ScummVM IRC Log - 20 Feb 2012

[Back to Index]

[00:11] <-- Kaidane_ left irc: Read error: Connection reset by peer
[00:12] --> Kaidane_ joined #scummvm.
[00:26] <CIA-71> Johannes Schickel master * r723a38c / (3 files): IPHONE: Rename screen related buffers a bit. - http://git.io/8XmOig
[00:26] <CIA-71> Johannes Schickel master * r1b9c4f3 / backends/platform/iphone/osys_video.cpp : IPHONE: Remove some dead code. - http://git.io/-kwvew
[00:26] <CIA-71> Johannes Schickel master * r65cda4c / backends/platform/iphone/osys_main.cpp : IPHONE: Fix some mismatching malloc/delete calls. - http://git.io/5AeLHA
[00:26] <CIA-71> Johannes Schickel master * r8102e7e / (6 files in 2 dirs): IPHONE: Implement cursor palette support. - http://git.io/icWjbg
[00:27] <-- Dominus left irc: Quit: Leaving.
[00:27] <-- Kirben left irc:
[00:40] <-- pok0j left irc: Remote host closed the connection
[00:41] <droid2727> LordHoto: \o/
[00:59] <-- Kaidane_ left irc: Read error: Connection reset by peer
[01:00] --> Kaidane_ joined #scummvm.
[01:00] sirlemonhead (bduncan22@86-43-169-164-dynamic.b-ras2.prp.dublin.eircom.net) left #scummvm.
[01:00] <-- Paaris left irc: Ping timeout: 260 seconds
[01:07] --> Paaris joined #scummvm.
[01:15] <-- DetectiveHoot left irc:
[01:26] <CIA-71> Johannes Schickel master * rd77253f / (6 files): IPHONE: Add support for unfiltered upscaling. - http://git.io/fvRMHA
[01:41] --> Kirben joined #scummvm.
[01:41] #scummvm: mode change '+o Kirben' by ChanServ!ChanServ@services.
[01:43] --> nutron|w joined #scummvm.
[01:44] <-- Javacat left irc: Quit: Please, try the fish
[01:55] <-- D0SFreak left irc: Ping timeout: 255 seconds
[01:58] <-- Kaidane_ left irc: Read error: Connection reset by peer
[01:59] --> Kaidane_ joined #scummvm.
[02:15] --> Vampire0_ joined #scummvm.
[02:18] <-- Vampire0 left irc: Ping timeout: 252 seconds
[02:20] --> cyco_ joined #scummvm.
[02:20] Nick change: cyco_ -> Guest40591
[02:22] <-- cyco left irc: Ping timeout: 260 seconds
[02:23] --> Tron_ joined #scummvm.
[02:25] <-- Tron left irc: Ping timeout: 248 seconds
[02:59] <-- Mataniko left irc: Read error: Connection reset by peer
[03:20] <-- droid2727 left irc: Ping timeout: 260 seconds
[03:21] --> droid2727 joined #scummvm.
[03:25] <-- droid2727 left irc: Ping timeout: 245 seconds

[03:48] --> risca joined #scummvm.
[03:48] <-- risca left irc: Read error: Connection reset by peer
[04:26] <-- LordHoto left irc: Quit: ...
[04:59] --> bgKa joined #scummvm.
[04:59] #scummvm: mode change '+o bgKa' by ChanServ!ChanServ@services.
[05:21] --> katycorp joined #scummvm.
[05:24] <katycorp> so
[05:24] <katycorp> scummvm on android
[05:24] <katycorp> is AWESOME
[05:25] <katycorp> that is all
[05:48] <-- Herrman left irc: Quit: yob tvoyu mat
[06:10] <-- SylvainTV left irc: Ping timeout: 240 seconds
[06:28] <-- Kaidane_ left irc: Read error: Connection reset by peer
[06:28] Nick change: Tron_ -> Tron
[06:29] <-- Smartnow left irc: Read error: Operation timed out
[06:30] --> Smartnow joined #scummvm.
[07:00] <-- bgKa left irc: Read error: Operation timed out
[07:09] <-- katycorp left irc: Ping timeout: 240 seconds
[07:10] --> katycorp joined #scummvm.
[07:10] <-- Guest40591 left irc: Ping timeout: 272 seconds
[07:24] <-- Strangerke left irc: Ping timeout: 252 seconds
[07:40] --> ny00123 joined #scummvm.
[07:42] --> pok0j joined #scummvm.
[08:03] <-- _sev left irc: Quit: This computer has gone to sleep
[08:04] --> Strangerke joined #scummvm.
[08:04] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services.
[08:04] <Strangerke> hi guys
[08:11] --> Schnaks joined #scummvm.
[08:12] <-- criezy left irc: Quit: criezy
[08:39] <wjp> morning
[08:39] <wjp> LordHoto has been busy I see
[08:41] --> dreammaster joined #scummvm.
[08:41] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[08:51] --> Dominus joined #scummvm.
[08:52] --> _sev joined #scummvm.
[08:52] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[08:59] <-- pablog left irc: Quit: No trees were killed in the sending of this message. However, a large number of electrons were terribly inconvenienced.
[09:08] <-- yaz0r left irc: Quit: leaving
[09:15] --> yaz0r joined #scummvm.
[09:15] #scummvm: mode change '+o yaz0r' by ChanServ!ChanServ@services.
[09:23] <fuzzie> yes, hooray for LordHoto \o/
[09:39] <-- pok0j left irc: Remote host closed the connection
[09:43] --> ajax16384 joined #scummvm.
[09:43] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[09:56] --> Kenryu joined #scummvm.
[09:56] Vaikungfu (UPP@c-71-193-60-169.hsd1.ca.comcast.net) got netsplit.
[09:56] lubyou (lubyou@unaffiliated/lubyou) got netsplit.
[09:56] Tomaz^ (tompsson@h-238-80.a199.priv.bahnhof.se) got netsplit.
[09:56] Nick change: Kenryu -> Vaikungfu
[09:56] Possible future nick collision: Vaikungfu
[10:01] Tomaz^ (tompsson@h-238-80.a199.priv.bahnhof.se) returned to #scummvm.
[10:01] lubyou (lubyou@unaffiliated/lubyou) returned to #scummvm.
[10:10] --> James|GlideM joined #scummvm.
[10:10] #scummvm: mode change '+v James|GlideM' by ChanServ!ChanServ@services.
[10:11] --> cyco joined #scummvm.
[10:19] --> Kenryu joined #scummvm.
[10:19] <-- Vaikungfu left irc: Disconnected by services
[10:19] Nick change: Kenryu -> Vaikungfu
[10:37] --> D0SFreak joined #scummvm.
[10:51] <-- D0SFreak left irc: Ping timeout: 252 seconds
[11:01] <-- qptain_Nemo left irc: Read error: Connection reset by peer
[11:01] --> qptain_Nemo joined #scummvm.
[11:03] --> D0SFreak joined #scummvm.
[11:24] <CIA-71> Paul Gilbert master * r0f3614b / (engines/tsage/resources.cpp engines/tsage/resources.h):
[11:24] <CIA-71> TSAGE: Implemented method for getting a resource section entry
[11:24] <CIA-71> This will be needed for streaming data directly from the resource file without fully loading the resource. - http://git.io/kMatMA
[11:24] <CIA-71> Paul Gilbert master * r7a54ba8 / (5 files in 2 dirs): TSAGE: Starting to implement animation player needed for R2R Title Screen - http://git.io/YO7P9Q
[11:27] --> giucam joined #scummvm.
[11:30] --> pok0j joined #scummvm.
[11:31] <-- D0SFreak left irc: Ping timeout: 240 seconds
[11:52] --> tsoliman joined #scummvm.
[11:52] #scummvm: mode change '+o tsoliman' by ChanServ!ChanServ@services.
[11:54] <-- pok0j left irc: Remote host closed the connection
[12:02] --> D0SFreak joined #scummvm.
[12:02] <-- dreammaster left irc: Quit: Leaving for the day
[12:04] --> LordHoto joined #scummvm.
[12:04] <-- LordHoto left irc: Changing host
[12:04] --> LordHoto joined #scummvm.
[12:04] #scummvm: mode change '+o LordHoto' by ChanServ!ChanServ@services.
[12:04] --> Adys_ joined #scummvm.
[12:06] <-- Adys left irc: Ping timeout: 276 seconds
[12:28] <-- ajax16384 left irc: Quit: KVIrc 4.1.3 Equilibrium http://www.kvirc.net/
[12:28] <LordHoto> fuzzie: any idea how we can get the android buildbot builds to work again?
[12:29] <fuzzie> sure, delete the debug key
[12:30] <LordHoto> and that would still let them be usable? ;-P
[12:30] <fuzzie> it'll make a new one
[12:31] <LordHoto> and why hasn't anyone done that yet?
[12:31] <fuzzie> well, I don't know
[12:31] <fuzzie> blame everyone with a buildbot account
[12:32] <LordHoto> I only blame people who have an idea about that!
[12:34] <-- Kirben left irc:
[12:35] Action: tsoliman wants a buildbot account
[12:37] --> pok0j joined #scummvm.
[12:40] <CIA-71> Johannes Schickel master * rae992be / backends/platform/iphone/iphone_video.m : IPHONE: More formatting fixes. - http://git.io/wMSa-A
[12:42] <-- Strangerke left irc: Ping timeout: 245 seconds
[12:48] <LordHoto> it seems I really need to get into ObjC again *sigh*
[12:48] <CIA-71> Tarek Soliman master * r3f6d549 / (5 files in 4 dirs): KEYMAPPER: Move F7 and F8 handling to DefaultEventMapper (+6 more commits...) - http://git.io/zb7eSw
[12:49] <tsoliman> LordHoto: there's ObjC in scummvm iOS code?
[12:49] <fuzzie> yes..?
[12:49] <LordHoto> obviously
[12:49] <LordHoto> I mean it's targetting iOS ;-)
[12:49] <LordHoto> it's like asking whether an OS X GUI application has ObjC code :-P
[12:50] <fuzzie> well once upon a time, Java was a supported language for that!
[12:50] <tsoliman> I guess it wasn't obvious when I was looking at the backend
[12:50] <LordHoto> you must have really been wondering what those msyterious ".m" files then
[12:50] <LordHoto> or what the strange syntax in there is about
[12:50] <LordHoto> files are*
[12:51] <tsoliman> oh I sort of filtered them out
[12:51] <LordHoto> you must've missed all the code parts which do most of the work then :-P
[12:51] <tsoliman> :)
[12:52] <fuzzie> they're also very careful to run stuff on the main thread when needed
[12:52] <fuzzie> which is sort of worrying
[12:52] <LordHoto> maybe, it's to allow the graphics code to be used from threads
[12:52] <tsoliman> the part I was looking at was that gesture code .. I guess I never really looked at all the code
[12:52] <fuzzie> yes, but we don't support graphics code being used from threads :P
[12:52] <LordHoto> but then again I think I saw that the iPhone timer code isn't threaded
[12:53] <fuzzie> oh drat
[12:53] <fuzzie> You know, I didn't think to look at it.
[12:54] <fuzzie> But look, it's doing it in pollEvent() and not in delayMillis().
[12:54] <LordHoto> yup
[12:55] <fuzzie> And thus Dreamweb's "call delayMillis repeatedly until a timer updates our state" wouldn't work, right?
[12:55] <fuzzie> So I have an excuse to kill it with fire?
[12:55] <LordHoto> not unless there's any event polling in between
[12:55] <LordHoto> then again only doing delayMillis without event polling is *really* bad on desktop platforms too :-)
[12:55] <fuzzie> Some of the other backends are careful to call the timer in delayMillis, I guess for this kind of madness.
[12:55] <LordHoto> yeah
[12:55] <fuzzie> Well, the timer is some absurdly fast rate, so. :P
[12:55] <LordHoto> I was wondering whether I make the timer code threaded
[12:56] <LordHoto> but right now I want to make this graphics code a bit more sane
[12:56] <fuzzie> well, I would prefer the timer code all be de-threaded
[12:56] <fuzzie> and also that we error("stop that") any time an engine tries calling it
[12:56] <fuzzie> so probably am bad to ask
[12:56] <LordHoto> it seems the mouse cursor offset is still wrong
[12:56] <LordHoto> at least the modern theme's cursor is off, but it doesn't use any hotspot offset as far as I can tell
[12:57] <LordHoto> so there must be some other bug there...
[12:57] <LordHoto> frankly, I don't think threaded timers are that bad ;-)
[12:57] <fuzzie> are they used legitimately *anywhere*, though
[12:57] <fuzzie> ?
[12:58] <fuzzie> every time I saw them, they were either completely non-thread-safe or just plain unnecessary
[12:58] <LordHoto> I think the SCUMM engine uses it for some audio cd timing, without too much doing non-sane stuff
[12:58] <LordHoto> but that might fall in the unecessary category :-P
[12:59] <fuzzie> well I don't know enough about the scumm engine here, but obviously at a glance it seems unnecessary
[12:59] <fuzzie> also of course completely mutex-less
[12:59] <LordHoto> can't remember much about that ;-)
[12:59] <LordHoto> I only looked briefly at it
[12:59] <fuzzie> yes, everyone has an excuse
[13:00] <fuzzie> having to remove them when they inevitably turn out to be broken is great fun, I love it
[13:00] <LordHoto> I loved that BASS timer usage :-P
[13:00] <fuzzie> so look, Saga is using it for the very important purpose of an fps counter
[13:00] <LordHoto> I think SAGA also uses it to make the options menu icon blink in case you didn't save in some time
[13:00] <LordHoto> or something insane like that
[13:01] <fuzzie> hugo is using it for pc speaker, which is sort of not terrible (after I fixed it)
[13:01] <fuzzie> dreamweb is using it completely pointlessly for this buggy vsync thing
[13:01] <fuzzie> and yes saga is also doing that
[13:01] <LordHoto> I think some MIDI audio code is also using the timers
[13:04] <LordHoto> but the MIDI code in general is one of the code, where a lot of deadlocks might occur, especially when exiting engines :-P
[13:04] <fuzzie> disappointingly saga's hintTimer stuff is only messing with un-mutexed graphics queues, rather than actually doing drawing
[13:05] <LordHoto> tsoliman: I think the C++ code only contains parts of the gesture support though, i.e. those which process some other data generated by the ObjC part :-P
[13:06] <LordHoto> I wonder whether it would make sense to switch the iPhone port to use OpenGL fixed point numbers instead of floats
[13:06] <tsoliman> :(
[13:07] <fuzzie> well, iPhone should be hardfp?
[13:07] <-- qptain_Nemo left irc: Read error: Connection reset by peer
[13:07] --> qptain_Nemo joined #scummvm.
[13:07] <LordHoto> oh it is hardfp, didn't know that, I thought I saw some softfp flags somewhere in the compiler toolchain build parameters
[13:09] <LordHoto> bleh then there isn't some other unimportant changes, I can distract myself from this whole screen coordinate mess :-P
[13:12] <-- qptain_Nemo left irc: Read error: Connection reset by peer
[13:12] --> qptain_Nemo joined #scummvm.
[13:18] <-- pok0j left irc: Remote host closed the connection
[13:19] --> Strangerke joined #scummvm.
[13:19] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services.
[13:37] <tsoliman> android looks like it is compiling now
[13:44] <LordHoto> fuzzie: seems like without the main threading stuff, ScummVM just "crashes"
[13:44] --> Hkz joined #scummvm.
[13:44] #scummvm: mode change '+o Hkz' by ChanServ!ChanServ@services.
[13:44] <LordHoto> fuzzie: so I guess there's some other nice threading going on...
[13:45] <fuzzie> well, does scummvm run in the main thread?
[13:46] <fuzzie> [NSThread detachNewThreadSelector:@selector(mainLoop:) toTarget:self withObject:nil];
[13:46] <fuzzie> ^- since mainLoop calls iphone_main, I am guessing not
[13:47] <LordHoto> ah right
[13:47] <LordHoto> I missed that
[13:47] <fuzzie> well, me too
[13:47] <LordHoto> but that makes me wonder why it doesn't
[13:48] <LordHoto> well one way to find out :-P
[13:49] <LordHoto> hm, doens't work then either
[13:49] <LordHoto> oh well, I'll just keep this threading then :-P
[13:51] <LordHoto> is there any good way to make iOS launch apps after you overwrote them with a different executable, except rebooting the device each time?
[13:51] <fuzzie> don't ask me :-p
[13:52] Action: LordHoto asks fuzzie
[13:53] <LordHoto> seems like we got so many nice GSoC task descriptions btw.
[13:53] <fuzzie> that would be 'none'?
[13:55] <LordHoto> looks like that
[13:57] <fuzzie> did anyone come up with any actual tasks?
[13:58] <wjp> hmm
[14:00] --> pok0j joined #scummvm.
[14:00] <fuzzie> I'm sure I could write a task description or two if anyone came up with a general idea that I didn't think was probably hopeless.
[14:01] <wjp> One or two touch-screen related ones? Properly RE SCI audio? Something AGS-related? ...?
[14:01] <wjp> something-with-SCI32?
[14:03] <Strangerke> I was about to type "/me puts on his Voyeur hat" .... But it sounds weird
[14:05] <fuzzie> wjp: but do we have anyone willing to mentor RE stuff?
[14:05] <fuzzie> it seemed to have gone fine previously
[14:05] <fuzzie> or do we have enough of a plan for SCI32 that it's all known?
[14:06] <wjp> I don't know
[14:07] <wjp> (about SCI32 that is; I could mentor SCI audio)
[14:09] <wjp> although I do wonder if it's big enough
[14:09] --> droid2727 joined #scummvm.
[14:09] <LordHoto> trust me, everything which seems big enough turns out to be much to big usually
[14:09] #scummvm: mode change '+o droid2727' by ChanServ!ChanServ@services.
[14:09] <LordHoto> :-P
[14:10] <tsoliman> what about MT-32 support for TsAGE?
[14:10] <fuzzie> have you *seen* tsage's audio code?
[14:10] <fuzzie> subjecting an innocent student to that might be cruelty
[14:11] <tsoliman> you mean the scummvm tsage audio code or the original tsage audio code?
[14:11] <fuzzie> they are roughly equivalent for these purposes :p
[14:11] <tsoliman> maybe "clean up TsAGE audio code"?
[14:12] --> kettuz joined #scummvm.
[14:12] <tsoliman> I have not seen it no
[14:12] <fuzzie> you're missing out.
[14:15] <LordHoto> maybe we can add some GSoC task to finish the remaining toltecs TODOs?
[14:15] <droid2727> LordHoto: mentor?
[14:16] <LordHoto> dunno!
[14:24] <fuzzie> an AGS task might be possible, but it's kind of annoying and horrible and not something to subject people to
[14:24] <-- droid2727 left irc: Quit: Statement: Good-bye, meatbags!
[14:24] <fuzzie> and no-one's actually seen Wintermute's code, right?
[14:24] --> droid2727 joined #scummvm.
[14:24] <-- droid2727 left irc: Client Quit
[14:26] <LordHoto> wjp: any estimates when your GUI fixes will make it to master?
[14:26] --> droid2727 joined #scummvm.
[14:26] #scummvm: mode change '+o droid2727' by ChanServ!ChanServ@services.
[14:26] Action: droid2727 kicks his phone
[14:27] <droid2727> somehow it got in a mode where hitting 'p' brought up the music player and 'l' would bring up the calendar, etc.
[14:27] <wjp> LordHoto: I think they're pretty much done
[14:28] <LordHoto> good
[14:28] <wjp> I did some initial clean up of the patches already last night
[14:28] <wjp> could you test them on ios?
[14:28] <wjp> fuzzie: and could you test on android, if you have some time?
[14:28] <LordHoto> I'm getting tired of these black pixels on the boarders, missing launcher background, when opening options menu, and ugly graphics glitches with the classic theme :-P
[14:29] <LordHoto> I'm sure I can try them
[14:29] <fuzzie> oh I see, there is *some* wintermute source
[14:29] <wjp> (C[C[C[C[C[C[C4 separate things need testing: modern/classic theme in launcher/in-game-menu)
[14:29] <wjp> uh
[14:29] <CIA-71> Johannes Schickel master * r76be031 / (2 files): IPHONE: Remove some unused variables. - http://git.io/xKPR-A
[14:29] <CIA-71> Johannes Schickel master * r46e622c / backends/platform/iphone/iphone_video.m : IPHONE: Slight cleanup. - http://git.io/AsEpxg
[14:29] <CIA-71> Johannes Schickel master * ra291679 / backends/platform/iphone/iphone_video.m : IPHONE: Slight game screen texture related variable renaming. - http://git.io/3XQ_0g
[14:30] <LordHoto> I'm also sure I can't show you any screens if something is wrong :-P
[14:30] <wjp> missing launcher background when opening options menu?
[14:30] <LordHoto> yeah, the orange background is missing
[14:30] <LordHoto> don't ask me why
[14:31] <LordHoto> everything around the options menu is just some odd color
[14:31] <LordHoto> probably some alpha bug
[14:31] <LordHoto> I hope nothing in the iOS backend
[14:31] <LordHoto> :-P
[14:31] <wjp> this would be the point where I would like to ask for screens anyway :-)
[14:31] <LordHoto> that's with *current* code mind you
[14:31] Action: wjp nods
[14:31] <LordHoto> I might check whether I can reproduce that with the OpenGL code and the alpha hack disabled
[14:39] <Strangerke> For info: I sent a mail to dead code about Wintermute
[14:40] <fuzzie> there's some source code on sf
[14:40] <fuzzie> it has all kinds of not-good looking stuff, like LoadLibrary to get the script compiler
[14:41] Schnaks (bc6dc0be@gateway/web/freenode/ip.188.109.192.190) left #scummvm.
[14:43] <LordHoto> hm why is the iPhone port enabling touchpad mode by default only for non high res devices...
[14:43] <fuzzie> Strangerke: also there's http://code.google.com/p/wmelite/source/browse/#svn%2Ftrunk%2Fsrc (which is MIT)
[14:44] <fuzzie> it seems a bit pointless to add support for a closed-source system though
[14:44] <fuzzie> although AGS is still not quite there
[14:46] <Strangerke> that's from the Lite version
[14:47] <Strangerke> they launched a project about making portable the 2D games by rewriting the thing
[14:47] <Strangerke> But it looks dead
[14:47] <fuzzie> less dead than Wintermute, though?
[14:48] <Strangerke> Not so sure
[14:49] <Strangerke> Would a task about the speech generation in Mortevielle be something interesting for a student?
[14:49] <fuzzie> only someone who knows it can say
[14:50] <Hkz> Strangerke, nah, we just link festival speech synth into ScummVM and be done ;-)
[14:51] <Strangerke> Hum...
[14:54] <tsoliman> LordHoto: probably because it assumes that the ipad is the only high res device
[14:54] <LordHoto> well
[14:55] <LordHoto> I guess it uses touchpad mode on the low res device, because it's horrible to click anything in the GUI otherwise
[14:55] <tsoliman> yeah that's what I meant .. but it is a bad assumption since iphone4 is higher res than ipad
[14:55] <LordHoto> sure
[14:55] <fuzzie> does that matter?
[14:56] <fuzzie> it's about dpi rather than res, no?
[14:56] <tsoliman> exactly
[14:56] <tsoliman> it should be measuring DPI not res
[14:56] <LordHoto> maybe
[14:56] <LordHoto> nothing to worry about for me right now :-P
[14:57] <fuzzie> the scummvm gui is pretty unusable on a high-dpi device
[14:57] <LordHoto> well, it's even unusable on the iPhone 3G I have
[14:57] <LordHoto> except for touchpad mode, there it's working somewhat fine
[14:58] <LordHoto> but it's still not a nice way to use it...
[14:58] <tsoliman> fuzzie: unless you use a stylus .. I am happy with maemo (same dpi as iphone) because of this
[14:58] <fuzzie> the iPhone 3G is 163ppi, that's pretty high
[14:58] <fuzzie> tsoliman: yes, I am admittedly ignoring stylus devices
[14:58] <tsoliman> of course the maemo stuff is all resistive so it is not exactly the same experience
[14:59] <LordHoto> it's also awesome that sometimes when trying to open the GMM it just closes agian
[14:59] <LordHoto> again*
[14:59] <tsoliman> I thought I fixed that
[14:59] <LordHoto> it seems there's some stray events added, which might accidentially hit the the Resume button
[14:59] <LordHoto> sometimes other menus are opened too
[14:59] <fuzzie> the galaxy tab 8.9 is about 170dpi and the GUI is only usable because the android backend does 2x scaling for it
[15:02] <tsoliman> LordHoto: what revision is your origin/master?
[15:02] <LordHoto> we definitly need either a platform specific GUI for iOS and Android or we need some real good custom layout for touchscreen devices :-P
[15:02] <LordHoto> tsoliman: up to date with some changes on top
[15:02] <LordHoto> tsoliman: but it's not related to any of your changes
[15:03] <tsoliman> oh yeah right .. you're not using F5 anymore either
[15:03] <LordHoto> when the mouse cursor is on a button spot and I do the gesture to open the menu, the button is automatically clicked
[15:03] <tsoliman> right because it is keying off of keydown
[15:03] <LordHoto> I guess there's some press event going on
[15:03] <tsoliman> not keyup
[15:04] <LordHoto> what?
[15:04] <tsoliman> keydown is enough to trigger buttons
[15:04] <LordHoto> ah you mean the handling is done on keydown
[15:04] <LordHoto> might be
[15:04] <tsoliman> yeah which sucks
[15:04] <tsoliman> this approach is done all over the place
[15:05] <tsoliman> so not necessarily in any specific one place
[15:06] <LordHoto> frankly, it rather sucks that there's some events sent after the gesture, which seem still to belong to the gesture :-P
[15:06] <tsoliman> the GUI buttons are on click-up only
[15:06] <LordHoto> well then it's clearly the backend's fault
[15:07] <tsoliman> well the GUI is sorta buggy as well .. if you click outside the button .. drag .. and "mouseup" on the button .. it activates
[15:07] <fuzzie> yeah, if the backend is leaking events from a gesture then that's surely a backend issue
[15:07] --> Kaidane joined #scummvm.
[15:08] <-- cyco left irc: Quit: cyco
[15:09] --> cyco joined #scummvm.
[15:14] Nick change: jweinberg -> A
[15:14] Nick change: A -> Guest8670
[15:15] Nick change: Guest8670 -> jweinberg
[16:03] <-- Strangerke left irc: Quit: Bbl
[16:34] <-- Kaidane left irc: Ping timeout: 240 seconds
[16:39] --> Kaidane joined #scummvm.
[16:41] jweinberg (jweinberg@unaffiliated/jweinberg) left #scummvm ("[Textual IRC Client: http://www.textualapp.com/]").
[16:46] <-- Kaidane left irc: Quit: Leaving
[16:52] katycorp (kaitlin@208.58.6.57) left #scummvm.
[16:53] <CIA-71> Tarek Soliman master * r6c4d65e / backends/keymapper/remap-dialog.cpp : (log message trimmed)
[16:53] <CIA-71> KEYMAPPER: Change how keymaps are displayed in the remap dialog
[16:53] <CIA-71> This makes it so that there's 2 groups:
[16:53] <CIA-71> 1 - The keymap stack
[16:53] <CIA-71> 2 - The loaded keymaps
[16:53] <CIA-71> The first group displays which keymaps are active and which keymaps
[16:53] <CIA-71> are completely blocked due to an opaque keymap on top.
[16:55] --> Strangerke joined #scummvm.
[16:56] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services.
[17:28] <tsoliman> I just discovered a game called Ripper .. FMV .. interesting cast :)
[17:29] <Strangerke> tsoliman: Do you like it? Last ime i checked it, I found it quite ugly and ininteresting
[17:29] <tsoliman> I am watching a sort of quick-look
[17:30] <LordHoto> Strangerke finds a game ugly and uninteresting?
[17:30] <LordHoto> boy that game must suck :-P
[17:30] <Strangerke> LordHoto: Or the opposite ;)
[17:30] <tsoliman> Strangerke: how does it compare to Urban Runner?
[17:30] <Strangerke> My opinion is biased
[17:31] <tsoliman> right .. I've seen both so I know the answer .. I wanted to know your opinion :P
[17:31] <Strangerke> After Fascination and Bargon Attack, Urban Runner is the gob title I prefer
[17:32] <CIA-71> Johannes Schickel master * r7957cc9 / backends/platform/iphone/iphone_video.m :
[17:32] <CIA-71> IPHONE: Use render buffer size instead of application frame size for video size calculations.
[17:32] <CIA-71> These match as far as I can tell, but in case they don't match, the render
[17:32] <CIA-71> buffer size should be the correct thing to use. (+5 more commits...) - http://git.io/RbTlLQ
[17:32] <tsoliman> I actually have Urban Runner and haven't played through it yet
[17:37] <droid2727> Guys, we shouldn't make fun of Ripper when DrMcCoy is around :P
[17:37] <DrMcCoy> I love Ripper!
[17:37] <DrMcCoy> It got Christopher Walken!!!!
[17:38] <DrMcCoy> "Dis guy is unfuckingbelievable"
[17:38] <droid2727> censor yourself, there could be children reading this!
[17:39] <DrMcCoy> ...
[17:39] <LordHoto> right droid2727 is around, don't use adult speach please :-P
[17:39] --> bgKa joined #scummvm.
[17:39] #scummvm: mode change '+o bgKa' by ChanServ!ChanServ@services.
[17:39] Action: droid2727 slaps LordHoto
[17:39] --> WooShell joined #scummvm.

[17:42] <tsoliman> It has a ton of other folks too
[17:42] <droid2727> yes, but Christopher Walken is DrMcCoy's idol
[17:42] <droid2727> (after William Shatner)
[17:43] --> SylvainTV joined #scummvm.
[17:43] #scummvm: mode change '+o SylvainTV' by ChanServ!ChanServ@services.
[17:43] <DrMcCoy> I'm not that much of a Shatner fanboi
[17:43] <droid2727> that's what they all say
[17:43] <DrMcCoy> I actually prefer Picard to Kirk :P
[17:43] <DrMcCoy> And Sisko to Picard
[17:43] <droid2727> and archer trumps all?
[17:44] <tsoliman> who's the doctor in the next generation?
[17:44] <droid2727> Crusher or Pulaski
[17:45] <tsoliman> Almost everyone in the ripper is now a famous person
[17:46] <tsoliman> compare to GK2 :P
[17:46] --> Kaidane joined #scummvm.
[17:46] <droid2727> hey, that old guy from the GK2 intro was in a Seinfeld episode, so psssssssssh
[17:47] <tsoliman> the guy who plays Gabriel was the waiter
[17:47] <tsoliman> but then he became a CEO of some scam company
[17:48] <tsoliman> s/scam/investment advisory/
[17:49] <tsoliman> he did Frasier *before* GK2
[17:49] --> Mataniko joined #scummvm.
[17:49] <tsoliman> http://www.imdb.com/name/nm0259039/
[17:50] <tsoliman> I actually saw some of the actors from the German part of GK2 (the big bad and the cop) in other TV shows
[17:50] <tsoliman> they're more "extras" that stars though
[17:57] <tsoliman> DrMcCoy: what engine is it using? it looks a lot like the 7th guest
[17:57] <droid2727> director
[17:57] <droid2727> (I think)
[17:57] <DrMcCoy> What, Ripper?
[17:57] <tsoliman> yes
[17:57] <DrMcCoy> I have no clue
[17:58] <tsoliman> director as in macromedia?
[17:58] <droid2727> i know of no other
[17:59] <tsoliman> didn't realize people made games with it
[17:59] <fuzzie> o.O
[17:59] <tsoliman> I have bad memories of authoring stuff with it
[17:59] <fuzzie> there are an awful awful lot of games using it :p
[17:59] <DrMcCoy> I don't know about Ripper, but The Dark Eye is Director for sure
[17:59] <-- _sev left irc: Quit: This computer has gone to sleep
[18:00] <droid2727> ah, it was DOS, so I take it back
[18:00] <fuzzie> tsoliman: so, what does 'Effective' mean?
[18:00] <droid2727> tsoliman: You didn't know Director was used for games? O_o
[18:01] <tsoliman> fuzzie: if you pick the topmost keymap .. the keymapper shows its actions but it also shows inherited actions from keymaps below it
[18:01] --> _athrxx joined #scummvm.
[18:01] #scummvm: mode change '+o _athrxx' by ChanServ!ChanServ@services.
[18:01] <fuzzie> tsoliman: am asking because it needs a comment for translations
[18:01] <fuzzie> tsoliman: but also asking because I can't see where on earth the code does that :P
[18:02] <tsoliman> I guess it can be translated as "primary" or "main"
[18:02] <tsoliman> or "most active"
[18:02] <tsoliman> remap dialog
[18:02] <fuzzie> yes, I know that much ;p
[18:02] <fuzzie> but the terminology in general is a bit confusing
[18:03] <tsoliman> I know .. I want to come up with better terms
[18:03] <fuzzie> but I am just passing on the request re: translations, so comment would be nice if possible
[18:04] <tsoliman> ok I will add some comments
[18:04] <tsoliman> fuzzie look at the if block starting at line 316
[18:05] <tsoliman> where it says // loop through remaining finding mappings for unmapped keys
[18:05] <fuzzie> ok
[18:05] <fuzzie> so 'load active keymaps' = 'display all effective actions, including inherited ones'?
[18:06] <tsoliman> don't look at it too much :P
[18:07] <tsoliman> it should really say something like //populate _currentActions
[18:07] <tsoliman> but it was here from before and I plan to rewrite this mess
[18:08] Nick change: Vampire0_ -> Vampire0
[18:09] <tsoliman> I will fix the comments now at least
[18:12] <fuzzie> sigh, I just spent a while trying to work out how on earth AGS's String class worked
[18:12] --> criezy joined #scummvm.
[18:12] #scummvm: mode change '+o criezy' by ChanServ!ChanServ@services.
[18:13] <fuzzie> when I notice that it is annotated with 'internalstring autoptr' in the editor
[18:13] <fuzzie> which is presumably some magic meaning "don't actually use a class here"
[18:15] <fuzzie> yes indeed it is. sigh.
[18:18] <-- Mataniko left irc: Ping timeout: 240 seconds
[18:18] <CIA-71> Tarek Soliman master * r56e517f / backends/keymapper/remap-dialog.cpp : KEYMAPPER: Add some comments - http://git.io/RQJGxg
[18:19] <fuzzie> thanks
[18:20] <tsoliman> np .. the whole things needs a rewrite
[18:21] --> sirlemonhead joined #scummvm.
[18:29] <criezy> Oh I am already connected :O
[18:29] <criezy> Good evening
[18:29] <-- SylvainTV left irc: Ping timeout: 252 seconds
[18:29] --> Sylvain joined #scummvm.
[18:30] <-- Kaidane left irc: Quit: Leaving
[18:30] <criezy> tsoliman: thanks for the comment
[18:30] <criezy> but I still don't see the need to differentiate between the primary effective most active keymap and the inherited ones for the user
[18:31] <fuzzie> did you look at the loadKeymap comment?
[18:32] --> Mataniko joined #scummvm.
[18:32] <fuzzie> there must be a better word for it, but I can't think of one right now
[18:33] <tsoliman> primary?
[18:33] <criezy> Oh sorry I missed that comment. So effective includes the inherited mappings?
[18:33] <fuzzie> that doesn't make much sense either
[18:33] <fuzzie> criezy: yes
[18:33] <tsoliman> criezy: from what the user sees, when the topmost keymap is selected, a special view is activated where you see more actions at the bottom (with grey-ed out labels)
[18:33] <criezy> But for what is the Active label used then?
[18:34] <tsoliman> I think a screenshot is better for this :) give me a min
[18:35] <fuzzie> criezy: all the keymaps which are in use, and thus might be inherited from
[18:35] <fuzzie> i think 'Active' and 'Enabled' makes more sense, honestly
[18:35] <fuzzie> instead of 'Effective' and 'Active'
[18:35] <fuzzie> 'operative' doesn't really work either..
[18:36] <CIA-71> Johannes Schickel master * rcf44d49 / (2 files): IPHONE: Precalculate vertex/tex coordinates for game screen/overlay. - http://git.io/SZrJ4w
[18:37] <criezy> So you have active and blocked keymaps and the effective mapping is build from all the active ones?
[18:37] <fuzzie> yep
[18:38] <tsoliman> criezy: http://bashasoliman.com/dump/screenshot7.png http://bashasoliman.com/dump/screenshot8.png
[18:38] <tsoliman> in the first image you are in LoL
[18:38] <tsoliman> in the second image you are in LoL but the GMM is displayed
[18:38] <fuzzie> okay that makes even less sense in the screenshots :-p
[18:39] <tsoliman> ignore what's below the separator in the screenshot
[18:39] <fuzzie> 'Active', 'Enabled', 'Unused'?
[18:39] <tsoliman> eveything below the separator is just a dump of all the known ones
[18:40] <tsoliman> above the separator the top one is always called "Effective" because it is the active one
[18:40] <fuzzie> but it's not the active one it's the effective one! :-p
[18:40] <tsoliman> I mean it is Active but in the effective view
[18:40] <tsoliman> I should rename this to be like
[18:40] <tsoliman> lol (Active) - Effective View
[18:41] <tsoliman> and then the ones below it are global (Active)
[18:41] <tsoliman> Effective is really describing a special view
[18:41] <tsoliman> not a kind of keymap
[18:42] <tsoliman> and re: unused .. I don't attempt to filter out "used vs unused" below the separator
[18:42] <fuzzie> yes, but you have 'Blocked' in the second sshot
[18:42] <tsoliman> yeah because the GUI blocks all below it
[18:42] <tsoliman> it is not transparent
[18:42] <fuzzie> i mean, i guess my point is: this terminology makes no sense
[18:42] <tsoliman> agreed
[18:42] <criezy> I agree. And I am still confused...
[18:42] <fuzzie> and if you use 'Active' for the topmost, 'Enabled' for the active ones and 'Unused' for the blocked ones, at least it makes sense to me at a glance
[18:42] <tsoliman> but now you understand the situation and are welcome to suggest stuff :)
[18:43] <fuzzie> although I am not suggesting that really
[18:43] <criezy> In the top inage does the 'lol (Effective)' includes inherited mapping from the 'Global (Active)' ?
[18:43] <tsoliman> yes
[18:43] <tsoliman> I didn't scroll down
[18:43] <tsoliman> but I should have
[18:44] <tsoliman> let me show you more screenshots (hopefully to *remove* some confusion instead of adding)
[18:45] <tsoliman> http://bashasoliman.com/dump/screenshot9.png
[18:45] <tsoliman> the global keymap has 8 or so actions
[18:45] <tsoliman> but only 2 of them are visibile
[18:46] <tsoliman> 4 of the 8 are bound to keys
[18:46] <tsoliman> but not all 4 are visible because some keys are "overridden" by the lol keymap
[18:46] --> Javacat joined #scummvm.
[18:46] <tsoliman> the 2 that are visible are visible because they are "reachable"
[18:47] <criezy> I think I get it now.
[18:47] <tsoliman> the way the keymapper works is that it looks at this stack of keymaps (that you see above the separator)
[18:47] <tsoliman> and stops when it finds an action that has it
[18:47] <tsoliman> but then some keymaps (like GUI) sort of block all keymaps below them
[18:47] <tsoliman> because they aren't transparent
[18:48] <criezy> But calling it 'lol (something)' is a bit confusing because there is more than the lol mapping there.
[18:48] <tsoliman> right
[18:48] <tsoliman> I am open to suggestions
[18:48] <tsoliman> it should be lol (Active) + Visible Actions?
[18:49] <criezy> Wouldn't it be better to have one 'Effective Mapping' entry and then one 'Active' or 'Inactive' entry for each keymap (lol, gui, global)?
[18:49] --> _sev joined #scummvm.
[18:49] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[18:49] <tsoliman> sure I can do that
[18:50] <tsoliman> but what would the topmost entry (effective) have?
[18:50] <tsoliman> just what is has now?
[18:50] <-- cyco left irc: Quit: cyco
[18:50] <tsoliman> (meaning all actions of the topmost keymap + reachable mapped actions from the ones below)?
[18:50] <fuzzie> well, that makes sense as an overview
[18:51] <criezy> Yes.
[18:51] <tsoliman> ok cool
[18:51] <criezy> And the lol one would only have the lol mapping without the inherited ones
[18:51] <tsoliman> ok
[18:52] <tsoliman> so the topmost one wouldn't mention lol at all?
[18:53] <tsoliman> I mean what label should it have?
[18:53] <criezy> Correct. I would not used lol in the label since it is only one of the keymap used (even though it is the primary one).
[18:54] --> Kaidane joined #scummvm.
[18:54] <criezy> But maybe some people would prefer to have an indication what the primary keymap is.
[18:56] <criezy> Thus my suggestion 'Effective Mapping' or maybe just 'Effective'.
[18:59] <criezy> You could have the 'Effective' for the first item, then 'lol (Primary Active)' then 'Global (Active)'...
[19:00] <tsoliman> I tried implementing separators with text on them
[19:00] <LordHoto> All that makes me wonder: Why do we expose all the layouting to the user anyway?
[19:01] <fuzzie> what else would you do?
[19:01] <fuzzie> I mean, the user wants to be able to edit different layers, right..?
[19:02] <tsoliman> before I started it was only 2 things .. the effective view .. and then a list of known keymaps
[19:02] <LordHoto> well quite frankly, I would've thought we want to allow editing non-active keymaps in the future too
[19:02] <tsoliman> this made it unclear what was active and what wasn't
[19:02] <fuzzie> you know it occurs to me that if engines register their own keymaps, then the options dialog is going to have a painful time getting it
[19:02] <-- Vampire0 left irc: Ping timeout: 240 seconds
[19:02] <fuzzie> someone tell me that you already thought of that and it's all solved
[19:02] <tsoliman> fuzzie: it is already happening and working
[19:03] <tsoliman> the lol keymap is registered inside the engine
[19:03] <LordHoto> yeah but you can ever change it, when it's active
[19:03] <LordHoto> which is not that nice ;-)
[19:03] <tsoliman> I don't understand
[19:03] <fuzzie> tsoliman: how does 'inside the engine' help?
[19:04] <fuzzie> i mean that sounds like *not* good
[19:04] <LordHoto> well some short recap: what's displayed currently?
[19:04] <LordHoto> all registered keymaps + the current stackthingy?
[19:04] <tsoliman> + an "effective view"
[19:05] <fuzzie> and indeed, initKeymap is on Engine, so it's useless
[19:05] <fuzzie> so I take it I can not be reassured that anyone thought about this :P
[19:05] <LordHoto> fuzzie: well that could be changed
[19:05] <tsoliman> LordHoto: http://bashasoliman.com/dump/screenshot8.png
[19:05] <fuzzie> LordHoto: well I don't really care. I will just cry.
[19:05] <tsoliman> fuzzie: what is the case you're trying to do?
[19:06] <LordHoto> fuzzie probably wants a sane feature, like editing a keymap before you start the game
[19:06] <fuzzie> tsoliman: I open my shiny futuristic ScummVM UI, I tap the game, I tap 'keymapping', and there's no keymap because the engine hasn't been instantiated.
[19:06] <fuzzie> yes, that.
[19:06] <tsoliman> ugh
[19:06] <tsoliman> fuzzie: I see your point
[19:07] <LordHoto> quit frankly, all the stack etc. doesn't make this nicely possible
[19:07] <LordHoto> since you never know which combination might end up on the stack and thus whatever map conflicts you might run into
[19:07] <tsoliman> and the current answer is a cop-out .. let the user figure it out
[19:07] <LordHoto> one of the reasons why I still dislike the stack :-P
[19:08] <tsoliman> in fact the stack's strength which was to add a maemo keymap on top of global .. fails miserably when LoL pushed the global keymap in its save dialog
[19:08] <fuzzie> this all sounds terrible.
[19:08] <LordHoto> I mean inherited keymaps could be modelled explicitly, by just adding a parent to a keymap
[19:09] <tsoliman> LordHoto: yes and there's already a parent keymap concept that I fail to understand
[19:09] <tsoliman> it has to do with the broken automapper code
[19:09] <LordHoto> of course maybe some games require uberflexibility like the stack would feature, but that will be quite some fun for the users out there, who then need to sort out possible conflicts on their own
[19:09] <tsoliman> it sounds very simple if you dont think about it
[19:10] <fuzzie> well
[19:10] <tsoliman> I mean sev has some really interesting use cases
[19:10] <fuzzie> you need more than one keymap in the stack right?
[19:10] <tsoliman> and the stack makes perfect sense there
[19:10] <LordHoto> like?
[19:10] <fuzzie> because you need the backend one, obviously
[19:10] <tsoliman> there's always a global one yeah
[19:10] <fuzzie> but I don't understand how this works so I don't understand if that is included
[19:10] <LordHoto> well if you make the backend one the global one and you have some fancy way to inherit from that, you don't need a stack ;-P
[19:11] <fuzzie> oh/
[19:11] <fuzzie> ?
[19:11] <tsoliman> yeah exactly
[19:11] <fuzzie> i mean, you're going to want 'backend' then 'game' then game-specifics on top, although actually I hate that idea due to being too complex
[19:11] <tsoliman> there just needs to be a "current" one
[19:11] <LordHoto> I just want a backend and frontend one really
[19:11] <fuzzie> it's kind of tricky due to in-game UIs and the such
[19:12] <tsoliman> the stack solves the problem of "where was I before this"
[19:12] <tsoliman> so I say keep the stack
[19:12] <tsoliman> but dont use inheritance as stack transparency
[19:12] <tsoliman> meaning everything on the stack is opaque
[19:12] <fuzzie> but then is the stack good?
[19:12] <tsoliman> yeah it lets you pop back to a previous state
[19:12] <LordHoto> right, but it also brings the nice joy of "how do I handle mapping conflicts?"
[19:13] <tsoliman> and inheritance is like what LordHoto said ... _parent
[19:13] <LordHoto> I mean if the game secretly pushes keymaps with conflicts on the stack, will the dialog be displayed, so the user can resolve it?
[19:13] <fuzzie> seems like you should never be in a position where the stack can't be automatically recreated
[19:13] <LordHoto> quit frankly, we can have a stack for saving the current frontend keymap and maybe pushing custom ones on top, but doing inheritance through that... I don't know ;-)
[19:13] <LordHoto> quite*
[19:14] <fuzzie> anyway
[19:14] <LordHoto> I mean the GUI is definitly the big use case for that
[19:14] <fuzzie> while you are doing boring stuff
[19:14] <tsoliman> when you say conflicts you mean input conflicts yeah not output conflicts.. I mean two HardwareActions that are conflicting
[19:14] <fuzzie> I think I worked out how AGS's dynamic objects work!
[19:14] <-- kettuz left irc: Quit: Leaving
[19:14] <LordHoto> if we bring op our GUI it should push its own keymap and afterwards restore the old one
[19:14] <fuzzie> You have a *real* object, which is stored inside a *dynamic* object, and then you have a global mapping with *handles* which reference both.
[19:14] <LordHoto> tsoliman: well the user can't fix action conflicts anyway
[19:14] <fuzzie> And then you have a Dispose(realobject*) on the dynamic object which deletes the real object and then does 'delete this;'.
[19:14] --> Vampire0 joined #scummvm.
[19:14] <LordHoto> fuzzie: sounds sane
[19:15] <fuzzie> you're volunteering to code it now, right?
[19:21] <criezy> I have a question regarding demos: do we want any demos for the supported games or only those that actually use the engine of the game?
[19:21] <criezy> I am asking this because I found a bunch of SCI game 'demos' on an old CD but they are actually QuickTime videos :P
[19:23] <criezy> I also found a few SCUMM mac demos. They are the same than the ones on the demo website except that I have the original executable while it is missing (or incomplete in one case) in the zip files from the demo website.
[19:23] <criezy> Is there any value in having the original executable?
[19:24] --> cyco joined #scummvm.
[19:33] <-- bgKa left irc: Ping timeout: 240 seconds
[19:35] <fuzzie> criezy: engine only I think
[19:35] <fuzzie> also I personally wanted the mac demo binaries and was not pleased to find them removed
[19:36] <fuzzie> so I would appreciate complete demos
[19:36] <sirlemonhead> hmm. i have some classes created with their own .h and .cpp files, as you do. and i'm creating instances of them within the main engine class. i'm including the headers for said classes within the main engine header
[19:36] <sirlemonhead> some of my instances are fine declared like normal, ie "Display *_display;"
[19:36] <sirlemonhead> but three of them require I put "class" in front of them, ie "class Inventory *_inventory;"
[19:37] <sirlemonhead> visual studio complains i'm missing a type specifier otherwise. any idea why that might be happening?
[19:37] <Mataniko> because VS warnings are weird
[19:37] <Mataniko> :)
[19:37] <sirlemonhead> it's an actual error, i cant compile until i add the 'class'
[19:38] <fuzzie> sirlemonhead: you're not including the headers for those classes before your engine class, presumably
[19:39] <sirlemonhead> i have all the includes in the same place, in my engine header, after the include guard, before the namespace declaration and before the engine class
[19:39] <fuzzie> and those headers aren't including the engine header?
[19:40] <sirlemonhead> they are
[19:40] <fuzzie> then there is your answer :)
[19:40] <fuzzie> *one* of them has to come first
[19:40] <sirlemonhead> oh yeah mean because ive included "scalpel.h" (my engine name) within the header for the class i'm trying to use within the engine class?
[19:40] <fuzzie> yep
[19:41] <fuzzie> usually people do a forward definition (e.g. 'class ScalpelEngine;') instead, if the header can be avoided
[19:41] <sirlemonhead> i'm doing that for one of the classes that DOESN'T require the 'class' type specification, so i didnt see it
[19:44] <droid2727> criezy: we usually like to package the demos with the binary if possible
[19:44] <droid2727> criezy: I ripped all my Mac SCUMM demos from the disc, so I never noticed the binaries missing from the packages online. which are missing?
[19:45] <criezy> Sam&Max and The Dig executable are missing.
[19:45] <criezy> Also DOTT executable is half the size of mine.
[19:52] <-- ajf|offline left irc: Ping timeout: 245 seconds
[19:55] --> chkr joined #scummvm.
[19:58] --> ajf|offline joined #scummvm.
[19:59] Nick change: ajf|offline -> Guest60087
[20:00] <-- James|GlideM left irc: Read error: Connection reset by peer
[20:05] <criezy> And full Throttle executable is also missing.
[20:05] <criezy> Should I replace the demos on the website by mine directly or upload somewhere else first so that the zip files can be checked?
[20:05] Nick change: Adys_ -> Adys
[20:16] <CIA-71> Johannes Schickel master * rd8531a9 / backends/platform/iphone/iphone_video.m : IPHONE: Move OpenGL ES context generation to its own method. - http://git.io/UB4daQ
[20:16] <CIA-71> Johannes Schickel master * r6cc5b83 / backends/platform/iphone/iphone_video.m : IPHONE: Default to portrait mode in case the orientation could not be determined. - http://git.io/urRqxA
[20:22] --> LunaVorax joined #scummvm.
[20:32] <-- _athrxx left irc: Read error: Connection reset by peer
[20:32] [md5] --> (~md5@ppp-2-84-27-136.home.otenet.gr) joined #scummvm.
[20:32] [md5] <-- (~md5@ppp-2-84-27-136.home.otenet.gr) left irc: Changing host
[20:32] [md5] --> (~md5@unaffiliated/md5/x-729473) joined #scummvm.
[20:32] #scummvm: mode change '+o [md5]' by ChanServ!ChanServ@services.
[20:32] --> _athrxx joined #scummvm.
[20:32] #scummvm: mode change '+o _athrxx' by ChanServ!ChanServ@services.
[20:33] <-- tsoliman left irc: Quit: bbl
[20:55] <-- KingGuppy left irc: Remote host closed the connection
[20:55] --> KingGuppy joined #scummvm.
[20:59] <-- D0SFreak left irc: Ping timeout: 252 seconds
[20:59] <-- KingGuppy left irc: Read error: Connection reset by peer
[20:59] <wjp> LordHoto, fuzzie: updated gui_rounded_rect branch with AAed tab corners, and some more cleanup
[20:59] --> KingGuppy joined #scummvm.
[21:00] <fuzzie> can you not just commit it? :P
[21:00] <wjp> I think I did everything I wanted to
[21:00] <-- DrMcCoy left irc: Quit: Server migration
[21:00] <LordHoto> wjp: that sounds like the fun part starts now, i.e. doing the things you didn't want to do :-P
[21:01] <wjp> fixing the annoying things I missed and only show up on ports with weird graphics formats? :-)
[21:01] --> DrMcCoy joined #scummvm.
[21:01] #scummvm: mode change '+o DrMcCoy' by ChanServ!ChanServ@services.
[21:02] <LordHoto> wjp: well I can make the OpenGL backend use the same format as iOS!
[21:02] <fuzzie> do we have ports *without* weird graphics formats?
[21:02] <fuzzie> I guess not, since the overlay mandates this weird 16bpp thing.
[21:02] <wjp> LordHoto: could you reproduce the weird glitches with the opengl backend?
[21:03] <LordHoto> frankly I didn't try yet
[21:03] <LordHoto> but let me do that now
[21:03] <LordHoto> shouldn't take long to hack the code :-P
[21:05] <LordHoto> well I can reproduce the missing launcher in the background when opening the options menu
[21:05] <LordHoto> but that's obviously because the blending isn't right
[21:05] <LordHoto> I can also reproduce odd artifacts on corners (same reason)
[21:05] <LordHoto> so if that's gone for you, I'm quite hopeful that it's also gone with iOS
[21:07] <LordHoto> wjp: so if you're done with your branch, please make our GUI usable for touchscreen devices. kthxbye
[21:07] Action: wjp mumbles something about GSoC
[21:08] Action: LordHoto mumbles something about no task description for that yet
[21:10] <wjp> what do you think about this commit? https://github.com/wjp/scummvm/commit/6ef7e4d433fca7436af128f5049fb2cf6ef4557b
[21:10] <-- pok0j left irc: Remote host closed the connection
[21:10] <wjp> I basically added it because scrolling to the right for the theme/renderer options was driving me crazy
[21:12] --> pok0j joined #scummvm.
[21:12] <LordHoto> I don't know what it looks like to be honest ;-), but in case all the tab button descriptions still fit, why not?
[21:12] <LordHoto> (and it might be wise to also check translations *cough*)
[21:14] <wjp> I'm only making the tabs smaller
[21:14] <LordHoto> well I can say for sure on my iPhone the graphics tab already only says "raphics" or something like that
[21:15] <wjp> ah
[21:15] <LordHoto> but, that's because it doesn't properly use the low res context I guess
[21:15] <LordHoto> it should display "GFX" there, IIRC
[21:15] <LordHoto> don't ask me why though :-P
[21:17] <CIA-71> Willem Jan Palenstijn master * r66d3def / (2 files): (log message trimmed)
[21:17] <CIA-71> GUI: Implement AAed tabs
[21:17] <CIA-71> The stroke effect isn't properly implemented yet since it isn't used
[21:17] <CIA-71> in the modern theme.
[21:17] <CIA-71> There are slight bleeding colours around the corners of the
[21:17] <CIA-71> tab_background when its corners overlap with the corners of the
[21:17] <CIA-71> underlying dialog. These are hard to avoid because the underlying dialog
[21:18] <wjp> there
[21:20] <LordHoto> it looks like you had some fun time with it
[21:21] <LordHoto> I'll test it soon on iOS
[21:21] <wjp> thanks
[21:21] <LordHoto> right now, I'm still doing some other changes :-P
[21:21] <wjp> I won't be fixing any more bugs tonight anyway :-)
[21:22] <wjp> (well, I suppose just maybe if buildbot complains...)
[21:22] <LordHoto> pff, if there's a bug, you will fix it!
[21:23] --> D0SFreak joined #scummvm.
[21:24] <-- LunaVorax left irc: Ping timeout: 245 seconds
[21:27] <wjp> oh look, build errors
[21:30] Action: LordHoto makes wjp fix it
[21:30] <fuzzie> fun
[21:30] <fuzzie> did someone not fix the android build yet?
[21:31] <CIA-71> Willem Jan Palenstijn master * rd0c655f / graphics/VectorRendererSpec.cpp : GUI: Fix mingw-w64 build errors - http://git.io/YlqqbQ
[21:31] <fuzzie> also, alas, we have no invasive shared pointer I guess?
[21:32] <LordHoto> wjp: seems like the shadow of the options dialog is wrong
[21:32] <LordHoto> wjp: it seems to remove some text from the launcher list here
[21:33] <LordHoto> wjp: can't say whether you broke it though, since before I couldn't see that at all :-P
[21:33] <ScummBot> Port build status changed with 66d3def0: Failure: master-mingw-w64
[21:34] <LordHoto> wjp: apart the issues seem fixed though, yay, thanks
[21:34] <wjp> modern, classic, platform?
[21:34] <LordHoto> modern
[21:34] <LordHoto> iOS
[21:34] <LordHoto> i.e. 480x320 overlay :-P
[21:35] <LordHoto> if you can hack the OpenGL backend to use that mode (probably just change the initSize call of the launcher to that resolution) you might be able to reproduce it
[21:36] <wjp> oh, interesting
[21:36] <fuzzie> :q
[21:36] <fuzzie> oh.
[21:36] <fuzzie> well stop being here, #scummvm.
[21:37] <wjp> it's LordHoto's fault. I was planning for #scummvm to be gone, but now he found a bug
[21:38] <wjp> LordHoto: was already there
[21:38] <LordHoto> guessed so
[21:38] <LordHoto> it's probably restoring too much background and since text isn't back buffered it's gone
[21:38] <wjp> yes
[21:38] <LordHoto> or something like that
[21:38] <LordHoto> the joy of our GUI
[21:38] <LordHoto> maybe someone can rework the whole rendering pipeline one day
[21:39] <LordHoto> fuzzie: maybe the android buildbot builder wasn't told we fixed it!
[21:39] <fuzzie> did you?
[21:39] <LordHoto> wjp: at any rate the GUI looks much nicer now
[21:39] <LordHoto> I didn't connect to buildbot at all today
[21:39] <wjp> great :-)
[21:40] <LordHoto> there's some odd other bug though
[21:40] <LordHoto> probably caused by the nice triple buffering...
[21:40] --> LunaVorax joined #scummvm.
[21:40] --> Kaidane_ joined #scummvm.
[21:40] <LordHoto> but that's not GUI related
[21:41] <wjp> right, mingw-w64 is happy again
[21:41] <-- Kaidane left irc: Ping timeout: 240 seconds
[21:48] <-- MetalSnake left irc: Quit: MetalSnake
[21:48] <-- ny00123 left irc: Ping timeout: 260 seconds
[21:52] --> ny00123 joined #scummvm.
[21:55] --> James|GlideM joined #scummvm.
[21:55] #scummvm: mode change '+v James|GlideM' by ChanServ!ChanServ@services.
[21:56] <ScummBot> Port build status changed with d0c655f1: Success: master-mingw-w64
[22:00] <-- _sev left irc: Read error: Connection reset by peer
[22:00] --> _sev joined #scummvm.
[22:00] <-- _sev left irc: Changing host
[22:00] --> _sev joined #scummvm.
[22:00] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[22:04] <-- ny00123 left irc: Quit: Leaving
[22:04] <eriktorbjorn> LordHoto: Had any time to look at the possible patch I attached to bug #3482459 ("GUI: Clear buttons not properly updated when switching theme")?
[22:04] <LordHoto> eriktorbjorn: well... let's say I totally forgot about that :-P
[22:13] --> _sev_ joined #scummvm.
[22:13] <-- _sev left irc: Read error: Connection reset by peer
[22:13] <-- _sev_ left irc: Changing host
[22:13] --> _sev_ joined #scummvm.
[22:13] #scummvm: mode change '+o _sev_' by ChanServ!ChanServ@services.
[22:13] <[md5]> LordHoto: I committed the changes you proposed
[22:14] <[md5]> but I did a force push for a conflicting commit and now github ate some of your comments :(
[22:17] <[md5]> I added them manually, hope that's ok
[22:17] <-- Kaidane_ left irc: Ping timeout: 244 seconds
[22:19] --> Kenryu joined #scummvm.
[22:19] <-- Vaikungfu left irc: Disconnected by services
[22:19] Nick change: Kenryu -> Vaikungfu
[22:24] <-- James|GlideM left irc: Read error: Connection reset by peer
[22:25] --> Kaidane joined #scummvm.
[22:26] Nick change: _sev_ -> _sev
[22:36] --> risca joined #scummvm.
[22:50] <-- _athrxx left irc: Quit: ChatZilla 0.9.88 [Firefox 10.0.2/20120215223356]
[22:53] <-- nutron|w left irc: Quit: I must go eat my cheese!
[22:53] <-- nutron left irc: Quit: I must go eat my cheese!
[22:55] --> Kirben joined #scummvm.
[22:55] #scummvm: mode change '+o Kirben' by ChanServ!ChanServ@services.
[22:56] <-- cyco left irc: Ping timeout: 240 seconds
[22:57] --> cyco joined #scummvm.
[23:00] Nick change: BaaaZen_ -> BaaaZen
[23:00] <-- BaaaZen left irc: Changing host
[23:00] --> BaaaZen joined #scummvm.
[23:04] [md5] <-- (~md5@unaffiliated/md5/x-729473) left irc:
[23:08] <-- WooShell left irc: Quit: svc.startd: The system is coming down. syncing file systems... done.
[23:31] <-- giucam left irc: Ping timeout: 240 seconds
[23:33] <-- pok0j left irc: Remote host closed the connection
[23:48] <-- LunaVorax left irc: Ping timeout: 260 seconds
[23:55] <-- cyco left irc: Quit: cyco
[00:00] --- Tue Feb 21 2012