by how recently I've gotten the album so it will favor stuff I haven't had the chance to listen to as much. The problem is I don't have a clean way of determining how recently I've gotten something. I have the spreadsheet that I've kept forever but there are formatting inconsistencies with how the folders are named in the filesystem (e.g. in my spreadsheet I have everything formatted for alphabetization, with last names first and articles at the end, etc., also folders in the filesystem are by disc so multidisc albums won't match up the same). I have a system of weighting by groups, as the spreadsheet does progress chronologically. At first I thought I'd have to enter everything manually, but then I realized I could at least get a good chunk of the albums that don't have any special formatting. Then I also took another step to at least transform some common mappings (like finding names/titles that end with ", The", ", An", or ", A" and moving that to the beginning...won't work in every case, like when there are stacked titles, and it's definitely not really possible to sort out the last name, first name stuff...but this reduces a lot of the manual work...and maybe I can do some data cleanup, I mean, that's manual too, but a one-time thing...I dunno, kinda blathering now so I'll shut up). Still have a bit of work to do on it altogether, but it's kind of a fun and nerdy project. At night Mary Beth and I went to Opera Memphis again. This was another of those more informal and not-so-much opera "mixes." It was themed on "Gershwin and friends" (mostly Gershwin and a good bit of Kurt Weill and some others), it leaned musical theater. Okay, now to continue this weekend.