[Back to Index]

[00:31] <snover> dreammaster: hey, do you happen to remember why there is this // Ensure that the same number of objects are registered now as when the scene started check in tsage/scenes.cpp around line 116
[00:31] <snover> ?
[00:31] <snover> oh
[00:31] <snover> i should check my emails when i come back
[00:31] <snover> then ask questions that already were answered :)
[00:32] <snover> thank you for answering my question!
[00:33] <dreammaster> No worries :)
[00:33] <dreammaster> Thanks for looking into the various outstanding issues.
[00:38] <snover> now i just have the problem of tracking down why there is an extra object i guess.
[00:42] <snover> deleting the assertion was way easier&
[00:56] <-- criezy left irc: Quit: criezy
[00:58] <snover> i am trying to get this release stuff prepared, as it turns out im not exactly sure all the places where the scummvm version would need to get updated in order for this to not be 1.10
[00:59] <snover> 1.10 is already on the wiki and in the forum in places, so i am a little concerned about where else it might be, and dont know that it is a good use of effort to try to figure it out or confuse users by suddenly changing it
[01:04] --> GitHub65 joined #scummvm.
[01:04] <GitHub65> [scummvm] csnover opened pull request #1065: TSAGE: Various fixes (master...tsage-fixes) https://git.io/vF5gd
[01:04] GitHub65 (GitHub65@192.30.252.34) left #scummvm.
[01:05] <dreammaster> Speaking of releases, we could use a clearer methodology for the compatibility list.. looking at 1.9.0, there's a bunch of entries as "Untested" still, and I'm honestly unsure whether they're actually all so
[01:06] <snover> indeed. i want to get this release out since it has been too long, and then i really want to expend a bit of effort to improve the release process so its not such a marathon.
[01:07] <snover> on my agenda is improving the IA so there is less duplication and more automation of things like that, as well as turning Buildbot into the generator of all release packages so at any moment we can just push a tag and theres a release
[01:08] <dreammaster> That'd be awesome
[01:17] <snover> it is one reason why i slyly suggested changing the versioning scheme, since if we can get rid of all these manual processes around releases then they can just happen every quarter or month or whatever with very little effort.
[01:56] <-- SylvainTV left irc: Read error: Connection reset by peer
[02:14] <-- dreammaster left irc:
[02:32] --> mortalius__ joined #scummvm.
[02:33] <-- mortalius_ left irc: Ping timeout: 240 seconds
[02:45] --> DominusExult joined #scummvm.
[02:46] --> GitHub88 joined #scummvm.
[02:46] <GitHub88> [scummvm] csnover pushed 1 new commit to master: https://git.io/vF5og
[02:46] <GitHub88> scummvm/master cf61174 Colin Snover: KYRA: Fix crash due to missing palette data for OldFloppy version...
[02:46] GitHub88 (GitHub88@192.30.252.42) left #scummvm.
[02:47] <-- Strangerke left irc: Ping timeout: 240 seconds
[02:50] <-- Dominus left irc: Ping timeout: 260 seconds
[02:50] Nick change: DominusExult -> Dominus
[02:55] Nick change: Stormkeeper -> Storm-AFK
[02:57] --> Strangerke joined #scummvm.
[03:19] --> vasc_ joined #scummvm.
[03:23] <-- vasc left irc: Ping timeout: 268 seconds
[03:57] <vasc_> awesome
[03:57] <snover> vasc_: did you see my comment in the channel log?
[03:57] <vasc_> ROFL
[03:57] Nick change: vasc_ -> vasc
[03:57] <vasc> not yet
[03:58] <vasc> you know something? the Space Quest V demo was done with Dynamix DGDS
[03:58] <snover> yes
[03:58] <vasc> and the binary for the demo has all the debug symbols on it.
[03:58] <snover> nice!
[03:58] <snover> (its listed at http://wiki.scummvm.org/index.php/Sierra_Game_Versions#Not_SCI)
[03:59] <vasc> you can download it from archive.org.
[03:59] <snover> we got partial debug symbols for sci32 from phantasmagoria mac
[04:00] <vasc> it mentions a LZSSHUFA.ASM. i guess that was the name of the decompressor.
[04:00] <vasc> although we can already unpack it.
[04:02] <snover> interesting. in sci, stac lzs didnt get used until sci32.
[04:02] <vasc> i hope that would help IDA understand it.
[04:03] <vasc> well the demo is like two years after the last major DGDS adventure game.
[04:04] <vasc> so it might be actually more advanced somehow.
[04:05] <vasc> planar mode?
[04:06] <snover> http://wiki.osdev.org/VGA_Hardware#Memory_Layout_in_256-color_graphics_modes
[04:08] <snover> also https://en.wikipedia.org/wiki/Planar_(computer_graphics)
[04:08] <vasc> dunno. which image?
[04:08] <vasc> the one of the VQT?
[04:09] <snover> whichever one showed graphics in four-pixel blocks
[04:13] <vasc> https://imgur.com/a/ksWcs ?
[04:14] <snover> yes
[04:16] <vasc> perhaps
[04:16] <vasc> it would be nice if it was that easy.
[04:17] <vasc> i'll tell you one thing:
[04:17] <vasc> the images uncompressed, are like 64000 bytes
[04:17] <vasc> the Mac version actually compresses them images with LZW
[04:17] <vasc> its like two 4-bit images blended together.
[04:18] <vasc> 9469+2001 bytes for the DYNAMIX.SCR file for example.
[04:18] <vasc> the VQT chunk is like
[04:18] <vasc> 19329 bytes for the same image
[04:18] <vasc> so it compresses even worse
[04:19] <vasc> i think the images are exactly the same
[04:19] <vasc> it's clearly compressed somehow
[04:19] <vasc> coz its less than 64000
[04:20] <snover> when you describe the uncompressed image, are you talking about an 8bpp image?
[04:20] <vasc> yes it's an 8bpp image, 320x200
[04:21] <vasc> the Macintosh Rise of the Dragon stores the image in two chunks: BIN: and VGA:
[04:21] <vasc> BIN: has the low 4-bits, VGA: the high 4-bits.
[04:21] <vasc> AFAIK it's legacy from when they had to support EGA mode in the PC.
[04:21] <vasc> they use the lower 4-bits for EGA.
[04:22] <vasc> the Heart of China for the Macintosh also has some MA8 format, which is like linear 8bpp.
[04:22] <vasc> you might be correct in that it uses planar mode.
[04:23] <vasc> the question is which kind and how the image is compressed.
[04:48] <Deledrius_> Ugh EGA planar images.
[04:49] <Deledrius_> Wow, Space Quest V demo didn't use SCI? That's surprising to me.
[04:49] Nick change: Deledrius_ -> Deledrius
[04:49] <snover> i imagine sq5 has an interesting history
[04:50] <snover> well. interesting.
[04:51] <snover> apparently it got sent from sierra to dynamix, then dynamix was having a financial crisis and pushed it out the door
[04:55] <Deledrius> Geez. I know I've heard bits and pieces about it being a really messed up development. Didn't know it involved Dynamix too.
[04:56] <snover> yeah, apparently they were responsible for development. i cant imagine what ended up happening where they got a demo running in DGDS and then they dumped it and went to SCI
[04:57] <snover> hopefully that person working on the history of sierra book will finish the book and it will say something.
[04:58] <vasc> it doesn't even pack the resources in volumes like usual. the files are all over the demo dir.
[05:00] --> GitHub193 joined #scummvm.
[05:00] <GitHub193> [scummvm] csnover pushed 2 new commits to master: https://git.io/vF51F
[05:00] <GitHub193> scummvm/master b89b909 Colin Snover: TESTS: Remove zero-length string and associated -Wformat massage...
[05:00] <GitHub193> scummvm/master 4045b9a Colin Snover: DISTS/DEBIAN: Update to current files from downstream, less patches...
[05:00] GitHub193 (GitHub193@192.30.252.45) left #scummvm.
[05:01] <vasc> i found out that RotD has some MT-32 patch files. that's probably why the instrument in the Dynamix intro doesn't sound ok
[05:01] <vasc> i don't parse and use those
[05:01] <vasc> SXTITLE.001 or something
[05:02] <vasc> seems to be in Sierra SCI format.
[05:02] <vasc> start bytes are 0x890e
[05:02] <snover> sci resource types are the first byte only, & ~0x80
[05:03] <snover> woo. down to 19 blockers!
[05:03] <snover> and i have patches for 6 of those, so its really like 13.
[05:03] <vasc> can you load a patch file on SCI to test it?
[05:04] <vasc> like the 7.PAK thing
[05:05] <snover> the midi driver is responsible for doing that, but you know what i am going to say next about working on the sound code right now :)
[05:05] <vasc> yes
[05:05] <vasc> i'll ask the other guy
[05:05] <vasc> i basically ripped your code (ahem) to bits, to get the MT-32 + Soundblaster PCM samples working on DGDS.
[05:06] <vasc> i think the MT-32 patches are the last thing to implement.
[05:06] <snover> its all just going to have to get reimplemented later if thats how its been done&
[05:06] <vasc> and maybe, soemthing to detect when a music stops playing.
[05:07] <vasc> i hope so. i hate maintaining code.
[05:07] <vasc> https://github.com/vcosta/scummvm/blob/master/engines/dgds/sound.cpp
[05:07] <vasc> https://github.com/vcosta/scummvm/blob/master/engines/dgds/music.cpp
[05:09] <snover> nobody loves maintaining code, its part of the job though.
[05:09] <vasc> i've tried decoding the Amiga images with a bunch of compressions and layouts, and it's always botched.
[05:12] <vasc> man, i really need to try to read that PLAY.EXE with all those debug symbols in IDA.
[05:12] <vasc> laterz.
[05:12] <-- vasc left irc: Quit: Leaving
[06:22] <-- Mia left irc: Read error: Connection reset by peer
[06:23] --> Axy joined #scummvm.
[06:23] <-- Axy left irc: Changing host
[06:23] --> Axy joined #scummvm.
[06:25] <-- balrog left irc: Ping timeout: 240 seconds
[06:28] --> balrog joined #scummvm.
[06:49] --> waltervn joined #scummvm.
[06:49] #scummvm: mode change '+o waltervn' by ChanServ!ChanServ@services.
[07:43] --> ajax16384 joined #scummvm.
[07:43] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[07:54] <-- Strangerke left irc: Ping timeout: 255 seconds
[08:30] --> girafe joined #scummvm.
[09:00] <-- Dark-Star left irc: Ping timeout: 240 seconds
[09:03] <-- LittleToonCat left irc: Remote host closed the connection
[09:08] --> Strangerke|work joined #scummvm.
[09:08] <Strangerke|work> hi guys
[09:09] --> Dark-Star joined #scummvm.
[09:12] <-- girafe left irc: Read error: Connection reset by peer
[10:14] <L0ngcat> helooo
[10:18] --> TMM joined #scummvm.
[10:18] <-- TMM left irc: Changing host
[10:18] --> TMM joined #scummvm.
[10:18] #scummvm: mode change '+o TMM' by ChanServ!ChanServ@services.
[10:30] <-- Strangerke|work left irc: Quit: Page closed
[10:31] --> Strangerke|work joined #scummvm.
[11:04] --> f2k joined #scummvm.
[11:09] --> ccawley2011 joined #scummvm.
[12:23] --> GitHub140 joined #scummvm.
[12:23] <GitHub140> [scummvm] ccawley2011 opened pull request #1066: RISCOS: Fix using filenames with spaces (master...riscos-fs) https://git.io/vFdcS
[12:23] GitHub140 (GitHub140@192.30.252.42) left #scummvm.
[12:25] --> dreammaster joined #scummvm.
[12:25] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[12:25] --> GitHub1 joined #scummvm.
[12:25] <GitHub1> [scummvm] dreammaster pushed 5 new commits to master: https://git.io/vFdCU
[12:25] <GitHub1> scummvm/master b20670e Colin Snover: TSAGE: Remove member access of null pointer
[12:25] <GitHub1> scummvm/master 7fc9e38 Colin Snover: TSAGE: Fix invalid C-style casts to unrelated classes
[12:25] <GitHub1> scummvm/master fe45300 Colin Snover: TSAGE: Fix deadlocks in audio code...
[12:25] GitHub1 (GitHub1@192.30.252.42) left #scummvm.
[12:25] --> GitHub194 joined #scummvm.
[12:25] <GitHub194> [scummvm] dreammaster closed pull request #1065: TSAGE: Various fixes (master...tsage-fixes) https://git.io/vF5gd
[12:25] GitHub194 (GitHub194@192.30.252.42) left #scummvm.
[12:26] <wjp> ccawley2011: I'm not entirely sure how the contents of that riscos/spaces patch correspond to the commit message
[12:28] <ccawley2011> wjp: RISC OS uses non-breaking spaces in filenames - ScummVM expects regual spaces.
[12:29] <wjp> and the other two/thirds of the patch?
[12:30] <wjp> s~/~ ~
[12:30] <ccawley2011> The commit also contains a change to better detect the presence of an absolute path. I can split the commit if needed.
[12:30] <wjp> A mention in the commit message would be good at least
[12:39] <wjp> Thanks
[12:51] <-- dreammaster left irc:
[13:45] --> jamm joined #scummvm.
[13:45] <-- jamm left irc: Changing host
[13:45] --> jamm joined #scummvm.
[14:12] --> Littleboy joined #scummvm.
[14:12] #scummvm: mode change '+o Littleboy' by ChanServ!ChanServ@services.

[14:44] <-- ajax16384 left irc: Quit: Leaving
[14:54] --> Littleboy joined #scummvm.
[14:54] #scummvm: mode change '+o Littleboy' by ChanServ!ChanServ@services.
[15:01] --> jam__ joined #scummvm.
[15:01] <-- jamm left irc: Ping timeout: 246 seconds
[15:44] <-- jam__ left irc: Quit: Leaving
[15:44] --> jamm joined #scummvm.
[15:44] <-- jamm left irc: Changing host
[15:44] --> jamm joined #scummvm.
[15:55] <-- f2k left irc: Quit: Leaving
[16:12] --> Henke37 joined #scummvm.
[16:21] <-- jamm left irc: Ping timeout: 252 seconds
[16:38] <-- TMM left irc: Quit: Ex-Chat
[17:34] --> LittleToonCat joined #scummvm.
[17:44] --> yegortimoshenko joined #scummvm.
[17:44] <yegortimoshenko> is it true that ScummVM now accepts RPG engines into the tree?
[17:45] <snover> yes
[17:47] <yegortimoshenko> snover: great! thanks. there is a RPG Maker 2000/2003 engine, EasyRPG Player, maybe i could work on getting it into ScummVM...
[17:47] <snover> no such thing exists.
[17:47] <snover> oh
[17:47] <snover> sorry, i misread your sentence, i thought you said is there.
[17:49] <snover> so, yes, that could potentially be a thing.
[17:51] <wjp> are you one of the EasyRPG Player devs?
[18:10] --> SylvainTV joined #scummvm.
[18:10] #scummvm: mode change '+o SylvainTV' by ChanServ!ChanServ@services.
[18:13] <yegortimoshenko> wjp: no, i'm not
[18:22] --> Strangerke joined #scummvm.
[18:22] #scummvm: mode change '+o Strangerke' by ChanServ!ChanServ@services.
[18:25] --> girafe joined #scummvm.
[18:27] <-- Strangerke left irc: Ping timeout: 258 seconds
[18:28] --> Farmboy0 joined #scummvm.
[18:28] <-- Farmboy0 left irc: Changing host
[18:28] --> Farmboy0 joined #scummvm.
[19:03] --> fdelapena joined #scummvm.
[19:04] --> GitHub179 joined #scummvm.
[19:04] <GitHub179> [scummvm] csnover pushed 1 new commit to master: https://git.io/vFFm6
[19:04] <GitHub179> scummvm/master ddbfb85 Cameron Cawley: CONFIGURE: Only enable ARM assembly on older ARM hosts...
[19:04] GitHub179 (GitHub179@192.30.252.34) left #scummvm.
[19:04] --> GitHub52 joined #scummvm.
[19:04] <GitHub52> [scummvm] csnover closed pull request #1067: CONFIGURE: Only enable assembly code on armv4 and armv5 devices (master...arm-asm) https://git.io/vFd8h
[19:04] GitHub52 (GitHub52@192.30.252.42) left #scummvm.
[19:08] <-- girafe left irc: Ping timeout: 248 seconds
[19:09] --> girafe joined #scummvm.
[19:13] --> vasc joined #scummvm.
[19:13] <vasc> https://pastebin.com/8x1wrKb4
[19:14] <vasc> Space Quest 5 Demo - Dynamix DGDS game engine - symbols dump with TDUMP.EXE
[19:14] <vasc> it even has the source code filenames in it.
[19:15] --> ajax16384 joined #scummvm.
[19:15] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[19:15] <vasc> this should help understand the script opcodes and things like that i think.
[19:17] <vasc> we should say thanks to DAVEM for keeping the symbols on the binary.
[19:17] <-- _sev left irc: Quit: Leaving
[19:17] Nick change: Storm-AFK -> StormWolf
[19:18] Nick change: StormWolf -> Stormkeeper
[19:22] <vasc> i also found a nice tool for disassembly
[19:23] <vasc> https://github.com/uxmal/reko
[19:24] --> Ghabry joined #scummvm.
[19:25] <wjp> Wow, even with variable names and struct definitions
[19:25] <vasc> they have binaries on SF.net.
[19:26] <vasc> yep.
[19:26] <vasc> shame i can't do disassembly worth a damn. no skills.
[19:27] <snover> i don't think any of us were born with disassembly skills.
[19:27] <vasc> VM2_DRWQ module seems to have draw code.
[19:27] <wjp> Well, not this kind of disassembly
[19:28] <snover> :)
[19:28] <vasc> well this should make it a lot easier to reverse engineer i think.
[19:28] <wjp> Full stack frame layouts...
[19:29] <wjp> This is quite an amazing amount of info
[19:29] <snover> thanks DAVEM, indeed!
[19:30] <snover> wjp: were you able to unbury yourself from the avalanche of tickets?
[19:31] <vasc> oh... SCR_UNPACK_RECTANGLE
[19:31] <Dark-Star> anyone have a sample disassembly from reko? I'm still looking for a good alternative to IDA
[19:31] <vasc> i guess that's the background image decoder.
[19:32] <vasc> i wonder how one can hook this info into IDA to get actual disassembled code.
[19:32] <vasc> for the bits that matter.
[19:32] <vasc> like the image decoder.
[19:32] <vasc> the rest i think it's more fun to guess.
[19:33] <Dark-Star> IDA has a simple scripting language that can do almost anything to your disassembly (define structures, labels, etc.) it just needs a converter from text (or anything) to IDA script, then you can import/run that script
[19:34] <vasc> you can get a reko binary for windows here: https://sourceforge.net/projects/decompiler/
[19:34] <vasc> and try it out.
[19:34] <vasc> i think it's pretty nice, when it works.
[19:34] <vasc> the developer is really responsive too.
[19:35] --> _sev joined #scummvm.
[19:35] <-- _sev left irc: Changing host
[19:35] --> _sev joined #scummvm.
[19:35] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[19:35] <yegortimoshenko> for cross-reference: asked EasyRPG devs to consider integration, https://github.com/EasyRPG/Player/issues/1310
[19:38] <snover> as far as guessing being more fun, guessing lead to a horrible long tail of hacks-upon-hacks in sci engine thats still being unravelled, so i have a pretty strong experiential reason to recommend doing the RCE
[19:43] <Dark-Star> hm, only available as MSI? I'd prefer a ZIP file with the binaries :)
[19:47] --> GitHub140 joined #scummvm.
[19:47] <GitHub140> [scummvm] csnover closed pull request #1060: COMMON: Add limited support for custom deleters to ScopedPtr (master...smart-ptr-deleter) https://git.io/vF9hq
[19:47] GitHub140 (GitHub140@192.30.252.42) left #scummvm.
[19:59] <snover> yegortimoshenko: what are you wanting to achieve with such a change?
[19:59] <yegortimoshenko> snover: playing RPG Maker 2000/2003 games in ScummVM
[20:00] <snover> whats wrong with playing them in easyrpg?
[20:00] <snover> or to put it another way, what will playing them in scummvm do that playing them in easyrpg doesnt?
[20:01] <vasc> i'll see if i can hook this dump file into IDA
[20:01] <vasc> perhaps ScummVM should split the engine from the games part.
[20:01] <yegortimoshenko> EasyRPG lacks a proper GUI
[20:01] <vasc> it's got a really nice engine for the most part. really useful.
[20:02] <vasc> but for RPGs like that you need more, like map drawing, and first person render, and things like that.
[20:02] <vasc> i think.
[20:02] <vasc> although prolly Lands of Lore and Eye of the Beholder likely already do that.
[20:03] <vasc> i used to be a Freeciv developer, we entertained the idea of doing a strategy game engine for a while.
[20:03] <yegortimoshenko> and also another major reason applies: it is very neat to have a game interpreter handle many game engines rather than just any particular one. imagine if ScummVM was broken down into SCUMM, SCI, etc.
[20:03] <vasc> but it's a really bad idea. the logic is really complicated. the engine ends up becoming a programming language, that no one knows how to write without learning it.
[20:04] <vasc> well, adventure games have a lot of similarities in render code and things like that so i guess it applies.
[20:04] <vasc> rpgs have things which are seldom used in adventure games like hps, and damage routines and things like that.
[20:05] <vasc> and strategy games are even worse.
[20:05] <vasc> but it's not like i can draw an easy separation border between those.
[20:05] <snover> yegortimoshenko: if you just want a gui, maybe you want to make a retroarch integration instead.
[20:05] <vasc> some games kinda fall in between.
[20:06] <vasc> why don't they just use the common libraries code? then if it becomes sensible to make a common base there, do it
[20:06] <vasc> you don't need to be in the tree to use the code.
[20:06] <vasc> i mean i'm developing outside the tree the dgds engine, and it can work
[20:06] <vasc> git is really good at that.
[20:07] <vasc> eventually it might become an issue though.
[20:09] <yegortimoshenko> snover: what's the problem with scummvm integration? considering that scummvm now accepts rpg engines, and the fact that easyrpg uses the same dependencies as scummvm, what's the problem with easyrpg getting integrated into scummvm? other than potential licensing issue
[20:10] <yegortimoshenko> (other than pixman, scummvm presumably has its own blitting code)
[20:10] <snover> well, one non-starter at the moment is that easyrpg is using c++11.
[20:10] <wjp> it'll involve a Serious(tm) amount of work
[20:10] <snover> getting beyond that hurdle, changing all of their library code to use the scummvm common code would be a bunch of work
[20:11] <snover> its not just like dumping the source code into the scummvm tree and changing a few API calls and its done.
[20:11] <snover> the save game system is going to be totally different, the way it reads files from the filesystem is going to be totally different
[20:12] <vasc> to be honest i didn't find the Archive abstraction particularly helpful for DGDS either.
[20:12] <vasc> File and DumpFile are ok.
[20:12] <snover> if it isnt designed to stop and start different games without restarting the entire process it will need to be overhauled so that it can unloaded and reloaded
[20:12] <vasc> DGDS games have an index file which lists filename hashes and volume names and offsets.
[20:12] --> Strangerke joined #scummvm.
[20:13] <wjp> easyrpg heavily uses the STL too
[20:13] <vasc> it consults that index file before actually opening a volume, which is what an archive is.
[20:13] <vasc> but it already knows the file offset in the archieve when it opens it.
[20:13] <vasc> because of the index.
[20:13] <snover> anyway, those are all just the things i can think of off the top of my head. as with any non-trivial software there are probably hundreds of other smaller issues.
[20:15] <wjp> vasc: and why does that make the Archive abstraction unhelpful?
[20:15] <vasc> the interface is mostly useless.
[20:15] <wjp> ...because?
[20:15] <vasc> let's say i want to open a file.
[20:16] <vasc> i go to the index, compute the filename hash, and then find the entry with that hash, and get the volume name and file offset inside that volume.
[20:16] <yegortimoshenko> snover: these reasons probably should have applied to any other major scummvm integrations in the past, though: Sarien, FreeSCI...
[20:16] <vasc> then when i open the volume (Archive) i want to just open a file at that offset and read it.
[20:17] <vasc> the interface wants me to pass the filename and then it wants to find the file in the archive.
[20:17] <vasc> with the filename.
[20:17] <vasc> it's all different.
[20:17] <vasc> it also has a file matcher pattern.
[20:18] <vasc> but the index doesn't store filename, it stores file hashes.
[20:18] <vasc> the filenames are in the volumes.
[20:18] <vasc> so to do a file match, it would take a lot of filesystem opens and closes
[20:20] <wjp> probably nothing will break if you don't implement matching as long as you never use it yourself
[20:21] <wjp> but the usual approach with our archives is to pre-cache the set of filenames
[20:22] <yegortimoshenko> snover: and you're raising a good point as to why i think libretro cores are inferior to scummvm integration: libretro cores offer very little in terms of consistency, meaning they are not much better than just a bunch of game engine executables: https://buildbot.libretro.com/docs/specs/api/
[20:23] <wjp> vasc: the main goal of the Archive/SearchSet classes is to combine files from multiple sources (directories, volumes, ...), by the way. Don't know if that is relevant for you
[20:26] <wjp> in any case, if it doesn't help you, there's no need to use it. We don't mind :-)
[20:26] <wjp> (for the volumes, I mean)
[20:28] <vasc> so to do a file match, it would take a lot of filesystem opens and closesaxpcbm85
[20:29] <vasc> oops
[20:29] <vasc> :-)
[20:29] <vasc> time to change the password again
[20:29] --> criezy joined #scummvm.
[20:29] #scummvm: mode change '+o criezy' by ChanServ!ChanServ@services.
[20:29] <vasc> haven't done this since i was an undergrad.
[20:35] <logix> make sure correctly staple a battery to a horse
[20:35] <logix> err, +"to"
[20:35] <-- _sev left irc: Quit: This computer has gone to sleep
[20:39] --> _sev joined #scummvm.
[20:39] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[20:51] --> GitHub185 joined #scummvm.
[20:51] <GitHub185> [scummvm] csnover pushed 1 new commit to master: https://git.io/vFFRh
[20:51] <GitHub185> scummvm/master 0c222d9 Cameron Cawley: CREDITS: Add credits for RISC OS port...
[20:51] GitHub185 (GitHub185@192.30.252.41) left #scummvm.
[20:51] --> GitHub0 joined #scummvm.
[20:51] <GitHub0> [scummvm] csnover closed pull request #1068: CREDITS: Add credits for RISC OS port (master...credits) https://git.io/vFdwz
[20:51] GitHub0 (GitHub0@192.30.252.42) left #scummvm.
[21:01] <-- ccawley2011 left irc: Ping timeout: 260 seconds
[21:02] <-- ajax16384 left irc: Quit: Leaving
[21:03] --> ccawley2011 joined #scummvm.
[21:06] <Ghabry> Was there any attempt made (any sourcecode available) to implement the m4 engine? http://wiki.scummvm.org/index.php/M4
[21:08] <criezy> snover: regarding your question about changing the version number, you are aware I suppose of devtools/update-version.pl ?
[21:08] <criezy> This should take care of most of the updates.
[21:09] <criezy> I don't think 1.10 is used anywhere on our website, and I just did a search on the wiki and it appears to be used in a single place that will be easy to update.
[21:09] <snover> criezy: hm, then the wiki search is broken, since it is used on multiple pages (i know since i put it there :))
[21:09] <snover> criezy: yes, it is mentioned in HOWTO-Release. what i dont know is about any of the newer things like weblate
[21:09] <snover> or even other-other things that i dont even know i dont know about
[21:10] <criezy> Weblate is fine I think, although I will take a look to double check. But I don't think there is anything to change there.
[21:11] <criezy> The one I found on the wiki is the one you added for the audio_buffer_size in the User Manual.
[21:11] <snover> its on a bunch of the game pages too for the games newly supported in 1.10
[21:12] <snover> so if the search is not finding those, that is a&problem
[21:12] <criezy> There are two other hits, but I assumed they were related to game versions and not ScummVM versions.
[21:12] <yegortimoshenko> Ghabry: i think it has been dropped: https://github.com/scummvm/scummvm/commit/ef8b106055c638377806e32e24922380118d03c7
[21:13] <criezy> One for the "AGI Mouse Demo" and a bunch on the Sierra_Game_Versions page,
[21:13] <snover> its at e.g. http://wiki.scummvm.org/index.php/RAMA
[21:13] <yegortimoshenko> (also, https://github.com/scummvm/scummvm/search?q=m4)
[21:13] <snover> wildcard search seems to be doing a bit better
[21:13] <snover> search for 1.10*
[21:14] <Ghabry> okay, so the development started but it was dropped.
[21:14] <criezy> Ah interesting. Searching for 1.10 does not include 1.10.0, which gives a lot more hit.
[21:14] <Ghabry> already 6 years ago
[21:14] <criezy> Ha, wildcards work?
[21:14] <snover> apparently! i dont know anything about this crazy thing, i just tried it on a whim.
[21:15] <criezy> But most of those look like something I would update when the release is done anyway (as per http://wiki.scummvm.org/index.php/HOWTO-Release#Updating_the_wiki )
[21:17] <criezy> But I can take care of updating them before the release if we want to do that.
[21:18] <-- Strangerke left irc: Ping timeout: 240 seconds
[21:18] <snover> considering the duration between releases it is a little scary to me that making those updates is currently deferred until release time. maybe you have a memory to remember all the things, but i sure dont :)
[21:19] <snover> i will run the update-version and get the testing spreadsheet and news post finished up, if you will take care of the wiki.
[21:19] <criezy> Update of the game page is listed on the HOWTO-Release page, because I don't really have the memory to remember.
[21:20] <criezy> I will add update of the new ports as well, since that is missing from the list of things to update.
[21:20] <criezy> And yes, I can take care of the wiki updates.
[21:20] --> Strangerke joined #scummvm.
[21:21] <snover> awesome. thank you!
[21:29] <snover> criezy: i moved the release testing page on the wiki since i needed the link for the news, otherwise ill leave it in your hands
[21:30] <criezy> OK. I am working on something else right now but I will do the remaining updates once I have finished.
[21:32] <vasc> since you're talking about Microprose games, what about Bloodnet?
[21:33] <vasc> does it use an engine?
[21:41] <vasc> hey guys, how can i get a ScummVM wiki account? i wanted to edit the DGDS entries.
[21:41] <vasc> it says to talk to sev or Strangerke
[21:42] <snover> i feel like you just answered your own question
[21:46] <vasc> oh right the SDS parser.
[21:47] <snover> hold onto your butts.
[21:47] --> GitHub180 joined #scummvm.
[21:47] <GitHub180> [scummvm] csnover pushed 1 new commit to master: https://git.io/vFFVK
[21:47] <GitHub180> scummvm/master c13defd Colin Snover: RELEASE: 1.10 will henceforth be known as 2.0...
[21:47] GitHub180 (GitHub180@192.30.252.42) left #scummvm.
[21:47] --> GitHub134 joined #scummvm.
[21:47] <GitHub134> [scummvm] criezy pushed 1 new commit to master: https://git.io/vFFVi
[21:47] <GitHub134> scummvm/master 96fab1d Thierry Crozat: I18N: Update translations templates
[21:47] GitHub134 (GitHub134@192.30.252.42) left #scummvm.
[21:47] <vasc> *clap* *clap*
[21:48] <snover> hm, the po-files still dont have a correct Project-Id-Version
[21:48] <snover> but it seems to have been that way for a while
[21:50] --> GitHub48 joined #scummvm.
[21:50] <GitHub48> [scummvm] csnover tagged desc/2.0.0git at 68d1f71: https://git.io/vFFwU
[21:50] GitHub48 (GitHub48@192.30.252.40) left #scummvm.
[21:50] <snover> i wonder if that desc tag might actually should go back to 1.10.0 instead
[21:51] <snover> probably it should&
[21:51] <snover> lets never do this again :)
[21:57] <-- kurtwr left irc: Read error: Connection reset by peer
[21:58] --> kurtwr joined #scummvm.
[22:00] <somaen> wjp: Hrm, Hamlet is very much not supported
[22:00] <somaen> (replying to what you said on the evening of the 16th)

[22:01] <snover> somaen: thanks, thats all i need to know to get rid of https://bugs.scummvm.org/ticket/6568
[22:01] <somaen> We do not officially support any WME games beyond chivalry
[22:01] <somaen> That's at least the only one I officially announced support for
[22:01] Action: vasc works at demangling the function signatures.
[22:02] <snover> somaen: sure, though nothing was ever done at http://forums.scummvm.org/viewtopic.php?t=13062
[22:02] <somaen> Ghost in the Shell I never got to test
[22:02] <somaen> Hmm, albadross has been busy without me noticing
[22:03] <somaen> What? Oknytt is completable?
[22:03] <somaen> If I find some time, I guess I'll have to do some announcing if this is correct
[22:03] <somaen> Oh, and sorry about not having taken the time to follow up recently.
[22:04] <snover> no worries, life gets busy.
[22:04] <snover> ghost in the sheet would be nice to correct if that one crash at the end of the game really is the only problem.
[22:05] <-- Henke37 left irc: Quit: ERR_SHUTDOWN
[22:05] <somaen> That would require me to own the game, unless it's reproducible in the demo
[22:05] <snover> yeah, the publisher seems to not be& publishing it any more
[22:05] <snover> there are a couple copies on ebay.
[22:06] <-- waltervn left irc: Quit: Leaving
[22:08] <snover> going through that list of tested wintermute games is on my list of todos for the next release
[22:08] --> GitHub127 joined #scummvm.
[22:08] <GitHub127> [scummvm-web] csnover pushed 1 new commit to master: https://git.io/vFFo8
[22:08] <GitHub127> scummvm-web/master 76d393a Colin Snover: WEB: Add release testing news for the next version
[22:08] GitHub127 (GitHub127@192.30.252.40) left #scummvm.
[22:09] <somaen> East Side Story and Dirty Split are fairly safe, having been completed by at least 2 people at this point
[22:09] <somaen> I think there's glitches and known bugs in the helga-games
[22:09] --> ec2-user4 joined #scummvm.
[22:10] Nick change: ec2-user4 -> peterkohaut
[22:38] <-- yegortimoshenko left irc: Remote host closed the connection
[22:40] <vasc> hmm
[22:56] <-- Farmboy0 left irc: Remote host closed the connection
[22:56] <vasc> the scripts have loops? hmm
[22:58] <-- Cheeseness left irc: Ping timeout: 248 seconds
[22:58] <criezy> Is the news post ready to go live or should I wait before updating the web site?
[22:59] <criezy> I see the date is November 22, which is tomorrow (I think?), so I suppose it is meant to go live only tomorrow?
[23:00] <Lightkey> it's.. now! :-)
[23:02] <criezy> It's not yet tomorrow for either snover (the author of the news post) or myself :P
[23:02] <-- _sev left irc: Quit: This computer has gone to sleep
[23:03] --> Cheeseness joined #scummvm.
[23:08] <criezy> snover: was there any reason why you left the Release Testing/1.10.0 page on the wiki with a redirect link to the 2.0.0 page?
[23:08] <criezy> Should I remove it?
[23:13] <vasc> it seems to have line and circle drawing routines. i guess that's how the text balloons are drawn.
[23:29] <vasc> hah, they used the same variable name i used.
[23:32] --> Stanto joined #scummvm.
[23:33] <Stanto> Does anyone know what the turnaround time is for getting forum accounts verified?
[23:34] <criezy> I don't, but you can try to ping _sev or Strangerke in here to get it done faster.
[23:40] <snover> criezy: i guess i just dont want to confuse anyone
[23:41] <snover> in my mind a redirect seemed clearer about this is what happened than just not having a version
[23:41] <criezy> OK, that makes sense. I will leave the redirect then.
[23:43] <criezy> snover: also, just to make sure, the news post is meant to go live tomorrow, right? Not today?
[23:43] <Stanto> I was wanting to help out with some posts about Cyan's Riven and the graphical dithering.
[23:46] <snover> criezy: i dont really know. _sev said he didnt have time to write it, so i did. i have really no idea what is going on beyond that. i was not blessed with ownership over the release process.
[23:46] <snover> i am just trying to keep things moving forward.
[23:47] <criezy> And thank you for that.
[23:47] <criezy> I was asking because I saw that it's data is tomorrow and I didn't want to update the web site too early ;)
[23:47] <snover> i figured there would be some amount of lag so i just gave it a tomorrow-date since i had to give it *some* date.
[23:47] <snover> originally it said the 20th ;)
[23:49] <snover> i guess the daily builds are not going to be updated with the correct version number until tomorrow anyway, so.
[23:49] <snover> it would probably not be a great idea to send it tonight.
[23:50] <criezy> Indeed.
[23:58] <snover> i guess i forgot to look for the word lock when i was searching for tickets, i found https://bugs.scummvm.org/ticket/5862 and it makes me wonder if the audio mixer mutex ought to just be exposed, since it seems like a few times now there are situations where the game is holding its own separate mutex for its audio code and then puts in an OPL callback and now youre deadlocked
[00:00] --- Wed Nov 22 2017