#gemrb@irc.freenode.net logs for 19 Aug 2014 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[01:15:36] <-- PKodon has left IRC (Quit: The Rodent Tracker 8000, just like on TV ... "Because household pests never build up an immunity to bullets." (Tex Murphy))
[02:29:20] --> kpedersen has joined #gemrb
[02:32:34] <-- kpederse1 has left IRC (Ping timeout: 240 seconds)
[02:37:26] --> PKodon has joined #gemrb
[03:13:24] --> Eli2 has joined #gemrb
[03:15:52] <-- Eli2_ has left IRC (Ping timeout: 260 seconds)
[04:36:34] <-- turtleman_ has left IRC (Ping timeout: 264 seconds)
[06:12:54] <-- PKodon has left IRC (Read error: Connection reset by peer)
[06:13:10] --> PKodon has joined #gemrb
[06:51:54] <-- Lightkey has left IRC (Ping timeout: 250 seconds)
[07:00:01] <-- PKodon has left IRC (Quit: The Rodent Tracker 8000, just like on TV ... "Because household pests never build up an immunity to bullets." (Tex Murphy))
[07:04:53] --> Lightkey has joined #gemrb
[07:15:42] --> raevol has joined #gemrb
[07:22:41] --> PKodon has joined #gemrb
[07:23:06] <-- PKodon has left IRC (Client Quit)
[08:20:02] --> PKodon has joined #gemrb
[10:01:40] <-- raevol has left IRC (Quit: Leaving)
[13:45:03] --> brada has joined #gemrb
[13:45:12] <brada> anybody have a better fix: https://github.com/bradallred/gemrb/commit/eaed597abfa98cf9b4cc89cc2daed99ef751295e
[13:45:14] <Pepelka13> Tooltips: fix (sort of) palette regression · eaed597 · bradallred/gemrb · GitHub
[13:45:15] <Pepelka13> »gemrb - Game Engine Made with preRendered Background«
[13:59:33] <lynxlynxlynx> i forgot what gcc complained about, but my glibc supports hh
[14:01:12] <brada> yeah thats why i figured thats not a good fix, but the change to “c” was obviously wrong
[14:01:50] <brada> i mean i could read them in as “c” then subtract whatever to convert...
[14:02:12] <brada> er no
[14:02:25] <brada> cuz that would only read 4 of the 8 chars…
[14:03:25] <-- brada has left IRC (Quit: brada)
[14:04:44] <fuzzie> um
[14:05:07] <fuzzie> so confused
[14:05:40] <fuzzie> %02x really doesn't work on mac?
[14:07:20] <fuzzie> hm
[14:07:31] <fuzzie> so the 0 specified doesn't make much sense for sscanf
[14:07:34] <fuzzie> specifier
[14:07:49] <fuzzie> but you'd expect just 0x%2x%2x%2x%2x to be ok
[14:10:21] <fuzzie> and for %02x to be equivalent
[14:11:39] <fuzzie> oh I see
[14:11:58] <fuzzie> the problem is that %02x is *horribly wrong*
[14:12:50] <fuzzie> should just change the code to use intermediate ints there, then
[14:14:16] <wjp> or just scan the whole thing into a uint32?
[14:14:57] --> turtleman_ has joined #gemrb
[14:16:03] <fuzzie> an even better idea
[14:20:06] --> brada has joined #gemrb
[14:21:16] <brada> so there really isnt a way to scan bytes in hex format without “hh”?
[14:21:40] <brada> i guess thats all i needed to know :p
[14:26:01] <fuzzie> msvc doesn't do it
[14:26:05] <fuzzie> so no
[14:30:39] <brada> what 32 bit type do we have to use everywhere?
[14:30:54] <brada> iedword i guess :p
[14:35:17] <wjp> it may be safer to use unsigned long
[14:35:27] <wjp> then you can just use the unsigned long size specifier for scanf as well
[14:36:00] <wjp> specific size types and scanf/printf don't mix well
[15:08:02] <lynxlynxlynx> as a general rule, we use iedword and co when dealing with the original file formats
[15:13:03] <brada> let me guess; thats an unwritten rule :D
[15:18:13] <-- |Cable| has left IRC (Remote host closed the connection)
[15:19:04] <brada> so I only have one more blocker before i feel ready to merge. tho some text may still be missing newlines… and i havent tested in any other games
[15:20:00] <brada> unfortunately ill probably be too busy to fix that for a while...
[15:20:27] --> |Cable| has joined #gemrb
[15:21:37] <brada> oh, but then there are still issues with opengl :/
[15:21:45] <brada> like the text not being clipped...
[15:22:19] <brada> the performace stuff im not too worried about since it still seems to perform better that non-opengl
[15:23:10] --> Cable_ has joined #gemrb
[15:26:40] <-- |Cable| has left IRC (Ping timeout: 260 seconds)
[15:37:24] <lynxlynxlynx> i'll be here for another two weeks, so just let me know when you need more extensive testing
[15:45:03] <lynxlynxlynx> 353eb03661c43dc219be16883a848ac3174b875b has a suspicious bit
[15:45:21] <lynxlynxlynx> - text = CommonTables.Classes.GetValue (text, "DESC_REF")
[15:45:21] <lynxlynxlynx> + text += CommonTables.Classes.GetValue (text, "DESC_REF")
[15:53:06] <brada> sure,it was suspicious before too
[15:53:18] <brada> assign to text then next line reassign something diffrent
[15:53:28] <brada> not sure what the intent was there
[15:55:01] <lynxlynxlynx> just variable reuse
[15:55:27] <lynxlynxlynx> first text is the column name and then we fetch the strref with it
[15:56:01] <brada> it wasnt reuse
[15:56:10] <lynxlynxlynx> but you also changed the indentation for this and the next block, so the results may differ further
[15:56:41] --> YamatoHD has joined #gemrb
[15:56:47] <YamatoHD> GUYS!
[15:56:53] <brada> look at old lines 895 and 896
[15:56:57] <YamatoHD> Despite all the struggling
[15:57:07] <brada> also 890 and 89
[15:57:10] <brada> 891
[15:57:11] <YamatoHD> i ALLMOST beat Planescape
[15:57:23] <YamatoHD> on GEMRB
[15:57:29] <lynxlynxlynx> can't count, but going off the diff http://paste.debian.net/116391/
[15:57:30] <Pepelka13> debian Pastezone
[15:57:33] <brada> oh? now that is impressive!
[15:57:33] <lynxlynxlynx> text = ... text
[15:57:46] <lynxlynxlynx> and fast too
[15:57:48] <YamatoHD> i won't lie it was really hard
[15:57:57] <YamatoHD> im at baator now
[15:58:43] <YamatoHD> to be honest, just a few minor problems are yet to be fixed
[15:58:54] <YamatoHD> the biggest one i already wrote about
[15:59:21] <YamatoHD> and how are you doing?
[16:00:03] <lynxlynxlynx> you didn't have problems with broken scripts?
[16:00:38] <YamatoHD> me?
[16:00:44] <lynxlynxlynx> ye
[16:00:56] <YamatoHD> well i remember some quest didn't trigger
[16:01:02] <YamatoHD> i skipped them
[16:01:06] <lynxlynxlynx> no crashes?
[16:01:12] <lynxlynxlynx> tymora smiles on you
[16:01:16] <YamatoHD> a TON of crashes
[16:01:28] <YamatoHD> yet im almost done
[16:01:50] <lynxlynxlynx> are you the nameless one?
[16:02:08] <lynxlynxlynx> remarkable persistence
[16:02:18] <YamatoHD> no, im Butthurting one -_-
[16:03:02] <YamatoHD> well game is increddible
[16:03:21] <brada> we have some animation crashes that i thought would prevent from finishing
[16:03:22] <YamatoHD> i am at home too little to play it on PC
[16:03:29] <YamatoHD> and ipad is always with me
[16:03:42] <brada> i guess you can cheat to bypass things sometimes
[16:04:18] <YamatoHD> oh i had a permacrash on robot
[16:04:23] <YamatoHD> in robotic maze
[16:04:31] <YamatoHD> he was impossible to speak to
[16:04:45] <YamatoHD> Nordom
[16:05:04] <lynxlynxlynx> do you have any saves around such problems?
[16:05:42] <YamatoHD> i was overwriting all the time
[16:06:01] <YamatoHD> well i can make some on second run if you need them
[16:06:13] <YamatoHD> oh wait
[16:06:19] <brada> :/ saves would surely be useful for us
[16:06:19] <YamatoHD> that chat problem...
[16:06:27] <brada> i think i have one for an animation crash
[16:06:36] <brada> and most likely that one crash would fix the rest
[16:06:46] <YamatoHD> that chat bug was the most frustrating thing in a whole game...
[16:06:54] <brada> oh that is fixed
[16:06:59] <brada> it wasnt really a bug tho
[16:07:02] <YamatoHD> REALLY?
[16:07:15] <brada> jsut unimplemented feature i guess
[16:07:27] <YamatoHD> nice to hear
[16:07:36] <brada> if its what i think you are talking about
[16:07:48] <YamatoHD> that was like 25% of game and i almost dropped playing it due to it
[16:08:14] <YamatoHD> http://forums.gibberlings3.net/index.php?showtopic=26544
[16:08:16] <Pepelka13> Planescape torment chat/log problem - GemRB - The Gibberlings Three Forums
[16:08:17] <Pepelka13> »Planescape torment chat/log problem - posted in GemRB: HelloSo basicly i have this problem when having a dialogue in a crowded place like market it gets spammed and autoscrolled bottomSomehow all the tips above NPC heads are being typed in log which makes it unreadable and it is barely possible to click any answer Is there a workaround like disabling autoscroll or getting rid of whatsoever?game itself is perfctly playable aside from this «
[16:08:29] <YamatoHD> yeye
[16:08:34] <YamatoHD> thats the one
[16:09:03] <YamatoHD> oh
[16:09:12] <brada> well i havent tested in PST, but it is either fixed or very simple to fix now
[16:09:20] <YamatoHD> and one more thing i would really like
[16:09:28] <YamatoHD> minimap has flags
[16:09:39] <YamatoHD> but its impossible to read them
[16:09:59] <YamatoHD> on pc you hover mouse over them to see what is it
[16:10:11] <YamatoHD> on ipad there is no way to hover
[16:10:36] <YamatoHD> *map, not minimap
[16:14:14] <YamatoHD> by the way, is log fixed in 0.8.2?
[16:15:48] <brada> you should be able to hover in theory on ipad
[16:15:55] <brada> just touch where you want to hover
[16:16:20] <brada> its fixed in my branch which is unmerged
[16:17:58] <YamatoHD> i tried
[16:18:28] <YamatoHD> oh and a funny bug with pillar of sculls
[16:18:31] <brada> oh lynx i see what you are saying now. it *is* variable reuse, jsut mighty confusing :/
[16:18:37] <brada> so yeah, ill fix that
[16:18:40] <YamatoHD> em you guys dont mind spoilers?
[16:18:59] <brada> im sure lynx has played pst already
[16:19:07] <brada> and i probably wont ever find time :(
[16:19:34] <YamatoHD> well pillar answers questions but you have to pay for them
[16:20:00] <YamatoHD> pillar says that it will need a drop of blood for answer
[16:20:13] <YamatoHD> but it takes anna instead
[16:20:24] <YamatoHD> or says that she wants morte
[16:20:31] <YamatoHD> takes a drop of blood
[16:21:03] <YamatoHD> not a biggie, just funny
[16:21:11] <YamatoHD> like russian roulette
[16:22:03] <brada> I wouldnt know how to fix things like that :)
[16:22:08] <brada> lynx or fuzzie would
[16:23:03] <YamatoHD> to be honest id leave this one as is
[16:23:14] <YamatoHD> after all pillar is chaotic evil
[16:23:33] <lynxlynxlynx> heh
[16:23:39] <brada> you are sure that that isnt how the original is?
[16:23:43] <lynxlynxlynx> i don't see chiv making any notes about it
[16:23:54] <brada> or is there a PST fixpack?
[16:23:54] <lynxlynxlynx> do you remember which area it is in?
[16:25:37] <YamatoHD> there is fix pack, i didnt use it tho
[16:25:45] <YamatoHD> i play vanilla 1.1
[16:26:21] <YamatoHD> i wanted to use high res patch over it, but it made UI glitch
[16:28:49] <YamatoHD> is the next version any time soon?
[16:28:59] <YamatoHD> im kinda ready for a second run
[16:29:17] <YamatoHD> i can provide sav games or something if needed
[16:32:00] <lynxlynxlynx> wait, you want to do a second run already?
[16:32:22] <lynxlynxlynx> where did you get stuck for not being able to finish the first one?
[16:32:32] <lynxlynxlynx> and since it's the last thing, you must have a save
[16:32:45] <YamatoHD> im not stuck
[16:32:50] <YamatoHD> im at baator
[16:33:03] <YamatoHD> thats like 85% of the game
[16:34:37] <brada> as for the “UI glitch” that is most likely due to invalid cfg setting
[16:35:21] <brada> also i think maybe a way to increase utility on ipad/touchscreen is to use larger labels on the map controls so “hovering” is easier
[16:35:43] <brada> i added newlines between dialog options too :p
[16:35:57] <brada> maybe we want to make that configurable, but i think its nicer with a mouse too
[16:37:26] <lynxlynxlynx> if it looks different, people will complain
[16:37:31] <lynxlynxlynx> that's a given :s
[16:37:44] <brada> well i dont have to listen! ;)
[16:38:10] <brada> it will likely always look diffrent. if nothing else then because diffrent games are diffrent, no?
[16:38:30] <brada> im too lazy to add the little extra color to dialog options (useless too IMO)
[16:40:38] <YamatoHD> there is a Videos button in menus in PST
[16:41:04] <YamatoHD> was it intended that all cut scenes are open from start?
[16:41:12] <YamatoHD> there are massive spoilers
[16:47:17] <lynxlynxlynx> no, it's just missing functionality
[16:47:36] <lynxlynxlynx> in bg2 the menu item was converted in a jukebox :>
[16:59:48] <brada> lynx: if you could just do basic testing with other games I think that would be good. maybe do a quick valgrind wouldnt be a bad idea too. if you want to make notes of what line endings are missing anywhere that would be nice too.
[17:00:54] <brada> the big known bug is with dialog being incorecctly scrolled when the text area is simultaniously deleting lines from the top just FYI if you encounter that
[17:12:27] <lynxlynxlynx> other games make little sense while there are still bugs in bg2
[17:12:59] <lynxlynxlynx> most of the gui code is shared, but yeah, I expect some problems
[17:13:10] <brada> ug… ok. there is just that one really
[17:16:13] <lynxlynxlynx> oh, did you check the journal? it was missing newlines the last time i did
[17:16:31] <lynxlynxlynx> hopefully all the caps issues are gone with your latest work
[17:16:55] <brada> ah i knew i was forgetting something
[17:17:08] <brada> what other caps issues were there?
[17:17:18] <brada> chargen/guirec are the only things i fixed
[17:18:00] <brada> tho the chargen fix would likely apply to other areas
[17:18:02] <lynxlynxlynx> there were several in guirec, but i think you encapsulated all the text in a paragraph
[17:18:31] <brada> yeah that should be fixed then
[17:18:54] <brada> there are possibly some wrapping issues with caps tho, since caps are considered a separate span from the rest of their word…
[17:19:05] <brada> we could wrap them in a container if we need to
[17:19:27] <brada> but really as long as teh data is sane it shouldnt manifest
[17:24:56] <lynxlynxlynx> [Python/ERROR]: File "../gemrb/GUIScripts/GUIREC.py", line 898, in OpenKitInfoWindow
[17:24:56] <lynxlynxlynx> [Python/ERROR]: TextArea.SetText ("[p]" + text + "[/p]")
[17:24:56] <lynxlynxlynx> [Python/ERROR]: TypeError: cannot concatenate 'str' and 'int' objects
[17:25:19] <lynxlynxlynx> same dodgy commit
[17:26:26] <lynxlynxlynx> + caps issues are still there in guirec
[17:26:52] <lynxlynxlynx> same first letter problem for a DEAD pc and a Conjurer
[17:27:46] <lynxlynxlynx> and the scrollbar doesn't reset on portrait change, which i'm pretty sure is new
[17:28:37] <lynxlynxlynx> tooltip font confirmed fixed
[17:28:52] <brada> is that with the tip? text shouldnt be messed up anymore
[17:30:16] <brada> i guess i didnt check a dead PC. im guessing a dead PC has text above the class?
[17:31:51] <brada> oh barf, ok i see the guirec thing is still messed up
[17:33:00] <lynxlynxlynx> Font.cpp:310: size_t GemRB::Font::RenderText(const String&, GemRB::Region&, GemRB::Palette*, GemRB::ieByte, GemRB::Point*, GemRB::ieByte**, bool) const: Assertion `charCount <= string.length()' failed.
[17:33:12] <lynxlynxlynx> ^ trying to guirec a dualclassed char
[17:33:49] <lynxlynxlynx> journal problem is still there
[17:34:30] <lynxlynxlynx> the new console font is better, but worse than the one we had first
[17:39:50] <brada> not sure its worth it to make a new ini setting for console font…
[17:41:30] <lynxlynxlynx> now it's hardcoded?
[17:41:33] <brada> no
[17:41:38] <brada> not exactly anywya
[17:42:03] <brada> just uses whatever the “textfont” is
[17:42:32] <brada> and that assert surely has nothing to do with dual class
[17:43:06] <lynxlynxlynx> NORMAL?
[17:43:30] <brada> couldnt be
[17:43:44] <brada> normal is what the text area uses
[17:43:50] <brada> doesnt look the same to me
[17:44:04] <brada> that was probably what it was using before tho
[17:44:15] <brada> iirc it just used the first font loaded for console
[17:45:12] <lynxlynxlynx> charCount flips over into gargantuan teritorry
[17:45:24] <brada> underflow for sure
[17:45:49] <brada> i did jsut recently split that gargantuan method into two for nicer code
[17:45:57] <brada> so i probably caused a regression
[17:46:27] <lynxlynxlynx> it's breaking on printing a space
[17:48:54] <brada> you mean for that underflow thing?
[17:50:02] <lynxlynxlynx> yes
[17:50:19] <lynxlynxlynx> well, at least at that point what remains of the string is a space
[17:51:31] <brada> im sure its an easy fix
[17:51:34] <brada> no time atm
[17:53:38] <brada> seems like it would have to be the linePos-- line tho. i dont see anywhere else that decrements to count
[17:53:56] <brada> probably should have a similar assert in RenderLine
[17:56:04] <brada> on the plus side its easy to reproduce. jsut open console and type a space...
[17:59:04] <brada> so yeah, messing up on any text taht begins with a space
[17:59:08] <brada> and im sure i know why
[18:32:23] --> Yoshimo has joined #gemrb
[18:32:23] <-- Yoshimo has left IRC (Changing host)
[18:32:23] --> Yoshimo has joined #gemrb
[18:48:58] <-- Yoshimo has left IRC (Quit: Yoshimo)
[20:06:21] --> edheldil_ has joined #gemrb
[21:41:11] <-- brada has left IRC (Quit: brada)
[21:55:46] <-- edheldil_ has left IRC (Ping timeout: 264 seconds)