#gemrb@irc.freenode.net logs for 20 Oct 2014 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage

[00:00:59] --> Mechanim1l has joined #gemrb
[00:00:59] <-- Mechanim1l has left IRC (Client Quit)
[03:12:19] --> Eli2_ has joined #gemrb
[03:15:26] <-- Eli2 has left IRC (Ping timeout: 250 seconds)
[04:19:29] --- tomprinc1 is now known as tomprince
[05:57:39] <Textmode> is there any information on emulating the graphical style of IE games, for creating new areas?
[06:36:17] <-- Lightkey has left IRC (Ping timeout: 272 seconds)
[06:48:11] --> Lightkey has joined #gemrb
[07:24:16] --> lynxlynxlynx has joined #gemrb
[07:24:16] --- ChanServ gives channel operator status to lynxlynxlynx
[07:28:59] --> edheldil has joined #gemrb
[07:28:59] --- ChanServ gives channel operator status to edheldil
[07:39:43] <-- edheldil has left IRC (Remote host closed the connection)
[07:42:25] --> edheldil has joined #gemrb
[07:42:37] --- ChanServ gives channel operator status to edheldil
[09:04:27] --> edheldil_ has joined #gemrb
[09:51:22] <-- edheldil_ has left IRC (Ping timeout: 272 seconds)
[11:05:51] <-- edheldil has left IRC (Remote host closed the connection)
[11:08:58] --> edheldil has joined #gemrb
[11:08:58] --- ChanServ gives channel operator status to edheldil
[11:26:51] <-- Textmode has left IRC (Quit: "It was one dev, naked in a room with a carton of cigarettes, a thermos full of coffee and bourbon, and all his summoned angels.")
[11:32:48] <-- lynxlynxlynx has left IRC (Ping timeout: 272 seconds)
[11:42:58] --> lynxlynxlynx has joined #gemrb
[11:42:58] <-- lynxlynxlynx has left IRC (Changing host)
[11:42:58] --> lynxlynxlynx has joined #gemrb
[11:42:58] --- ChanServ gives channel operator status to lynxlynxlynx
[14:16:49] --> brada has joined #gemrb
[17:30:51] <brada> lynx: pushed a fix for whitespace at beginning of line
[17:30:57] <brada> tested only iwth ring of gaxx
[17:31:18] <brada> if you are wondering why it looks like puke at the botttom, its how the original looks too.
[17:35:58] <lynxlynxlynx> i'll test later, still working
[17:37:14] <brada> yesh, me too :p no rush anyhow
[19:12:11] --> edheldil_ has joined #gemrb
[19:33:26] <lynxlynxlynx> ok, let's see
[19:33:44] <lynxlynxlynx> didn't find any mentions in the fixpack, so no idea what you're talking about yet
[19:34:32] <lynxlynxlynx> not so fast, said the compiler
[19:37:44] <lynxlynxlynx> seems fine
[19:40:39] <lynxlynxlynx> and matches original (ring of fire control is a good test with all its indents): http://www.gamebanshee.com/showshot.php?/baldursgateii/equipment/images/ringoffirecontrol.jpg
[19:40:41] <Pepelka90> Screenshots @ GameBanshee
[19:40:42] <Pepelka90> »Daily news, reviews, interviews, previews, screenshots, walkthroughs, databases, cheats, and more for PC, Xbox, GameCube, Playstation, and Game Boy Advance.«
[19:41:17] <brada> oh nice.
[19:41:26] <brada> im a bit suprised something didnt blow up...
[19:41:57] <brada> i canged quite few lines in less than 20 min… usually that means im not thinking though things good enough
[19:42:32] <brada> were we printing those indents before?
[19:42:44] <brada> i dont seem to recall seeing that sort of formatting before
[19:43:15] <brada> it also might explain why the newcode was trimming the spaces too, but that could also be laziness on my part :)
[19:43:59] <lynxlynxlynx> don't worry, while every item in the inventory seemed to be fine, pst still blows up
[19:44:19] <lynxlynxlynx> there are several problems in the journal, but in the end i triggered the same assert
[19:44:49] <lynxlynxlynx> it wasn't immediate though, it sounded like it went into a long loop (fan speed increased) then boom
[19:46:19] <lynxlynxlynx> printing only string=L"Day 1" no less
[19:47:24] <lynxlynxlynx> point is set, but 0,0
[19:47:37] <lynxlynxlynx> you probably wanted isempty() there?
[19:47:46] <brada> not sure
[19:48:01] <lynxlynxlynx> hmm, no, that's at -1,-1 iirc
[19:48:17] <lynxlynxlynx> yeah
[19:48:20] <brada> ill wager tho, that adding the missing newlines would at least make it stop exploding. not that i shoudlnt fix it properly
[19:48:48] <lynxlynxlynx> isnull checks for 0,0
[20:00:14] <brada> you should turn on the font debug flag and see if you get log messages that are useful
[20:00:31] <brada> or if the visual feedback maybe sheds some light on what is going on
[20:01:56] <brada> the PST journal was exploding before too right? just hoping this isnt a regression :)
[20:09:55] <lynxlynxlynx> it was, but immediately
[20:10:32] <lynxlynxlynx> i know point is passed as a reference, but if (point) is still dodgy
[20:10:37] <lynxlynxlynx> without looking at the code
[20:12:54] <lynxlynxlynx> the flag provides no new info
[20:13:47] <lynxlynxlynx> ohh, but plain gdb does
[20:14:37] <lynxlynxlynx> new assert - assert(charsPrinted < text.length());
[20:14:48] <lynxlynxlynx> text has a trailing whitespace char
[20:14:55] <lynxlynxlynx> both lengths are the same
[20:16:25] <lynxlynxlynx> so either i'm lucky or the debug code has unwanted side effects
[20:17:08] <brada> thats the same assert i thought
[20:17:46] <brada> debug flag should jsut add logging and graphical feedback
[20:19:51] <lynxlynxlynx> no, this is in TextSpan::DrawContentsInRegions
[20:20:07] <lynxlynxlynx> the other one was > 0
[20:20:08] <brada> thats where its always been
[20:20:21] <brada> so what/where is the other assert?
[20:21:02] <lynxlynxlynx> ok, we're talking about two things
[20:21:13] <lynxlynxlynx> i thought i was hitting the one you changed before, but not so
[20:21:28] <brada> right, i added a new one that wjp pointed out
[20:21:38] <brada> valgrind complined about it
[20:21:50] <brada> and i was trying to figure out when it occurs
[20:21:56] <brada> but i had it logically wrong
[20:22:02] <brada> and you fixed it
[20:23:11] <lynxlynxlynx> anyway, the gist is that the values are equal at that point
[20:23:23] <lynxlynxlynx> maybe the ending spaces is not counted as printed?
[20:24:10] <brada> and to answer your earlier question, no, we dont want iszero or isnull or any such thing there
[20:24:34] <brada> that is just a check to see if the optional parameter should be assigned to
[20:25:15] <brada> i am jsut trying to seen when charchount == 0 there
[20:25:36] <brada> i dont think it shoudl ever be 0, but you are clearly triggering it
[20:26:26] <brada> but yes possibly is a trailing space
[20:26:31] <brada> i will have to poke at it later on
[20:31:00] <lynxlynxlynx> it's not 0, it's 127 in this case
[20:34:54] <lynxlynxlynx> i am actually hitting both asserts
[20:35:28] <lynxlynxlynx> 0 with the new one and "Day 1" and the old one with the actual journal entry
[20:37:02] <lynxlynxlynx> it's odd that it's not completely consistent
[20:38:17] <lynxlynxlynx> ah, different saves; i can reproduce both 100% if you believe in N=3
[20:40:17] <brada> im surious the conditions that trigger the assert(charCount > 0)
[20:40:21] <brada> curious
[20:40:25] <lynxlynxlynx> and the 0 produces a warning under debug
[20:40:28] <lynxlynxlynx> [Font/WARNING]: Line (23 width) > 6; for text "Day 1"!
[20:43:35] <brada> ah, there we go
[20:44:39] <brada> ok, so it seems it *is* valid for charcount to be 0 sometimes
[20:44:48] <brada> and that should jsut be handled with a simple check
[20:45:12] <lynxlynxlynx> linePos = RenderLine(line, lineRgn, color, linePoint, canvas) <-- linepoint is 0 here if that helps
[20:46:49] <brada> nah, line point 0 just means the beginning of the line. the reall issue is that the line is only 6 px wide. im assuimg this is a tiny remainder of line left over from another span
[20:47:22] <brada> so this shoudl rightfully return 0 chars printed
[20:47:46] <brada> and the text should be rendered on the next complete line
[20:49:56] <lynxlynxlynx> $7 = {x = 262, y = 606, w = 17, h = 13}
[20:50:02] <lynxlynxlynx> linergn is not that small
[20:51:12] <lynxlynxlynx> i guess the title comes first
[20:55:26] <lynxlynxlynx> also seems pst does not force all dialog states to start on top
[20:55:36] <lynxlynxlynx> http://lparchive.org/Planescape-Torment/Update%2014/10-PST010-05.JPG http://www.significant-bits.com/wp-content/uploads/2011/04/Torment-2011-04-02-11-27-28-83.jpg
[21:06:08] <brada> the warning wasy it is 6px…
[21:06:30] <brada> as for dialog, i would question if that is the desired behavior
[21:06:51] <brada> i havent played pst so im not sure, but is there a problem with using the same behavior as BG2 et al
[21:07:44] <brada> presumably we can easily control that with the auto scroll flag
[21:08:01] <brada> because PST doesnt put feedback in that window right?
[21:08:09] <brada> it ahs the overhead text stuff
[21:08:57] <-- brada has left IRC (Quit: brada)
[22:34:09] <-- edheldil_ has left IRC (Ping timeout: 246 seconds)
[22:34:26] --> edheldil_ has joined #gemrb
[22:51:57] --> Textmode has joined #gemrb
[22:54:38] <-- Textmode has left IRC (Read error: Connection reset by peer)
[22:54:54] --> Textmode has joined #gemrb