Open GoogleCodeExporter opened 8 years ago
We actually have automated tests for exactly this situation, and I couldn't
repro the issue. I will dig a bit deeper, it could be AI specific, or maybe our
test does not take into account some specific case.
Original comment by wagic.the.homebrew@gmail.com
on 4 Sep 2011 at 9:10
The errors I get on the command line are (seemingly randomly) one of the below:
wagic: ../../../../src/projects/mtg/src/ActionLayer.cpp:170: void
ActionLayer::setCurrentWaitingAction(ActionElement*): Assertion `!ae ||
!currentWaitingAction' failed.
Aborted
wagic: ../../../../src/projects/mtg/src/AIPlayer.cpp:933: virtual int
AIPlayer::chooseTarget(TargetChooser*, Player*, MTGCardInstance*): Assertion
`tc->source->controller() == this' failed.
Aborted
Original comment by hutchinsfairy
on 19 Oct 2011 at 12:04
kudzus hardcode was causing this, it is now softcoded.
Original comment by omegabla...@gmail.com
on 27 Aug 2012 at 5:23
Just tried this on Windows and it still errors. The land is tapped and
destroyed and then then Wagic seems to get stuck in a loop with the screen
flickering. I tried hitting "a" and "k" keys (default mapping) and after about
4 seconds kudzu is sent to my discard pile. I no longer have an N900 (sob) to
test this with.
Original comment by hutchinsfairy
on 6 Sep 2012 at 9:39
re-opening to investigate. We do have automated tests for kudzu so I am really
surprised...
Original comment by wagic.the.homebrew@gmail.com
on 6 Sep 2012 at 9:45
Just recreated and it is only an issue when the ai opponent has to select which
land to assign Kudzu to. I played evil twin with just kudzu and forests and
when the ai played kudzu against me there was no problem except that I didn't
seem to get any choice about which land kudzu was assigned to after the first.
Let me know if I can do any specific testing etc.
Original comment by hutchinsfairy
on 6 Sep 2012 at 9:54
Original issue reported on code.google.com by
hutchinsfairy
on 15 Aug 2011 at 9:18