ScummVM IRC Log - 04 Oct 2019

[Back to Index]

[00:36] <Scummette> [scummvm] digitall pushed 1 new commits to master: https://git.io/JecPw
[00:36] <Scummette> scummvm/master 522c5b6 digitall: BLADERUNNER: Add Some Missing Switch Default Cases
[01:09] <-- Asterisk left irc: Ping timeout: 246 seconds
[01:13] --> Asterisk joined #scummvm.
[01:16] <Scummette> [scummvm] digitall pushed 1 new commits to master: https://git.io/JecXL
[01:16] <Scummette> scummvm/master 40337b6 digitall: BLADERUNNER: Add More Missing Switch Default Cases in ESPER Script
[01:25] <-- ldevulder left irc: Remote host closed the connection
[01:26] --> ldevulder joined #scummvm.
[01:29] <-- eriktorbjorn left irc: Ping timeout: 245 seconds
[01:31] --> eriktorbjorn joined #scummvm.
[01:32] <Scummette> [scummvm] digitall pushed 1 new commits to master: https://git.io/JecXm
[01:32] <Scummette> scummvm/master 8a96a18 digitall: RECORDER: Add Missing Default Switch Cases
[01:37] --> DominusExult joined #scummvm.
[01:37] <-- DominusExult left irc: Changing host
[01:37] --> DominusExult joined #scummvm.
[01:41] <-- Dominus left irc: Ping timeout: 245 seconds
[01:41] Nick change: DominusExult -> Dominus
[01:44] --> dreammaster joined #scummvm.
[01:44] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[02:32] <Scummette> [scummvm] digitall pushed 1 new commits to master: https://git.io/JecXH
[02:32] <Scummette> scummvm/master 50b0d62 digitall: SCI: Really Fix MSVC Warning
[02:53] <Scummette> [scummvm] digitall pushed 1 new commits to master: https://git.io/JecXp
[02:53] <Scummette> scummvm/master 4d91101 digitall: MT32: Add Missing Default Switch Cases
[02:57] <-- yuv422 left irc: Ping timeout: 240 seconds
[03:00] --> yuv422 joined #scummvm.
[03:00] #scummvm: mode change '+o yuv422' by ChanServ!ChanServ@services.
[03:13] <Grogbot> <Henke37> https://pastebin.com/qc9EBLfk
[03:29] <Scummette> [scummvm] digitall pushed 2 new commits to master: https://git.io/Jec1n
[03:29] <Scummette> scummvm/master cee1b56 digitall: COMMON: Add Missing Default Switch Cases in DCT Class
[03:29] <Scummette> scummvm/master 4a2cbd9 digitall: COMMON: Add Missing Switch Default Case in ZLIB Class
[04:49] <-- dreammaster left irc:
[05:51] <-- eriktorbjorn left irc: Quit: Leaving.
[07:03] <Scummette> [scummvm] digitall pushed 2 new commits to master: https://git.io/JecMF
[07:03] <Scummette> scummvm/master ee79aba digitall: SCUMM: Fix GCC Duplicated Condition Warning
[07:03] <Scummette> scummvm/master 50f71c8 digitall: STARTREK: Fix GCC Duplicated Condition Warning
[07:06] --> Begasus joined #scummvm.
[07:08] --> Begas_VBox joined #scummvm.
[07:51] <LePhilousophe> hey!
[07:51] <LePhilousophe> if some user has an unknown version of a game we support and he doesn't know how to add it and there is operations to do on the game binary to get informations
[07:51] <LePhilousophe> how should I proceed?
[07:57] <Grogbot> <antoniou79> Is this the process you are referring to? I am not sure the article is up to date
[07:57] <Grogbot> <antoniou79> https://wiki.scummvm.org/index.php?title=Reporting_unknown_MD5_checksums
[07:59] <Grogbot> <antoniou79> On an unrelated note, my google search brought me to this ScummVM wiki page for Android Guide: https://wiki.scummvm.org/index.php/Android/Guide
[08:00] <Grogbot> <antoniou79> Which is outdated and also mentions donations to ScummVM project via SourceForge. Is this still a thing?
[08:07] <LePhilousophe> antoniou79, it's not just the md5. I need to extract information from the binary
[08:08] <-- timofonic left irc: Remote host closed the connection
[08:09] <Grogbot> <SupSuper> @antoniou79 well scummvm isn't using SF anymore but all the donation links seem to lead to the correct place at least
[08:10] <Grogbot> <SupSuper> might be worth scouring the wiki for any outdated/broken links, but they all seem to be on old pages
[08:11] <Grogbot> <antoniou79> @LePhilousophe other than them sending you the binary, I suppose another solution would be for them to use a hex viewer to send values to use at specific addresses if that's what you need
[08:12] <LePhilousophe> antoniou79: is it legal to proceed like this with him sending me the binary?
[08:12] <Grogbot> <SupSuper> LePhilousophe: if i need to reverse a specific version of a game i can't acquire otherwise, i usually just ask for the file(s) in private. you don't need the whole game so should be fine
[08:12] <LePhilousophe> ok
[08:12] <LePhilousophe> I will do this so
[08:12] <LePhilousophe> thanks
[08:12] <Grogbot> <antoniou79> Yes I am with SupSuper here.
[08:38] --> vliaskov joined #scummvm.
[08:42] --> ny00123 joined #scummvm.
[08:55] --> ldevulder_ joined #scummvm.
[08:58] <-- ldevulder left irc: Ping timeout: 252 seconds
[09:29] <Grogbot> <antoniou79> So as far as I can tell, our buildbot builds armv5 apk (master-android_arm worker). In our downloads page we distribute a separate v7a package and also an apk with multi-eabi (all target architectures bundled). This was/is set up for the Android SDL port, which was what was used for the 2.0.0 release, but not for the plain port. I suppose we need at least the separate package for v7a.
[09:30] <Grogbot> <antoniou79> The rest of the packages (arm-v8a, arm-x86, arm-x86_64) are already being build for the plain port
[09:45] <LePhilousophe> why can't you bundle v7a?
[09:52] <-- jamm left irc: Quit: Leaving
[09:53] --> jamm joined #scummvm.
[10:19] --> timofonic joined #scummvm.
[10:20] --> ST joined #scummvm.
[10:20] #scummvm: mode change '+o ST' by ChanServ!ChanServ@services.
[10:27] <-- jamm left irc: Read error: Connection reset by peer
[10:28] --> jamm joined #scummvm.
[10:30] <Grogbot> <antoniou79> @LePhilousophe, what I mean is our buildbot does not build a separate package for v7a as far as I can tell, and now that I think about it, it also does not build a separate package for arm-x86_64 either.
[10:30] <Grogbot> <antoniou79> Should we build them as well?
[10:33] <Grogbot> <antoniou79> About a multi-architecture bundle, I am not sure if it makes sense to continue distributing it. It seems that ScummVM does this only for releases -- the buildbot does not build such a thing daily, and the resulting package is quite huge.
[10:59] <Scummette> [scummvm] antoniou79 pushed 2 new commits to master: https://git.io/JecHt
[10:59] <Scummette> scummvm/master e5a65c2 ShadowNate: CONFIGURE: Add clarification of possible target android hosts
[10:59] <Scummette> scummvm/master 8ee774f ShadowNate: CONFIGURE: Add android-mips64 in clarification for target hosts
[10:59] <-- jamm left irc: Ping timeout: 245 seconds
[11:13] --> jammm joined #scummvm.
[11:13] <-- jammm left irc: Changing host
[11:13] --> jammm joined #scummvm.
[11:20] --> cd joined #scummvm.
[11:21] <LePhilousophe> @antoniou79: is there an alternative to distribute a multi architecture apk? or is this only through play store?
[11:21] <LePhilousophe> maybe we could automate buildbot package distribution through store too
[11:27] <Grogbot> <antoniou79> As far as I can tell the alternative is to distribute separate apks per target architecture, as the buildbot does now for the daily builds (we just add the ones missing (v7a and x86_64). From a quick google search it seems that google supported uploading separate APKs for the same application, but they are sort of moving away from that (more accurately: they seem to be replacing that) with the app bundles they introduced.
[11:28] <Grogbot> <antoniou79> Pertinent links should be: https://stackoverflow.com/questions/42049253/publish-multiple-apks-to-google-play-store https://developer.android.com/google/play/publishing/multiple-apks.html#Rules
[11:28] <Grogbot> <antoniou79> and https://developer.android.com/guide/app-bundle
[11:40] <Grogbot> <Henke37> my view on game binaries is that ideally the game engine code shouldn't need them. but if it does, then it knows enough to scoop out the information no matter the exact version
[11:41] Nick change: ldevulder_ -> ldevulder
[11:53] <LePhilousophe> @antoniou79: yeah the bundle stuff depends on Google Play
[11:54] <LePhilousophe> there is intersting stuff: for example we could download only engines user want but that would complexify the platform code
[11:55] <Grogbot> <Henke37> and require a well working plugin system
[11:56] <LePhilousophe> in fact that would be useless as we need all engines to check for games
[11:57] <LePhilousophe> we would need a screen with the list of all engines and let user check which plugin he wants
[11:57] <LePhilousophe> that seems quite complex
[11:58] <Grogbot> <Henke37> that is a problem yes. but that is a solvable problem
[11:58] <Grogbot> <Henke37> it requires separating the real engine code from the metaengines.
[12:00] --> P2E_ joined #scummvm.
[12:04] <-- P2E left irc: Ping timeout: 276 seconds
[12:39] <-- yuv422 left irc: Remote host closed the connection
[12:39] --> yuv422 joined #scummvm.
[12:39] #scummvm: mode change '+o yuv422' by ChanServ!ChanServ@services.
[13:23] <-- timofonic left irc: Ping timeout: 265 seconds
[13:24] --> timofonic joined #scummvm.
[13:29] <timofonic> What about releasing ScummVM nightly builds as beta in Google Play? Many apps do that, there's Nightly as the nightly builds of the Firefox browser. This way, there would be a lot more betatesting in the ScummVM Android port.
[13:36] <-- yuv422 left irc: Remote host closed the connection
[13:37] --> yuv422 joined #scummvm.
[13:37] #scummvm: mode change '+o yuv422' by ChanServ!ChanServ@services.
[13:50] --> mlk joined #scummvm.
[13:54] <Scummette> [scummvm] sev- pushed 4 new commits to master: https://git.io/Jec5Y
[13:54] <Scummette> scummvm/master 72a5d9d sev-: JANITORIAL: Whitespace fixes
[13:54] <Scummette> scummvm/master dbf56a3 sev-: COMMON: Add methods to U32String to match String
[13:54] <Scummette> scummvm/master 48b261c sev-: GRAPHICS: MACGUI: Implement Windows-style shortcut underlining
[14:07] <-- timofonic left irc: Ping timeout: 240 seconds
[14:08] --> timofonic joined #scummvm.
[14:14] --> BeefEats joined #scummvm.
[14:35] <-- timofonic left irc: Read error: Connection reset by peer
[14:35] --> timofonic joined #scummvm.
[14:55] --> girafe joined #scummvm.
[15:18] <Scummette> [scummvm] sev- pushed 6 new commits to master: https://git.io/JecFp
[15:18] <Scummette> scummvm/master 15b44e9 sev-: GRAPHICS: MACGUI: Remove extra debug output
[15:18] <Scummette> scummvm/master 28ac30b sev-: GRAPHICS: MACGUI: Properly process action menu ids from PE exe
[15:18] <Scummette> scummvm/master 02ff887 sev-: GRAPHICS: MACGUI: Enhande printMenu output
[15:36] --> eriktorbjorn joined #scummvm.
[15:36] #scummvm: mode change '+o eriktorbjorn' by ChanServ!ChanServ@services.
[15:43] <-- ny00123 left irc: Quit: Leaving
[16:31] <Scummette> [scummvm] tsoliman pushed 1 new commits to branch-2-1: https://git.io/JecNu
[16:31] <Scummette> scummvm/branch-2-1 8507b7e tsoliman: MAEMO: Update debian/changelog
[16:34] <Scummette> [scummvm] tsoliman pushed 1 new commits to master: https://git.io/JecNw
[16:34] <Scummette> scummvm/master f7a1961 tsoliman: MAEMO: Update debian/changelog
[16:47] --> ajax16384 joined #scummvm.
[16:47] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[16:53] <-- borosky left irc:
[17:09] <-- Lightkey left irc: Ping timeout: 264 seconds
[17:09] <Scummette> [scummvm] sev- pushed 2 new commits to master: https://git.io/JecAC
[17:09] <Scummette> scummvm/master 93e8109 sev-: GRAPHICS: MACGUI: Add possibility to pass borders as a surface
[17:09] <Scummette> scummvm/master c5f3e12 sev-: PINK: Add built-in empty border and use for texts
[17:21] --> Lightkey joined #scummvm.
[17:21] <Grogbot> <Henke37> https://pastebin.com/kMSK14eL
[17:21] <Grogbot> <Henke37> 74>D:\scummvm\scummvm\graphics\macgui\macmenu.cpp(339,13): warning C4018: '>=': signed/unsigned mismatch
[17:40] --> SylvainTV joined #scummvm.
[17:40] #scummvm: mode change '+o SylvainTV' by ChanServ!ChanServ@services.
[18:07] --> ink0gnito joined #scummvm.
[18:21] <-- Begas_VBox left irc: Quit: Vision[0.10.3]: i've been blurred!
[18:21] <-- Begasus left irc: Quit: Ex-Chat
[18:25] <Scummette> [scummvm] sev- pushed 2 new commits to master: https://git.io/JecpZ
[18:25] <Scummette> scummvm/master 94e543c sev-: GRAPHICS: MACGUI: Implement enforcing built-in fonts
[18:25] <Scummette> scummvm/master 15ac5da sev-: PINK: Do not use Mac fonts for menu rendering
[18:40] <-- vliaskov left irc: Remote host closed the connection
[18:46] <-- jammm left irc: Ping timeout: 240 seconds
[18:49] --> P2E joined #scummvm.
[18:52] <-- P2E_ left irc: Ping timeout: 252 seconds
[18:52] <Grogbot> <sev> @everyone I want to start tagging soon
[18:52] <Grogbot> <sev> any objections?
[18:53] <Grogbot> <Henke37> "tagging"? as in git tags?
[18:53] <Grogbot> <sev> I give 1-2 hours for objections
[18:53] <Grogbot> <sev> yes, release tagging
[18:53] <Grogbot> <sev> so no changes to branch-2-1 will be accepted
[18:53] <Grogbot> <rootfather> @sev all systems GO from my point of view.
[19:00] <Grogbot> <DreamMaster> No problems from me
[19:01] <Grogbot> <tsoliman> You still intend to branch-2-1-0 from branch-2-1 and then tag off of that, right?so it's not that changes won't be accepted, but rather that they won't end up in the release. They'll catch the ride on to the next 2.1.x
[19:02] <Grogbot> <tsoliman> I'm building one last build right now (taking forever) to make sure it still builds - it did yesterday
[19:03] <Grogbot> <sev> @tsoliman that is correct, but I would like to get things in 2.1.0, and preferably avoid 2.1.1
[19:03] <Grogbot> <tsoliman> Titanic is still too titanic for the official maemo SDK (but not for the unofficial one) so it won't be in the release (again)
[19:04] <Grogbot> <DreamMaster> Pity, but understandable. From past discussions, Blade Runner is not going to be available for all systems either
[19:05] <Scummette> [scummvm] ccawley2011 pushed 1 new commits to branch-2-1: https://git.io/JechE
[19:05] <Scummette> scummvm/branch-2-1 8f47927 ccawley2011: BLADERUNNER: Remove use of C99 math functions
[19:05] <Grogbot> <antoniou79> I thought only psp would be missing out on blade runner
[19:06] <Grogbot> <antoniou79> Due to os limitations of concurrent open files
[19:06] <Grogbot> <tsoliman> I'd be actually tempted to make a special package with the new compiler (it is just the compiler and not the rest of the SDK) if I got even one active maemo user in 2019
[19:07] <Grogbot> <tsoliman> currently the situation is "install the most recent package, then overwrite the binary from the new compiler" which is like how the old iOS buildbot releases were like
[19:07] <Grogbot> <DreamMaster> Right.. the PSP. I was trying to remember which system it was, but couldn't.
[19:08] <Grogbot> <tsoliman> the PSP is actually pretty popular .. it's found a resurgence in popularity because of recent .. ahem .. advances
[19:09] <Grogbot> <DreamMaster> https://www.youtube.com/watch?v=PLRZ0dIvwHY
[19:09] <Grogbot> <tsoliman> I'm actually half-seriously considering a switch lite
[19:10] <Grogbot> <antoniou79> Too late to do anything about the psp situation now. Not sure of it s worth making it a priority
[19:10] <Grogbot> <tsoliman> all clear from me 👍
[19:12] <Grogbot> <antoniou79> It is an a OK for me too
[19:24] <Lightkey> While I don't understand how the game flags work, it looks to me like at least hdb is still using ADGF_TESTING and also prince?
[19:27] <Grogbot> <tsoliman> When I asked this a long time ago, I was told that that a game's readiness is different from the engine's readiness is different from the release's readiness.
[19:28] <Grogbot> <tsoliman> I don't know if that's still the prevailing thought
[19:29] <Lightkey> Well, in both cases they are single-game engines announced for this release.
[19:31] <Grogbot> <tsoliman> touche
[19:33] <Grogbot> <tsoliman> @sev both of these TESTING entries will make a noise when launched post-release
[19:34] <Grogbot> <tsoliman> Lightkey: the way it works is that TESTING will silence the warning unless the build was built with --enable-release (i.e. a release build)
[19:35] <Grogbot> <tsoliman> this is to allow an engine that has multiple games in various states of supportedness
[19:38] <-- mlk left irc: Quit: The Lounge - https://thelounge.chat
[19:39] --> mlk joined #scummvm.
[19:52] <Grogbot> <GeekOB> Hi @sev , how long does it usually take for the new releases to be available on other platforms? I'm interested in 2.1.0 for the Raspberry Pi.
[19:59] <Lightkey> _sev_: Also cryomni3d/versailles is still using ADGF_TESTING.
[19:59] <_sev_> that is part of the release checklist, to remove those
[19:59] <Grogbot> <tsoliman> cool
[20:00] <Grogbot> <sev> @GeekOB there is no guarantee and no expectations
[20:00] <Grogbot> <sev> @GeekOB we hope that at least major platforms will be ready by time of the release
[20:01] <Grogbot> <sev> which is 13th of October, and Raspberry Pi is not a major platofrm. It depends how fast the port maintainer provides the build
[20:03] <Grogbot> <GeekOB> Understood and much appreciated
[20:03] --> criezy joined #scummvm.
[20:03] #scummvm: mode change '+o criezy' by ChanServ!ChanServ@services.
[20:19] <-- timofonic left irc: Ping timeout: 240 seconds
[20:25] --> timofonic joined #scummvm.
[20:32] <Scummette> [scummvm] sev- pushed 2 new commits to master: https://git.io/JeCe0
[20:32] <Scummette> scummvm/master c17800c sev-: GRAPHICS: MACGUI: Add possibility to setup engine redraw callback
[20:32] <Scummette> scummvm/master 090bb44 sev-: PINK: Properly redraw game screen when browsing menus
[20:50] --> ldevulder_ joined #scummvm.
[20:54] <-- ldevulder left irc: Ping timeout: 265 seconds
[21:06] <-- ajax16384 left irc: Quit: Leaving
[21:13] <Grogbot> <sev> okay, release time, people
[21:13] <Scummette> [scummvm] sev- pushed 3 new commits to branch-2-1: https://git.io/JeCej
[21:13] <Scummette> scummvm/branch-2-1 1587a74 sev-: CRYOMNI3D: Remove ADGF_TESTING flag
[21:13] <Scummette> scummvm/branch-2-1 ce98494 sev-: HDB: Remove ADGF_TESTING flag
[21:13] <Scummette> scummvm/branch-2-1 5e55b09 sev-: PRINCE: Remove ADGF_TESTING flag
[21:14] <Scummette> [scummvm] sev- pushed 3 new commits to master: https://git.io/JeCve
[21:14] <Scummette> scummvm/master 0b5ea11 sev-: CRYOMNI3D: Remove ADGF_TESTING flag
[21:14] <Scummette> scummvm/master 9e2bcfb sev-: HDB: Remove ADGF_TESTING flag
[21:14] <Scummette> scummvm/master 2d0cfc7 sev-: PRINCE: Remove ADGF_TESTING flag
[21:14] <Grogbot> <rootfather> Woohoo
[21:15] <Lightkey> Now, now, release is in nine days, this is how misunderstandings in news happen. ;-)
[21:19] <Scummette> [scummvm] sev- pushed 1 new commits to branch-2-1-0: https://git.io/JeCvY
[21:19] <Scummette> scummvm/branch-2-1-0 857a8eb sev-: RELEASE: This is 2.1.0
[21:19] <Grogbot> <sev> argh
[21:20] <Grogbot> <sev> create_projects is changed, so the release flow is broken
[21:28] <Grogbot> <SupSuper> what changed
[21:28] <Scummette> [scummvm] sev- pushed 1 new commits to branch-2-1-0: https://git.io/JeCvW
[21:28] <Scummette> scummvm/branch-2-1-0 f9f4400 sev-: DISTS: Generated Code::Blocks and MSVC project files
[21:29] <Scummette> [scummvm] sev- pushed 1 new commits to master: https://git.io/JeCv8
[21:29] <Scummette> scummvm/master 4137a7e sev-: BUILD: Update ideproject target to the current create_project behavior
[21:29] <Scummette> [scummvm] sev- pushed 1 new commits to branch-2-1: https://git.io/JeCv4
[21:29] <Scummette> scummvm/branch-2-1 3d8b4b5 sev-: BUILD: Update ideproject target to the current create_project behavior
[21:36] <-- timofonic left irc: Ping timeout: 252 seconds
[21:38] <Scummette> [scummvm] sev- pushed 0 new commits to refs/tags/v2.1.0: https://git.io/JeCvV
[21:39] <Scummette> [scummvm] sev- pushed 1 new commits to branch-2-1: https://git.io/JeCv6
[21:39] <Scummette> scummvm/branch-2-1 e35ed3b sev-: RELEASE: This is 2.1.1pre
[21:39] --> timofonic joined #scummvm.
[21:39] <Scummette> [scummvm] sev- pushed 0 new commits to refs/tags/desc/2.1.1pre: https://git.io/JeCvi
[21:42] Topic changed on #scummvm by _sev_!sev@scummvm/undead/sev: ScummVM 2.1.0 tagged || https://www.scummvm.org/faq/ || No warez, no Game Requests, read the FAQ! || Channel logs: https://logs.scummvm.org
[22:10] <_sev_> Done
[22:10] <Grogbot> <sev> DONE
[22:10] <Strangerke> Nice :)
[22:10] <Grogbot> <rootfather> Woohoo
[22:10] <Grogbot> <rootfather> Windows port is building right now 😛
[22:10] <Grogbot> <sev> nice, because I need to sign it for Sparkle
[23:12] <-- cd left irc: Quit: cd
[23:17] <Grogbot> <rootfather> Windows: DONE! (yay, I'm first!)
[23:31] <-- criezy left irc: Quit: criezy
[23:33] <Scummette> [scummvm] digitall pushed 1 new commits to master: https://git.io/JeCJt
[23:33] <Scummette> scummvm/master 452b380 digitall: GRAPHICS: MACGUI: Fix GCC Signed vs. Unsigned Warnings
[23:35] <Grogbot> <rsn8887> Why am I getting so many compilation warnings?
[23:35] <Grogbot> <rsn8887> ../scummvm/engines/scumm/players/player_sid.cpp: In member function 'void Scumm::Player_SID::update()': ../scummvm/engines/scumm/players/player_sid.cpp:474:21: warning: array subscript 4 is above array bounds of 'uint8 [3]' {aka 'unsigned char [3]'} [-Warray-bounds] attackReg[channel] = ptr1[y]; ~~~~~~~~~~~~~~~~~^ ../scummvm/engines/scumm/players/player_sid.cpp:476:22: warning: array subscript 4 is above array bounds of 'uint8 [3]'
[23:35] <Grogbot> {aka 'unsigned char [3]'} [-Warray-bounds] sustainReg[channel] = ptr1[y];
[23:35] <Grogbot> <rsn8887> ../scummvm/engines/scumm/imuse/imuse_player.cpp: In member function 'virtual void Scumm::Player::sysEx(const byte*, uint16)': ../scummvm/engines/scumm/imuse/imuse_player.cpp:411:52: warning: 'snprintf' output truncated before the last format character [-Wformat-truncation=] snprintf((char *)&buf[a * 3], 3 * sizeof(char), " %02X", p[a]); ^~~~~~~
[23:35] <Grogbot> ../scummvm/engines/scumm/imuse/imuse_player.cpp:411:12: note: 'snprintf' output 4 bytes into a destination of size 3 snprintf((char *)&buf[a * 3], 3 * sizeof(char), " %02X", p[a]);
[23:36] <Grogbot> <rsn8887> Should I ignore these or be worried?
[23:36] <Lightkey> I think digitall's warning fixes did not get into 2.1.0-
[23:40] <-- _sev_ left irc: Quit: This computer has gone to sleep
[23:42] --> _sev_ joined #scummvm.
[23:42] #scummvm: mode change '+o _sev_' by ChanServ!ChanServ@services.
[23:42] <Scummette> [scummvm] digitall pushed 1 new commits to master: https://git.io/JeCJc
[23:42] <Scummette> scummvm/master 3911b56 digitall: GRAPHICS: MACGUI: Fix Possible Out of Bounds Read with Trailing Ampersand
[23:43] --> digitall joined #scummvm.
[23:43] #scummvm: mode change '+o digitall' by ChanServ!ChanServ@services.
[23:43] <digitall> Lightkey: Not all of them in any case... but there are a lot of them to get fixed so the compilers are less noisy and thus real issues don't get hidden in the noise of warnings
[23:43] <digitall> or the warnings get turned off :|
[23:45] <Grogbot> <rsn8887> Some of these look serious? ../scummvm/engines/lab/engine.cpp: In member function 'void Lab::LabEngine::processMainButton(uint16&, uint16&, uint16&, bool&, uint16, uint16&)': ../scummvm/engines/lab/engine.cpp:828:13: warning: 'void* memcpy(void*, const void*, size_t)' accessing 508 bytes at offsets 656 and 660 overlaps 504 bytes at offset 660 [-Wrestrict] memcpy(&_breadCrumbs[0], &_breadCrumbs[1], _numCrumbs * sizeof
[23:45] <Grogbot> _breadCrumbs[0]); ~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
[23:45] <digitall> rsn8887: scummvm/engines/scumm/imuse/imuse_player.cpp has not been subject to any recent changes by myself for checking warnings...
[23:46] <Grogbot> <rsn8887> Maybe I just always ignored these warnings. Hmm.
[23:46] <digitall> I don't know about serious... They should get fixed, but it is hard if some of the engines may actually be dependent on that for correct behaviour...
[23:46] <Grogbot> <rsn8887> But memcpy with overlapping regions is UB, or not?
[23:46] <-- SylvainTV left irc: Read error: Connection reset by peer
[23:46] <digitall> Apart from that, I would guess the warnings have always been there... or rather older compiler toolchains had less warnings / were more permissive or had them disabled by default
[23:46] <Lightkey> _sev_: I'm guessing you mean HOWTO-Release, checking the ADGF flags is a step of release testing there, so it should have been already done before preparing for tagging, which it wasn't. I agree it would be better to let the one preparing the source do that at tagging time instead of relying on every engine developer not to forget at release testing time but please only one or the other.
[23:46] <Grogbot> <rsn8887> I guess this is the first time I actually paid attention to warnings...
[23:47] <digitall> It certainly is problematic ... hence why the warning.
[23:47] <digitall> It is always a good idea to kill as many as possible so you can find the "real" ones... or new ones before they cause major bug report headaches
[23:47] <Lightkey> _sev_: Also it looks to me like the sections Tagging a version and Creating source balls should go before Binaries.
[23:49] <digitall> Overall, I would really like to blacklist memset, memcpy, sprintf etc. in the engines code as too damn likely to cause issues... but we have alot of cases of usage which need to be refactored before that could be considered.
[23:50] <Grogbot> <rsn8887> Yeah almost all the warnings I get are re: sprintf and snprintf
[23:50] <Grogbot> <rsn8887> And maybe-uninitialized
[23:50] <Grogbot> <rsn8887> Probably ok.
[23:53] <digitall> rsn8887: If you drop them into a pastebin, I will take a look at patching them so we don't have the same warnings at the next release.
[23:54] <Grogbot> <rsn8887> Ok let me finish the build, and then I will put the full output into pastebin.
[23:54] <digitall> Oh hell... that memcpy in LAB engine is horrible. I am not sure how easy that would be to replace without refactoring all of the breadcrumb maze logic to ensure that it is not buggy as heck
[23:56] <digitall> Some warnings are pretty easy to supress / fix... those tend to be either minor or slightly spurious, but it is good to get rid of those so you can find the nasty ones like that
[23:56] <digitall> which really do indicate code which is really problematic. The HDB engine and PRINCE engine still have pointer issues, but hard to rework without making major invasive changes to the
[23:56] <-- JohnnyonFlame left irc: Ping timeout: 276 seconds
[23:56] <digitall> engine and needing a bunch of playtesting to revalidate.
[23:57] <Lightkey> What a cliffhanger.
[23:57] <digitall> Lightkey?
[23:58] <Lightkey> Waiting for each line to materialise, I was biting my fingernails. ;-)
[00:00] --- Sat Oct 5 2019