Developing Bugs

[Sloth]

I still can't start a game, but i don't (and didn't) have any compiling errors. So where's the problem?

[Chris H.]

I still can't start a game, but i don't (and didn't) have any compiling errors.

 
I see that same thing as Sloth, no compiling errors and I am able to run forge from within Eclipse, but starting a game will return me to the home view without starting the match.

[friarsol]

I definitely don't have any compilation errors and no new files were added, maybe your files aren't refreshed? Specifically, GuiChoose which will now take a reference Card parameter.

[Max mtg]

I definitely don't have any compilation errors and no new files were added, maybe your files aren't refreshed? Specifically, GuiChoose which will now take a reference Card parameter.

Yet, I do have them.
Have a workaround too: cast nulls to Object[].

Code: Select all

final List<Object> firstPile = GuiChoose.getOrderChoices("Place into two piles", "Pile 1", -1, pool.toArray(), (Object[])null, card);

Causes no error

[friarsol]

That doesn't make any sense. That parameter that youre casting has been there for a week or two already

[Max mtg]

That doesn't make any sense. That parameter that youre casting has been there for a week or two already

Would you like me to make a screenshot to proove that there emerges a compile error?

Look, the whole DualListBox without parameters and accepting arrays of objects has 2 problems:
1. You have to cast the lists you are passing as parameters to arrays. (and null passed for parameter of type Object[] leads to compilation error)
2. After you've received a list from the routine, you have to cast it to a list of appropriate type, which in turn leads to code duplication - mind the efforts you spend to make a card array

Code: Select all

                List<Object> ordered = GuiChoose.getOrderChoices("Choose Blocking Order", "Damaged First", 0, attackers.asArray(), null, blocker);
               
                orderedAttacker = new ArrayList<Card>();
                for(Object o : ordered) {
                    orderedAttacker.add((Card)o);
                }


I have already refactored see r17372.

Please do not hesistate to use generic templates, even if current runtime is discouraging you to do so.

Code: Select all

class UnsortedListModel<T> extends AbstractListModel { // Java 7 has a generic version. In 6 we have to cast types


[Max mtg]

Using 17364, i can't start a game. Clicking Start with a random deck, fixed deck or quest mode will always return to the home screen after the loading window appears briefly.

Max Mtg's r17360 is the culprit. Max could you please at least try to start a game and play a little before you check in code.

It is now possible to start a new game.

[Doublestrike]

null


Version:
Forge version SVN

OS: Windows 7 Version: 6.1 Architecture: amd64

Java Version: 1.6.0_34 Vendor: Sun Microsystems Inc.

Detailed error trace:
java.lang.NullPointerException
at forge.card.spellability.SpellAbility.chooseTargetAI(SpellAbility.java:211)
at forge.game.player.ComputerUtil.handlePlayingSpellAbility(ComputerUtil.java:160)
at forge.game.player.ComputerUtil.playSpellAbilities(ComputerUtil.java:111)
at forge.game.player.ComputerAIGeneral.playSpellAbilitiesStackEmpty(ComputerAIGeneral.java:85)
at forge.game.player.ComputerAIGeneral.playSpellAbilities(ComputerAIGeneral.java:340)
at forge.game.player.ComputerAIInput.think(ComputerAIInput.java:113)
at forge.game.player.ComputerAIInput.showMessage(ComputerAIInput.java:74)
at forge.gui.GuiInput.setInput(GuiInput.java:74)
at forge.gui.GuiInput.update(GuiInput.java:60)
at java.util.Observable.notifyObservers(Unknown Source)
at java.util.Observable.notifyObservers(Unknown Source)
at forge.util.MyObservable.updateObservers(MyObservable.java:41)
at forge.game.GameNew.newGame(GameNew.java:60)
..... et cetera

Getting quite a few of these (during match play).

[friarsol]

forge.card.spellability.SpellAbility.chooseTargetAI(SpellAbility.java:211)
at forge.game.player.ComputerUtil.handlePlayingSpellAbility(ComputerUtil.java:160)

Oh god, chooseTargetAI? I think only hardcoded cards use that garbage.

[Doublestrike]

Yeah, I don't do much work in that department so I can't give a helpful report, unfortunately, but it was when I was running various games to test the ante problem and multiblocker UI. I hadn't changed any code yet.

They came up quite often in several different games...that's about as narrow as I can get for now.

[Max mtg]

forge.card.spellability.SpellAbility.chooseTargetAI(SpellAbility.java:211)
at forge.game.player.ComputerUtil.handlePlayingSpellAbility(ComputerUtil.java:160)

Oh god, chooseTargetAI? I think only hardcoded cards use that garbage.

That was mine. Already fixed

[Sloth]

Oh god, chooseTargetAI? I think only hardcoded cards use that garbage.

Challenge accepted.

[RumbleBBU]

For some reason, when you try to edit a Quest deck now, you always get an empty deck instead of the deck you are trying to edit. It worked on Friday, and the relevant V&CSubmenus haven't been changed, so the problem is somewhere else.

(...and no, I don't think I caused this either.)

[Max mtg]

For some reason, when you try to edit a Quest deck now, you always get an empty deck instead of the deck you are trying to edit. It worked on Friday, and the relevant V&CSubmenus haven't been changed, so the problem is somewhere else.

(...and no, I don't think I caused this either.)

The reason is removal of global variable HUMAN_DECK (because global variables are evil and there won't be an only human in game)

Suggested solution: keep the current quest player's deck somewhere in quest-related structures.
A transient field in QuestAssets may be an option.

[RumbleBBU]

Could that also be the underlying cause of the Draft game problem observed by Sloth?
viewtopic.php?f=52&t=8029&start=45#p99790