[00:23] blondebeard (~derp@195.204.123.225) joined #scummvm. [00:25] does anyone in here have a scummvm compatible savegame from curse of monkey island right before or after you have to insert the second cd? [00:25] where would that be around [00:26] i have a dozen of savegames.... tell me where in the story [00:28] guybrush in the big chicken container on the sea cucumber ship.. [00:33] uuh after the sea battles that is... [00:33] alright will check for [00:35] its before the battle part, other scenes are... uhm.. guybrush as el pollo diablo.. brimstome beach with the sunbather [00:37] i think i mostly finished my savedir path check [00:37] http://www.abload.de/img/savewarner0dzq.png [00:40] still not sure if its of quality enough [00:41] blondebeard> im checking my save dir ... [00:42] thanks starg [00:45] http://rapidshare.com/files/346024914/comi-de.s05.html [00:45] i hope i did not miscalculate [00:45] you might have to rename it to comi-en or the like [00:49] you might be a bit closer if you take this save: http://rapidshare.com/files/346025810/comi-de.s06.html [00:49] thanks alot! :) that'll work.. as long as i dont have to start from the beginning.. [00:50] thanks again.. you're a lifesaver [00:50] the numbering is a bit of a hassle... inside savegame dialog nr0 is autosave, in filesystem nr0 is prolly the first savegame [00:50] np :) [00:50] well, then im off to play! adios [00:51] blondebeard (~derp@195.204.123.225) left irc: Quit: I can't use the skeleton arm with that. [00:58] Javacat (~ghsgh@unaffiliated/javacat) left irc: Ping timeout: 260 seconds [01:02] Javacat (~ghsgh@unaffiliated/javacat) joined #scummvm. [01:11] anyone for discussing the savegame directory stuff? [01:14] howardazathoth (~Administr@c-98-215-207-17.hsd1.il.comcast.net) joined #scummvm. [01:14] howardazathoth (Administr@c-98-215-207-17.hsd1.il.comcast.net) left #scummvm. [01:15] ill be off for today [01:40] lisppaste4 (~lisppaste@common-lisp.net) left irc: Quit: Want lisppaste4 in your channel? Email lisppaste-requests AT common-lisp.net. [01:44] Ceriand|work (~Ceriand@unaffiliated/ceriand) left irc: Quit: Leaving. [01:48] lisppaste4 (~lisppaste@common-lisp.net) joined #scummvm. [01:51] Robin_Watts (~Javacat@82.152.255.89) left irc: Quit: Pop! [02:06] Nevyn5 (jhgjfgl@81-236-19-16-no39.tbcn.telia.com) left irc: [02:12] Nevyn3 (jhgjfgl@81-236-19-16-no39.tbcn.telia.com) joined #scummvm. [02:28] jvprat (~jvprat@unaffiliated/jvprat) left irc: Quit: Leaving. [02:41] Javacat (~ghsgh@unaffiliated/javacat) left irc: Quit: ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ [03:12] raorn (~raorn@altlinux/developer/raorn) left irc: Read error: Connection reset by peer [03:16] raorn (~raorn@altlinux/developer/raorn) joined #scummvm. [03:28] Adys (~Adys@unaffiliated/adys) left irc: Ping timeout: 264 seconds [03:38] Kirben (kirben@c58-107-187-179.brasd3.vic.optusnet.com.au) left irc: Ping timeout: 240 seconds [03:39] Adys (~Adys@unaffiliated/adys) joined #scummvm. [03:43] D0SFreak (~D0SFreak@c-24-99-111-221.hsd1.ga.comcast.net) left irc: Quit: Leaving [03:58] mthreepwood * r47894 /scummvm/trunk/ (5 files in 4 dirs): [03:58] Have SCI look through the patches directory (except in KQ6 where the patches in [03:58] that directory are broken) to fix some GK1 Windows problems. Add the MSRLE codec [03:58] to AVI (from DrMcCoy's Dark Seed II engine, with permission). GK1 Windows now [03:58] will play the credits video. [04:03] ScummBot (~ScummBot@beta.switchlink.se) left irc: Read error: Operation timed out [04:10] qvist (qvist@switchlink.se) left irc: Ping timeout: 256 seconds [04:10] qvist (qvist@switchlink.se) joined #scummvm. [04:10] #scummvm: mode change '+v qvist' by ChanServ!ChanServ@services. [04:39] Nevyn3 (jhgjfgl@81-236-19-16-no39.tbcn.telia.com) left irc: [04:45] mthreepwood * r47895 /scummvm/trunk/engines/sci/graphics/screen.cpp: SCI2 Windows games are scaled from 320x200 to 640x480, not 640x400. Aspect ratio correction will eventually have to be applied. This fixes a possible segfault when playing the credits video in GK1 (which is 640x480). [04:57] mthreepwood * r47896 /scummvm/trunk/graphics/video/avi_decoder.cpp: Silence a valgrind warning. The sound buffer should be created with malloc and not new[]. [04:58] eriktorbjorn (~d91tan@81-233-98-9-no108.tbcn.telia.com) left irc: Remote host closed the connection [05:27] waltervn (~walter@82-171-47-198.ip.telfort.nl) left irc: Quit: night! [05:40] |Pixel| (pixel@nobis-crew.org) left irc: Ping timeout: 256 seconds [05:50] |Pixel| (pixel@nobis-crew.org) joined #scummvm. [06:00] eriktorbjorn (~d91tan@81-233-98-9-no108.tbcn.telia.com) joined #scummvm. [06:00] #scummvm: mode change '+o eriktorbjorn' by ChanServ!ChanServ@services. [06:40] Strangerke (~a@cable-85.28.89.119.coditel.net) left irc: Quit: Bbl [06:49] clone2727 (~clone2727@nat10-resnet-ext.rutgers.edu) left irc: Quit: later [07:27] _sev|work_ (~sev@scummvm/undead/sev) joined #scummvm. [07:27] #scummvm: mode change '+o _sev|work_' by ChanServ!ChanServ@services. [07:30] _sev|work__ (~sev@kharkov.intetics.com) left irc: Ping timeout: 240 seconds [07:36] eriktorbjorn * r47897 /scummvm/trunk/engines/sci/sci.cpp: Use strcmp() to test the game ID, not the != operator. [08:03] syke (~matt@76-191-154-20.dsl.dynamic.sonic.net) joined #scummvm. [08:03] #scummvm: mode change '+o syke' by ChanServ!ChanServ@services. [08:18] syke (~matt@76-191-154-20.dsl.dynamic.sonic.net) left irc: Quit: Leaving. [08:27] johndoe123 (~john_doe@p5DE7B313.dip.t-dialin.net) joined #scummvm. [08:27] #scummvm: mode change '+o johndoe123' by ChanServ!ChanServ@services. [08:32] |Cable| (~cable@237.130.117.91.dynamic.mundo-r.com) left irc: Remote host closed the connection [08:46] Sho_ (~EHS1@kde/hein) joined #scummvm. [08:48] LordHoto (~loom@s0921.vpn.hrz.tu-darmstadt.de) joined #scummvm. [08:48] #scummvm: mode change '+o LordHoto' by ChanServ!ChanServ@services. [08:49] Echidna (~a@77-20-58-117-dynip.superkabel.de) left irc: Ping timeout: 265 seconds [08:49] hi [08:52] Echidna (~a@77-20-58-117-dynip.superkabel.de) joined #scummvm. [08:58] Lars2 (~lh@0x573f1066.cpe.ge-1-1-0-1101.hjnqu1.customer.tele.dk) joined #scummvm. [09:18] m_kiewitz (~m_kiewitz@kons-5f7104a2.pool.mediaWays.net) joined #scummvm. [09:18] #scummvm: mode change '+o m_kiewitz' by ChanServ!ChanServ@services. [09:23] [md5] (~d4fb49fe@gateway/web/freenode/x-jyejdeiyojoetimc) joined #scummvm. [09:23] #scummvm: mode change '+o [md5]' by ChanServ!ChanServ@services. [09:23] <[md5]> hello [09:24] hi [md5] [09:24] <[md5]> found out what the broken "patches" folder is inside some versions of KQ6CD... [09:25] <[md5]> it's for a demo included in the disk :P [09:25] haha [09:25] <[md5]> so it seems that the interpreter of KQ6CD was supposed to ignore that folder.... odd [09:25] <[md5]> anyway [09:25] Anyone know why in a template function, I can instantiate a Common::HashMap, but not a Common::HashMap::iterator? [09:26] DrMcCoy: what exactly is the error your compiler gives you? [09:26] LordHoto: error: expected ; before it [09:27] ah it's a function which takes template arguments? [09:27] you might need to use typename in front of the iterator declaration then [09:27] i.e. "typename Common::HashMap::iterator it;" [09:28] LordHoto: Yeah, that helped [09:28] LordHoto: Thanks [09:28] you're welcome [09:28] <[md5]> hm [09:28] <[md5]> may I ask what that typename does? [09:28] Action: [md5] googles it up... [09:28] LordHoto: Hmm, I remember reading about that years ago and why that's necessary :P. But I forgot again :P [09:28] actually you can scrap that "might" in my sentence, you always need that in template functions ;-P [09:30] [md5]: http://pages.cs.wisc.edu/~driscoll/typename.html [09:30] [md5]: our your copy of the "The C++ Programming Language" :-) [09:30] or* [09:31] thebluegr * r47898 /scummvm/trunk/engines/sci/sci.cpp: Added information on the patches folder included in KQ6CD [09:33] LordHoto: Ah, good to know. I bet I have forgotten it again in 2 weeks :P [09:37] that article I linked seems *ok* [09:38] at least it should be good enough to remember that you need it in certain cases ;-) [09:39] Hkz (~noname@host46-141-dynamic.104-80-r.retail.telecomitalia.it) joined #scummvm. [09:39] #scummvm: mode change '+o Hkz' by ChanServ!ChanServ@services. [09:39] hi [09:39] Hkz: Hej hej [09:39] LordHoto: Or not, if you've got a memory like a sieve. Like me :P [09:39] DrMcCoy \o/ [09:40] DrMcCoy, i bought darkseed II, i'll start buggering you :P [09:40] DrMcCoy: you can still bookmark it then ;-P [09:40] Hkz: \o/ [09:40] LordHoto: Meeeeh. Too lazy :P [09:40] haha [09:42] <[md5]> LordHoto: yeah, read that page, makes things a bit clerer [09:42] <[md5]> clearer * [09:42] well you need it for dependent unqualified types ;-P [09:42] <[md5]> DrMcCoy: I have Dark Seed II, I can start buggering you too! :D [09:42] Action: [md5] grins [09:42] Yeah, everyone do that :) [09:43] <[md5]> Urb... ah nevermind :P [09:43] DrMcCoy: fix the ADPCM issue!!111 [09:43] oh you did :-P [09:43] Action: DrMcCoy slaps LordHoto [09:43] Action: DrMcCoy slaps [md5] [09:43] meh, lesson time. [09:43] Hkz (~noname@host46-141-dynamic.104-80-r.retail.telecomitalia.it) left irc: Client Quit [09:44] noname? [09:45] <[md5]> haha [09:47] <[md5]> I don't know why... but bugging DrMcCoy is pretty funny sometimes :P Have to stop uttering the forbidden words ("Urban", "Runner", "Dark", "Seed", "II", "Inca", "FIX IT OMG OMG WTF ITZ NOT WORKING!!!!1111") [09:47] ... [09:47] <[md5]> ok the last one isn't a word :P [09:47] <[md5]> :D [09:49] <[md5]> ...hate spam bots [09:49] <[md5]> "I have a Nintendo DS and tons of...that can connect the DS to the internet if anyone... an access point, which then lets your DS connect to the internet. Unfortunately, I don't...one is from ebay or maybe directly from Nintendo." [09:49] <[md5]> meh [09:49] <[md5]> loads of text that doesn't make sense [09:57] dreammaster (~dreammast@c114-77-95-66.lowrp3.vic.optusnet.com.au) joined #scummvm. [09:57] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services. [10:09] <[md5]> http://www.theregister.co.uk/2010/02/05/apple_slaps_iphone_app_for_mentioning_google_android/ [10:09] <[md5]> "While your application has not been rejected, it would be appropriate to remove 'Finalist in Googles Android Developers Challenge!' from the Application Description." [10:09] <[md5]> ... [10:10] And people look at me strangely when I say that Apple isn't any better than M$ [10:11] <[md5]> I don't, I think that both Apple and M$ are evil [10:11] <[md5]> so I don't look at you strangely :) [10:12] <[md5]> actually, all big corporations are evil... look at Oracle, for example, and what they're doing with Java now... [10:12] :P [10:13] <[md5]> http://www.theregister.co.uk/2010/02/04/schwartz_exit_tweet/ [10:14] <[md5]> and... [10:14] <[md5]> http://www.theregister.co.uk/2010/02/04/oracle_lights_out_for_project_darkstar/ [10:19] Google wants to be good :P [10:19] s/wants/tries [10:19] :D [10:21] Hmmm, anyone seen the "A Question About scummvm" mail on -devel? [10:22] Depending on his definition of "major", the answer is either none or too fricking much, I'd say :P [10:23] Or was there ever an all-encompassing restructuring, one not localized to one subsystem? [10:24] DrMcCoy: ScummVM is a FOSS project, it has the hallmarks of many hands and many tweaks ;-). I think early on it got massive restrcture work (back in the ScummVM/SimonVM days) and most/all subsystems have been gutted at some point if you look back. [10:25] DJWillis: Yeah [10:25] That's why I think the question on -devel is rather nonapplicable :P [10:26] not applicable? [10:26] Something like that :P [10:26] Its one of the things you don't notice when your working on things but take the codebase 12 months apart and it's scary ;-). Yep, reading -devel that question strikes me as poorly considered research. [10:27] It seems to disregard the churn that a constant changing set of eyes bring to a project over a closed project with fixed end goals etc. [10:29] wtf that guy asks us to supply him with all the information, without any gain for us? oO [10:30] Well, he is a student and we are "in the know" about the codebase (theoretically at least :P) [10:30] So that there's no "gain" for us, I don't mind much [10:30] he doesn't even offer the results of "his research" [10:31] It's just that ScummVM isn't really a good example for that sort of thing. It would only lead to much tedious work for us and unusable results for him :P [10:32] :-) [10:32] Well, it's academia, so the results should be relatively open in the end [10:33] sure, if we invest time to supply the information to him and take the time to search for the results ;-) [10:33] While a more fitting word for my internet connection today would be "gutted", or at least partly [10:34] it would be at least "nice" from him to offer us his results [10:35] :P [10:36] <[md5]> LordHoto: true, we might ask him for the results [10:36] <[md5]> as for restructuring/redesign... ho hum [10:36] [md5]: No [10:36] <[md5]> No? [10:36] <[md5]> no what? :P [10:36] [md5]: We should rather tell him that ScummVM is a bad idea for such research :P [10:37] <[md5]> well, it is a big project, and a lot of things have changed since the beginning... [10:38] The number of times any one subsystem has changed fundamentally would be in the ballpark of thousands, I'd wager [10:38] <[md5]> that's correct [10:38] <[md5]> e.g. I can think of the recent changes to the way sound streams are looped... [10:39] => Lots of work for us to write up the before/after revisions and far too much data for hi [10:39] him* [10:39] Hkz (~noname@host46-141-dynamic.104-80-r.retail.telecomitalia.it) joined #scummvm. [10:39] #scummvm: mode change '+o Hkz' by ChanServ!ChanServ@services. [10:39] <[md5]> true [10:39] wtf [10:39] <[md5]> so yes, I can't imagine that there's a single revision for any major change... all of them were done over several revisions [10:39] Your account has been temporarily blocked from sending messages. This block can be caused by sending messages that trigger our spam filters, or by having too many recipients in one email. We encourage you to review the contents and recipient list of your message, and try sending it at after an hour or two. Doing so will usually resolve the matter. [10:40] damn yahoo [10:40] => ScummVM is not a prime example for the research he does [10:40] => He should move to some other project [10:40] :P [10:40] <[md5]> meh :P [10:40] <[md5]> but all projects are like that [10:40] <[md5]> I can't imagine any project having major changes done in one revision [10:40] <[md5]> it's just not feasible [10:40] or we just tell him what changed and let him search through our revisions ;-P [10:40] <[md5]> yeah [10:40] [md5]: All? I wouldn't be so sure about that. Most, sure [10:40] LordHoto: That's still bloody much work :P [10:41] <[md5]> DrMcCoy: I can't think of any project that would do major restructuring of its code in one revision [10:41] <[md5]> it's just too much work [10:41] [md5]: Me neither. That doesn't mean there no such animal out there, though :P [10:41] [md5]: Well, not necessarily one revision [10:41] <[md5]> so it can be a span of revisions, like we do [10:41] [md5]: Just a set of continous revisions [10:42] <[md5]> yes [10:42] <[md5]> DrMcCoy: so, his question is something not feasible, at least not observable directly [10:42] [md5]: Well, the range we'd give him would r1-r47898 :P [10:42] <[md5]> but he can look at the code before and after [10:42] <[md5]> haha :D [10:42] <[md5]> evil :P [10:42] LordHoto (~loom@s0921.vpn.hrz.tu-darmstadt.de) left irc: Quit: Verlassend [10:42] [md5]: No, what I mean: [10:43] DrMcCoy: he could do research on urban runner :P [10:43] [md5]: There's surely a project out there, running from r1-r10000, that had a major restructuring thing form r1234-r1312 or something [10:43] <[md5]> hahahahaha [10:43] <[md5]> DrMcCoy: yeah [10:43] <[md5]> DrMcCoy: we did so too [10:43] [md5]: No [10:44] <[md5]> e.g. we could mention the 16-bit color backend, for example [10:44] [md5]: We changed one subsystem in r1234, then worked on features, then another subsystem in r1250, worked on features, ... [10:44] <[md5]> so that's probably an example of how code looked before (i.e. 256 colors only) and after (both 256 colors and 16-bit are possible) [10:45] [md5]: Not one big all-encompassing rewrite in one cluster of commits, unbroken by new features/bugfixes [10:45] <[md5]> yeah we didn't have that at all in our history [10:45] <[md5]> as far as I know [10:45] [md5]: Like I said, whether 16bit-code counts, is a question of his definition of "major" [10:46] [md5]: And if that /would/ count, we've got thousands of these changes in our history [10:46] <[md5]> yeah... I don't know what "major" stands in this context, but imho 16-bit code is a bit major change [10:46] <[md5]> true [10:46] [md5]: So would we just give him one example? Or all of them? [10:46] <[md5]> I don't know [10:46] [md5]: All of them is too much work for us, too much work for him [10:47] <[md5]> but still, his question is quite hard to answer reliably [10:47] <[md5]> for *any* project [10:47] <[md5]> unless.... [10:47] <[md5]> hm [10:47] [md5]: Only one is stupid, since there's no one change that's representational for all of them. So which to choose? The choice will influence his research [10:47] <[md5]> unless we're talking about a project which is quite well segmented [10:47] [md5]: And, like I said, I'm sure there is a project out there that would fit his template :P [10:48] <[md5]> a pure emulation project would sound a more likely candidate [10:48] Hkz (~noname@host46-141-dynamic.104-80-r.retail.telecomitalia.it) left irc: Ping timeout: 264 seconds [10:48] <[md5]> e.g. MAME, for example [10:48] <[md5]> or DOSBox? [10:48] <[md5]> for MAME, his question could be answered in a better manner [10:49] I dunno, I've never looked "under the hood" of MAME [10:49] <[md5]> I remember that quite recently they added support for dual CPUs [10:49] <[md5]> that needed restructuring of several code functionality, from what they said [10:50] <[md5]> in any case, perhaps we could point him to a project where things are more segmented [10:51] <[md5]> a platform emulation project might perhaps have several independent subsystems, so things could be more clearly defined there (though I'm not sure) [10:51] If you know of such a project, sure. But I'd advise against bouncing him off to a project that might answer the same :P [10:51] This will only lead to frustration [10:51] <[md5]> true [10:51] <[md5]> I don't know of the inner workings of such projects [10:52] <[md5]> I'm just saying that probably a pure emulation project could in theory answer his questions more reliably [10:52] Yeah [10:53] Anyway, I'll wait what Fingolfin has to say about that, maybe he knows something of a good case early in the history of ScummVM that would be useful for that guy :P [10:54] it wouldn't surprise me if projects with smaller teams (and fewer subteams) have such periods semi-regularly [10:54] Or _sev. Whoever of those is less busy these days :) [10:55] hashier (~hashier@jwchat.org) left irc: Read error: Operation timed out [10:55] wjp: Might be, yeah [10:55] hashier (~hashier@jwchat.org) joined #scummvm. [11:05] LordHoto (loom@s0283.vpn.hrz.tu-darmstadt.de) joined #scummvm. [11:05] #scummvm: mode change '+o LordHoto' by ChanServ!ChanServ@services. [11:06] [md5]: I don't think 16bpp qualifies at least he said: "in which major changes were applied to the code base to improve software architecture/design while little or no functional enhancement was made" [11:06] and 16bpp support was basically only functional enhancement :-) [11:08] <[md5]> true [11:08] <[md5]> hm [11:08] <[md5]> perhaps the engine plugin restructuring could qualify as something like that... [11:25] Sho_ (~EHS1@kde/hein) left irc: Remote host closed the connection [11:25] Sho_ (~EHS1@kde/hein) joined #scummvm. [11:25] Robin_Watts (~Javacat@82.152.255.89) joined #scummvm. [11:25] #scummvm: mode change '+o Robin_Watts' by ChanServ!ChanServ@services. [11:25] Schnaks (~Schnaks@p5486FD2A.dip.t-dialin.net) joined #scummvm. [11:26] LordHoto (loom@s0283.vpn.hrz.tu-darmstadt.de) left irc: Quit: Verlassend [11:37] Kirben (kirben@c58-107-187-179.brasd3.vic.optusnet.com.au) joined #scummvm. [11:37] #scummvm: mode change '+o Kirben' by ChanServ!ChanServ@services. [12:04] Nevyn4 (jhgjfgl@81-236-19-16-no39.tbcn.telia.com) joined #scummvm. [12:06] dreammaster (~dreammast@c114-77-95-66.lowrp3.vic.optusnet.com.au) left irc: [12:32] waltervn (~walter@82-171-47-198.ip.telfort.nl) joined #scummvm. [12:32] #scummvm: mode change '+o waltervn' by ChanServ!ChanServ@services. [12:32] hi [12:33] hi [12:35] peres (~peres@unaffiliated/peres) joined #scummvm. [12:35] #scummvm: mode change '+o peres' by ChanServ!ChanServ@services. [12:43] m_kiewitz * r47899 /scummvm/trunk/engines/sci/graphics/ (frameout.cpp paint16.cpp paint32.cpp picture.cpp picture.h): SCI: renamed SciGuiPicture to GfxPicture [12:46] Lubomyr (~Miranda@194.44.255.155) joined #scummvm. [12:57] D0SFreak (~D0SFreak@72.1.88.158) joined #scummvm. [13:00] m_kiewitz * r47900 /scummvm/trunk/engines/sci/graphics/ (7 files): SCI: renamed class Font to GfxFont [13:26] waltervn * r47901 /scummvm/trunk/engines/sci/engine/ (kernel.cpp kpathing.cpp): SCI: Add implementation for Intersections(). [13:30] Strangerke (~51f60aa1@gateway/web/freenode/x-lplrwypkxiifptsm) joined #scummvm. [13:30] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services. [13:30] hi guys [13:34] m_kiewitz * r47902 /scummvm/trunk/engines/sci/ (11 files in 2 dirs): SCI: renamed class Cursor to GfxCursor [13:37] wjp: thanks for helping out with Intersections() (I forgot to put it in the commit message) [13:39] Schnaks (~Schnaks@p5486FD2A.dip.t-dialin.net) left irc: Ping timeout: 248 seconds [13:44] Schnaks (~Schnaks@p5486BFA2.dip.t-dialin.net) joined #scummvm. [13:52] Kirben (kirben@c58-107-187-179.brasd3.vic.optusnet.com.au) left irc: [14:00] _sev|work__ (~sev@kharkov.intetics.com) joined #scummvm. [14:01] _sev|work__ (~sev@kharkov.intetics.com) left irc: Read error: Connection reset by peer [14:02] _sev|work__ (~sev@kharkov.intetics.com) joined #scummvm. [14:03] _sev|work_ (~sev@scummvm/undead/sev) left irc: Ping timeout: 258 seconds [14:52] ok i have some time... anyone want me to do something? [14:53] (i have a working patch about the save path issue i had however im not perfectly content with) [15:06] StarG_ : paste it there, people will take a look [15:06] or better, create a new thread on the patch tracker on sf.net [15:10] my sf account is somewhat lost :/ [15:11] i used to have one but neighter cant seem to remember, could not find the acc details in notes nor in browser password storage =/ (although i'd admit i still have a location to look for) [15:11] use the pastebox then [15:11] pastebot* [15:12] will do, i had a small idea how to make it better so i dont have the feeling its too ugly :D [15:12] will be ironing it a bit [15:13] uhm... *caugh* *caugh* [15:13] how do i do a diff between directories? [15:13] use dif [15:13] svn should be a help as it contains hashes [15:13] diff [15:13] i modified 2 files [15:14] i know how to diff a single file... [15:14] diff also diffs directories [15:14] i see i still need to figure that out [15:16] diff -urN dir1 dir2 [15:16] thx :) [15:18] m_kiewitz * r47903 /scummvm/trunk/engines/sci/ (13 files in 3 dirs): SCI: calling most of the cursor functions directly via _gfxCursor instead of SciGui/32 [15:21] _sev|work__ (~sev@kharkov.intetics.com) left irc: Read error: Connection reset by peer [15:21] _sev|work__ (~sev@kharkov.intetics.com) joined #scummvm. [15:23] i had the strange issue that a gui dialog would not be shown if the interpreter plugin instance has just been created [15:23] (tested with the gob plugin) [15:31] can you be more specific? [15:33] sure [15:33] during engine initalisation i added a check about a valid savegame path... and if found to be invalid prompt the user [15:34] i found that i would not see the dialog if it were after the actual createInstance of the engine happened earlier [15:34] or i had to display the dialog before the createInstance [15:35] at least thats what i've perceived yesterday [15:43] so I suggest displaying the dialog before the createInstance :-) [15:44] hehe... [15:45] will do.. i hopefully cleaned up a bit so i think its better now ... i'll test it again today and if it works as expected ill diff it and upload it to the pastebot [15:46] the mailing list or patch tracker would probably be more appropriate if you want people to discuss about it [15:46] if that is i will post it to the mailing list... [15:50] [md5] (~d4fb49fe@gateway/web/freenode/x-jyejdeiyojoetimc) left irc: Quit: later [16:01] LordHoto (~loom@p54B1DF67.dip.t-dialin.net) joined #scummvm. [16:01] #scummvm: mode change '+o LordHoto' by ChanServ!ChanServ@services. [16:13] StarG (~John@g224194255.adsl.alicedsl.de) joined #scummvm. [16:16] StarG_ (~John@g224197151.adsl.alicedsl.de) left irc: Ping timeout: 256 seconds [16:18] m_kiewitz * r47904 /scummvm/trunk/engines/sci/graphics/ (16 files): SCI: adding some minor documentation for the various graphic classes [16:22] m_kiewitz * r47905 /scummvm/trunk/engines/sci/ (6 files in 2 dirs): SCI: screen specific debug functions are now called directly w/o SciGui/32 [16:24] m_kiewitz * r47906 /scummvm/trunk/engines/sci/graphics/ (gui32.cpp gui32.h): SCI: removing debugShowMap from SciGui32 [16:26] Hkz (~noname@host46-141-dynamic.104-80-r.retail.telecomitalia.it) joined #scummvm. [16:26] #scummvm: mode change '+o Hkz' by ChanServ!ChanServ@services. [16:26] Hi Hkz :) [16:26] hi! [16:32] Kerbox (~Kernox@ti0106a340-dhcp0478.bb.online.no) joined #scummvm. [16:32] m_kiewitz * r47907 /scummvm/trunk/engines/sci/ (6 files in 3 dirs): SCI: frameout sci32 stuff now gets called directly w/o SciGui/32 [16:46] Sho_ (~EHS1@kde/hein) left irc: Read error: Connection reset by peer [16:47] Sho_ (~EHS1@kde/hein) joined #scummvm. [17:06] peres (peres@unaffiliated/peres) left #scummvm ("Leaving"). [17:10] jvprat (~jvprat@unaffiliated/jvprat) joined #scummvm. [17:10] #scummvm: mode change '+o jvprat' by ChanServ!ChanServ@services. [17:13] Lars2 (~lh@0x573f1066.cpe.ge-1-1-0-1101.hjnqu1.customer.tele.dk) left irc: Read error: Connection reset by peer [17:19] Ceriand|work (~Ceriand@unaffiliated/ceriand) joined #scummvm. [17:38] clone2727 (~clone2727@nat02-njh-ext.rutgers.edu) joined #scummvm. [17:38] #scummvm: mode change '+o clone2727' by ChanServ!ChanServ@services. [17:40] eriktorbjorn: Heh. Thanks for that gameID fix. I (wrongly) assumed it was a Common::String. [17:46] ok im pasting the diff now... hope it works :) [17:47] StarG pasted "savegame path check proposal" at http://paste.lisp.org/display/94427 [17:53] if anyone wants to comment this check... [18:01] jvprat (~jvprat@unaffiliated/jvprat) left irc: Quit: Leaving. [18:05] Strangerke (~51f60aa1@gateway/web/freenode/x-lplrwypkxiifptsm) left irc: Quit: Bbl [18:10] syke (~matt@76-191-154-20.dsl.dynamic.sonic.net) joined #scummvm. [18:10] #scummvm: mode change '+o syke' by ChanServ!ChanServ@services. [18:11] mornin [18:12] hey syke [18:23] clone2727: I wouldn't even have noticed it if GCC hadn't warned. [18:24] eriktorbjorn: What version do you have? [18:25] clone2727: gcc (Debian 4.4.3-2) 4.4.3 [18:26] eriktorbjorn: I don't think I got a warning ("gcc version 4.0.1 (Apple Inc. build 5490)"). I could have just missed it though ;) [18:26] hi syke [18:26] gcc 4.4 added a lot of warnings [18:27] At least if you use -Wall [18:28] johndoe123 (~john_doe@p5DE7B313.dip.t-dialin.net) left irc: [18:32] clone2727: I was surprised the SCI engine used the game id string like that. Most other engines define numerical constants for it instead. [18:33] eriktorbjorn: Yeah, not sure why that's done either. [18:34] eriktorbjorn: I think in the beginning, the game ID's weren't used at all in game [18:36] s/in game/in the source [18:38] Schnaks (~Schnaks@p5486BFA2.dip.t-dialin.net) left irc: Ping timeout: 258 seconds [18:44] talking about game id's... i wonder if it were possible to have proper foreign names in the detection [18:45] (the foreign names of the titles) [18:46] Ceriand|work: yea -- gcc 4.4 has been a godsend @ work [18:46] especially with megacompilation and -O3 [18:46] StarG: That's a separate issue dealing with translation. There should be a patch on the patch tracker with related code. In the meantime, you can change the name in your config. [18:47] WooShell (~Markus@woo.li) joined #scummvm. [18:49] clone2727 of course. if i can help about feel free to instruct me how. [18:50] i noticed a few misdetections with the 1.0.0 release, minor stuff like a engine mistaking a english with a spanish version [18:51] Schnaks (~Schnaks@p5486E68C.dip.t-dialin.net) joined #scummvm. [18:56] m_kiewitz: hey man, how's it going? [19:03] nobbend [19:16] _athrxx (~athrxx@e181077083.adsl.alicedsl.de) joined #scummvm. [19:16] #scummvm: mode change '+o _athrxx' by ChanServ!ChanServ@services. [19:18] digitall (~digitall@cpc1-hitc2-0-0-cust861.lutn.cable.ntl.com) joined #scummvm. [19:18] #scummvm: mode change '+o digitall' by ChanServ!ChanServ@services. [19:22] TAS_2012x (~TAS_2012x@c-dccbe253.1710-7-64736c12.cust.bredbandsbolaget.se) left irc: Ping timeout: 260 seconds [19:25] m_kiewitz * r47908 /scummvm/trunk/engines/sci/ (14 files in 3 dirs): SCI: GfxCoordAdjuster class added, local2Global and global2Local use that class directly, kGraph / RedrawBox is now using GfxPaint16 directly [19:28] TAS_2012x (~TAS_2012x@c-dccbe253.1710-7-64736c12.cust.bredbandsbolaget.se) joined #scummvm. [19:28] servus (~servus@71.107.57.176) left irc: Quit: :wq [19:29] servus (~servus@71.107.57.176) joined #scummvm. [19:31] Sho_ (~EHS1@kde/hein) left irc: Remote host closed the connection [19:32] Ge0rG (georg@bender.boerde.de) left irc: Ping timeout: 260 seconds [19:33] Ge0rG (georg@bender.boerde.de) joined #scummvm. [19:35] m_kiewitz: ./engines/sci/graphics/coordadjuster.h:41: warning: class Sci::GfxCoordAdjuster has virtual functions but non-virtual destructor [19:37] clone2727: msvc again...argh [19:37] m_kiewitz: You could switch to MinGW :P [19:37] i could switch to linux completely :P [19:38] Or that :P [19:38] m_kiewitz * r47909 /scummvm/trunk/engines/sci/graphics/coordadjuster.h: SCI: fix warning inside GfxCoordAdjuster header [19:40] scummvm compiles with mingw? [19:40] i guess there isnt a devcpp project file ready? [19:45] nope [19:45] (i.e. there isn't, just an very outdated one) [19:48] hey LordHoto... remember we spoke about a savegame directory check yesterday? [19:49] yes and your patch does warn always when there's no explicit savepath set and this is definitly wrong [19:49] since UNIX and OS X does have a sensible default setting described as in the README [19:51] alright removed [19:52] and it is a bit strange that you called the variable warn, but all variables, in the if block which checks for warnings, err* [19:52] does unixoid hosts permit to have the global domain var "savedir" to be blank? [19:53] savedir? [19:53] well i originally thought about a minor error constant but dumped the idea in favor of a minor error state which i dubbed warn [19:53] eh yes in the scummvmrc [19:54] it's blank for me when I create a new .scummvmrc [19:54] when it's blank under Windows it just uses "." [19:55] still reading the docs but i guess it uses a default dir for savegames too when under unix ? [19:55] like the executable dir [19:55] yes ~/.scummvm [19:56] hm it seems not to be described in the README [19:56] but it should use ~/.scummvm on UNIX... [19:57] let me try.. [19:57] yup [19:57] and for OS X it should use "~/Documents/ScummVM Savegames" [19:59] works just fine without any savepath setting for me [19:59] anyways i dumped the blank check [20:02] no i need to wait ages so i can try a savegame :) [20:03] m_kiewitz * r47910 /scummvm/trunk/engines/sci/graphics/frameout.cpp: SCI: initializing _highPlanePri inside GfxFrameout [20:04] strangely it does not create a ~/.scummvm folder here... [20:04] does it happen upon the first actual game saved? [20:06] might be [20:07] there are 2 things im not content with... thats first warnings might be not just a single state but should be a stack of [20:08] if there are multiple it should be looped through [20:09] and 2nd the point of the check. i'd rather have it checked upon entering gui so the user can adjust another directory instead of quitting after he might found out that he used a "corrupted" save path [20:09] I think the place is fine there too, at least the user can have per game save paths [20:10] and thus he should be notified definitly on game launch about this [20:10] mthreepwood * r47911 /scummvm/trunk/engines/sci/engine/kgraphics.cpp: Cleanup kShowMovie; merge the SCI1.1/SCI2 and SCI2.1 AVI code. [20:12] sounds reasonable [20:12] bbl [20:12] m_kiewitz: it would be really nice if thoes qfg1 issues got worked out so I could test over the weekend [20:12] heading into work [20:12] syke (~matt@76-191-154-20.dsl.dynamic.sonic.net) left irc: Quit: Leaving. [20:13] Ge0rG (georg@bender.boerde.de) left irc: Ping timeout: 260 seconds [20:14] Ge0rG (georg@bender.boerde.de) joined #scummvm. [20:19] StarG: anyway I just checked, it seems our SaveFileManager automatically sets an error in case there's something wrong with the savepath (it seems the PSP implementation of that is a bit more lax there though... and the default implementation doesn't check for write access, but that could be changed easily I guess) [20:19] StarG: so maybe the way would be to extend SaveFileManager to correct setup its error flags, in case something goes wrong and use that for the user display [20:20] Ge0rG (georg@bender.boerde.de) left irc: Ping timeout: 260 seconds [20:20] StarG: and of course a way to check the savepaths state, currently that is only happening on read/write/list access [20:22] StarG annotated #94427 "addentum" at http://paste.lisp.org/display/94427#1 [20:22] sounds a better place for those checks [20:23] actually the POSIXSaveFileManager does check for quite a few things ;-) [20:23] ive experienced problems only on pc yet.. but i'd assume handhelds with changable media (like sd cards) might be somewhat prone to that too [20:23] well if you remove the sd card while running a game saved on it you have a problem anyway ;-) [20:23] lol _D [20:24] also I guess we can get rid of "kInvalidSavePathError" too in case we use SaveFileManager directly [20:25] Ge0rG (georg@bender.boerde.de) joined #scummvm. [20:25] think so :) [20:28] anyway if you have patches, which should be reviewed they should definitly sit in our patch tracker too and not on our paste site ;-) [20:28] when is the savefilemanager invoked.. at actual savegame use? [20:28] clone2727 (~clone2727@nat02-njh-ext.rutgers.edu) left irc: Quit: bbl -- heading home [20:28] ehm... agreed... need to find my sf account details again =/ [20:29] actually on everything, which needs savegames [20:29] (means --list-saves etc. too) [20:29] there is a manual check in the gui code [20:29] when you select a new savegame directory [20:30] might be outdated then too [20:30] Ge0rG (georg@bender.boerde.de) left irc: Read error: Operation timed out [20:33] TAS_2012x (~TAS_2012x@c-dccbe253.1710-7-64736c12.cust.bredbandsbolaget.se) left irc: Ping timeout: 260 seconds [20:34] Javacat (~ghsgh@unaffiliated/javacat) joined #scummvm. [20:34] syke (~matt@cesium.clock.org) joined #scummvm. [20:34] #scummvm: mode change '+o syke' by ChanServ!ChanServ@services. [20:34] hi [20:39] TAS_2012x (~TAS_2012x@c-dccbe253.1710-7-64736c12.cust.bredbandsbolaget.se) joined #scummvm. [20:40] Ge0rG (georg@bender.boerde.de) joined #scummvm. [20:41] the defaultsavefilemanager could need a writable check [20:42] Schnaks (~Schnaks@p5486E68C.dip.t-dialin.net) left irc: Ping timeout: 258 seconds [20:42] StarG (John@g224194255.adsl.alicedsl.de) left #scummvm. [20:42] StarG (~John@g224194255.adsl.alicedsl.de) joined #scummvm. [20:43] StarG: yup said that :-) [20:43] oh sry i overlooked that [20:51] m_kiewitz: were you able to fix those kq5 valgrind errors? [20:51] syke: im just cleaning up classes [20:52] ok [20:53] output looks the same; let me know if you need more info [20:53] Schnaks (~Schnaks@p5486F0EB.dip.t-dialin.net) joined #scummvm. [20:54] Ge0rG (georg@bender.boerde.de) left irc: Ping timeout: 248 seconds [20:57] Ge0rG (georg@bender.boerde.de) joined #scummvm. [20:58] info is still here: http://pastie.org/808379 [21:12] Dounet (~massoubr@88-107-190-112.dynamic.dsl.as9105.com) joined #scummvm. [21:12] Dounet (massoubr@88-107-190-112.dynamic.dsl.as9105.com) left #scummvm. [21:13] m_kiewitz * r47912 /scummvm/trunk/engines/sci/ (11 files in 2 dirs): SCI: moved onControl etc. into GfxCompare, now getting called directly. also fixed loading saved games due Gfx* changes [21:14] D0SFreak (~D0SFreak@72.1.88.158) left irc: Quit: Leaving [21:17] m_kiewitz * r47913 /scummvm/trunk/engines/sci/ (5 files in 2 dirs): SCI: drawMenuBar, drawStatus now called directly and w/o SciGui [21:23] Deepa (~lisa@wrongplanet/deepa) joined #scummvm. [21:38] m_kiewitz * r47914 /scummvm/trunk/engines/sci/ (9 files in 2 dirs): SCI: remaining cursor functions now also directly called [21:39] Lubomyr (~Miranda@194.44.255.155) left irc: Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org [21:46] LordHoto (~loom@p54B1DF67.dip.t-dialin.net) left irc: Quit: bbl [21:46] Strangerke (~a@cable-85.28.89.119.coditel.net) joined #scummvm. [21:46] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services. [22:02] m_kiewitz * r47915 /scummvm/trunk/engines/sci/graphics/ (controls.h gui.h): SCI: moved control type enum [22:17] m_kiewitz: do you get any memory errors from the debug CRT when saving/loading a game? [22:19] syke: nope [22:19] syke: i tried kq5 and lsl1, you got those with another game? [22:20] D0SFreak (~D0SFreak@c-24-99-111-221.hsd1.ga.comcast.net) joined #scummvm. [22:27] Kirben (kirben@c58-107-187-179.brasd3.vic.optusnet.com.au) joined #scummvm. [22:27] #scummvm: mode change '+o Kirben' by ChanServ!ChanServ@services. [22:28] digitall (digitall@cpc1-hitc2-0-0-cust861.lutn.cable.ntl.com) left #scummvm. [22:31] m_kiewitz: kq5 [22:32] do you have the runtime checks turned on in VC++? [22:32] i got the ones enabled by default [22:32] odd [22:33] you want to enable both with /RTCu, if memory serves [22:33] it's in the project settings under C/C++ (I think) [22:33] where do you get those errors? [22:33] Hum... The project files are now generated by a tool [22:34] i mean do you get those when saving or loading? [22:34] i just tried loading [22:34] salty-horse (~ori@bzq-79-183-15-245.red.bezeqint.net) joined #scummvm. [22:34] saving [22:34] hi salty-horse :) [22:34] I figure if saving has this many errors, loading might be broken [22:34] hey :) [22:38] syke: i just tried right at the beginning - no problems on saving [22:38] im getting WARNING: Could not find local script from segment 4744! [22:38] WARNING: cmdDummy invoked! on load [22:38] I get no errors when saving, do get some when loading [22:38] actually loading segfaults ;) [22:39] oh [22:39] interesting [22:40] flood of valgrind errors on save happens only after I invoke the About dialog [22:40] looks like invoking the about dialog causes some memory corruption that is then exacerbated by saving [22:41] _sev|work_ (~sev@scummvm/undead/sev) joined #scummvm. [22:41] #scummvm: mode change '+o _sev|work_' by ChanServ!ChanServ@services. [22:42] first one os [22:42] ==5592== by 0xE9F390: Common::GZipWriteStream::processData(int) (zlib.cpp:225) [22:42] ==5592== by 0xE9F913: Common::GZipWriteStream::write(void const*, unsigned int) (zlib.cpp:302) [22:42] ==5592== by 0x4274F5: Common::WriteStream::writeUint16LE(unsigned short) (stream.h:111) [22:42] ==5592== by 0x680D63: void Common::Serializer::syncAsUint16LE(unsigned short&, unsigned int, unsigned int) (serializer.h:101) [22:42] ==5592== by 0xB99757: Sci::sync_reg_t(Common::Serializer&, Sci::reg_t&) (savegame.cpp:81) [22:42] ==5592== by 0xB99B3D: void Sci::syncWithSerializer(Common::Serializer&, Sci::reg_t&) (savegame.cpp:219) [22:42] ==5592== by 0xB9FA32: Sci::DefaultSyncer::operator()(Common::Serializer&, Sci::reg_t&) const (savegame.cpp:174) [22:42] ==5592== by 0xB9EF6E: Sci::ArraySyncer >::operator()(Common::Serializer&, Common::Array&) const (savegame.cpp:204) [22:42] Newa (~Neva@vpn16-138.csc.fi) joined #scummvm. [22:42] Newa2 (~Neva@dhcp-40.csc.fi) joined #scummvm. [22:43] let me svn up first because I'm not up-to-date (things are moving so fast....) [22:44] btw, sci guys, gk1 dos crashed when I went to the park [22:44] Grachev (~grachev@149-144-31.oke2-bras7.adsl.tele2.no) joined #scummvm. [22:44] |Cable| (~cable@237.130.117.91.dynamic.mundo-r.com) joined #scummvm. [22:45] Newachu (~Neva@vpn16-138.csc.fi) left irc: Ping timeout: 240 seconds [22:45] NewaWrk (~Neva@dhcp-40.csc.fi) left irc: Ping timeout: 240 seconds [22:45] _sev|work__ (~sev@kharkov.intetics.com) left irc: Ping timeout: 240 seconds [22:45] syke: it would be better to fix the about box issue first, then the other thing might go away automatically [22:45] Nick change: Newa2 -> NewaWrk [22:47] waltervn: that's what I'm thinking [22:50] m_kiewitz * r47916 /scummvm/trunk/engines/sci/ (6 files in 2 dirs): SCI: priority band specific stuff now gets called directly w/o SciGui [22:55] Ceriand|work (~Ceriand@unaffiliated/ceriand) left irc: Quit: Leaving. [22:57] ScottT (~ScottT@92.55.233.220.static.exetel.com.au) joined #scummvm. [22:57] #scummvm: mode change '+o ScottT' by ChanServ!ChanServ@services. [22:59] scummvm palm pre port: http://forums.precentral.net/webos-apps-software/223524-scummvm.html [23:01] m_kiewitz * r47917 /scummvm/trunk/engines/sci/ (5 files in 2 dirs): SCI: moving kDisplay into GfxPaint16, now getting called directly as well [23:04] Fingolfin (~Fingolfin@p548577ED.dip.t-dialin.net) joined #scummvm. [23:04] #scummvm: mode change '+o Fingolfin' by ChanServ!ChanServ@services. [23:16] waltervn: any idea on the about box issue? [23:17] haven't looked at it [23:17] ok [23:17] that also gets worse each time you do it [23:17] the first time you get a single valgrind warnings [23:18] the second time, you get screens full [23:19] Deepa (~lisa@wrongplanet/deepa) left irc: Ping timeout: 240 seconds [23:19] syke: you dont get valgrind errors when looking at the about box? [23:20] m_kiewitz: I do. the first time I bring up the about box, there's one valgrind error. the next time Ibring up the about box, there are screens full of valgrind errors. the first time appear sto corrupt memory, then everything after that is horked as a result [23:21] syke: what valgrind error do you get? [23:21] syke, don't paste here please :) [23:22] salty-horse: I just pasted this message [23:22] salty-horse: and this one [23:22] highlight to copy [23:22] middle click to paste [23:23] m_kiewitz: same URl as 2 days ago: @syke> salty-horse: I just pasted this message [23:23] 14:53 < [23:23] http://pastie.org/808379 [23:23] 14:53 <@syke> salty-horse: I just pasted this message [23:23] 14:53 <@syke> salty-horse: and this one [23:23] 14:53 <@syke> highlight to copy [23:23] 14:53 <@syke> middle click to paste [23:23] which one do you get when opening the about box? [23:23] m_kiewitz: the first one is the first time I open it [23:24] this one? at 0xB6C324: Sci::SegManager::getString(Sci::reg_t, int) (seg_manager.cpp:817) [23:24] waltervn (~walter@82-171-47-198.ip.telfort.nl) left irc: Quit: have fun [23:24] fingolfin * r47918 /scummvm/trunk/engines/sci/ (19 files in 4 dirs): SCI: Add a 'SELECTOR' macro [23:25] fingolfin * r47919 /scummvm/trunk/engines/sci/graphics/compare.cpp: SCI: Add comment to GfxCompare::canBeHereCheckRectList [23:25] m_kiewitz: yup [23:25] gouchi (~Karim@ivr94-8-88-162-27-162.fbx.proxad.net) joined #scummvm. [23:25] fingolfin * r47920 /scummvm/trunk/engines/sci/graphics/ (compare.cpp compare.h): SCI: Document GfxCompare::canBeHereCheckRectList; pass Common::Rect by ref [23:25] fingolfin * r47921 /scummvm/trunk/engines/sci/graphics/compare.cpp: SCI: Get rid of the mysterious sign_extend_byte() and replace it by a simple cast ;) [23:29] hm, this valgrind error could be triggered by a bug in SegManager::memcpy() if I understand it correctly? [23:29] that method looks innocent enough, though, probably elsewhere [23:30] Fingolfin: the actual problem is further up the callchain [23:30] ? [23:30] the Uninitialised value comes from a memory block allocated by "new", though, right? [23:31] allocated in kFileIO to be precise [23:31] so I thought that maybe this block never get inited properly, or gets copied incorrectly [23:32] like, if the read() call in K_FILEIO_READ_RAW fails... in theory I guess the calling script should detect this by checking the return value of read(), but if the script for some reasons doesn't do that, then it would copy uninited data, wouldn't it? [23:32] syke: but apparently I am missing something? [23:34] m_kiewitz (~m_kiewitz@kons-5f7104a2.pool.mediaWays.net) left irc: Quit: leaving [23:36] Fingolfin: I think you're right, that it comes down to not checking the return of read() [23:39] syke: with which game does that error occur? can one reproduce it well? [23:39] KQ5 CD [23:39] it's easy for me to reproduce [23:39] not sure if others have had luck [23:39] m_kiewitz uses VC++, so he doesn't have valgrind access [23:39] but it seems like /RTC1 would catch it as well [23:40] Grachev (~grachev@149-144-31.oke2-bras7.adsl.tele2.no) left irc: Quit: leaving [23:40] then again, it doesn't look like /RTC1 is being turned on for the generated VC++ files anymore [23:41] I have valgrind on OS X [23:41] lskovlun (~lars@port1.cvx3-ejb.ppp.cybercity.dk) joined #scummvm. [23:41] #scummvm: mode change '+o lskovlun' by ChanServ!ChanServ@services. [23:41] hey [23:42] syke: Someone mentioned that READ_RAW is in some cases by called with a zero (or non-existent) size parameter [23:42] you may want to check for it, everything else seems to follow from that [23:42] (apart from some of the last bits that don't mention FileIO at all) [23:43] Kirben (kirben@c58-107-187-179.brasd3.vic.optusnet.com.au) left irc: [23:44] m_kiewitz * r47922 /scummvm/trunk/engines/sci/graphics/compare.cpp: SCI: removing int8 inside kernelSetNowSeen as well, didnt make any sense :P [23:47] m_kiewitz * r47923 /scummvm/trunk/engines/sci/graphics/compare.cpp: SCI: removing int8 from kernelSetNowSeen() as well, didnt make any sense :P [23:49] Grachev (~grachev@149-144-31.oke2-bras7.adsl.tele2.no) joined #scummvm. [23:53] lskovlun (lars@port1.cvx3-ejb.ppp.cybercity.dk) left #scummvm ("Client exiting"). [23:57] Fingolfin (~Fingolfin@p548577ED.dip.t-dialin.net) left irc: Quit: Fingolfin [00:00] --- Sat Feb 6 2010