[00:47] GloKidd (~GloKidd@d50-98-254-218.bchsia.telus.net) joined #scummvm. [00:49] Hello all, I used to be very active here back in the day, just popped in to say hi :-) so much has been added for support since I was a regular contributor. hi sev, kirben, drmccoy, clone2727 etc if any of you are still around :-) [00:54] not as active of a room as I remember it but then again it's almost 10 years later ;-) [00:59] Still Europe-centric, it's 3 am here. [01:14] yeah that would be a barrier lol [01:37] DominusExult (~dominus@port-92-194-4-209.dynamic.qsc.de) joined #scummvm. [01:37] DominusExult (~dominus@port-92-194-4-209.dynamic.qsc.de) left irc: Changing host [01:37] DominusExult (~dominus@unaffiliated/dominus) joined #scummvm. [01:39] GloKidd (~GloKidd@d50-98-254-218.bchsia.telus.net) left irc: Ping timeout: 248 seconds [01:40] Dominus (~dominus@unaffiliated/dominus) left irc: Ping timeout: 268 seconds [01:40] Nick change: DominusExult -> Dominus [01:47] GloKidd (~GloKidd@2605:b100:90d:278a:55cd:61d2:5380:1d9c) joined #scummvm. [01:58] [scummvm] digitall pushed 1 new commits to master: https://git.io/fjFYX [01:58] scummvm/master 1083ff6 digitall: KYRA: LOL: Fix GCC Compiler Warning from memset Usage [02:13] GloKidd (~GloKidd@2605:b100:90d:278a:55cd:61d2:5380:1d9c) left irc: Ping timeout: 250 seconds [02:26] dreammaster (~dreammast@c-73-241-66-252.hsd1.ca.comcast.net) joined #scummvm. [02:26] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services. [03:20] [scummvm] bluegr pushed 1 new commits to master: https://git.io/fjFO3 [03:20] scummvm/master ea9e33b ccawley2011: GUI: Check if fullscreen is available at runtime [03:20] [scummvm] bluegr closed pull request #1810: GUI: Check if fullscreen is available at runtime (master...gui-only-fullscreen) https://git.io/fjdNo [04:08] What madman gets up before 6 am on a Sunday? [04:32] dreammaster (~dreammast@c-73-241-66-252.hsd1.ca.comcast.net) left irc: [04:57] SylvainTV (~Sylvain@alille-658-1-80-233.w92-147.abo.wanadoo.fr) left irc: Ping timeout: 258 seconds [05:03] SupSuper (~SupSuper@openxcom/dev/supsuper) left irc: Quit: Rip [05:13] GloKidd (~GloKidd@199.127.174.53) joined #scummvm. [05:39] Begasus (~begasus@ptr-4qba4eb2kskz5kjgm8t.18120a2.ip6.access.telenet.be) joined #scummvm. [05:46] Begas_VBox (~Begasus@d54C05BEA.access.telenet.be) joined #scummvm. [06:17] JohnnyonFlame (~quassel@131.0.31.185) left irc: Ping timeout: 272 seconds [06:55] cd (~cd@unaffiliated/cd) left irc: Quit: cd [07:45] ajax16384 (~User@109.60.130.33) joined #scummvm. [07:45] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services. [08:03] nutron|w (~nutron@66.51.134.118) left irc: Ping timeout: 268 seconds [08:04] nutron|w (~nutron@66.51.134.118) joined #scummvm. [08:33] timofonic (~timofonic@unaffiliated/timofonic) left irc: Read error: Connection reset by peer [08:35] timofonic (~timofonic@unaffiliated/timofonic) joined #scummvm. [09:31] @Lightkey: I got back from an all-night party 😃 was just getting to sleep at that point [09:51] Guest42 (523106dd@gateway/web/cgi-irc/kiwiirc.com/ip.82.49.6.221) joined #scummvm. [09:52] Guest42 (523106dd@gateway/web/cgi-irc/kiwiirc.com/ip.82.49.6.221) left #scummvm. [10:17] SylvainTV (~Sylvain@alille-658-1-80-233.w92-147.abo.wanadoo.fr) joined #scummvm. [10:17] #scummvm: mode change '+o SylvainTV' by ChanServ!ChanServ@services. [10:40] ajax16384 (~User@109.60.130.33) left irc: Quit: Leaving [10:41] @ccawley2011 and others interested: as I promised here it is: https://github.com/lephilousophe/dockerized-bb [10:42] a first try at making a dockerized version of buildbot service [10:54] LePhilousophe: Looks good so far. [10:54] not finished yet though :) [10:55] hum, just noticed I didn't give credit to csnover for the work [10:57] whiterandrek (~CVoltage@ppp-109-104-179-20.wildpark.net) joined #scummvm. [10:57] #scummvm: mode change '+o whiterandrek' by ChanServ!ChanServ@services. [10:58] One question, though: Is it still possible to use the Docker images stand-alone, similar to what's described here: https://github.com/csnover/scummvm-buildbot#using-workers-as-stand-alone-compilers [11:08] ccawley2011 (~ccawley20@host109-149-11-197.range109-149.btcentralplus.com) joined #scummvm. [11:08] #scummvm: mode change '+o ccawley2011' by ChanServ!ChanServ@services. [11:09] hum I don't think so [11:10] I don't know how easy it would be to add this functionality [11:13] timofonic (~timofonic@unaffiliated/timofonic) left irc: Read error: Connection reset by peer [11:14] timofonic (~timofonic@unaffiliated/timofonic) joined #scummvm. [11:15] What I've been wondering with both snover's work and yours is wether the buildbot worker files need to be in the docker image. Would it be possible to have docker images that contain simply the toolchain for the target platform, and insert the buildbot worker through a volume? [11:15] hum [11:16] but we would need dependencies in the worker image too [11:16] yeah [11:16] I think it's doable [11:16] but we would have to specify entry point in DockerLatentWorker on master side [11:16] I will check that [11:21] seems possible to specify a command [11:22] anyway I am still not sure it's a good idea as currently pip checks dependencies when installing worker [11:22] that lets ensure that nothing is missing in the image to run it [11:23] and dependencies can change when buildbot is upgraded [11:23] bgK: why would you want to have worker outside of the worker? [11:29] [scummvm] antoniou79 pushed 1 new commits to master: https://git.io/fjFZQ [11:29] scummvm/master e263207 antoniou79: BLADERUNNER: Restore bits of Guzza dialogue [11:30] LePhilousophe: The way see things, the toolchains are images we push to DockerHub, people can use them to do their own builds easily. Buildbot simply reuses the images. But then again, I don't plan to contribute heavily on the system, so feel free to ignore me. (However I plan to make sure there are images for my platforms of interest and that the windows and mac toolchains work well with ResidualVM) [11:35] ccawley2011 (~ccawley20@host109-149-11-197.range109-149.btcentralplus.com) left irc: Remote host closed the connection [11:36] ccawley2011 (~ccawley20@host109-149-11-197.range109-149.btcentralplus.com) joined #scummvm. [11:36] #scummvm: mode change '+o ccawley2011' by ChanServ!ChanServ@services. [12:35] JohnnyonFlame (~quassel@131.0.31.185) joined #scummvm. [12:35] ccawley2011_ (~ccawley20@host109-149-11-197.range109-149.btcentralplus.com) joined #scummvm. [12:35] #scummvm: mode change '+o ccawley2011_' by ChanServ!ChanServ@services. [12:37] bgK: I see [12:38] ccawley2011 (~ccawley20@host109-149-11-197.range109-149.btcentralplus.com) left irc: Ping timeout: 248 seconds [12:39] in this case we should split the job in two parts: one where we build toolchains images and one which extends from first part to create buildbot workers [12:41] if we do this, that solves ccawley2011 question about using images out of buildbot [12:52] GloKidd (~GloKidd@199.127.174.53) left irc: Read error: Connection reset by peer [13:27] ajax16384 (~User@109.60.130.33) joined #scummvm. [13:27] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services. [13:58] timofonic (~timofonic@unaffiliated/timofonic) left irc: Ping timeout: 248 seconds [13:59] timofonic (~timofonic@unaffiliated/timofonic) joined #scummvm. [14:03] timofonic (~timofonic@unaffiliated/timofonic) left irc: Ping timeout: 245 seconds [14:04] timofonic (~timofonic@unaffiliated/timofonic) joined #scummvm. [16:14] ny00123 (~ny00123@85-250-102-120.bb.netvision.net.il) joined #scummvm. [16:47] girafe (~girafe@lfbn-lyo-1-470-122.w2-7.abo.wanadoo.fr) joined #scummvm. [16:53] I wonder whats the main reason for the dockerized buildbot? The toolchains need to be maintained anyway... [16:53] Just asking because I never really dealt with docker and not being the biggest fan of its hype [16:54] rootfather: the reason about why I looked at it in the first place is because when mataniko upgraded the server, it broke several toolchains [16:55] using containerized toolchains let you be immune against upgrades on the host [16:55] in addition it will let us have a repository where we have reproductible rules to build toolchains [16:56] Ah I see, thats a valid point [16:57] that is that if tomorrow you want to do a build for a some random platform you just install docker, run some make command and you are ready to build ScummVM for it [17:01] containers are overhyped, but build toolchains are one of not so many things where they are actually pretty useful ;) [17:02] rootfather: note that I wasn't a huge of fan of containers (and still isn't) [17:03] that takes bytes on disk, increase CPU and RAM consumption [17:04] same here; used them at work for "microservices", used in some other places as well and wasn't convinced... but then I've used it for packaging cross-compilation toolchains while configuring CI for my games for all supported platforms (as it was the easiest way to do that for GitLab CI) and it actually rocks [17:06] I am still not sure it rocks :) [17:11] it sure rocks compared to manually installed and configured toolchains for 10+ platforms that I had before [17:12] and I really appreciated that work I did for CI when I decided to reinstall my OS after SSD failure :D [17:12] Yeah now that I think of the number of platforms we have it makes sense 😁 [17:13] Speaking of toolchains, my win32 toolchain based on MXE is now in it 5th month of testing, so mayybe... [17:15] Concerning this: https://github.com/scummvm/scummvm/commit/ea9e33bd1096389f8f11614d080d16f0005b1e10 [17:15] GUI_ONLY_FULLSCREEN was used to indicate that there's ONLY fullscreen. [17:16] Not that there's no fullscreen. [17:17] After this commit, there's now no way at all to treat this case. [17:17] rootfather: I didn't know MXE and I used recipes from csnover which I will have to use for all other platforms [17:17] too bad MXE is only Windows oriented [17:18] MXE is for building for win32 only so I guess it wont help the official buildbot that much 😉 [17:19] I guess the fullscreen thing is ok, because functionally it is the same as before. [17:19] rsn8887: if(g_system->hasFeature(OSystem::kFeatureFullscreenMode)) ... [17:19] But if everything works fine we could use it for the win32 builds since its always up to date regarding the libs [17:19] I suppose that if you don't have the feature it will act in fullscreen only [17:20] Yeah. I guess that is fine. [17:20] criezy (~criezy@host86-133-88-198.range86-133.btcentralplus.com) joined #scummvm. [17:20] #scummvm: mode change '+o criezy' by ChanServ!ChanServ@services. [17:20] Although I would have thought that would mean there's only windowed modes. [17:21] THere's definitely still something missing. One needs to be able to tell ScummVM if there are ONLY windowed modes or if there is ONLY fullscreen, or both are allowed. [17:21] I don't know [17:21] do you think it makes sense to not have fullscreen? [17:21] Not for any platform I have worked on, so maybe not. [17:22] So probably all is ok. [17:22] well, maybe the case could be treated in due time [17:23] Yeah. It is just it WAS treated before the commit. [17:23] Even if maybe not used. [17:23] The description for kFeatureFullscreenMode is "If supported, this feature flag can be used to switch between windowed and fullscreen mode" [17:23] https://github.com/scummvm/scummvm/blob/master/common/system.h#L269 [17:24] GUI_ONLY_FULLSCREEN was only used to disable the checkbox in the GUI, so it served the same purpose. [17:24] hum, I don't see how windowed-only mode was supported before [17:24] Ah, so it never defined a default. [17:24] windowed-only? By not setting the FEATUREFULLSCREEN. [17:25] AFAIK. [17:25] But I agree windowed-only is useless. Every platform I know that supports windowed also supports fullscreen. [17:25] But there are many platforms that ONLY support fullscreen. [17:26] So it is better now than it was before 😃 [17:27] in fact I am not sure to see how it prevents windowed mode only. The only thing implied by the feature in commit is that it disables setting in UI [17:28] maybe we shouldn't do the _fullscreenCheckbox->setState(true); when kFeatureFullscreenMode is not active [17:30] GetFeatureState is used with fullscreen all the time. [17:31] what does it return if kFeatureFullscreen flag is not set? [17:31] I suppose false, but that doesn't make sense, because then fullscreen should be ON? [17:31] Why do features have a state anyways? [17:31] Shouldn't the state be handled by confman? [17:32] cd (~cd@unaffiliated/cd) joined #scummvm. [17:32] maybe it is time to invert it and just call it WindowedMode [17:32] the feature can be disabled during lifetime? [17:32] I think so, because GetFeatureState is a thing. [17:32] it really doesn't make sense that features have state. [17:32] smells more like polymorphism [17:33] ah no g_system->getFeatureState(OSystem::kFeatureFullscreenMode); means fullscreen or not [17:34] How do I force vita to be always fullscreen? [17:34] hum ok there is hasFeature and getFeatureState [17:35] Is it enough to unset kFeatureFullScreenMode, and make GetFeatureState return true? [17:35] how do you do currently? [17:35] I don't there's a tickbox and if unchecked, the gates of hell open. [17:35] which backend do you use? [17:35] Because Window mode is not supported, so the graphics are in the top left corner, the touch coordinates are wrong etc. [17:36] SurfaceSDL [17:39] dreammaster (~dreammast@c-73-241-66-252.hsd1.ca.comcast.net) joined #scummvm. [17:39] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services. [17:41] beresky (~borosky@095160157148.dynamic-ra-10.vectranet.pl) left irc: Ping timeout: 268 seconds [17:41] beresky (~borosky@095160157148.dynamic-ra-10.vectranet.pl) joined #scummvm. [17:46] well, I don't see how ccawley2011 changes would affect the behaviour but I am not sure the code is robust enough [17:46] each time we do a setFeatureState we should check hasFeature [17:48] because if someone copies a config file with fullscreen setting in it, it seems to modify fullscreen mode in SurfaceSdlGraphicsManager [17:52] SupSuper (~SupSuper@openxcom/dev/supsuper) joined #scummvm. [17:59] Yes. I think I found a way for Vita, but need to test. It is clunky: [18:00] rsn8887 did you test with just changes from ccawley2011 ? [18:00] void OSystem_PSP2::initBackend() { ConfMan.setBool("fullscreen", true); (...) } bool OSystem_PSP2::hasFeature(Feature f) { if (f == kFeatureFullscreenMode) return false; return (f == kFeatureKbdMouseSpeed || f == kFeatureJoystickDeadzone || f == kFeatureShader || f == kFeatureTouchpadMode || OSystem_SDL::hasFeature(f)); } void OSystem_PSP2::setFeatureState(Feature f, bool enable) { [18:00] switch (f) { case kFeatureTouchpadMode: ConfMan.setBool("touchpad_mouse_mode", enable); break; case kFeatureFullscreenMode: break; default: OSystem_SDL::setFeatureState(f, enable); break; } } bool OSystem_PSP2::getFeatureState(Feature f) { switch (f) { case kFeatureTouchpadMode: return ConfMan.getBool("touchpad_mouse_mode"); break; case kFeatureFullscreenMode: [18:00] return true; break; default: return OSystem_SDL::getFeatureState(f); break; } } [18:00] ouch [18:00] Have pity on us IRC users :) [18:01] Oh sorry [18:01] rsn8887: I don't understand how it worked before to ensure fullscreen state [18:01] It didn't. [18:01] But now I would like it to work. [18:01] So I am making changes. [18:01] oh! [18:02] I thought you did this just because of ccawley2011 changes [18:02] Before, the user could disable fullscreen tickmark, and got horrible results. [18:02] No [18:02] I am now convinced that ccawley2011 changes are good. [18:02] And will not change existing behavior. [18:03] But on Vita and Switch, it was never done right. The Fullscreen tickbox should have been disabled and permanently on on those platforms. Because the SDL2 port on those does not support window mode correctly. [18:03] well, I think changes could be done way more generic [18:03] Or better: SDL works, but the OS doesn't provide window resize etc. buttons. [18:03] in SurfaceSdlGraphicsManager::setFeatureState we should check hasFeature before applying the setting [18:03] So on those systems, fullscreen mode should be strictly enforced regardless of exisiting .ini files etc. [18:04] this way you don't have to override the setFeatureState [18:04] But then also we have to ensure that fullscreen mode is actually on, currently users can still override via .ini etc. [18:05] nah because setting is applied using setFeatureState [18:05] ? [18:05] fullscreen or windowed? AFAIK ScummVM defaults to windowed mode. [18:05] GUI does this: g_system->setFeatureState(OSystem::kFeatureFullscreenMode, ConfMan.getBool("fullscreen", _domain)); [18:06] And what is the default? [18:06] read from settings it seems :/ [18:06] So that's not good. [18:06] yep :( [18:07] if the feature is NOT present, the default should be fullscreen mode. [18:07] anyway there is already an uglay ifdef [18:07] It is a mess. [18:07] #if !defined(_WIN32_WCE) && !defined(__SYMBIAN32__) in surfacesdl-graphics.cpp [18:07] to force fullscreen at true [18:08] I hope I won't have to use that on Vita to make it work the way I want. [18:08] And since when was that allowed? [18:08] I would bet long time ago [18:08] I will blame [18:08] 😃 [18:08] Probably the former maintainer of these platforms. [18:09] 2010-06-07 03:14:42 +0000 [18:09] commit 280c18f96e3 [18:09] Back 9 years [18:09] [scummvm] antoniou79 pushed 2 new commits to master: https://git.io/fjFBn [18:09] scummvm/master a584919 antoniou79: BLADERUNNER: McCoy should face Clovis in KP07 (ClickedOnActor) [18:09] scummvm/master b7c089f antoniou79: BLADERUNNER: Restore bits of dialogue for McCoy, Chew, Walls [18:34] Lightkey (~Darklock@p200300F693D1236522CF30FFFE083718.dip0.t-ipconnect.de) left irc: Ping timeout: 264 seconds [18:39] dreammaster (~dreammast@c-73-241-66-252.hsd1.ca.comcast.net) left irc: [18:43] Drenn (~Drenn@toroon0717w-lp130-05-184-145-50-104.dsl.bell.ca) joined #scummvm. [18:47] Lightkey (~Darklock@p200300F693D1239222CF30FFFE083718.dip0.t-ipconnect.de) joined #scummvm. [19:38] Begas_VBox (~Begasus@d54C05BEA.access.telenet.be) left irc: Quit: Vision[0.10.3]: i've been blurred! [19:39] Begasus (~begasus@ptr-4qba4eb2kskz5kjgm8t.18120a2.ip6.access.telenet.be) left irc: Quit: Ex-Chat [20:14] [scummvm] rsn8887 pushed 1 new commits to master: https://git.io/fjFRp [20:14] scummvm/master 073d833 rsn8887: PSP2: Force fullscreen always on [20:17] [scummvm] whiterandrek opened pull request #1811: COMMON: added support for reading ini files with different 8bit encondings (master...ini2) https://git.io/fjF0J [20:29] [scummvm] whiterandrek closed pull request #1811: COMMON: added support for reading ini files with different 8bit encodings (master...ini2) https://git.io/fjF0J [20:31] _sev_ (~sev@scummvm/undead/sev) joined #scummvm. [20:31] #scummvm: mode change '+o _sev_' by ChanServ!ChanServ@services. [20:31] _sev (~sev@scummvm/undead/sev) left irc: Ping timeout: 248 seconds [20:35] ajax16384 (~User@109.60.130.33) left irc: Quit: Leaving [20:42] _athrxx (~athrxx@5-157-124-110.v4.ngi.it) joined #scummvm. [20:45] [scummvm] whiterandrek opened pull request #1812: COMMON: added support for reading ini files with different 8bit encod& (master...ini) https://git.io/fjF0Z [20:53] whiterandrek (~CVoltage@ppp-109-104-179-20.wildpark.net) left irc: Ping timeout: 268 seconds [21:20] [scummvm] bluegr pushed 3 new commits to master: https://git.io/fjF09 [21:20] scummvm/master 6a9969e bluegr: STARTREK: Fix loading of some texts, and add the "text" console command [21:20] scummvm/master 8e69e9e bluegr: STARTREK: Fix failed action texts [21:20] scummvm/master 0add13e bluegr: STARTREK: Read most of the TUG mission texts from RDF files [21:32] Zaarin (~dfh@84-52-248.235.3p.ntebredband.no) left irc: Quit: If I see you again back in detox, put my remains in my snuff box. [21:34] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services. [21:35] md5: Ah, so like me, not a good time to review right before going to sleep IMHO. Anyway, as I've added you to the engine wiki page recently, I was looking at AUTHORS and Startrek seems to be completely missing there. [21:38] Ah, is it because only complete engines are added there? Lastexpress threw me off, since it's not supported. [21:41] dreammaster (~dreammast@c-73-241-66-252.hsd1.ca.comcast.net) joined #scummvm. [21:41] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services. [21:44] @Lightkey: yeah, I believe we only add finished engines there, and Startrek is far from finished [21:45] As in, half the game is missing (the parts inside the USS Enterprise) [21:46] The space battle parts, in 3D [21:46] timofonic (~timofonic@unaffiliated/timofonic) left irc: Ping timeout: 245 seconds [21:47] timofonic (~timofonic@unaffiliated/timofonic) joined #scummvm. [21:47] [scummvm] ccawley2011 pushed 1 new commits to master: https://git.io/fjFEc [21:47] scummvm/master 475eafc ccawley2011: ANDROID: Don't deinitialize the EGL surface when pausing the application [21:56] timofonic (~timofonic@unaffiliated/timofonic) left irc: Ping timeout: 245 seconds [21:56] timofonic (~timofonic@unaffiliated/timofonic) joined #scummvm. [22:02] Deledrius_ (~Deledrius@68.118.139.214) left irc: Read error: Connection reset by peer [22:02] Deledrius_ (~Deledrius@68.118.139.214) joined #scummvm. [22:03] [scummvm] rsn8887 pushed 1 new commits to master: https://git.io/fjFEg [22:05] ny00123 (~ny00123@85-250-102-120.bb.netvision.net.il) left irc: Quit: Leaving [22:09] scummvm/master f14a7f1 rsn8887: SWITCH: Force fullscreen always on [22:09] [scummvm] ccawley2011 opened pull request #1813: ANDROID: Add a button to show the virtual keyboard (master...android-osc) https://git.io/fjFEw [22:18] [scummvm] dreammaster pushed 3 new commits to master: https://git.io/fjFEP [22:18] scummvm/master 34f842e dreammaster: GLK: FROTZ: Fix creating text vs buffer windows [22:18] scummvm/master 1f38a82 dreammaster: GLK: FROTZ: Whitespace fixes [22:18] scummvm/master 73a4f22 dreammaster: XEEN: Implement autosave support [22:20] _sev_ (~sev@scummvm/undead/sev) left irc: Ping timeout: 258 seconds [22:40] dreammaster (~dreammast@c-73-241-66-252.hsd1.ca.comcast.net) left irc: [23:32] ccawley2011_ (~ccawley20@host109-149-11-197.range109-149.btcentralplus.com) left irc: Quit: Leaving [23:40] _athrxx (~athrxx@5-157-124-110.v4.ngi.it) left irc: Quit: Leaving [23:54] dreammaster (~dreammast@c-73-241-66-252.hsd1.ca.comcast.net) joined #scummvm. [23:54] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services. [00:00] --- Mon Aug 19 2019