ScummVM IRC Log - 23 Jul 2017

[Back to Index]

[00:10] <-- Henke37 left irc: Quit: ERR_SHUTDOWN
[00:14] <-- girafe left irc: Quit: Leaving
[00:22] <snover> tsoliman: yeah! this is exactly why i said that thing about exposing SDL_FlushEvents.
[00:25] <-- WooShell left irc: Quit: If you understand or if you don't, if you believe or if you doubt - There's a universal justice, and the eyes of truth are always watching you.
[00:35] <snover> now if i can just figure out what is causing composited videos to play back too slowly when frameout throttling is disabled, i can actually close some tickets&
[01:01] --> Polynomial-C joined #scummvm.
[01:02] --> GitHub90 joined #scummvm.
[01:02] <GitHub90> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/v7IkT
[01:02] <GitHub90> scummvm/master 371a594 Paul Gilbert: VIDEO: Fix AVIDecoder skipping alternate frames in reverse playback
[01:02] GitHub90 (GitHub90@192.30.252.45) left #scummvm.
[01:04] <-- Poly-C left irc: Ping timeout: 255 seconds
[01:06] <-- Joefish left irc: Ping timeout: 240 seconds
[01:06] --> Joefish_ joined #scummvm.
[01:06] #scummvm: mode change '+v Joefish_' by ChanServ!ChanServ@services.
[01:10] <-- Joefish_ left irc: Ping timeout: 246 seconds
[01:18] <-- _sev left irc: Ping timeout: 268 seconds
[01:24] <tsoliman> sorry what does "exposing" mean in this context? Put a method in eventMan (or some other event object) that calls SDL_FlushEvents?
[01:25] --> TMM joined #scummvm.
[01:25] #scummvm: mode change '+o TMM' by ChanServ!ChanServ@services.
[01:26] <-- Pantheon left irc: Read error: Connection reset by peer
[01:30] --> omer_mor_ joined #scummvm.
[01:32] <-- omer_mor left irc: Ping timeout: 260 seconds
[01:34] <-- SylvainTV left irc: Read error: Connection reset by peer
[01:41] --> DominusExult joined #scummvm.
[01:46] <-- Dominus left irc: Ping timeout: 260 seconds
[01:46] Nick change: DominusExult -> Dominus
[01:56] --> GitHub161 joined #scummvm.
[01:56] <GitHub161> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/v7IIz
[01:56] <GitHub161> scummvm/master 96a1b70 Paul Gilbert: TITANIC: Fixes for CIdleSummoner to summon Bellbot or Doorbot
[01:56] GitHub161 (GitHub161@192.30.252.45) left #scummvm.
[01:57] <snover> tsoliman: yes, creating some flushEvents method that calls to the appropriate thing in the backend
[01:57] <tsoliman> ok cool - this is going to be a bit more intrusive code-wise
[01:59] <tsoliman> the leaf method is probably going to be in SdlEventSource
[02:10] <tsoliman> wow .. I totally forgot how all these event classes were wired up
[02:28] <-- dreammaster left irc:
[02:30] --> omer_mor joined #scummvm.
[02:32] <-- omer_mor_ left irc: Ping timeout: 240 seconds
[03:15] --> omer_mor_ joined #scummvm.
[03:17] <-- omer_mor left irc: Ping timeout: 260 seconds
[03:35] <-- Lightkey left irc: Ping timeout: 246 seconds
[03:38] <snover> derp. spent way too much time being confused why i kept dropping frames; i was skipping the rendering call. :|
[03:39] <snover> how programming works????!
[03:48] --> Lightkey joined #scummvm.
[04:36] --> kurtwr joined #scummvm.
[04:38] <-- kurtwr2 left irc: Ping timeout: 255 seconds
[06:13] --> _sev joined #scummvm.
[06:13] <-- _sev left irc: Changing host
[06:13] --> _sev joined #scummvm.
[06:13] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[06:24] <logix> oh nintendo... http://www.businessinsider.de/nintendo-64-classic-edition-console-trademark-2017-7
[06:50] <tsoliman> wjp, _sev: I am trying to call SDL_FlushEvents() (which belongs in backend code) from GUI code to clear pending events that shouldn't reach ScummVM (macOS specific problem) .. I am looking for ideas on the right way is to reach backend code from GUI code
[06:51] <tsoliman> I was thinking the entry point is eventManager
[06:52] <tsoliman> so something like g_system->getEventManager()->clearEventQueue() .. but I am sort of stuck trying to figure out how eventManager should tell a backend about it ..
[06:58] <tsoliman> it can't just push an artificial event to the queue with a backend observer because the point is to actually call SDL_FlushEvents before any poll() is called
[07:17] --> t0by joined #scummvm.
[07:17] #scummvm: mode change '+o t0by' by ChanServ!ChanServ@services.
[07:20] --> evil-t0by joined #scummvm.
[07:20] #scummvm: mode change '+o evil-t0by' by ChanServ!ChanServ@services.
[07:22] <-- t0by left irc: Ping timeout: 260 seconds
[07:24] <tsoliman> I guess the bigger question is: is it acceptable if a call to the SDL function is made from browser_osx.mm .. I didn't think so but I can see a point of view that considered that file as part of the macOS backend :)
[07:25] --> DDR_ joined #scummvm.
[07:42] --> GitHub166 joined #scummvm.
[07:42] <GitHub166> [scummvm] tsoliman opened pull request #974: GUI: Stop leaking input events from macOS file browser dialog (master...fix-macos-browser-event-leak) https://git.io/v7Icn
[07:42] GitHub166 (GitHub166@192.30.252.34) left #scummvm.
[07:44] --> Joefish joined #scummvm.
[07:44] #scummvm: mode change '+v Joefish' by ChanServ!ChanServ@services.
[08:06] --> m_kiewitz joined #scummvm.
[08:06] #scummvm: mode change '+o m_kiewitz' by ChanServ!ChanServ@services.
[08:13] --> Henke37 joined #scummvm.
[08:19] <-- LittleToonCat left irc: Remote host closed the connection
[08:19] <wjp> is that an SDL bug?
[08:20] <tsoliman> that's a good question - I don't know
[08:22] <tsoliman> Linux can't have this problem because it uses the ScummVM file browser
[08:27] <-- _sev left irc: Quit: This computer has gone to sleep
[08:28] --> _sev joined #scummvm.
[08:28] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[08:28] <-- _sev left irc: Client Quit
[08:30] --> _sev joined #scummvm.
[08:30] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[08:35] --> girafe joined #scummvm.
[08:39] <tsoliman> wjp: Is this a reasonably clear explanation? (updated the PR)
[08:41] <-- _sev left irc: Quit: This computer has gone to sleep
[08:42] <tsoliman> I am not sure that this is a bug in SDL itself, it seems reasonable that SDL captures user input in an SDL application and puts it in a queue, until someone (SDLEventSource) calls SDL_PollEvent to consume from that queue
[08:43] <tsoliman> the fact that the SDL application (ScummVM) stopped calling SDL_PollEvent because it launched a native macOS dialogue .. is not SDL's concern .. it keeps those events until the SDL application calls SDL_PollEvent
[08:58] <tsoliman> which happens soon after the dialog is closed .. especially if you move the mouse
[09:06] <m_kiewitz> tsoliman: so one should call the dialog from another thread then?
[09:06] <tsoliman> I am honestly not sure what is going on in that *.mm file
[09:08] --> waltervn joined #scummvm.
[09:08] #scummvm: mode change '+o waltervn' by ChanServ!ChanServ@services.
[09:09] <waltervn> going from kryoflux stream files to something that an Apple II emulator can run, seems to be a mission impossible
[09:10] <tsoliman> m_kiewitz: I admit I have a very shallow understanding of what is going on. I am going by observations made using debug() calls .. in poll() for example
[09:13] <evil-t0by> waltervn, wow.
[09:13] <evil-t0by> Whops, I'm evil?
[09:13] Nick change: evil-t0by -> t0by
[09:14] <m_kiewitz> waltervn: you can't simply write a regular dump file using kyroflux?
[09:14] <t0by> waltervn, out of curiosity, why do you need that?
[09:15] <waltervn> m_kiewitz: it's a copy-protected game. I have gotten it running by patching it, but I'd really like to get an original image working as well
[09:16] <waltervn> t0by: it's often useful to be able to run the games we're working on, right? :)
[09:18] <waltervn> the closest I've gotten is a .nib image, but it doesn't pass the copy protection check. I'm not sure if that's simply due to the .nib format not being strong enough, or an issue with the converter
[09:18] <t0by> Oh, sorry, I misunderstood you completely.
[09:22] <tsoliman> m_kiewitz: Ah I think I understand what you meant by another thread ... did you mean SDL wouldn't capture input from that thread?
[09:28] <tsoliman> it's 4:30 AM here ... time for bed ... I'll read the logs in the morning :)
[09:39] --> WooShell joined #scummvm.
[09:40] <m_kiewitz> tsoliman: yes
[09:40] <m_kiewitz> but removing the native dialog probably makes more sense
[09:41] <m_kiewitz> waltervn: is it a hires game?
[09:42] --> ajax16384 joined #scummvm.
[09:42] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[09:42] <WooShell> good meowning =^.^=
[09:47] <waltervn> m_kiewitz: yes, the Cranston Manor we got a month ago or so
[09:48] <m_kiewitz> ah it worked fine? nice
[09:49] <waltervn> seems to be in good condition, but as I said, I haven't managed to get an unpatched version working yet
[09:52] <m_kiewitz> i mean the floppy was readable, that's the most important part
[10:04] <Henke37> https://archive.org/details/apple_ii_library_4am?sort=-publicdate
[10:07] <m_kiewitz> i don't think archive.org was actually allowed by the rights holder to put those on the internet for free
[10:08] <Henke37> don't worry, they hacked copyright law
[10:08] <m_kiewitz> which means archive.org has to be seen as a pirate site w/ US government approval
[10:08] <Henke37> they have an excemption from copyright.
[10:15] <Henke37> and each title comes with a cracking log. this is one of my favorites:
[10:15] <Henke37> https://ia801601.us.archive.org/16/items/PinballConstructionSet4amCrack/Pinball%20Construction%20Set%20%284am%20and%20san%20inc%20crack%29.txt
[10:34] <m_kiewitz> as i said - pirates w/ US government approval
[10:34] <m_kiewitz> otherwise they would have asked all the companies instead
[10:35] <Henke37> good luck hunting down companies that were around before I learned how to walk
[10:36] --> GitHub70 joined #scummvm.
[10:36] <GitHub70> [scummvm] bgK pushed 1 new commit to master: https://git.io/v7IBU
[10:36] <GitHub70> scummvm/master bc95280 Bastien Bouclet: TITANIC: Adjust the switch/case fall-through comments...
[10:36] GitHub70 (GitHub70@192.30.252.45) left #scummvm.
[10:36] <Henke37> and good luck finding originals even if the company exists and is cooperative
[10:36] <m_kiewitz> that's no excuse. Then you simply can't publish their games.
[10:37] <m_kiewitz> And it's a joke anyway, because on the other hand Disney gets copyright extensions all the time, to stop works from going to public domain
[10:38] <m_kiewitz> if they would simply keep a private archive from pirated games and would simply wait for copyright to expire, then it wouldn't be as bad
[10:39] <m_kiewitz> Don't they even specify somewhere on their site, that companies who do not want this should get in contact with them?
[10:40] <m_kiewitz> then i think about https://en.wikipedia.org/wiki/Megaupload_legal_case and I can't take any of that seriously
[10:53] <wjp> 9990... getting closer
[10:54] <logix> Henke37: heh, when reading "apple ii" and "copy protection", 4am's writeups also was the first thing I thought of
[10:57] <-- ajax16384 left irc: Read error: Connection reset by peer
[11:03] <logix> although his goal is rather running the games on original hardware (where it's probably harder to write disks with deliberately "broken" sectors etc.) and not in an emulator
[11:05] <logix> waltervn: https://github.com/a2-4am/passport (you probably know this already...)
[11:13] --> Strangerke joined #scummvm.
[11:13] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services.
[11:13] <Strangerke> hi guys
[12:02] --> SylvainTV joined #scummvm.
[12:02] #scummvm: mode change '+o SylvainTV' by ChanServ!ChanServ@services.
[12:29] <-- omer_mor_ left irc: Ping timeout: 248 seconds
[12:43] <-- chadj left irc: Ping timeout: 240 seconds
[13:06] Action: criezy wonders how likely it is that bug 10000 will be created by dafioram
[13:15] --> GitHub53 joined #scummvm.
[13:15] <GitHub53> [scummvm] wjp pushed 1 new commit to master: https://git.io/v7IgO
[13:15] <GitHub53> scummvm/master 7fcfb42 Willem Jan Palenstijn: MOHAWK: Remove unnecessary include that was breaking some builds...
[13:15] GitHub53 (GitHub53@192.30.252.45) left #scummvm.
[13:26] --> chadj joined #scummvm.
[13:52] --> K4T joined #scummvm.
[13:52] <-- t0by left irc: Remote host closed the connection
[13:52] --> t0by joined #scummvm.
[13:52] #scummvm: mode change '+o t0by' by ChanServ!ChanServ@services.
[14:02] --> dreammaster joined #scummvm.
[14:02] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[14:03] --> GitHub1 joined #scummvm.
[14:03] <GitHub1> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/v7IaS
[14:03] <GitHub1> scummvm/master 807fecf Paul Gilbert: TITANIC: Add missing break lines to switch statements...
[14:03] GitHub1 (GitHub1@192.30.252.34) left #scummvm.
[14:57] <-- t0by left irc: Remote host closed the connection
[14:57] --> t0by joined #scummvm.
[14:57] #scummvm: mode change '+o t0by' by ChanServ!ChanServ@services.
[15:16] --> omer_mor joined #scummvm.
[15:22] --> ny00123 joined #scummvm.
[15:34] <-- ny00123 left irc: Read error: Connection reset by peer
[15:34] --> ny00123 joined #scummvm.
[15:40] --> GitHub89 joined #scummvm.
[15:40] <GitHub89> [scummvm] csnover pushed 11 new commits to master: https://git.io/v7I6R
[15:40] <GitHub89> scummvm/master 970c312 Colin Snover: SCI32: Detect multi-disc audio by RESSCI files instead of RESAUD...
[15:40] <GitHub89> scummvm/master 9f33f2b Colin Snover: SCI32: Don't warp the mouse when it doesn't need to be warped
[15:40] <GitHub89> scummvm/master 5b166a5 Colin Snover: SCI32: Avoid extra cursor paints when the cursor has not moved
[15:40] GitHub89 (GitHub89@192.30.252.42) left #scummvm.
[15:45] --> ajax16384 joined #scummvm.
[15:45] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[15:49] <snover> whoops! i fixed some code in one of those commits but i forgot to change the commit message to match. at least the code in the commit has comments that make it clear what is going on.
[15:52] <snover> it would sure be nice if git got some capability to create commit message amendments in the future for when these things inevitably happen.
[15:55] <-- K4T left irc:
[15:55] <dreammaster> Amen for that.
[15:55] <dreammaster> I've had my own small share of oopsy commit messages. :)
[15:56] <Strangerke> ... that's the definition of a force push, no?
[15:57] <dreammaster> Right, but that requires everyone to do a force pull if they'd previously pulled the incorrect commit, and worry about their local subsequent commits, and all sorts of nastiness. A nice incremental prior update to commit messages at least would be nice
[15:59] <Strangerke> yep indeed :)
[16:07] <-- ny00123 left irc: Quit: Leaving
[16:24] --> LittleToonCat joined #scummvm.
[16:27] <-- dreammaster left irc:
[16:39] --> f-a joined #scummvm.
[16:44] f-a (f-a@151.21.94.244) left #scummvm.
[16:45] --> GitHub3 joined #scummvm.
[16:45] <GitHub3> [scummvm] csnover pushed 1 new commit to master: https://git.io/v7I10
[16:45] <GitHub3> scummvm/master 580f387 Colin Snover: SCI32: Fix explicit mouse position changes...
[16:45] GitHub3 (GitHub3@192.30.252.45) left #scummvm.
[16:46] --> f-a joined #scummvm.
[16:49] f-a (f-a@151.21.94.244) left #scummvm.
[17:00] --> travis-ci joined #scummvm.
[17:00] <travis-ci> scummvm/scummvm#3885 (master - 580f387 : Colin Snover): The build has errored.
[17:00] <travis-ci> Change view : https://github.com/scummvm/scummvm/compare/dabcacb0ca21...580f3875c099
[17:00] <travis-ci> Build details : https://travis-ci.org/scummvm/scummvm/builds/256630663
[17:00] travis-ci (travis-ci@ec2-54-159-223-41.compute-1.amazonaws.com) left #scummvm.
[17:05] <snover> no, YOUR an error.
[17:29] <Henke37> *you're
[18:03] <-- Strangerke left irc: Ping timeout: 255 seconds
[18:19] --> dreammaster joined #scummvm.
[18:19] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[18:38] <-- t0by left irc: Remote host closed the connection
[18:38] --> t0by joined #scummvm.
[18:38] #scummvm: mode change '+o t0by' by ChanServ!ChanServ@services.
[18:49] <t0by> well, you could (ew) revert and recommit with the right message...
[19:00] <dreammaster> But that doesn't help, unfortunately, if the misnamed commit has already been pushed to master
[19:04] --> GitHub67 joined #scummvm.
[19:04] <GitHub67> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/v7I5c
[19:04] <GitHub67> scummvm/master d3f3da4 Paul Gilbert: TITANIC: Fix delay before Bellbot & Doorbot appears
[19:04] GitHub67 (GitHub67@192.30.252.40) left #scummvm.
[19:11] <-- t0by left irc: Read error: Connection reset by peer
[19:20] <snover> it helps in the sense that the blame will point to the correct commit, but reverting code just to mess with the commit history doesnt seem very appropriate
[19:23] --> ny00123 joined #scummvm.
[19:37] <wjp> the updateScreen commit?
[19:38] <snover> no, but if i did it twice thats not good! it was 2ed18b86db51caacd123b5e95f4672685dc49799
[19:39] <wjp> that's the one I meant
[19:39] <wjp> (it adds an updateScreen call)
[19:39] <snover> ah, ok
[19:39] <snover> in that case. yes. that one.
[19:39] <wjp> I guess I should maybe have used the commit's id to id the commit :-)
[19:50] <snover> it is becoming quite difficult to understand how frame throttling & rendering is happening, since there are so many different places that take control of it.
[20:04] --> Strangerke joined #scummvm.
[20:19] --> Vampire0 joined #scummvm.
[20:20] <-- Vampire0_ left irc: Ping timeout: 255 seconds
[20:20] --> omer_mor_ joined #scummvm.
[20:22] <-- m_kiewitz left irc: Quit: technology isn't intrinsically good or evil. It's how it's used. Like the Death Ray.
[20:23] <-- omer_mor left irc: Ping timeout: 260 seconds
[20:51] <-- ny00123 left irc: Quit: Leaving
[20:55] <-- ajax16384 left irc: Read error: Connection reset by peer
[21:05] <-- Henke37 left irc: Quit: ERR_SHUTDOWN
[21:09] --> GitHub161 joined #scummvm.
[21:09] <GitHub161> [scummvm] csnover pushed 3 new commits to master: https://git.io/v7Ipz
[21:09] <GitHub161> scummvm/master 4d52b01 Colin Snover: SCI: Keep audio maps out of the LRU cache
[21:09] <GitHub161> scummvm/master d38704e Colin Snover: SCI: Blacklist certain audio map patch files...
[21:09] <GitHub161> scummvm/master 2005ed7 Colin Snover: SCI32: Emit a warning when an audio resource cannot be found...
[21:09] GitHub161 (GitHub161@192.30.252.34) left #scummvm.
[21:15] <snover> now i am just waiting for that resource blacklisting to blow up in my face when some localised version of phant1 uses patch maps legitimately&
[21:17] <-- waltervn left irc: Quit: Leaving
[21:22] <tsoliman> what I sometimes do for post-push commit annotation is: add a comment to the commit in github .. it's far from ideal and isn't visible easily, but sometimes I run into my comments when getting the commit on github.com so I can send the link, and find that someone commented on it
[21:28] <snover> yeah. i would do that if it was particularly egregious, in this case the comments in the code for that commit speak for themselves enough that it is probably not worth it
[21:38] <-- snover left irc: Ping timeout: 240 seconds
[21:44] --> _sev joined #scummvm.
[21:44] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[21:44] --> snover joined #scummvm.
[21:44] #scummvm: mode change '+o snover' by ChanServ!ChanServ@services.
[21:48] <-- _sev left irc: Ping timeout: 240 seconds
[21:53] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services.
[21:57] --> Strangerke_ joined #scummvm.
[21:58] <-- Strangerke left irc: Ping timeout: 260 seconds
[21:58] Nick change: Strangerke_ -> Strangerke
[22:37] <-- girafe left irc: Read error: Connection reset by peer
[22:47] <-- Strangerke left irc: Ping timeout: 240 seconds
[23:10] <-- WooShell left irc: Quit: If you understand or if you don't, if you believe or if you doubt - There's a universal justice, and the eyes of truth are always watching you.
[23:52] <tsoliman> criezy: I welcome a patch that actually compiles inside the *.mm file (I really know zero about ObjectiveC++, I got by by writing regular C++ in there) - I am not sure what the best way to solve this problem is and both ideas work more or less.
[23:53] <tsoliman> the way I see it: the whole "separate browser for macOS" is not clean in itself - but I don't think I have a good sense of what is clean code and what isn't yet
[23:54] <-- Joefish left irc: Ping timeout: 276 seconds
[23:54] <tsoliman> as such my opinion should matter the least in this .. several devs here use macOS regularly and maybe they can weigh in on this?
[23:55] <tsoliman> I only care that the user experience isn't the current "question your sanity after adding a game"
[00:00] --- Mon Jul 24 2017