Bug Reports (snapshot builds)

[Agetian]

@ drdev: I'm on Linux, I do not use Windows on any of my computers. It might be platform-specific, yes, I recall there was a number of issues with the card zoom mode in the past (on Linux) until they were solved, this seems to be a new one that also affects Linux (and possibly Mac, not sure) only. I can try looking into it myself but I admit my understanding of how the UI works in general (and this part in particular) is severely limited, I'm more likely to break more things than to fix this one without affecting anything else inadvertently... I can test on my OS though if you'd like to try something out.

EDIT 1: The fix that takes care of a similar, yet more serious issue while in the match is on lines 189-193 of CardZoomer.java. Without those lines, this same issue happens on the match screen as well (very annoying). The bug in the match mode fixed by these lines was much more serious than the deck editor mode bug though - the game used to fully hang in the card zoomer mode if you inadvertently wheeled while middle clicking, and the only option was to force-quit Forge.

EDIT 2: At least I've found a way to unlock it! If the mouse "hangs" after wheeling while middle clicking in the deck editor, it's possible to get the deck editor to function normally again by middle-clicking the card image in the card picture panel in the bottom right, the mouse functionality is restored to normal then (after the card zoomer is closed once again).

- Agetian

[drdev]

RumbleBBU, you're not doing anything wrong. I marked certain columns with a fixed size so that Swing would resize other columns like name as is ideal. I think what I will do is make it so, when you mouse down to begin a resize, the size restriction is removed temporarily then restored after you finish resizing. That should improve usability while still supporting Swing in good autosizing.

The quest deck issue I'll look into, and I can add a setting to not show confirmation prompts for buying and selling cards, though the default will remain to show them.

How's that sound?

All of the above sounds great to me.

...but you probably knew I was going to say that.

Quest deck issue fixed in r24781. I still need to fix the column resize issue.

All columns now support resizing as of r24820. I also added a fix so the resize mouse cursor properly appears when hovering over the border between two columns on the header and while resizing.

[friarsol]

r24831

The tap X Type cost is no longer cancellable. Not sure when that happened because it definitely used to be. This is the cost found on cards like Field Surgeon and Tradewind Rider.

[xbon]

Latest snapshot; Sphere of the Suns is still bugged, when playing it you get two options, when you should not get any options @ all. (it comes into battlefield tapped, or it comes into battlefield tapped with 3 counters. It should always come into play with counters.)

[Marek14]

Is there any difference in result based on which option you choose? It's supposed to have two replacement effects and you can choose which to apply first, but both should lead to the same result.

[swordshine]

Latest snapshot; Sphere of the Suns is still bugged, when playing it you get two options, when you should not get any options @ all. (it comes into battlefield tapped, or it comes into battlefield tapped with 3 counters. It should always come into play with counters.)

I'll convert it to use a single replacement effect.

[Agetian]

r24837: Got this exception when passing a turn to the AI. Battlefield position attached. Right before I made Polukranos monstrous and attacked with everything I had. The attack went fine, but passing in M2 caused the crash below. Could not continue playing after the crash, had to force-quit Forge.

NullPointerException | Open

Code: Select all

Forge Version:    1.5.13-SNAPSHOT-r${forge.revision}${forge.specialStatus} (mixed revisions detected; please update from the root directory)
Operating System: Linux 3.8.0-26-generic amd64
Java Version:     1.7.0_51 Oracle Corporation

java.lang.NullPointerException
   at forge.ai.ComputerUtilCombat.getLifeThreateningCommanders(ComputerUtilCombat.java:293)
   at forge.ai.ComputerUtilCombat.lifeInSeriousDanger(ComputerUtilCombat.java:379)
   at forge.ai.ability.StoreSVarAi.canPlayAI(StoreSVarAi.java:71)
   at forge.ai.SpellAbilityAi.canPlayAIWithSubs(SpellAbilityAi.java:23)
   at forge.ai.AiController.canPlaySa(AiController.java:619)
   at forge.ai.AiController.canPlayAndPayFor(AiController.java:606)
   at forge.ai.AiController.chooseSpellAbilyToPlay(AiController.java:1115)
   at forge.ai.AiController.getSpellAbilityToPlay(AiController.java:1096)
   at forge.ai.AiController.choooseSpellAbilityToPlay(AiController.java:1047)
   at forge.ai.PlayerControllerAi.chooseSpellAbilityToPlay(PlayerControllerAi.java:382)
   at forge.game.phase.PhaseHandler.startFirstTurn(PhaseHandler.java:965)
   at forge.game.GameAction.startGame(GameAction.java:1487)
   at forge.game.Match.startGame(Match.java:80)
   at forge.control.FControl$4.run(FControl.java:433)
   at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
   at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
   at java.lang.Thread.run(Thread.java:744)


- Agetian

[friarsol]

r24837: Got this exception when passing a turn to the AI. Battlefield position attached. Right before I made Polukranos monstrous and attacked with everything I had. The attack went fine, but passing in M2 caused the crash below. Could not continue playing after the crash, had to force-quit Forge.

Was it a Commander game? Looks like the AI is considering if it should activate it's Tree of Redemption.

[Agetian]

r24837: Got this exception when passing a turn to the AI. Battlefield position attached. Right before I made Polukranos monstrous and attacked with everything I had. The attack went fine, but passing in M2 caused the crash below. Could not continue playing after the crash, had to force-quit Forge.

Was it a Commander game? Looks like the AI is considering if it should activate it's Tree of Redemption.

No, it was a regular Quest match vs. a hard opponent (Caterpillar whatever, the name is in that screenshot XD).

- Agetian

[1224Deko]

AI using Bond of Faith on own nonhuman creatures...

[swordshine]

Offering is broken. When I offers a Meloku the Clouded Mirror to Patron of the Moon, the moonfolk is not sacrificed.

[swordshine]

Mandatory CopySpellAbility is broken (Hive Mind). Maybe something is not correct in forge.ai.PlayerControllerAi.playSpellAbilityForFree(SpellAbility, boolean).

[friarsol]

r24863
I have a scripted Perplexing Chimera, but it seems to reveal a similar problem with the triggers hitting the stack at the right time that was resolved in other circumstances.

To repro this with cards already in Forge try this:

1. Give the AI an Angelic Protector
2. Cast Shock on Angelic Protector.

Expected: Shock on stack with Protector trigger on top of it.
Actual: Shock on stack.

Passing priority adds the Protector trigger and resolves it immediately, which is more a symptom of the main problem of the trigger not showing up on the stack at the right time.

[Max mtg]

You must be using devmode cheats to get that buggy behavior.
Try casting Shock from hand.

UPD: Nevermind, now (r24874) the game should add triggers properly after you play Shock either way

[moomarc]

When changing an avatar in match setup by either right clicking to get a random one or left clicking then selecting one, you can't change the avatar again without first making that player lose focus clicking on another avatar or adding a player etc.