6-15-24 Update - The AI Update

Skip to the next paragraph if you don't want to hear more excuses, but have you ever had something you really need to do, so you say "I'll do that tomorrow", and suddenly almost 2 weeks have passed by?  The main issue that held back the previous update a few days was that I was trying to get the AI to chain its actions together, which had taken me a few days of working on up to that point, only to finally get it figured out within an hour after I had posted the update, so I was pretty annoyed about that.  Turns out there were still quite a few more issues anyways, like the AI calling the Evaluate function about a dozen times when it should have only been called once, and then about a dozen cards that were broken that I found while testing the AI's capabilities, so it was a few frustrations that caused this delay.

So the day has finally come, the AI is finally ready!  Mostly.  Sort of.  The AI of any game has 2 main parts, the Artificial, and the Intelligence.  The Artificial is making sure the AI can actually play the game by itself, and the Intelligence part is making sure the AI actually behaves like a player.  I've got the Artificial part mostly done, I'm putting this update out now because I said I'd have it up by the middle of the month, but there's almost no logic currently implemented, so the AI won't be playing very well.  I probably should hold off on posting this until the AI is actually more ready, but whatever, it's not like I'm going to stop now.  I have pages of notes for the AI logic, it's just going to take more time to really get it going.  Right now many decisions the AI does is random, which obviously isn't great, but that's the next step.

The AI is (or at least should be) capable of handling any card you give it, so unlike the AI for PTCGO/Live, you can give the AI any deck you want and it can use it decently well.  The ultimate goal is to have the AI play like a player of medium-high skill, which of course is going to take quite some time to accomplish.  But with the Artificial portion basically done, the next chunk of work I'm going to be doing will be implementing the logic.  I'll try to remember to mention if/when bigger AI changes happen, but a lot of it will be behind the scenes changes, so a lot of it won't be overly noticeable for a while.

To use the AI, all you need to do is select the 'Play With AI' checkbox when selecting the decks you want to play with.  Player 2 will always be the AI controlled player.  There's also an 'AI Pause Timer' selection that lets you choose how long the AI will 'think'.  It's really more if you want to watch what the AI does, since you can select a 0 timer and have the AI's turn done in just a few seconds.  If you don't select anything, it defaults to a timer of 2.  During the AI's turn, an 'Unstuck AI' button will appear, which if the AI seems stuck, click it to restart the AI's processing.  

I think that should be everything.  And unlike the previous updates, I actually kept a changelog this time, if anybody cares.  It's the rest of this log, so if you don't want to read it all, thank you for checking out this project.  It's been a ton of time, but it's been a fun project.  The next update should be the next set in a few months.

Base AI (almost no logic yet)

Unstuck AI button (and 'thinking' swirl)

In game Log and related trackers

Win Panel end of game stats

Win Panel 'Deck Manager' and 'Edit This Deck' buttons working

In game Help button to describe certain intricacies of how I've set up the PDT

Deck Select Panel Extra Settings: Play With AI, Always Show Players' Hands, AI Pause Timer, Custom Rulesets (and a Help button too)

Hiding player hands: either always shown, only active player's is shown, or yours always shows and the AI never shows.

Bugfix: reflip effects were pretty off.  A single reflip effect wasn't actually reflipping, and when used with PRC Torchic you could skip the other player's turn and damage was being applied incorrectly for the second attack.

Bugfix: cards that requires multiple copies to play (Cross Switcher/Crossceiver) were playing with just 1 in hand, but would just discard it without using it, and would crash if left clicked on in the discard pile.

Bugfix: Versatile Mew EX (and others with attack borrowing abilities) could attack turn 1 going first.

Bugfix: apparently Agatha broke at some point. no idea how, but it's fixed now.

Bugfix: typo fix for Bill's Analysis and Trainers' Mail that prevented them from working.

Bugfix: Greedy Dice wasn't working at all for Player 2, and it was able to be played from the hand, which would only discard it.

Bugfix: Grant was completely broken, and I'd love to know how I managed that one, considering it wasn't before.

Bugfix: speaking of Grant, apparently a lot of trainer cards could be used from the Discard, so this fix may have broken other cards. Isn't programming great?

Bugfix: Hisuian Heavy Ball and other prize swapping cards weren't actually swapping cards. Another case of making 1 seeming unrelated change for a card breaking another.

Bugfix: Lost Vacuum was also broken.

Bugfix: Professor Sada's Vitality was missing the Draw 3 effect.

Bugfix: there were a few cards that I had as 'second, first turn' rather than 'first turn, second', which is the 'you can only use this effect if you go second on your first turn' effect.

Bugfix: Escape Rope and Guzma were being returned to the hand if the player who played the card only had 1 benched mon.