[17:42:23] Dominus: Marzo, I should remind you of https://github.com/exult/exult/issues/159 :)
[17:42:39] Dominus: I don't really dare removing the dragging check...
[18:22:23] Marzo: Is there any object in the original games that calls UI_close_gumps when readying/unreadying it?
[18:22:55] Dominus: I don't recall one at the moment
[18:32:45] Marzo: Ah, there is the black sword in SI
[18:41:03] Marzo: Huh
[18:41:24] Marzo: The black sword is enough to prove that ready/unready events are delayed in SI
[18:42:40] Marzo: Oh, I get it
[18:42:57] Marzo: With that check, Exult kind of matches the behavior in SI when dragging the black sword
[18:43:28] Marzo: With the check removed, as soon as you drag the black sword, gumps close immediately
[18:43:50] Marzo: Hm
[18:44:42] Marzo: Delaying the ready/unready usecode to after gummps being closed leaves room for some exploits that happen in the original which I don't think work in Exult
[18:47:08] github-actions: exult/refs/heads/master 540ca8cc36ffc5d9c24fcb36a915835456d20825 https://github.com/exult/exult/actions/runs/1938781401 The Windows build failed.
[18:54:19] github-actions: exult/refs/heads/master 540ca8cc36ffc5d9c24fcb36a915835456d20825 https://github.com/exult/exult/actions/runs/1938781399 The Linux build succeeded.
[19:06:56] github-actions: exult/refs/heads/master 0707b3e025250a2787627284dd0f50eddd84e325 https://github.com/exult/exult/actions/runs/1938804677 The Mac OSX build succeeded.
[19:07:22] github-actions: exult/refs/heads/master 0707b3e025250a2787627284dd0f50eddd84e325 https://github.com/exult/exult/actions/runs/1938804678 The Linux build succeeded.
[19:08:55] github-actions: exult/refs/heads/master 0707b3e025250a2787627284dd0f50eddd84e325 https://github.com/exult/exult/actions/runs/1938804679 The Windows build succeeded.
[20:08:17] Dominus: ugh, so I guess you want to do it properly. We *could* have added an extra exult intrinsic only for that, right? (UI_close_gumps_while_dragging)
[20:25:55] Marzo: I added a long post in the issue explaining my findings, and proposing a possible solution
[20:31:41] Marzo: It is funny that the first thing I tested was the one that would be broken by removing the check
[20:44:10] Dominus: funny and good that you thought of that first
[20:44:20] Dominus: so, it's complicated...
[20:45:12] Dominus: I think, stuff currently works and without breaking the game somewhere, I think it would be best to add a parameter as you suggested
[20:45:41] Dominus: and backward compatible, so that the parameter is not required?
[20:57:44] Marzo: Yes, the parameter would be assumed to be false if it is not there
[21:07:30] Marzo: On second thought, I think I will add another intrinsic with the new behavior. It will be easier on the documentation
[22:19:13] github-actions: exult/refs/heads/master 4f22f4f9a983b8d5092d4affb1063017e4b51491 https://github.com/exult/exult/actions/runs/1939224749 The Windows build succeeded.
[22:21:06] github-actions: exult/refs/heads/master 4f22f4f9a983b8d5092d4affb1063017e4b51491 https://github.com/exult/exult/actions/runs/1939224750 The Mac OSX build succeeded.
[22:22:33] github-actions: exult/refs/heads/master 4f22f4f9a983b8d5092d4affb1063017e4b51491 https://github.com/exult/exult/actions/runs/1939224751 The Linux build succeeded.