#gemrb@irc.freenode.net logs for 27 Jan 2012 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:02:21] --> Drakkar has joined #gemrb
[00:04:37] <-- PixelScum has left IRC (Ping timeout: 252 seconds)
[00:04:39] --> brad_a has joined #gemrb
[01:17:03] --> wrotek_ has joined #gemrb
[01:19:36] <-- wrotek has left IRC (Ping timeout: 240 seconds)
[01:39:29] <-- edheldil_ has left IRC (Ping timeout: 252 seconds)
[02:06:52] --> PixelScum has joined #gemrb
[02:07:17] <-- Drakkar has left IRC (Ping timeout: 252 seconds)
[02:12:38] <brad_a> hazah! some sdl update made it so i dont need to hack my own run loop!
[02:17:35] --> Drakkar has joined #gemrb
[02:20:19] <-- PixelScum has left IRC (Ping timeout: 260 seconds)
[03:31:09] --> joneirik has joined #gemrb
[03:55:13] <CIA-36> GemRB: 03bradallred * ra39e8bea9860 10gemrb/apple/CocoaWrapper/ios/GEM_ConfController.m:
[03:55:13] <CIA-36> GemRB: CocoaWrapper (iOS): fix broken progress indicators on iOS 5
[03:55:13] <CIA-36> GemRB: in iOS cannot update GUI classes from threads other than main (you werent supposed to prior anyway, but now it is enforced).
[03:55:21] <CIA-36> GemRB: 03bradallred * r24bc2219c360 10gemrb/apple/CocoaWrapper/ios/ (7 files in 2 dirs):
[03:55:21] <CIA-36> GemRB: CocoaWrapper (iOS): remove runloop hack (fixes scrollview bugs)
[03:55:21] <CIA-36> GemRB: Previously SDL would exit after SDL_main returned; it now leaves the main
[03:55:21] <CIA-36> GemRB: runloop running. This factors the old main code into our new AppDelegate class
[03:55:22] <CIA-36> GemRB: and removes the old CFRunLoop hack we have been using to run the wrapper in.
[03:55:22] <CIA-36> GemRB: 03bradallred * rdcbcf09d531e 10gemrb/apple/CocoaWrapper/ios/GEM_ConfController.m: CocoaWrapper (iOS): no need to log the archive contents. this just slows the unarchiving process and bloats the log.
[03:55:23] <CIA-36> GemRB: 03bradallred * rf7efa890fbb7 10gemrb/gemrb/plugins/SDLVideo/SDLVideo.cpp: SDLVideo: apparently an update to sdl 1.3 added a new event field.
[03:55:23] <CIA-36> GemRB: 03bradallred * r948aba0db76e 10gemrb/gemrb/plugins/SDLVideo/SDLVideo.cpp: SDLVideo: use a more appropriate path for this include so we compile more universally (probably only compiles for me prior)
[04:07:13] <gembot> build #75 of osx-xcode-binary is complete: Failure [failed compile] Build details are at http://buildbot.gemrb.org/builders/osx-xcode-binary/builds/75 blamelist: bradallred@me.com
[04:08:09] <brad_a> o_O
[04:09:07] <brad_a> tomprince: once i figure out what went wrong how do i tell buildbot to recompile?
[04:09:21] <brad_a> just that specific one
[04:09:43] <tomprince> You ask me if I am around. :)
[04:11:13] <brad_a> ok
[04:11:18] <brad_a> the problem is the sdl update
[04:12:34] <tomprince> I haven't implemented auth on the webstatus, and the irc doesn't support it.
[04:15:35] --> PixelScum has joined #gemrb
[04:16:56] <-- Drakkar has left IRC (Ping timeout: 240 seconds)
[04:19:16] <brad_a> well it looks like i will need to change an include. SDL changed the header for some virtual keyboard crap. so ill have to push an update anyway
[04:23:22] <CIA-36> GemRB: 03bradallred * re83c0c96f2b9 10gemrb/apple/CocoaWrapper/ios/GEM_SDL_Keyboard.m: GEM_SDL_Keyboard (iOS): an SDL update changed the required header.
[04:37:05] --> Drakkar has joined #gemrb
[04:38:47] <-- brad_a has left IRC (Quit: brad_a)
[04:39:19] <-- PixelScum has left IRC (Ping timeout: 245 seconds)
[04:47:58] <gembot> build #76 of osx-xcode-binary is complete: Success [build successful] Build details are at http://buildbot.gemrb.org/builders/osx-xcode-binary/builds/76
[05:00:35] --> PixelScum has joined #gemrb
[05:03:02] <-- Drakkar has left IRC (Ping timeout: 252 seconds)
[05:10:06] <gembot> build #435 of mingw32 is complete: Failure [failed install minimal test] Build details are at http://buildbot.gemrb.org/builders/mingw32/builds/435 blamelist: bradallred@me.com
[05:28:45] --> Drakkar has joined #gemrb
[05:31:08] <-- PixelScum has left IRC (Ping timeout: 240 seconds)
[05:41:55] --> PixelScum has joined #gemrb
[05:43:54] <-- Drakkar has left IRC (Ping timeout: 245 seconds)
[06:06:09] --> Drakkar has joined #gemrb
[06:07:38] <-- PixelScum has left IRC (Ping timeout: 252 seconds)
[06:20:39] <-- Drakkar has left IRC (Ping timeout: 260 seconds)
[06:22:00] --> Drakkar has joined #gemrb
[06:24:13] <-- Drakkar has left IRC (Read error: Connection reset by peer)
[06:24:21] <-- joneirik has left IRC (Remote host closed the connection)
[06:26:29] --> Drakkar has joined #gemrb
[07:39:22] --> lynxlynxlynx has joined #gemrb
[07:39:22] <-- lynxlynxlynx has left IRC (Changing host)
[07:39:22] --> lynxlynxlynx has joined #gemrb
[07:39:22] --- ChanServ gives channel operator status to lynxlynxlynx
[08:28:19] --> edheldil_ has joined #gemrb
[08:36:41] <-- edheldil_ has left IRC (Ping timeout: 272 seconds)
[09:30:20] <CIA-36> GemRB: 03lynxlupodian * rdd974b5a528c 10gemrb/gemrb/GUIScripts/LevelUp.py: fixed odd typo in dualclass code
[10:19:05] --> SiENcE has joined #gemrb
[10:23:58] <gembot> build #436 of mingw32 is complete: Success [build successful] Build details are at http://buildbot.gemrb.org/builders/mingw32/builds/436
[10:55:05] <-- Kiranos_ has left IRC (Quit: No Ping reply in 180 seconds.)
[10:55:20] --> Kiranos has joined #gemrb
[12:31:28] --> Baldurer has joined #gemrb
[13:09:41] --> Ahtu has joined #gemrb
[14:00:08] <-- Baldurer has left IRC (Ping timeout: 276 seconds)
[14:00:09] <-- Ahtu has left IRC (Ping timeout: 264 seconds)
[15:08:53] --> kettuz has joined #gemrb
[16:40:12] --> Yoshimo has joined #gemrb
[16:47:40] --> Baldurer has joined #gemrb
[16:50:47] <-- Yoshimo has left IRC (Quit: Yoshimo)
[16:51:58] --> Yoshimo has joined #gemrb
[17:09:15] <-- SiENcE has left IRC (Read error: Connection reset by peer)
[17:41:35] --> edheldil_ has joined #gemrb
[17:46:25] <-- edheldil_ has left IRC (Ping timeout: 272 seconds)
[18:07:30] --> kingron has joined #gemrb
[18:27:56] <-- Kiranos has left IRC (*.net *.split)
[18:28:09] --> Kiranos has joined #gemrb
[18:34:42] <-- Yoshimo has left IRC (Quit: Yoshimo)
[18:52:29] --> SiENcE has joined #gemrb
[18:56:11] <-- kettuz has left IRC (Quit: Leaving)
[18:58:50] --> brad_a has joined #gemrb
[19:24:26] --> brad_a_ has joined #gemrb
[19:27:10] <-- brad_a has left IRC (Ping timeout: 244 seconds)
[19:27:10] --- brad_a_ is now known as brad_a
[19:30:49] --> Yoshimo has joined #gemrb
[19:44:43] <-- wrotek_ has left IRC (Ping timeout: 252 seconds)
[19:54:28] --> edheldil_ has joined #gemrb
[20:25:34] <-- kingron has left IRC (Remote host closed the connection)
[21:15:28] --> kettuz has joined #gemrb
[21:19:35] <-- kettuz has left IRC (Client Quit)
[21:21:01] <-- Yoshimo has left IRC (Quit: Yoshimo)
[21:39:38] <CIA-36> GemRB: 03tom.prince * rbfbcbfd0204a 10gemrb/gemrb/core/Scriptable/Scriptable.cpp:
[21:39:38] <CIA-36> GemRB: Scriptable: Handle case where GlobalID overflows.
[21:39:38] <CIA-36> GemRB: This should never happen, 640k should be enough for everyone.
[22:23:36] <lynxlynxlynx> http://forums.gibberlings3.net/index.php?showtopic=23916&st=0
[22:23:55] <lynxlynxlynx> as if they overrid the newline char
[22:24:07] <lynxlynxlynx> it works in guirec, since there we add them manually
[22:27:43] <edheldil_> to me it looks like the text was anchored left instead of center
[22:31:54] <edheldil_> and the NL thing too
[22:33:33] <edheldil_> maybe some botched conversion? or the strings contain cr instead of cr-lf
[22:36:28] <CIA-36> GemRB: 03avenger_teambg * rfc84f0361c6d 10gemrb/gemrb/ (override/iwd2/effects.ids plugins/IWDOpcodes/IWDOpcodes.cpp): new opcodes to simulate feats
[22:36:28] <CIA-36> GemRB: 03avenger_teambg * r9de88d979fcd 10gemrb/gemrb/override/iwd2/ (28 files): spell files for a few feats
[22:36:29] <CIA-36> GemRB: 03avenger_teambg * r20f28192b996 10gemrb/ (11 files in 5 dirs): Merge branch 'master' of ssh://gemrb.git.sourceforge.net/gitroot/gemrb/gemrb
[22:38:03] <wjp> hm, I still wonder why many button texts are drawn very misaligned here
[22:39:12] <edheldil_> maybe the tinkered with Font()? Sabotaging alignment and multiline texts at once
[22:39:16] <CIA-36> GemRB: 03avenger_teambg * r5892076224ca 10gemrb/gemrb/plugins/IWDOpcodes/IWDOpcodes.cpp: fixed a small problem with previous commit
[22:40:58] <wjp> let's see if bisect and/or valgrind get me anywhere
[22:53:16] <lynxlynxlynx> yeah, need to check that
[22:53:41] <lynxlynxlynx> only the gog version seems to be problematic, but i doubt they'd touch the chus
[22:54:10] <wjp> I get it everywhere
[22:54:28] <wjp> and valgrind is going crazy when printing text
[22:54:36] <-- SiENcE has left IRC (Quit: cya)
[22:54:43] <wjp> still tracking it down
[22:55:14] <wjp> (by everywhere I mean not just the gog version)
[22:57:03] <wjp> aha, it thinks the space character is 47 pixels high
[22:57:19] <wjp> (where 47 is an uninitialized value)
[22:58:17] <brad_a> i thought we fixed that...
[22:59:50] <wjp> when?
[22:59:57] <brad_a> also i dont think the height of the character or line should cause horizontal issues…
[23:00:06] <brad_a> right before merging the TTF code
[23:00:56] <wjp> there are no horizontal issues
[23:01:04] <wjp> (at least not in what I'm seeing)
[23:01:43] <wjp> Font::whiteSpace isn't initialized it seems?
[23:01:43] <brad_a> i still thought we fixed that :-P
[23:02:34] <brad_a> it looks like its initialized on line 63 to me...
[23:02:48] <brad_a> very end of the initializer list
[23:03:08] <wjp> I see that
[23:03:41] <wjp> but that doesn't change my remark :-)
[23:03:47] <brad_a> it could be a compiler thing for you
[23:04:08] <brad_a> but poster is using ipad version and i know that gets initialized to 0
[23:04:20] <wjp> poster?
[23:04:24] <brad_a> the russian guy
[23:04:32] <wjp> I'm missing something
[23:04:52] <brad_a> oh i thought you were talking about the link lynx posted :-P
[23:05:06] <brad_a> russian who is also having text issues
[23:05:10] <wjp> ah
[23:05:17] <brad_a> just diffrent ones apparently ;-)
[23:05:21] * wjp clicks
[23:05:41] <brad_a> what compiler are you using?
[23:05:57] <wjp> some old g++
[23:06:00] <wjp> 4.3.4 it seems
[23:06:41] <brad_a> tomprince: input?
[23:07:44] <brad_a> we should probably just change the way we initialize whitespace
[23:07:47] <fuzzie> so whitespace.yPos is bad?
[23:08:33] <wjp> assert(whiteSpace[0].yPos == 0);
[23:08:38] <wjp> in the Font constructor is triggered
[23:08:43] <fuzzie> or whiteSpace[1].size only?
[23:08:46] <wjp> so something later?
[23:09:16] <brad_a> i remember this issue from working on ttf when the problem was whitespace wasnt initialized to 0
[23:09:17] <fuzzie> it is not after any suspicious-looking structures or anything..
[23:09:24] * wjp will look some further
[23:09:38] <brad_a> and we allegedly fixed that by adding whitespace() to the init list
[23:09:54] <brad_a> but it appears your compiler is treating that diffrently
[23:10:06] <wjp> um, wait
[23:10:10] <brad_a> so the solution is to probably initialize it diffrently
[23:10:17] <wjp> the assert _is_ triggered. Ugh. I need sleep
[23:10:54] <fuzzie> go sleep!
[23:11:01] <brad_a> yes :)
[23:11:12] <wjp> anyway, an array isn't POD or scalar, so I see no reason why it would be zeroed by this initializer entry
[23:11:56] <brad_a> isnt it a pointer array?
[23:12:21] <fuzzie> an array should be POD
[23:12:27] <brad_a> thats what i thought
[23:12:59] <brad_a> clearly something disagreees :-P
[23:13:05] <wjp> whee, C++ standard hunting, I suppos :-)
[23:13:26] <fuzzie> 3.9, "arrays of such types"
[23:13:55] <fuzzie> you really think it's not initialised by compiler?
[23:14:13] <wjp> I'm not sure if I'm really thinking at all at this point, to be honest
[23:14:20] <brad_a> well its an array of GlyphInfo and not GlyphInfo pointers like i thought
[23:14:31] <fuzzie> "if T is an array type, then each element is value-initialized"
[23:14:38] <brad_a> but from what i recall when we had this problem before that was the cause
[23:14:42] <fuzzie> wjp: am fairly sure this can wait :)
[23:14:45] <brad_a> ive been know to forget quickly tho
[23:14:59] <wjp> let me just try a more modern gcc
[23:15:23] <fuzzie> unfortunately I only have 4.6 here
[23:15:45] <brad_a> well there isnt a reason we cant just initialize it differently that works for everybody
[23:15:56] <fuzzie> if that is the problem
[23:16:14] <brad_a> well i can test if its a cause by removing it real quick :-P
[23:16:27] <wjp> it works in 4.6.2
[23:16:34] <brad_a> heh
[23:16:53] <fuzzie> haha.
[23:16:59] <fuzzie> yes, ok, broken in 4.3
[23:17:14] <fuzzie> 4.4 release notes: "G++ now properly implements value-initialization"
[23:17:20] <brad_a> nice!
[23:17:40] <brad_a> thought i was breaking standards :-P
[23:18:01] <wjp> fuzzie: aha
[23:18:19] <wjp> don't have any 4.4.x here, but it does indeed work in 4.5.2 too
[23:18:39] <fuzzie> i would like to highly recommend using 4.6+ for all your non-broken-compilation needs though
[23:18:48] <brad_a> yes
[23:18:56] <brad_a> or clang ;-)
[23:19:03] <fuzzie> i would not highly recommend clang for that :-p
[23:19:04] <wjp> never had a particularly pressing reason to upgrade, but I suppose this counts :-)
[23:19:28] <brad_a> wel clang 3 is stable! 2.x used to crash a bunch tho
[23:19:29] <fuzzie> but of course the bugs in both compilers are enough to make me cry for days and days
[23:20:12] <brad_a> i mostly care about the better errors from clang i dont really care after that :-P
[23:20:17] <fuzzie> yeah
[23:20:40] <fuzzie> was saying earlier, clang is wonderful as long as you don't let it generate code
[23:20:47] <fuzzie> (as opposed to gcc which is never wonderful)
[23:21:30] <brad_a> :-P
[23:23:39] * wjp will now go and fall asleep in an actual bed
[23:23:41] <wjp> good night
[23:23:46] <fuzzie> go!
[23:23:49] <fuzzie> ninight.
[23:42:42] <-- lynxlynxlynx has left IRC (Remote host closed the connection)