#gemrb@irc.freenode.net logs for 14 Feb 2015 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage

[01:27:15] --> Eli2 has joined #gemrb
[01:27:25] <-- Eli2_ has left IRC (Ping timeout: 264 seconds)
[01:35:27] <-- NoseGoth has left IRC (Read error: Connection reset by peer)
[01:35:39] --> Drakkar has joined #gemrb
[02:31:35] <-- turtleman_ has left IRC (Ping timeout: 256 seconds)
[05:05:30] <-- raevol has left IRC (Quit: Leaving)
[05:16:29] --> brada has joined #gemrb
[05:17:32] <brada> https://www.dropbox.com/s/mlii2hpt5dsomh6/portraiticons.mov?dl=1
[05:17:49] <brada> little sample of my latest work
[05:18:02] <brada> ignore the messed up scrollbar tho :)
[06:02:28] <-- Lightkey has left IRC (Ping timeout: 250 seconds)
[06:15:16] --> Lightkey has joined #gemrb
[06:47:14] <-- brada has left IRC (Quit: brada)
[09:58:53] <lynxlynxlynx> no flickering?
[13:50:19] --> turtleman_ has joined #gemrb
[14:53:48] <-- Drakkar has left IRC (Ping timeout: 244 seconds)
[14:57:25] --> Drakkar has joined #gemrb
[16:28:30] --> brada has joined #gemrb
[16:28:46] <brada> lynx: correct, no flickering.
[16:29:13] <brada> also animated borders, but understandable if nobody cares about that :)
[16:29:29] <brada> they arent as pretty as original anyway
[16:35:54] <Lightkey> https://swordcoast.com/ that's a strange one, so apparently former Bioware game director Dan Tudge is now leading n-Space who until now did Nintendo conversions of games and they are teaming up with Digital Extremes, known for doing contract work on FPSs to build a 5th edition RPG that is supposed to come out in Q3 2015 already (only for Windows), with a 4-player coop modus plus a dungeon master
[16:35:56] <Pepelka> Home | Sword Coast Legends
[17:38:24] --> kpedersen has joined #gemrb
[17:41:12] <-- kpederse1 has left IRC (Ping timeout: 244 seconds)
[18:25:51] <-- brada has left IRC (Quit: brada)
[18:49:02] --> brada has joined #gemrb
[19:31:13] --> edheldil_ has joined #gemrb
[19:40:47] <-- brada has left IRC (Quit: brada)
[19:42:20] <-- turtleman_ has left IRC (Ping timeout: 250 seconds)
[20:12:28] --> brada has joined #gemrb
[20:15:19] --> fizzle has joined #gemrb
[20:26:36] --> turtleman_ has joined #gemrb
[20:49:54] <fizzle> another regression: scrollbar behaviour is broken if scrollbar height is smaller than uparrow + downarrow + trough (e.g. the small message window)
[20:54:31] <lynxlynxlynx> my short term goal is to keep brad off shiny new stuff, so we can tie up all the loose ends
[20:55:11] <lynxlynxlynx> about 10 regressions remain
[20:57:11] <fizzle> I'm also wondering why it has separate code for 'pixel-perfect' scrolling when you could just scale up the value in the TA and use the regular code path
[20:59:55] <brada> probably due to how text areas were originally coded? we used to have to fake the per-pixel scrolling entirely using % lineheight
[21:03:03] <fizzle> maybe if the text area didn't know its own text height
[21:03:18] <fizzle> we should be able to get rid of it now, though
[21:06:33] <brada> yes, now you can surely. and obviously feel free to do so
[21:07:40] <fizzle> I'm trying not to touch it while there are still regressions lingering in there :)
[21:08:52] <brada> you can just fix those too :)
[21:10:31] <fizzle> phhhfff, where's the fun in that ;-P
[21:16:20] <brada> fizzle: what game can i reproduce that with?
[21:16:24] <brada> BG2 works fine
[21:16:45] <fizzle> I see the problem in both bg1 and bg2
[21:16:55] <lynxlynxlynx> which scrollbar though
[21:16:57] <brada> im trying in BG2...
[21:17:12] <brada> i dont see a problem
[21:17:14] <lynxlynxlynx> and are you using the same resolution
[21:17:21] <fizzle> if you look at the constructor, you'll get an undeflow in SliderRange
[21:18:00] <brada> im on 640x480
[21:18:05] <brada> so cant go any smaller
[21:18:20] <fizzle> I'm on 800x600 in bg2
[21:18:48] <fizzle> to reproduce, start a dialog with the message window minimized
[21:19:14] <fizzle> after the dialog ends, the message window should scroll to the bottom but it stays at the top
[21:19:29] <brada> oh ok, so you have to do more than just inimize it
[21:19:32] <brada> minimize
[21:19:45] <brada> because that works fine :p
[21:19:48] <fizzle> if you already have text in it, just minimizing does it as well
[21:19:55] <brada> not for me..
[21:20:32] <brada> even after dialog it works
[21:21:47] <brada> are you just talking about how the text scrolls to the top instead? because it sounded like the actual scrollbar was not functional
[21:22:12] <fizzle> well, it's functional (sort of) but its internal values are all messed up
[21:22:31] <brada> and im not sure what that has to do with the specific height of up+down+trough
[21:22:48] <fizzle> look at the ScrollBar constructor
[21:23:02] <fizzle> SilderRange will underflow if height is smaller than the graphics
[21:24:06] <brada> oh sure. so easy fix right?
[21:24:29] <brada> just cap that at 0
[21:24:57] <fizzle> no, with SliderRange 0 the scrollbar won't work anymore
[21:25:10] <fizzle> I'm not sure what the expected behaviour is without a trough, though
[21:25:41] <brada> well you said this is a regression so it should be easy to look at the log
[21:25:59] <fizzle> also, the gfx offets aren't taken into account in other methods like SetPosForY
[21:26:21] <brada> i doubt that is needed
[21:26:25] <fizzle> so it's likely that numbers are all subtly off
[21:27:19] <fizzle> but I haven't looked at it that closely
[21:29:00] <brada> im pretty sure slider range should be 0 in the mentioned case
[21:30:48] <fizzle> try it :)
[21:31:35] <brada> i did. it doesnt fix the bug, but the scrollbar still works
[21:32:10] <brada> the bug with it scrolling to the top may be elsewhere of course
[21:33:28] <brada> i may have not done this exactly right
[21:36:34] <brada> has no effect, and that should be expected from what i can tell. sliderrange isnt used by the up/down button path so it didnt matter that it was underflowing
[21:38:45] <fizzle> it matters if you try to set position by code (as TA does)
[21:44:08] <brada> sure, but the result will always be the same anyway (end up at the top). 0 is always the top :/
[21:45:18] <fizzle> sure, but that's not where it should be
[21:46:07] <fizzle> anyway, need to go now
[21:46:10] <fizzle> gn
[21:46:14] <brada> right, agreed. so what will work here is instead of clamping it to 0, try some positive value
[21:46:24] <brada> probably > 1
[21:46:56] <-- fizzle has left #gemrb
[21:47:15] <brada> oh, no, 1 seems to work :D
[21:47:25] <brada> so yeah, ill commit that
[21:51:42] <Pepelka> [commit] bradallred: Scrollbar: insist on a positive value https://github.com/gemrb/gemrb/commit/d7b00eed96b911c824079995cfb93b157463850b
[21:53:32] <-- brada has left IRC (Quit: brada)
[22:17:26] <-- Lightkey has left IRC (Remote host closed the connection)
[22:44:14] --> brada has joined #gemrb
[22:45:32] <Pepelka> [commit] bradallred: addendum to last commit: should be <= 0 https://github.com/gemrb/gemrb/commit/7f1062d93ad3c530bb360ed4e844455a870cb026
[23:14:52] <-- turtleman_ has left IRC (Ping timeout: 240 seconds)
[23:33:59] <-- edheldil_ has left IRC (Ping timeout: 245 seconds)
[23:43:24] <Pepelka> [commit] bradallred: Fix character importing https://github.com/gemrb/gemrb/commit/7358f46993ea59171a382bd326671eff9a9b2292
[23:49:13] --> edheldil_ has joined #gemrb