site banner

Tinker Tuesday for October 27th, 2026

This thread is for anyone working on personal projects to share their progress, and hold themselves somewhat accountable to a group of peers.

Post your project, your progress from last week, and what you hope to accomplish this week.

If you want to be pinged with a reminder asking about your project, let me know, and I'll harass you each week until you cancel the service.

1
Jump in the discussion.

No email address required.

Not a particularly good week for Parsefinder: I'm really bogged down in class and subclass handling, which has a lot of weird corner cases I need to get into. I realized there's a weird spectrum of what defines a subclass versus just a class feature with selections:

  • A class like Barbarian gets one choice, which gives multiple features over the course of character progression. Fine. That's easy.
  • A class like Investigator gets one choice but gets all its benefits up front. Still pretty easy.
  • Thaumaturge gets to pick three implements over the course of its levels, and also gets to pick ones to upgrade (you'll get the basic effect of all 3, two will get a second effect, and one will get a third effect). Weird in that the selected subclasses may mean different things in terms of abilities from the order/level they're picked, but not that bad.
  • Commander picks a half dozen tactics, which it can change at will, but they're class features. They're not subclasses but they aren't really distinct from subclasses in terms of how they're worded, just how they work mechanically
  • Champion has up to four subclasses, depending on how you look at it: the Cause that determines their reaction, a choice of Blessing that's one of three small buffs, a choice of focus spell, and a holy/unholy sanctification. They're all single choices with a single effect, but all have follow-up feats that are dependent on them, and I'm not sure which ones I'd consider subclasses and which ones I wouldn't: Causes clearly are, but the others are much more shallow.

I need to spend some time evaluating what would make the most sense: in the end Pathfinder doesn't really have an official subclass designation, it's just about what features I let the user select on the left side and have text in the main frame change based on (as opposed to just showing all options in the main frame). I think it's useful to show these selections to make it clear what choices players have to make apart from their class-independent selections like ancestry and attributes. But this is very much "down the rabbit hole" that I should look into later, I have way more important features to be adding. So for now I just define each class's selections as subclass or not in a config file (classes are rare enough that I'm not worried about this scaling, I might have to do 4 a year, and I can set reasonable defaults too), and am fixing smaller bugs.

Luckily, I've found an in-hindsight-obvious fix for my issue of needing "clean" pf2etools data to put Parsefinder on: we're running Git, so just do it on a different branch! Switch PF2etools to prod, hard-reset to match dev, run a Parsefinder run and copy the data to the repo, force-push the whole thing up, and now Cloudflare's serving Parsefinder's best on dev's latest, but pf2etools's dev branch still has "clean" data I can reset to for a future run. I'll eventually need to do an intermediate branch with a preview to make sure everything doesn't break, but for now only I have access to it, and I'm not sure this will ever be widespread enough to matter.

I've been submitting a lot of bug reports to Nethys as well as a result of this, which hopefully makes up for using their data as a source. They're making some more decisions that I personally find ridiculous, so once again it's a great time to be "blackjack and hookers"-ing it.