Time Nick Message 02:42 v-rob Oh gee whiz, Irrlicht doesn't even have keycodes for some keyboard things so we have to hack around it in KeyPress. 02:42 v-rob I really don't want to touch the keycode names for fear of breaking anything. 02:42 v-rob Ugh 07:07 sfan5 which "keyboard things" would that be? 08:51 MTDiscord specing, AFAIK you cannot just change the license of MT since every single contributor ever would need to agree to it first...that's what a core dev once said 08:53 MTDiscord Unless you were talking about a license that is directly compatible with LGPL...not sure there 08:57 specing Yes, GPL would, upon linking, overrule the LGPL 08:58 specing and the whole would become GPL 08:58 specing I think... 11:52 pgimeno yes, the LGPL has provisions for relicensing as GPL specifically, but it's not a change I as a user would like 13:00 MTDiscord https://github.com/nextcloud/ios#contribution-guidelines--license 13:00 MTDiscord https://github.com/nextcloud/ios/blob/master/COPYING.iOS 13:00 MTDiscord It looks like GPL applications can actually be available on the AppStore 13:00 MTDiscord (https://github.com/audacity/audacity/discussions/932#discussioncomment-781558) 13:08 MTDiscord ironic that the last link is from discussion on one of the most disastrous software, that spun so many forks when the disaster happened 13:11 MTDiscord GPL apps cannot be on the app store. NextCloud is not GPL, it's dual-license. 13:11 MTDiscord It's distributed under either the terms of the GPL, or under GPL plus the App Store exception. 13:13 MTDiscord Minetest can make a similar promise not to enforce the GPL specifically against the App Store, but it does not have the power to make that promise on behalf of each individual contributor. 13:13 specing What's the apple hostility against the GPL about? 13:14 specing Also, my fork of MT is AGPLv3 13:14 specing Apple can go suck it 13:39 MTDiscord bullshit 13:39 MTDiscord https://apps.apple.com/us/app/adguard-adblock-privacy/id1047223162 / https://github.com/AdguardTeam/AdguardForiOS 13:40 MTDiscord it's GPL-3 13:42 MTDiscord In that case, feel free to report them. 13:43 MTDiscord unfortunately that's a 10 year old joke that needs to rightfully get fucked 13:43 MTDiscord GPL is fine 13:43 MTDiscord Haha, that damn license is so hard to dig through, it makes it hard to find out who actually owns the copyright... 13:45 MTDiscord COPYING already proves it's under GPL 13:46 MTDiscord It looks like they don't have any transparency about how to become a code contributor. It's sort of a closed team. That means we don't know what kind of CLA-type agreements are involved in contributing code to the core. 13:46 MTDiscord That means we can't be certain it's purely GPL. They can very easily have a dual license setup. 13:46 MTDiscord i doubt there's a CLA involved, see audacity on how to fuck that up 13:47 MTDiscord second the acknowledgements page in app links to all the GPL / oss licenses used in libraries 13:48 MTDiscord https://github.com/nextcloud/ios is even GPL3 13:48 MTDiscord Like I said, report them if you'd like. It's not really relevant to MT licensing discussions, or really to MT dev in general. 13:48 MTDiscord it means there really isn't any issue deploying to the App Store 13:49 MTDiscord ? NextCloud iOS is not GPL, says right at https://github.com/nextcloud/ios#contribution-guidelines--license, and this was mentioned right at the beginning. PLEASE try to follow along with the discussion before calling bullshit on anything. 13:50 MTDiscord we already do that in Minetest for new files 13:50 MTDiscord clouds.h, skyparams.h and other class headers 13:50 MTDiscord Yeah, that's fine. 13:51 MTDiscord we only need this boilerplate and we're fine 13:51 MTDiscord https://github.com/nextcloud/ios/blob/master/COPYING.iOS 13:51 MTDiscord I mean, I've considered MIT-licensing my contributions. At the very least, I'd relicense them if I were to publish them separately. 13:51 MTDiscord And as previously mentioned, yes, that boilerplate would work fine, but you'd have a hard time getting it ratified. 13:52 MTDiscord Especially since some past contributors are no longer even with us and their ownership of the copyrights to their contributions have been inherited and you'd have to get them to understand and accept those terms. 13:52 MTDiscord at the end of the day as long as people aren't using the IAP laden semi-closed forks 13:52 MTDiscord is it really a conscious problem 13:53 MTDiscord isn't it illegal to impose additional restrictions on GPL'ed software? 13:53 MTDiscord MT forks that are in the App store are already in violation of the GPL, the MT devs just don't have the time/energy to chase after them constantly. 13:53 MTDiscord (also, does the discord reply feature work over the bridge?) 13:54 MTDiscord I think replies only send the text, not which message you're replying to. 13:55 MTDiscord that's pretty bad (test) 13:56 Kimapr yep, doesn't even include the name of the person who i replied to 14:14 sfan5 Krock: how much would you mind if bitmap fonts stopped working? 14:14 rubenwardy can you still do bitmap fonts using freetype? 14:14 rubenwardy for a voxel game, seems fairly important to allow pixel fonts 14:15 sfan5 "pixel" != bitmap 14:15 sfan5 pixel fonts are doable anyway 14:15 MTDiscord there exists a TTF version of the MC font 14:16 sfan5 by bitmap fonts here I mean -DENABLE_FREETYPE=0 which nobody has been using seriously in years 14:16 rubenwardy yeah, removing support for no freetype makes sense to me 14:17 rubenwardy I believe that's required for colored chat 14:17 sfan5 it is 14:17 MTDiscord remove support for no freetype 14:18 rubenwardy [Delete] ? 14:19 sfan5 the context here is that I want to delete all XML code from irrmt but the png fonts we ship have an xml file 14:19 MTDiscord PNG fonts? 14:19 rubenwardy yeah I remember this discussion 14:19 rubenwardy appguru: bitmap fonts, for non-freetype 14:19 MTDiscord I saw them, but have never really seen them in use 14:19 MTDiscord (saw them as files) 14:35 MTDiscord I'd like to get a second approval for #11287! 14:35 ShadowBot https://github.com/minetest/minetest/issues/11287 -- Take advantage of IrrlichtMt target by JosiahWI 14:49 sfan5 > Drop XML implementation, related code and dependent features 14:49 sfan5 > 49 files changed, 10 insertions(+), 13412 deletions(-) 14:49 sfan5 *grin* 14:51 rubenwardy that's such a nice ratio 14:54 specing irrlicht had its own implementation of XML? 14:54 sfan5 of course 14:55 specing lol 14:56 rubenwardy sounds about right 14:56 sfan5 it's honestly not that bad if you need a subset and not a spec-compliant implementation, allows you to avoid security issues e.g. libxml2 is plagued with 17:54 celeron55_ that's one way of going about getting rid of irrlicht... 17:54 celeron55_ just delete it piece by piece 18:03 sfan5 @appguru it's always possible to fix a branch by force pushing, no need to open a new pr 18:03 sfan5 if you have git questions I can help 18:03 appguru I tried force pushing 18:03 appguru And I reset my master branch 18:03 appguru Unfortunately, I had origin set to my fork 18:03 appguru Sorry for the commotion 18:04 appguru I couldn't really find a way to get the commits I accidentally did on my master branch out of there though 18:05 appguru So I wanted to just delete & restore the branch, but the deletion closed the PR, and the restoring didn't reopen it... 18:05 celeron55_ git reset --hard HEAD~3; git push -f 18:05 celeron55_ assuming 3 commits you want to get rid of 18:05 appguru oh 18:06 celeron55_ i.e. just force push the branch without the commits 18:06 appguru I tried resetting to origin/master 18:06 celeron55_ well, of course you'll need to make sure you have the commits saved up in some branch so that they're not completely lost 18:07 celeron55_ but force pushing means "make the remote branch exactly like the one i'm pushing, including deleting commits" 18:08 appguru yeah 18:08 sfan5 secret tip: git reflog 18:12 MTDiscord I only use git reflog when one flogging wasn't enough. 18:15 appguru I should read some good comprehensive git guide somewhen 18:15 rubenwardy I've found that contributing to Minetest has made me better at git than most of my colleagues 18:16 appguru If I'm a rich burglar with 10k worth of stolen items, and you have 5k I can steal, why should I not do it? 18:16 appguru oof wrong channel 18:19 sfan5 appguru has revealed his plans to steal the Minetest source code 18:20 appguru rumor says I have it copied to my local drive 18:22 Krock he'll add telemetry to improve Minetest's look&feel, and also forward information to legal authorities if necessary 18:22 appguru Kinda like an Audacity fork but the other way around, I'm here to add the good stuff not remove it :P 18:24 rubenwardy https://content.minetest.net/privacy_policy/#who-has-access 18:24 rubenwardy ? 18:25 kilbith Krock: given the amount of pedophiles among the MT servers that make perfect sense 18:26 appguru That depends on whether you consider MT forks MT. 18:27 kilbith like... "Final Minetest", yes 18:28 rubenwardy All sites are required to share information with law enforcement. The problem is that Audacity was so vague about what was collected and they collected enough to forbid under 13s from using it 18:29 appguru Well, whenever there's some kind of chat, you will have predators... It's on the server staff to deal with this. 18:29 appguru rubenwardy: COPPA again, I assume? 18:29 kilbith server staff are not required to save logs meanwhile 18:30 appguru Saving logs can be problematic regarding data protection regulations 18:31 appguru How would you enforce that requirement BTW? 18:31 appguru How could staff "prove" that they are saving logs, without making them available to everyone? 18:32 MTDiscord Generally the only way that gets enforced is when there's an actual case that requires you to cough up the logs, and then they get to find out whether you were actually keeping them as required or not. 18:33 appguru I don't see why this belongs in #minetest-dev, there IMO is little that can be done on the engine/serverlist side of things about this. 18:33 MTDiscord You can't rally reconcile data retention requirements with RTBF anyway... 18:33 appguru Probably a "kid-friendly" badge for servers would be good. 18:33 MTDiscord A channel move sounds good to me. 18:41 MTDiscord online and user generated content not rated by content bureau 19:28 pgimeno hm, the license stuff that was commented earlier was pretty much nonsense. Posting a promise-not-to-sue can only be done if every copyright holder agrees to that, including the copyright holders of every work incorporated as part of the code (like the Chili/SpringRTS authors that started the discussion). That also includes future contributions; and despite being GPL, you can't incorporate more GPL code without asking all of the the copyright holders 19:28 pgimeno of the code you want to include if they agree. 19:32 pgimeno Please just keep avoiding the GPL, thanks. 19:32 MTDiscord ^ :-) 19:33 MTDiscord ^^ :^) 19:34 MTDiscord why double caret when ^3 20:30 rubenwardy there's some PRs here that have exceeded the 1 week to bypass the roadmap: https://github.com/minetest/minetest/pulls?q=is%3Apr+is%3Aopen+sort%3Aupdated-desc+label%3A%22Roadmap%3A+Needs+approval%22 20:32 MTDiscord Ooh, I didn't know about 11378; that sounds like something I could really use... 20:32 sfan5 I guess the rule doesn't count if no dev looks at the PRs 20:34 MTDiscord I like the concept, would need to review it to be sure about the implementation. Of particular concern to me is, if the checkbox is hidden, where does the value come from? Can a player-specific override exist but the player now can't see it, or does it allow the game to force a specific value? 20:34 MTDiscord I'm actually already puzzled enough about how the enable_damage setting is handled to begin with. 20:35 MTDiscord OIC it allows you to disable them BY forcing them to a specific value... 20:35 sfan5 that makes sense 20:36 MTDiscord Oh, or maybe not. It seems like you can specify default independently of disabling them, and it forces use of the default is if they're specified AND the setting is disabled. 20:36 MTDiscord Disabling a setting but not setting a default makes sense in the case of "don't care" settings that the game doesn't use at all anyway 20:37 MTDiscord Like NodeCore and creative_mode. 20:37 MTDiscord But being able to force a default is important for settings that have enginey meanings, like enable_damage. 20:50 sofar so set_sky() is broken for me on ITB. Even with API fixes the client never receives the skybox textures and on any client that doesn't have them cached, I get disco skyboxes (6 flat colors) 20:51 sofar Pretty sure this is easy to oversee since everyone working with skybox textures will be using a client that has lots of stuff in cache 20:51 sofar anyone mind wiping cache and e.g. using 5.4 to connect to insidethebox for a second to confirm? I think the client isn't requesting the textures 20:54 MTDiscord I joined but I hve to do a mandatory tutorial before I can reach the lobby 20:55 sofar the boxes are open on top so you can see the sky 20:55 sofar oh 20:55 MTDiscord I see a flat azure-colored sky in tutorial #2 20:55 sofar but it's the nonbroken plain one in there 20:55 ShadowBot https://github.com/minetest/minetest/issues/2 -- Burned wood 20:56 sfan5 this wouldn't be an issue with set_sky but rather the texture finding/sending 20:56 sofar yeah the tutorial doesn't use textures 20:56 sofar sfan5: exactly 20:57 sfan5 did you check if textures in the same folders as your skyboxes get sent? 20:57 sfan5 as in, try with another file 20:57 sofar the textures are under mods/skybox/textures 20:57 sofar just like they were when we launched the server multiple years back 20:57 sofar they haven't changed 20:58 sofar there's 6 sets of skyboxes in that mod, they're all not working 20:58 sofar all the other textures (from other non-skybox mods) are working, the code is working 20:59 sofar I can see it e.g. change light levels, turn off sun/clouds 21:00 sofar oh hey 21:00 sofar if I copy the jpg files to another mod/textures folder, they're still broken 21:00 sofar I wonder if it's jpg vs. jpeg 21:01 sfan5 uhh yea I think .jpeg might not work 21:03 MTDiscord I can see this sky which is new to me... 21:03 MTDiscord https://cdn.discordapp.com/attachments/747163566800633906/862438607669952512/screenshot_20210707_170235.png 21:03 sofar it was jpg 21:03 sofar jpeg doesn't work either 21:04 sofar warr1024: can you clear your cache, log out, try again? 21:04 sofar warr1024: what MT client version? 21:05 sofar geesh maybe my MT is built without jpg support? is that even possible? 21:05 MTDiscord 5.5-dev 21:05 MTDiscord Works fine. Clouds keep getitng in the way of a good screenshot tho 21:05 MTDiscord https://cdn.discordapp.com/attachments/747163566800633906/862439253798682644/screenshot_20210707_170519.png 21:05 sofar no, libjpeg linked dynamically 21:06 sfan5 it's technically possible but it'd throw an error 21:06 sofar it throws "can't build texture" errors in my client 21:06 sfan5 does the file hash appear in client/cache/? 21:07 sofar good point 21:09 sofar no JPG whatsoever in the cache 21:09 sofar only PNG and OGG, and a few OBJ 21:11 sofar warr1024: ty for trying 21:12 sofar generateimage(): COuld not load image: ... 21:20 sofar sfan5: or I run mt from a terminal and read: 21:20 sofar 2021-07-07 14:19:45: ERROR[Main]: Client: Cannot create image from data of file "TropicalSunnyDayBack.jpg" 21:20 sofar 2021-07-07 14:19:45: ERROR[Main]: Irrlicht: JPEG FATAL ERROR in _tempreadfile: Wrong JPEG library version: library is 62, caller expects 80 21:20 sofar so um... wth 21:21 sofar guess I need to rebuild, when I get to it 21:22 sofar ok, flatpak version works 21:22 * sofar calls it solved, moves on to actual ITB code issues 21:24 specing sofar: your server is cool 22:14 MTDiscord rubenwardy: Re: #11378, do I need to act on it soon or can I wait a few days? I wanted to finish getting #11405 merged before I put additional irons in the fire... 22:14 ShadowBot https://github.com/minetest/minetest/issues/11378 -- Add disable_settings to game.conf to get rid of "Enable Damage"/"Creative Mode"/"Host Server" checkbox by Wuzzy2 22:14 ShadowBot https://github.com/minetest/minetest/issues/11405 -- Add API for mods to hook liquid transformation events by Warr1024 22:14 rubenwardy well "Supported by core dev" is more like you think the PR could be a good idea, doesn't mean you've looked into it exactly 22:15 rubenwardy like, you should be able to determine whether you "support" a PR just by reading the title and description, without going into technical details 22:15 rubenwardy maybe we need a different label idk 22:15 rubenwardy that's how I see it anyway 22:20 MTDiscord Okay, I wasn't even sure what that label meant, since it's seems kind of oblique to the "approval" labels. 22:20 MTDiscord I can add that one. 22:47 Wuzzy #11431 22:48 ShadowBot https://github.com/minetest/minetest/issues/11431 -- Add paramtype2s for horizontal rotation and 64 colors by Wuzzy2 22:49 MTDiscord Oh heck yeah thats useful 22:52 MTDiscord Haha, I swear it was like only a few days ago when I was playing around with facedirs and thinking "it would be a lot easier to manage this if there were an only-4-way version of this" 22:53 MTDiscord With furnaces and chests ironically I do enjoy flipping them around in weird directions so I'd rather those stay full facedir, but in NodeCore I recently merged raking sand, and sand only makes sense to stay raked on the top surface... 22:55 MTDiscord Eh, that reminds me that for 11405 I probably need to look into what's necessary to include regression testing tools for it to devtest...