Bug Reports (snapshot builds)
Moderators: timmermac, Blacksmith, KrazyTheFox, Agetian, friarsol, CCGHQ Admins
Re: Bug Reports (snapshot builds)
by Hellfish » 11 May 2013, 22:04
Hmm, The reason for this bug looks simple. While the copy token gets moved to the battlefield, it is still "nowhere" or in "null", so it doesn't get included in Player.getAllCards() (I forget the exact name) that's used to round up all cards to look for replacement effects in. I'll take a look tomorrow if noone beats me to it.friarsol wrote:r21433
CopyPermanent does not fire static ETB abilities.
1. Bring a Spike Feeder into play.
2. Cast Cackling Counterpart on it.
3. Watch as nothing seems to happen.
I just noticed this with Prime Speaker Zegana not ETB with counters when testing Momir Basic stuff, but the above is an easier repro.
EDIT: Done in r21442
Last edited by Hellfish on 12 May 2013, 08:53, edited 1 time in total.
So now you're
Screaming for the blood of the cookie monster
Evil puppet demon of obesity
Time to change the tune of his fearful ballad
C is for "Lettuce," that's good enough for me
Screaming for the blood of the cookie monster
Evil puppet demon of obesity
Time to change the tune of his fearful ballad
C is for "Lettuce," that's good enough for me
-
Hellfish - Programmer
- Posts: 1297
- Joined: 07 Jun 2009, 10:41
- Location: South of the Pumphouse
- Has thanked: 110 times
- Been thanked: 169 times
Re: Bug Reports (snapshot builds)
by Diogenes » 11 May 2013, 22:08
I'm going to Unburial Rites a very old issue here, forgive me. This one has a bit of history, and several users and devs had different behavior on their ends, so not everyone might be experiencing this. The problem at the time was that we couldn't get any of the devs to reproduce this, so I'd really appreciate confirmation from at least one other person for whom this is true.
Only one image is used per basic land per game on the human side unless the "randomize card art" option is checked (it isn't the same image every game, however.) If that option is checked, basic lands will use all the images from one set (other cards will draw from multiple sets in which they appear.)
This first cropped up around 12.9, as a fix for an issue where all lands from a set would be used *except* for the very first. Some users experienced this after that fix, some did not. The theory floated at the time was that it had something to do with using HQ images, although I've tested with both LQ and HQ images and have the exact same behavior (at the time, and again with today's daily.)
Anyway, I imagine that the expected behavior should be that without "randomize card art" checked all the island/etc. images from one set should be seen in play, and if the option is checked then a random assortment of mountains/etc. from across all sets should be displayed.
Just to confirm, this has been present at all times for me since 12.10. I'm bringing it up now in part because Xitax posted a 32-bit compatible version of his set-assignment tool for decklists which mostly negates the issue for me (and brought it to mind,) and because I repro'ed it with LQ images earlier in the day (I do this occasionally to check if anything has changed.)
Only one image is used per basic land per game on the human side unless the "randomize card art" option is checked (it isn't the same image every game, however.) If that option is checked, basic lands will use all the images from one set (other cards will draw from multiple sets in which they appear.)
This first cropped up around 12.9, as a fix for an issue where all lands from a set would be used *except* for the very first. Some users experienced this after that fix, some did not. The theory floated at the time was that it had something to do with using HQ images, although I've tested with both LQ and HQ images and have the exact same behavior (at the time, and again with today's daily.)
Anyway, I imagine that the expected behavior should be that without "randomize card art" checked all the island/etc. images from one set should be seen in play, and if the option is checked then a random assortment of mountains/etc. from across all sets should be displayed.
Just to confirm, this has been present at all times for me since 12.10. I'm bringing it up now in part because Xitax posted a 32-bit compatible version of his set-assignment tool for decklists which mostly negates the issue for me (and brought it to mind,) and because I repro'ed it with LQ images earlier in the day (I do this occasionally to check if anything has changed.)
Re: Bug Reports (snapshot builds)
by Max mtg » 12 May 2013, 07:56
Fixed by 21441Diogenes wrote:The AI played a Tablet of the Guilds, resulting in the following NPE when it had to choose colors:
- Code: Select all
Forge Version: 1.3.16-SNAPSHOT-r21419
java.lang.NullPointerException at forge.game.ai.ComputerUtilCard.getColorByProminence(ComputerUtilCard.java:835)
Single class for single responsibility.
- Max mtg
- Programmer
- Posts: 1997
- Joined: 02 Jul 2011, 14:26
- Has thanked: 173 times
- Been thanked: 334 times
Re: Bug Reports (snapshot builds)
by krevett » 12 May 2013, 09:07
Hi, about Sylvan Library, if I got a Dark Confidant out the additional card it gives me also appear as drawn with Sylvan Library (I can choose between the one from Dark Confidant, the one from my draw step and two more from Sylvan Library, wich is not true as Dark Confidant doesn't draw cards.)
- krevett
- Posts: 109
- Joined: 21 Feb 2012, 22:24
- Location: France
- Has thanked: 18 times
- Been thanked: 9 times
Re: Bug Reports (snapshot builds)
by Salmiaka » 13 May 2013, 11:03
I couldn't use Conjurer's Closet ability on Master of the Pearl Trident which I took to an opponent with Overtaker . I do not know if it is a bug.
Conjurer's closet : At the beginning of your end step, you may exile target creature you control, then return that card to the battlefield, under your control.
Conjurer's closet : At the beginning of your end step, you may exile target creature you control, then return that card to the battlefield, under your control.
Re: Bug Reports (snapshot builds)
by Diogenes » 13 May 2013, 23:46
Planechase related: Very frequently when you planeswalk away from a plane, it won't be removed from the command zone. The game continues with players having access to both planes and their abilities, until planeswalk is rolled on the planar die (of either plane). At that point, three planeswalking triggers will go on the stack and the game will crash.
The following crash was recorded without using rigged rolls (just lots and lots of mana - it took me 19 rolls to get the first planeswalk.)
The following crash was recorded without using rigged rolls (just lots and lots of mana - it took me 19 rolls to get the first planeswalk.)
- UnsupportedOperationException | Open
- UnsupportedOperationException | Open
Re: Bug Reports (snapshot builds)
by Diogenes » 14 May 2013, 00:28
Sorry for the double-post, but this is a completely distinct issue.
I just realized that the card-detail text is generated from the scripted abilities on cards, and is not using the Oracle text field in the card's file. This means that certain cards display "Unblockable" and "Indestructible" as keyworded abilities (they are not keywords,) rather than the proper "X is unblockable" or "X is indestructible", while others do not. On a minor aesthetic note, older-set or coreset creatures with landwalk appear under a search for "unblockable" in the deck editor, which does not appear anywhere in their details (Oracle retains reminder text for these cards.)
Is there a reason not to use the Oracle text in the card detail field? I feel it would be superior to the current system.
[Edit:] Here's as good a place as any to bring up a UI issue, I guess.
There was talk over in the feature requests thread to make the AI's spell casting more obvious. I think the number one reason I don't realize the AI is doing something is because I have my stack and combat tabs sharing a pane, and the UI doesn't return focus to the stack when the AI does anything during combat - I only realize when I attempt to pass to the next step, only to see that the step-tracker hasn't advanced (and I have allowed the AI's action resolve without having responded.) Can it be changed so that focus returns to the stack tab when anything goes on the stack?
On a related note, I've been hunting down a bug where the AI can play spells during its untap step (between turns, no step is highlighted on the track.) I also believe I've seen abilities auto-resolve, I think when the AI puts multiple on the stack during untap, but I haven't been able to repro this (it's too easy to pass priority habitually from CU to UP.) Just figure I'd mention this in case I never find repro steps.
I just realized that the card-detail text is generated from the scripted abilities on cards, and is not using the Oracle text field in the card's file. This means that certain cards display "Unblockable" and "Indestructible" as keyworded abilities (they are not keywords,) rather than the proper "X is unblockable" or "X is indestructible", while others do not. On a minor aesthetic note, older-set or coreset creatures with landwalk appear under a search for "unblockable" in the deck editor, which does not appear anywhere in their details (Oracle retains reminder text for these cards.)
Is there a reason not to use the Oracle text in the card detail field? I feel it would be superior to the current system.
[Edit:] Here's as good a place as any to bring up a UI issue, I guess.
There was talk over in the feature requests thread to make the AI's spell casting more obvious. I think the number one reason I don't realize the AI is doing something is because I have my stack and combat tabs sharing a pane, and the UI doesn't return focus to the stack when the AI does anything during combat - I only realize when I attempt to pass to the next step, only to see that the step-tracker hasn't advanced (and I have allowed the AI's action resolve without having responded.) Can it be changed so that focus returns to the stack tab when anything goes on the stack?On a related note, I've been hunting down a bug where the AI can play spells during its untap step (between turns, no step is highlighted on the track.) I also believe I've seen abilities auto-resolve, I think when the AI puts multiple on the stack during untap, but I haven't been able to repro this (it's too easy to pass priority habitually from CU to UP.) Just figure I'd mention this in case I never find repro steps.
Re: Bug Reports (snapshot builds)
by lsdmaster » 14 May 2013, 09:23
Description: AI used cycling on Rejuvenate.
This crash seems to happen when AI used cycling (verified with other cards) on my turn.
This crash seems to happen when AI used cycling (verified with other cards) on my turn.
- NullPointerException | Open
Re: Bug Reports (snapshot builds)
by Pux » 14 May 2013, 15:44
If you play modes like constructed or vanguard and select some decks/colours and hover over the selected deck option (for example "Custom User deck" it will reset selected deck or colour (which is annoying if you switch tabs in vanguard).
Re: Bug Reports (snapshot builds)
by slowe » 14 May 2013, 17:42
I can confirm that the AI will cast spells during its untap step. As no one's supposed to get priority, it resolves immediately and you can't respond. Simple steps to repro: give AI a Doom Blade in hand and enough mana sources in play to cast it, making sure they're tapped. Give yourself a nonblack creature. Pass the turn.
Another weird bug: whenever I use any Deathrite Shaman ability, AI gets the benefit (I haven't tried against multiple opponents; not sure what would happen). Interestingly, the stack description correctly says who will gain/lose life or add mana. I've checked, and it doesn't matter whose cards are exiled.
Both r21481.
Another weird bug: whenever I use any Deathrite Shaman ability, AI gets the benefit (I haven't tried against multiple opponents; not sure what would happen). Interestingly, the stack description correctly says who will gain/lose life or add mana. I've checked, and it doesn't matter whose cards are exiled.
Both r21481.
Re: Bug Reports (snapshot builds)
by Sloth » 14 May 2013, 18:57
@Max: It looks like you've removed the activatingPlayer variable of SpellAbilityStackInstances in r21468. How should the game know who activated the spell or ability later?slowe wrote:Another weird bug: whenever I use any Deathrite Shaman ability, AI gets the benefit (I haven't tried against multiple opponents; not sure what would happen). Interestingly, the stack description correctly says who will gain/lose life or add mana. I've checked, and it doesn't matter whose cards are exiled.
-
Sloth - Programmer
- Posts: 3498
- Joined: 23 Jun 2009, 19:40
- Has thanked: 125 times
- Been thanked: 507 times
Re: Bug Reports (snapshot builds)
by myxibrium » 15 May 2013, 00:34
Description: Modular artifact creature dies without any other creatures or artifacts in play.
- IllegalArgumentException | Open
- myxibrium
- Posts: 2
- Joined: 12 May 2013, 22:29
- Has thanked: 0 time
- Been thanked: 0 time
Re: Bug Reports (snapshot builds)
by Diogenes » 15 May 2013, 01:32
It took me a few minutes to realize that this is probably done so that the UI properly reflects when a creature loses abilities. Since this is legitimately useful, I guess the only bug is that indestructible and unblockable appear as keywords - but I'd still like to be able to read the original Oracle text somehow.Diogenes wrote:Is there a reason not to use the Oracle text in the card detail field? I feel it would be superior to the current system.
Would it be possible to change the appearance of these two to "CARDNAME is unblockable" and "CARDNAME is indestructible" in the card details, and also make hovering over or clicking on the card detail panel display the card's unmodified Oracle text?
Re: Bug Reports (snapshot builds)
by myxibrium » 15 May 2013, 02:52
Description: Restarted the game using Karn Liberated's ability. Modular artifacts were added to the battlefield including an Arcbound Crusher which get's +1+1 for all the artifacts now added to the battlefield from Karn's exile. Then this happened.
(Apparently my post looks too spamy with off-site URLs. So I removed some details which may have resembled a url. Buggy forum.)
Forge Version: 1.3.15-SNAPSHOT-r21296
Operating System: Mac OS X 10.8.3 x86_64
Java Version: 1.6.0_45 Apple Inc.
RuntimeException: Trying to unlock input which is not locked! Do check when your threads terminate!
at forge-control-input-InputControl-unlock(InputControl-java:208)
(Apparently my post looks too spamy with off-site URLs. So I removed some details which may have resembled a url. Buggy forum.)
Forge Version: 1.3.15-SNAPSHOT-r21296
Operating System: Mac OS X 10.8.3 x86_64
Java Version: 1.6.0_45 Apple Inc.
RuntimeException: Trying to unlock input which is not locked! Do check when your threads terminate!
at forge-control-input-InputControl-unlock(InputControl-java:208)
- myxibrium
- Posts: 2
- Joined: 12 May 2013, 22:29
- Has thanked: 0 time
- Been thanked: 0 time
Who is online
Users browsing this forum: No registered users and 21 guests